Citation
Exploring the Relationship between Wildfire Education Programs and Social Capital in Communities at Risk of Wildfires

Material Information

Title:
Exploring the Relationship between Wildfire Education Programs and Social Capital in Communities at Risk of Wildfires
Creator:
AGRAWAL, SHRUTI
Copyright Date:
2008

Subjects

Subjects / Keywords:
Community associations ( jstor )
Community life ( jstor )
Community resource files ( jstor )
Disasters ( jstor )
Educational programs ( jstor )
Neighborhoods ( jstor )
Retirement communities ( jstor )
Social capital ( jstor )
Social perception ( jstor )
Wildfires ( jstor )
City of Palm Coast ( local )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Shruti Agrawal. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
5/31/2007
Resource Identifier:
496599827 ( OCLC )

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

EXPLORING THE RELATIONSHIP BE TWEEN WILDFIRE EDUCATION PROGRAMS AND SOCIAL CAPITAL IN COMMUNITIES AT RISK OF WILDFIRES By SHRUTI AGRAWAL A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2006

PAGE 2

Copyright 2006 by Shruti Agrawal

PAGE 3

This document is dedicated to my family for their love and support.

PAGE 4

ACKNOWLEDGMENTS First and foremost, I would like to thank my advisor, Dr. Martha Monroe, for her guidance and support during my years at the University of Florid a. Her expertise, enthusiasm, and dedication for her work were a constant source of motivation to perform and excel. I would also like to thank my committee members, Dr. Glenn Israel, Dr. Karen Kainer, Dr. Alan Long, and Dr. Anthony-Oliver Smith, for their s upport and valuable contributions to my research. I am extremel y grateful to my colleagues and group mates Andrea Albertin, Kristy Bender, Kelly Bi edenweg, Chris Demers, Bebette DeVera, Janice Easton, Julie Ernst, Holly Johnson, Lauren McDonell, Richard Plate, Lisa Penissi, Jenny Seitz, and Judith Cheng for their support and helpful discussions. I would especially like to thank my friend Soum ya Mohan for her invaluable support and encouragement throughout my program and continued friendship. I am extremely grateful to the North Central Research St ation, USDA Forest Service, for funding my research, which made this research possible. Special thanks go to Rachel Hudson, Pa mela Jakes, Linda Kruger, Erica Lang, Sarah McCaffrey, Kristen Nelson, and Vicky Stur tevant, for their frui tful discussions and help defining this research project. Thanks are due to the helpful staff at the School of Forest Resources and Conservation who have ea sed my way at the University of Florida. My friends Aditi Jariwala, Tarun Kushwaha, Shonan Norohna, and Swati Pant have enhanced the last four years. iv

PAGE 5

I would like to thank my sister, Swati Ag rawal, and brother-in-law, Manoj Gupta, for their unconditional love and support. Without them I could have not made it this far. I would like to thank my dearest friend and br other Kunal Shah for always believing in me and being there for me through the happy and the tough times. And of course I thank Jairaj Payyapilly, my best friend whose affec tion and support has always been with me. Finally, words are not enough to thank my parents, Madhu and Binod Agrawal, who are not only my best friends and inspirati on in life but also a big reason for where I stand today. And my everlasting appreciati on goes to my husband, Devang Parikh, for his unconditional love, support, and appreciation through the mo st difficult stages of my graduate school. v

PAGE 6

TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................iv LIST OF TABLES ...............................................................................................................x LIST OF FIGURES .........................................................................................................xiii ABSTRACT .....................................................................................................................xiv CHAPTER 1 INTRODUCTION........................................................................................................1 Context of the Problem .................................................................................................1 Statement of the Problem ..............................................................................................3 Purpose of the Study and Objectives ............................................................................6 Research Questions .......................................................................................................6 Definitions ....................................................................................................................6 Community ............................................................................................................6 Community Preparedness ......................................................................................7 Neighborhood ........................................................................................................8 Participation in Wildfire Education Programs ......................................................8 Social Capital.........................................................................................................8 Wildfire ..................................................................................................................8 Wildfire Education Programs ................................................................................8 Wildland-Urban Interface ......................................................................................9 Scope and Limitations of the Study ..............................................................................9 Significance of the Study ............................................................................................10 2 LITERATURE REVIEW...........................................................................................13 Community.................................................................................................................13 Disasters......................................................................................................................15 Wildfires.....................................................................................................................1 7 Wildfire Reduction Methods...............................................................................17 Ecology of Wildfires...........................................................................................18 Disaster Education......................................................................................................19 vi

PAGE 7

Wildfire Education Programs.....................................................................................24 FireWise..............................................................................................................25 Wildland Fire Education Toolkit.........................................................................26 Florida Division of Forestry W ildfire Education Initiatives...............................26 Social Capital..............................................................................................................27 Theories of Social Capital...................................................................................28 Bourdieu and social capital..........................................................................29 Coleman and social capital...........................................................................29 Putnam and social capital.............................................................................31 Compare and Contrast Soci al Capital Perspectives.............................................31 Value of Social Capital........................................................................................33 Contrasting Views on Social Capital...................................................................34 Indicators and Measures of Social Capital..........................................................35 Independent variables...................................................................................36 Criticisms of social capital indicators..........................................................38 Role of Education Institutions in Building Social Capital..................................39 Role of Social Capital in Disaster Preparedness.................................................40 This Research......................................................................................................41 Summary.....................................................................................................................45 3 METHODOLOGY.....................................................................................................48 Research Questions.....................................................................................................49 Study Communities....................................................................................................49 Selection of Study Communities.........................................................................49 Description of Study Communities.....................................................................51 Wedgefield...................................................................................................51 Palm Coast....................................................................................................52 Lake Wales Ridge........................................................................................54 Organization of Study Communities...................................................................56 Information Gathering: In-depth Interviews at Community Level.............................57 Qualitative Research: 3CM in Each Neighborhood...................................................58 Data Sampling.....................................................................................................60 3CM Process and Data Collection.......................................................................60 Generating components................................................................................63 Organizing categories...................................................................................63 Labeling categories......................................................................................63 Quantitative Research: Mail Surveys.........................................................................65 Data Sampling.....................................................................................................65 Survey Development Process..............................................................................66 Social capital scale development..................................................................67 Survey administration..................................................................................72 Nonresponse Analysis.........................................................................................74 4 RESULTS AND DISCUSSION.................................................................................75 vii

PAGE 8

Qualitative Data Analysis : In-depth Interviews.........................................................75 Qualitative Data Analysis: 3CM.................................................................................80 Component Analysis...........................................................................................81 Thematic Analysis...............................................................................................82 Category Analysis...............................................................................................87 People’s Perception of Neighborhood Attributes................................................90 Comparative Neighborhood Analysis.................................................................94 Quantitative Data Analysis.........................................................................................96 Dealing with Missing Data..................................................................................98 Demographic Characteristics of Respondents.....................................................99 Section 1: Neighborhood Data Analysis...........................................................104 Tiger Creek Forest......................................................................................105 Indian Lake Estates....................................................................................106 Wedgefield.................................................................................................107 Cypress Knoll.............................................................................................108 Seminole Woods........................................................................................109 Placid Lakes...............................................................................................109 Leisure Lakes.............................................................................................110 Section 2: Aggregate Data Analysis..................................................................112 Part one: Wildfire education program s and perception of social capital...113 Part two: Wildfire preparedness and perception of social capital..............121 Part three: Wildfire education pr ograms and wildfire preparedness..........126 Part four: Participation in community activities and perception of social capital......................................................................................................128 Community demographic variables a nd perception of social capital.........132 Section 3: Interaction in Disaster Education Program and Perception of Social Capital.................................................................................................134 Viewing electronic media...........................................................................135 Receiving print information.......................................................................137 Attending events.........................................................................................138 Conclusion................................................................................................................141 5 IMPLICATIONS AND CONCLUSIONS...............................................................144 Results in Brief.........................................................................................................147 Implications..............................................................................................................149 Implications for Environmental Educators........................................................149 Implications for Resource Managers.................................................................150 Recommendations for Future Research....................................................................151 Conclusions...............................................................................................................154 APPENDIX A DESCRIPTION OF WILDFIRE EDUCATION PROGRAM IN STUDY COMMUNITIES......................................................................................................157 viii

PAGE 9

B SOCIAL CAPITAL AND WILD FIRE EDUCATION PROGRAM RELATIONSHIP DIAGRAM..................................................................................170 C 3CM INTERVIEW PROTOCOL AND INTERVIEW GUIDE..............................171 D SURVEY PROTOCOL AND MAIL SURVEY......................................................175 E 3CM INTERVIEWS AND ANALYSIS..................................................................192 F QUANTITATIVE DATA ANALYSIS TABLES....................................................231 LIST OF REFERENCES.................................................................................................286 BIOGRAPHICAL SKETCH...........................................................................................293 ix

PAGE 10

LIST OF TABLES Table page 2.1: Education’s role in s upporting community preparedness..........................................23 3.1: Community orga nization criteria...............................................................................57 3.2: 3CM participants in each neighborhood....................................................................60 3.3: Sample of components and ca tegories of a participant’s map...................................64 3.4: Factor analysis: Communalities.................................................................................69 3.5: Total variance explained............................................................................................70 3.6: Component matrix a...................................................................................................71 3.7: Rotated component matrix a.......................................................................................72 3.8: Mail survey response rate..........................................................................................73 4.1: Items relating to theme Social events........................................................................84 4.2: Items relating to them e Characteristics of people......................................................84 4.3: Items relating to theme Associations/clubs...............................................................85 4.4: Items relating to them e Neighborhood characteristics..............................................85 4.5: Items relating to th eme Negative perceptions............................................................86 4.6: Variation in participation in wildfire educational programs......................................97 4.7: Gender distribution of respondents by neighborhood................................................99 4.8: Age distribution of respondents by neighborhood...................................................100 4.9: Educational level of respondents by neighborhood.................................................101 4.10: Income level of respondents by neighborhood......................................................102 4.11: Employment status of respondents by neighborhood............................................103 x

PAGE 11

4.12: Ethnicity of respondents by neighborhood............................................................104 4.13: Mean social capital values fo r respondents of Tiger Creek Forest........................105 4.14: Mean social capital values for respondents of Indian Lake Estates......................106 4.15: Mean social capital values for respondents of Wedgefield...................................107 4.16: Mean social capital values for respondents of Cypress Knoll...............................108 4.17: Mean social capital values for respondents of Seminole Woods...........................109 4.18: Mean social capital values for respondents of Placid Lakes.................................110 4.19: Mean social capital values for respondents of Leisure Lakes...............................111 4.20: Mean social capital values for part icipants and non-participants in different educational media...................................................................................................114 4.21: Mean social capital values for part icipants and non-partic ipants for specific electronic media.....................................................................................................116 4.22: Mean social capital values for particip ants and non-participants for specific print information.............................................................................................................116 4.23: Mean social capital values for part icipants and non-partic ipants for specific events......................................................................................................................116 4.24: Mean social capital values for part icipants and non-participants in specific events......................................................................................................................117 4.25: Effect of learning, usi ng and interacting with the information on perception of social capital...........................................................................................................118 4.26: Coefficients a for regression analysis of pr ogram participation indices and perception of social capital.....................................................................................119 4.27: Coefficients a of regression analysis of event index and perception of social capital.....................................................................................................................120 4.28: Mean social capital values for part icipants and non-partic ipants of wildfires preparedness activities............................................................................................122 4.29: Coefficients a for regression analysis of elec tronic media and perception of social capital...........................................................................................................124 4.30: Coefficients b for regression analysis of print media and perception of social capital.....................................................................................................................124 xi

PAGE 12

4.31: Coefficients b for regression analysis of event and perception of social capital....125 4.32: Mean social capital values for part icipation in various community activities.......130 4.33: Mean social capital values for part icipation in various community activities.......130 4.34: Mean social capital values for part icipation in various community activities.......130 : 4.35: Coefficients b for regression analysis of attending meetings and perception of social capital...........................................................................................................131 : 4.36: Coefficients b for regression analysis of attending meetings and perception of social capital...........................................................................................................131 4.37: Coefficients b for regression analysis for elec tronic media participation and demographic variables with perception of social capital.......................................136 4.38: Coefficients b for regression analysis for print participation and demographic variables with perceptio n of social capital.............................................................137 4.39: Coefficients b for regression analysis for participation in events and demographic variables with perception of social capital.......................................139 xii

PAGE 13

LIST OF FIGURES Figure page 2.1: Model of relationship between wildfi re education programs and perception of social capital in communities at risk of wildfires.....................................................44 3.1: Map of study communities........................................................................................51 4.1: Distribution of a ll respondents by gender..................................................................99 4.2: Distribution of all respondents by age group...........................................................100 4.3: Educational level of all respondents........................................................................101 4.4: Income level of all respondents...............................................................................102 4.5: Employment status of all respondents.....................................................................103 4.6: Ethnicity of all respondents.....................................................................................104 xiii

PAGE 14

Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy EXPLORING THE RELATIONSHIP BE TWEEN WILDFIRE EDUCATION PROGRAMS AND SOCIAL CAPITAL IN COMMUNITIES AT RISK OF WILDFIRES By Shruti Agrawal May 2006 Chair: Martha C. Monroe Major Department: Forest Resources and Conservation Education is an important aspect of a ny community preparedness initiative and may operate at the individual, group, or institutional levels, in support of homeowner actions, community policy, or staff training. A variet y of communication strategies, such as electronic, print, and events, are used to e ducate people about the risk of wildfires and motivate them to take action. In addition to education, social capital is an important element of community preparedness. Communities rich in social capital are be tter able to reduce their vulnerability and solve problems as compared to communities with less social capital. The objective of this study was to use the context of community preparedness for one type of disaster—wildfire— to investigate the relationship between wildfire education programs and social capital. More specifica lly, this study was designed to determine whether there is a difference in percepti on of social capital am ong people who did and did not participate in various aspects of th eir local wildfire education programs. The xiv

PAGE 15

study also examined elements of the educati onal program that are more likely to be associated with enhanced social capital. Data were collected from individuals in seven neighborhoods in three communities in Florida; a ll were at risk of wildland fire and had experienced some level of w ildfire education program. In-depth interviews (n=20) helped ga ther information on wildfire education programs in each community. 3CM interviews (n=15) helped understand people’s perception of their community and social inte raction within their community. Results of the mail survey (n=1379, 37% response) indica te that people who participated in the wildfire education program perceived higher le vels of social capital than people who did not participate. Also, a s ignificant difference in perception of social capital was observed for people who did and did not take steps to reduce risk of wildfires around their homes. Those who reduced their risk credited talk ing to friends in the neighborhood and to community leaders for their actions. Engageme nt in events appears to make a greater contribution to predicting perception of soci al capital and is a ssociated with more participation in risk reduction activities. Clearly, social capital and ri sk reduction behaviors are associated with certain aspects of wildfire educati on programs; this study did not ascertain which causes the other. Managers and community leaders w ho wish to improve community preparedness could focus on delivering their messages th rough existing community groups (where social capital might already exist) and through formats that encourage community engagement (where social capital might be nurtured). xv

PAGE 16

CHAPTER 1 INTRODUCTION Context of the Problem Over the past few years, there have b een several natural disasters worldwide— major earthquakes in India and China and a tsunami in the Indian Ocean. The United States, too, has experienced disasters including severe winter storms, wildfires, landslides, and recently hurricanes Katrina a nd Rita. In 2005 alone, these disasters were responsible for losses of lives and property averaging more than $ 100 billion (National Climatic Data Center, 2006). The subcomm ittee of Disaster Reduction, part of the president’s National Science and Technology Counc il, reinforces the need to proactively prepare for a disaster by identifying potentia l hazard agents that cause disasters, enhancing resilience, and thus minimizing damage and disrup tion should a disaster occur (Wood et al., 2005). In the report Grand Challenges of Disaster Reduction , the subcommittee identifies six challenges to enhance resilience and reduce vulnerability from future natural and technological disast ers. These include (1) providing hazard and disaster information where and when it is needed; (2) understanding the natural processes that produce hazards; (3) developing hazard m itigation strategies and technologies; (4) recognizing and reducing vulnerability of inte rdependent critical infrastructure; (5) assessing disaster resilience using standa rd methods; and (6) promoting risk-wise behavior (Wood et al., 2005). The sixth challenge focu ses on the need to “develop and apply principles of economics and human behavior to enhance communications, trust, and understanding within the community to promote risk-wise behavior” (Wood et al., 1

PAGE 17

2 2005, p 3). It identifies the need to understand effective techniques to educate people and gain community support for taking actions to reduce the risk of di sasters and improving preparedness for a disaster. Disaster preparedness can be defined as a community’s ability to reduce its vulnerability to disasters. E ducation is an important aspect of a community preparedness initiative and may include steps such as co mmunicating knowledge of the disaster or motivating people to take action to reduce impacts caused by the disaster. This may include taking actions at the individual, collec tive, and the organizational level. Research suggests that in addition to education, “soc ial capital” is another important element of community preparedness (Jakes et al. , 2002). The structure of relationships existing among individuals in a community and the resulting norms of trust and reciprocity that arise from it are collec tively known as “social capital.” Literature suggests that social cap ital is associated with disaster preparedness ( Jakes et al., 2002). Communities that are better prepared for disasters like wildfires appear to have higher social capital and may be better able to work together to reduce their vulnerability (Jakes et al. , 2002). Community activit ies that enhance both information exchange and social capital w ould be efficient mechanisms for promoting disaster preparedness. This study focuses on one particular disa ster—wildfires in Florida. Forest and range ecosystems across the U.S. evolved with frequent, periodic fi res ignited either by lightning or on purpose by Native Americans (Long et al ., 2004), who used fire to clear grasslands and forests to create open spaces for agriculture or hunting. Most frequent in the South and the West, these fires promot ed the development of fire-dependent

PAGE 18

3 vegetation communities dominated by fire-toler ant pines, shrubs, and grasses. In the twentieth century, federal agencies across the U.S. initiated intensive fire exclusion and prevention efforts. No doubt these efforts help ed decrease the incide nce of wildfires and the amount of acreage burned, but at the same time they led to accumulation of vegetation in forest ecosystems. With mo re people moving into the wildland-urban interface, and with the heavy fuel loads, w ildfire is a major threat to homeowners and communities. Statement of the Problem The 1998 fires in Florida burned approximately 500,000 acres and damaged or destroyed more than 300 structures. Despite th e fact that wildfires are a real and present threat in Florida, not all homeowners are at equal risk of wildfires, but homeowners play a critical role in reducing risk. Factors that lead to increased risk of wildfires include nearby land use, amount and type of vegetati on within 30 feet of homes, and type of building materials used. Federal, state, and lo cal forest and fire agencies across the U.S. are taking steps to be more prepared to fight wildfires and protect human lives and property. After the 1998 fires, the Florida Divisi on of Forestry (DOF) and other state and local agencies in Florida increased availa ble wildfire-fighting equipment, improved communication systems, and provided wildfire training to structural fire fighters to improve fire fighter safety and emergency response (Jakes et al. , 2003). Although agencies and communities have some resources to help suppress wildfires, residents of wildland-urban in terface regions can help reduce their risk by accepting responsibility and taking actions to improve the chances to protect their homes during a wildfire (Nelson et al., 2005). For example, homeowners can reduce the fuel around their homes to reduce the ri sk of wildfires. They can al so facilitate access for fire

PAGE 19

4 fighters by providing enough space in their yards for fire trucks to turn around (Monroe and Long, 1999; Doran et al. , 2004). Educational programs to teach people about actions they can take to protect their property are and should be an important elem ent of community preparedness. A variety of programs have been initiated at the national, state, and county level, for example, FireWise, Fire Safe, FireFree, FireCAP, and Fire Safety Councils. Some wildfire educational programs also target school chil dren. Most wildfire education programs are aimed at educating residents a bout risk of wildfires, ecology of wildfires, and steps that individuals can take to reduce their vulnera bility to wildfires (FireWise, 2001; DOF, 2003; FireFree, 2004). Community education can al so be used to inform residents of new community policies and ordinances, gain suppor t for expenditures, and involve them in their community’s risk reduction initiative. Providing information to homeowners is one important objective of an education initiative, but such programs have other be nefits as well. Educational programs can provide a forum for people to talk about a c oncern, enable people to work together, or enhance partnerships. They can lead to the formation of new relationships among residents in the community. For example, educational opportun ities may introduce neighbors to each other, pull together individu als with similar jobs or interests, or reconnect old friends. As a result these pr ograms may help develop new bonds, improve and strengthen existing relationships, increase trust, and provide a reason for people to help each other. Putnam (2000) equates social capital w ith the level of community participation, engagement for collective action, and coopera tive problem solving. Thus, when people in

PAGE 20

5 the community interact with each other to a lleviate risk of wildfires, the formal and informal networks are a resource possessed by the individuals in the community. These networks can be used to achieve a va riety of objectives. Although success of a community-based educational program may be a result of existing networks in the community, it is also possible that a partic ipatory educational program may help develop or strengthen new relationships and networ ks—an added resource to individuals in a community. Research indicates that the more people in a community interact, share, participate, and trust each othe r, the better is the community ’s ability to solve common problems (Putnam, 2000). A variety of communication strategies are available to agencies that deliver wildfire educational programs. These include el ectronic media sources (e.g. websites and television), print media sources (e.g. handouts and brochures), and events (e.g. demonstrations and picnics). However, litt le information is available on the most effective options that engage people in risk reduction activities. Also, no information is currently available on the potential of thes e education programs to improve or build social capital. The programs that involve in teraction between the agency delivering the message and the audience may be more effec tive in generating action to reduce risk of wildfires and improving people’s pe rception of social capital. This dissertation explores th e relationship between wildfi re education programs and social capital. A better understanding a bout the interaction between interactive community programs and the perception pe ople have about the networks and relationships in their communities might enable more effective education programs to improve preparedness for various kinds of disasters.

PAGE 21

6 Purpose of the Study and Objectives The purpose of this study was to investigate the relationship between disaster education programs, specifically wildfire educ ation programs, and individual perceptions of social capital in commun ities at risk of wildfire. This study intends to: provide information to agencies that can be used to increase community preparedness for wildfires improve understanding of the relationshi p between participation in wildfire education programs and percep tion of social capital, and improve understanding of the elements of an educational program that are more likely to be associated with enhanced social capital. Results of this study could be used to design effective environmental education programs that provide support for both soci al capital and wildfire preparedness by improving community networks and capacity to resolve local problems. More effective programs could help reduce losses due to w ildfire and improve quality of life. Research Questions The study addresses the follo wing research questions: 1. What is the relationship between wildfire disaster education programs and social capital? 2. What elements of an educational program are more likely to be associated with enhanced social capital? Definitions Community For the purposes of this study a community is defined geographically as one that spans a larger area than a neighborhood and may consis t of several neighborhoods. A community often includes areas of commer ce as well as residential neighborhoods. In this study, only two neighborhoods are part of a community that fit the latter definition:

PAGE 22

7 Cypress Knoll and Seminole Woods in the city of Palm Coast. Wedgefield is identified as a community because it is isolated (one huge neighborhood with no commercial zone), and the four neighborhoods along the Lake Wales Ridge are part of the same ecological community, even though they span two c ounties and many small towns. For a more complete discussion of the term co mmunity, refer to Chapter 2, p 13. Community Preparedness Community preparedness is a community’s ability to reduce vulnerability to disasters, like wildfires. This includes act ions taken at the individual (homeowner), community (collective), and th e organizational level. Actions at the individual level reduce risk of wildfires on personal propert y. These may include trimming vegetation around structures, changing building materials, improving access, removing trees, clearing debris, and installing water resources. At the commun ity level, individuals may come together as a group to make decisions th at help reduce risk of wildfires. These may include establishing codes and covenants, passing fuel reduction ordinances, and increasing fire-fighting equipment to reduce ri sk of wildfire. Agencies and organizations like the USDA Forest Service, state forestry agencies, local fire agencies, The Nature Conservancy (TNC), and the local neighborhood association may make decisions at an organizational level that affect community preparedness. These may include carrying out prescribed burns to reduce fuel load in a hi ghly vulnerable area, obtaining equipment and training, and developing new ordinances and policies for fuel reduction. Overall, community preparedness is measured by how successful a community is in reducing its vulnerability to wildfires.

PAGE 23

8 Neighborhood A neighborhood is a geographical area where people live near each other. For the purposes of this study, neighborhoods were se lected to have population between 1002700 people. Of the seven neighborhoods in the study, most had unbuilt lots; only Cypress Knoll was completely built out . People in a neighborhood may be taking locality-oriented actions to increase wildfire preparedness. A neighborhood often shares some organized recreationa l and social activities. Participation in Wildfire Education Programs Participation in a wildfire education progr am is defined as receiving, engaging in, or being exposed to any aspect of a wild fire education initia tive—including viewing electronic media, receiving printed informa tion, or attending events. The neighborhood, the community, or the state may organize the program. Social Capital For this study, social capital is an in dividual’s perception of the quality of relationship he/she holds with his/her nei ghbors and other people in the community, and community characteristics like community participation. Wildfire Also referred to as forest fire, a wildfire is an uncontrolled fire in a natural area. It may be caused by lightning, human carelessness, or arson. Wildfire Education Programs A wildfire education program is an educational initiative that informs homeowners about their history and risk of wildfires, educates them about steps they can take at individual and community level to reduce their vulnerability to wildfires, informs them about new policies and ordinances, or gain s support for expenditures and policies. A

PAGE 24

9 variety of communication strate gies are used to deliver th ese messages through electronic media, print media, and community events. Wildland-Urban Interface The term wildland-urban interface has b een defined in several ways. The most common definitions are those based on spatial, fire, sociopolitical, and natural resource perspectives. In this study, wildland-urban interface is defined as “an area or zone where increased human influence and land use conve rsion are changing natu ral resource goods, services, and their management” (H ermansen and Macie, 2002, p 2). Scope and Limitations of the Study This study is based on residents in a sa mple of seven neighborhoods in three communities of Florida that are at risk of wildfires and have been exposed to a range of wildfire education programs. The study explores the relationship between wildfire education programs and social capital, determ ined by whether there is a difference in perception of social capital for people who did and did not particip ate in the wildfire education program. This research also examin es the elements of the educational programs that are more likely to be associated with enhanced social capital. So, for people who participated in a wildfire education program, th e research will investigate the elements of the educational program that are more able to enhance social capital. The unit of analysis for this research st udy is the individual. An inherent problem in this study is the lack of baseline data on social capital before the implementation of the wildfire educational programs. To treat this problem data was collected from partic ipants and non-participants of wildfire education program in each community and chose communities with indicators of high and low levels of social capital. Four basic assumptions of the study are that

PAGE 25

10 1. People in each of these communities have been exposed to different types and amounts of educational ma terials and messages. 2. Individuals perceive varying le vels of social coherence. 3. Individuals differ in their exposure to wildfires, wildfire preparedness, and mitigation activities. 4. Individuals differ in the numb er of years they have liv ed in Florida; which may affect their understand ing of wildfires. The nature of survey research is another limitation to the study. It is difficult to know how honest people are being when they complete the survey or if they are answering accurately. To mitigate this limitati on, questions on the survey were written with a neutral tone and asked in more than one way. Another limitation is response rate of th e participants. In the study, the Dillman (2000) method was used to improve the respon se rate. Overall, response rate was 36.83% (N=3744). An analysis of non-respondents, based on property values, was also done to improve the generalizabi lity of the responses. Significance of the Study Social capital helps communities achieve goals and solve problems together. Research indicates that the more people in a community interact, share, participate, and trust each other, the better is the comm unity’s ability to solve common problems (Putnam, 2000). Many communities that demonstrate preparedness for wildfire appear to have higher levels of social capital (Jakes et al. , 2002). Education programs vary in the degree to which they enhance community networks, relationships, and trust. One can im agine that a brochure left on the door would not have the same impact on the community norm as the same brochure distributed in a community meeting that may involve intera ction among people and agencies. Programs

PAGE 26

11 that involve some sort of communication between the agency delivering the message and the audience may be more eff ective in involving residents in risk reduction activities and improving social capital. A community with higher levels of social capital may participate in the educational activities and reduce wildfire risk. Altern ately, educational programs can help build social capital that enables communities to solve problems such as wildfire hazards. In either case, the net result should be increas ed preparedness and higher perceptions of social capital. This study focuses on residents of communitie s at risk of wildfires who have been exposed to an educational in itiative aimed at cr eating awareness or initiating action for reducing risk of wildfires. At a given in stance a community may be undergoing one or more of these educational in itiatives to educat e and involve residents in wildfire preparedness. In this study, the whole range of programs people were involved in was measured. Their experiences in different edu cation initiatives were clustered to explore the relationship between disaster education programs and social capital in communities at risk of wildfires. Programs were clustered on the basis of communication strategy used to deliver the message. This study attempts to delineate components of an educational program that may successfully involve the reside nts in taking actions to reduce the risk of wildfire. The study examined the relationships between part icipation in wildfire education program, perception of social capital, and taking acti on to reduce risk of w ildfire. Understanding these relationships will enable agencies a nd educators to design more effective and

PAGE 27

12 powerful programs. It may lead to additional work exploring the contribution of educational programs to e nhancing social capital.

PAGE 28

CHAPTER 2 LITERATURE REVIEW The objective of this study, as stated in Chapter 1, was to explore the relationship between wildfire education programs and per ception of social capital at the individual level. The study also explored elements of e ducational programs that are associated with enhanced social capital. Due to the interdisciplinary nature of the re search question, the study draws on literature from a large number of fields incl uding disasters, wildfire preparedness, social capital, and participa tion. This chapter is organized into six sections—community, disasters, wildfires, disaster education, wildfire education programs, and social capital. Each of these sections relates to the research questions under investigation. Community The term community has been defined in the literature in the several ways. The Greek word for fellowship is the source of the English word “community” (Flora et al. , 2004). Community can mean many things—a group of people who live in the same geographic area (Foth, 2003), a group of people with simila r interest groups (Roy and Bonacich, 1998; Flora et al. , 2004), or social groups (Wilkinson, 1991). Groups of people who live or stay in the same geographic area are referred to as a community of place, residential, or local communities (Foth, 2003). These people may not necessarily share any common characterist ic—age, occupation, interest, or trust. A community of interest, on the other hand, is defined as groups of people or corporations linked together because of a common inte rest or ownership— land, stock, or other 13

PAGE 29

14 material resources (Roy and Bonacich, 1998). Communities of interest are organized around trust, reciprocity, or mutual respect (Flora et al. , 2004). Although in the past people turned to their community of place fo r almost everything, today they associate themselves more with their community of interest. Wilkinson (1991) provides a historical back ground of the evolution and change of the concept of community in so cial sciences and suggests that there are three elements of community: where people live and meet thei r everyday needs together, network of associations for meeting common needs and interests, and pro cess of interrelated actions through which people express their needs and interests in the local society (Wilkinson, 1991). The author emphasizes that although some people believe ther e are problems with these conventional definitions of the community such as fuzziness of geographical boundaries, individual’s linkage s and associations outside the community, and reduced solidarity for collective behavior, community has not ceased to exist in rural America. Community still plays an important role in the life of the individual. An essential element of the community is social interaction (Wilkinson, 1991). Agrawal and Gibson (1999) explore the con cept of community in conservation and resource management. Their analysis reveals three aspects: community as a small spatial unit, as a homogeneous social structure, and as a set of shared norms. Those who believe that the community has a positive role in conservation and natural resource management share this view. For the most part, all th e seven neighborhoods in the study are small spatial units, have homogenous structures, a nd have created their own set of shared norms.

PAGE 30

15 This study explores social sense of community in neighborhoods. Using a geographic definition of neighborhood helps define wildfire risk. Using the word “community” to mean larger geographic or ganization of residents and businesses is suitable because some wildfire education ac tivities come from this level, not the neighborhood. Even though community is defined geographically, the social interaction at the neighborhood level helps to build a community of interest among those who actively reduce their risk of wildfires. Thus many elements of community are relevant in this study. Disasters This section on disasters helps explain the evolution of the definition of the term “disaster”—how it has changed over time and how that impacts disaster preparedness, response, and recovery. Early definitions of the term “disaster” focused on the physical agents that cause the event, labeling some “natural disaster” to emphasize the physical cause of fires, floods, and hurricanes. But this term ignores th e fact that even natural disasters are often a result of inappropriate huma n actions (Dynes, 1993). Many su ch disasters are more of a “social” rather than a “natural” happening. Th ey do not simply happen because of natural forces; they are a result of pre-existing social and ecological forces in the society—socioeconomic status, historical process, and hu man-nature interacti ons (Quarantelli, 1978; Oliver-Smith and Hoffmann, 1999). For example, although the natural force of Hurricane Katrina hit the city of New Orleans, the en suing disaster was also a result of broken levees, poor emergency response, and poverty. Fritz (1961) defines “disaster” as events that are observable in time and space and that may result in physical damage, losses and/or disruption of normal functioning of

PAGE 31

16 community life. He emphasizes that social structure and human activities in the community may be related to the cause and consequence of a disaster ( Fritz, 1961; Kreps, 1984). Different communities might respond to a disa ster event or its threat in different ways. How they act might be a result of th eir past experiences with disasters, economic conditions, cultural orientations, awareness about the issues, and the na ture and extent of the threats (Kreps, 1984). Disaster responses vary, based on whether the action is taken prior to the disaster or as a protective measure during the disaster. For example, during most wildfires, immediate steps are taken by lo cal, state, and federa l agencies to suppress or control the fire so there is minimum damage and no loss of life—this is good disaster management (Quarantelli, 1997). There are also steps that a community can take prior to a disaster to be more prepar ed and to reduce the threat itself—this is good disaster preparedness (Quarantelli, 1997). Such st eps would likely involve the cooperation between both the individuals in the comm unity and the agencies responsible for mitigation. Thus, a disaster preparedness progr am that involves training and resources for the disaster managers as well as education for the landowners should be helpful in reducing the vulnerability of the disaster. Dynes (1993) argues that actions for reducing the threat of disasters should target the local community, as those who will be affected by disasters should be the ones to prepare for it. Giving residents a sense of responsibility about the issue as well as education leads to increased preparedness initiatives and a sense of ownership about the initiative. These strategies may likely im prove the chances for the success of the program.

PAGE 32

17 Wildfires This study focuses on wildfires and particip ation in wildfire preparedness programs to reduce risk of wildfires. Thus it is important to understa nd what a wildfire is, what causes it, and in what ways the risk of wild fires can be reduced. This section deals with the aspect of understa nding wildfires. Fire is a natural part of the ecosystem in most southern forests where plant species are adapted to frequent fires. During the tw entieth century, intensified fire suppression activities excluded fire from the ecosystem . This resulted in changes in forest ecosystems—replacement of pines with ha rdwoods, growth of dense understory, and accumulation of dead vegetation on the forest floor (Long, 1999). These changes created an increased wildfire risk. After several di sastrous fire seasons, people called for a reintroduction of fire in the forests. During the period of th is study, on average, 2 million acres of Florida were burned under prescrip tion every year; this helps reduce fuels and thus abate destructive high intensity wildfires. Wildfire Reduction Methods Prescribed burning (also known as controlled burning) is an economical and desirable practice to reduce the understory a nd debris that may accumulate and become a wildfire hazard. A single prescribed burn can produce a variety of be nefits like improving the quality of the forests and reducing the potential damage of wildfires ( Mobley, 1981; Wade, 1989). It is also used to dispose of debris, limbs, and stems, which may be left scattered at the logging site after harvest and may hinder in firebreak construction; prepare sites for planting and seeding; impr ove wildlife habitat; improve forage for livestock grazing; recycle nutrients; contro l pest problems; improve access; enhance appearance; and perpetuate fi re-dependent species (Wade, 1989). Prescribed burning is

PAGE 33

18 defined as “fire applied in a knowledgeable ma nner to forest fuels on a specific land area under selected weather c onditions to accomplish pre-determined, well defined management objectives” (Wade, 1989, p 2). Some other alternatives such as chemical and mechanical treatments are also being used to reduce heavy fuel loads. However, fire seems to be the most viable option in term s of effectiveness and cost (Wade, 1989). Ecology of Wildfires Fire is a dynamic process, predictable but uncertain, that varies over time and the landscape. Fire has shaped vegetative comm unities for as long as vegetation and lightning have existed on earth. Florida’s na tural ecosystems depend on periodic fire to maintain their health and diversity. Though lightning has been a major cause of fire, 90 percent of today’s ignitions are due to hu man carelessness (e.g., unattended campfires, cigarette butts) or arson (USDA Forest Service, 1995). In Flor ida, the heaviest fire season spreads from mid-April to July (Tanner et al. , 2002). Of any region in the United States, Florida experiences the greatest number of thunderstorm days. About 1,000 lightning-set fires are documented in Fl orida each year (Tanner et al. , 2002). An understanding of fire types and fire behavior is helpful for recognizing the value of activities that reduce risk of fire. Wildfires are classified as brush fire, crown fire, and firestorms based on the height of the forests in which they occur and the fire behavior. Brush fires are fires in vegetation less than six feet tall such as gr ass, grain, brush, and young saplings. Crown fires are generally charac terized as uncontrolled fires that spread in the tops of trees. Firestorms are violent surges in which fire behavior becomes very intense with rapid spread rates through both surface fuels and tree crowns, usually caused by very dry fuels and strong winds (Pyne, 1984).

PAGE 34

19 The general patterns of fire spread are ground fires, surface fires, and crown fires. During a ground fire the organic matter in the soil beneath surface litter is burned and such a fire is sustained by glowing com bustion. Surface fires generally spread with a flaming front and burn leaf litter, fallen branches, and other fuels that are at the ground level. Crown fires burn through the top layer of the forest foliage known as the canopy. These are the most intense of the forest fire s and most difficult to control. Three basic mechanisms of heat transfer in a fire are c onvection (transfer of h eat within an object by molecular activity), radiation (transfer of heat to fuel surface through electromagnetic waves), and conduction (vertical movement of gas or liquid as heat rises). For a fire to move it requires fuel, heat , and oxygen. Vegetation is the fuel in a wildfire, and its arrangement greatly influences the type and extent of wildfire. Oxygen is abundant in the atmosphere. Thus, strategies to reduce risk of wildfires deal with curtailing the other two requirements in the fire triangle—fuel and heat. Wildfires have both advantages to forest ecosystems and disadvantages to human communities. Fire plays an impor tant role in the ecosystem. Certain plant species depend on fire for sprouting and regeneration. Once the fire burns out the understory, the new sprouts may provide nourished food for the wi ldlife. Fire is also recognized as an instrument of change and a catalyst for promoting biological diversity and healthy ecosystems. But uncontrolled fire can lead to loss of human life and property. Dry and windy conditions might intensify the fire. Fire also produces smoke and ashes that may have negative impacts on human health and vehicle safety. Disaster Education An important element of this study is part icipation in disaster education programs, especially wildfire education programs. Thus this section views components of disaster

PAGE 35

20 education programs, role of education in community preparedness, and other elements that affect participation a nd disaster preparedness. Brown (1979) states that di saster preparedness programs should have a component of public education and awareness. He em phasizes the need for non-formal continuing education, an ongoing process that shows participants the consequences of not taking appropriate preventive measures (Brown, 1979) . To be effective and successful, these educational programs should be designed to be specific to the community. For example, landowners in a community at risk of wildfire s need to be told about the factors that contribute to risk of wildfi re in their community and ar ound their houses, steps they can take to reduce the threat around their homes, and ways they can obtain the resources and help from agencies. Luna’s (2001) study “Disaster Manageme nt and Preparedness: A Case of NGOs in the Philippines” describes the different approaches that nongovernmental organizations (NGOs) in Philippines have taken toward di saster mitigation and preparedness. This research found that some NGOs believe that preparedness for a disaster is the responsibility of the government because th ey have great financ ial and technological resources (Luna, 2001). Other NGOs take a mo re pro-active appro ach in involving the community in the preparedness initiatives —community-based disaster management (CBDM). Some NGOs facilitate CBDM th rough community organizing and capacity building. Important elements of preparing a community for disa sters through the CBDM process include training, orient ation on different topi cs like disaster management, disaster preparedness, community organizing, evacuat ion management, environmental education, and emergency response (Luna, 2001). The CBDM process is a proactive approach to

PAGE 36

21 disaster preparedness that involve s people at the risk of disast er to be better prepared for it. To improve community preparedness, step s to reduce the threat from disasters involve planning and action by various social units and should be integrated in people’s daily lives ( Quarantelli, 1978; Dynes, 1993). Th is is the logical outcome of the belief that disasters are social proce sses rather than natural happen ings. Research shows that if people’s perceptions of the probability of a disaster are low, people tend to avoid thinking about disaster (Kreps, 1984); hence people n eed constant reminders in order to take appropriate actions when necessary. For exam ple, repeated prompts to landowners about the need and advantages of mitigation on thei r property would be an important element of an educational program, especially in the s outheast where vegetation grows very quickly. A community’s vulnerability to disaster is directly related to its level of social and economic development. Communities that are socially and economically marginalized are less successful in community-based disa ster management (Bolin and Standford, 1998). One can expect that communities that have higher levels of physical, human, and social capital will be more successful in preparing for a disaster (Buckland and Rahman, 1999) . There are cases, of course, of economically impoverished communities that have a wealth of social capital, perhaps growing out of years of self-resilience. To some extent we might expect communities with high social capital and fewer resour ces to be able to plan for disaster. However, limited financial resources may hamper them. The study “Community Partnerships: Lands cape Level Strategies to Reduce the Risk and Loss from Catastrophic Fires” (Jakes et al. , 2002) , focused on actions by communities to increase wildfire preparedness and the social resources or conditions

PAGE 37

22 necessary to implement and support these pr eparedness activities. Ac tions to reduce the threat from wildfires are influenced by d ecisions made at the individual and the community level (Jakes et al. , 2002). At the individual leve l, decisions about selecting building material and creating defensible space around homes can help reduce the vulnerability from wildfires (Jakes et al. , 2002). Disaster education plays a vital role in educating people about the nature and extent of the disaster and about actions that they can take to reduce their vulnerability to wild fires. At the community level, resources to implement disaster mitigation measures include zoning, fuel reduction ordinances, increased equipment, and training. This study suggests the social foundation th at supports community preparedness for wildfire. This includes social capital, human capital, cultural capital, agency involvement, and landscape. Based on the social foundation, this study identified seven keys to community preparedness for wildfire. These include having a leadership group to coordinate activities; educating the pub lic; recruiting and recognizing leadership; adopting a new philosophy; increasing interaction, trust, ne twork, and support between agencies and residents; developing codes or plans; and building on local knowledge. Education has a role in many of these key preparedness activities. Education programs can support and foster each of these keys to wildfire preparedness. Table 2.1 lists the seven keys to wildfire preparedness along with the role education can play in supporting these keys.

PAGE 38

23 Table 2.1: Education’s role in supporting community preparedness No. Keys to Community Preparedness Role of Education 1 Form leadership group to coordinate activities For successful dissemination and implementation of wildfire education information and activ ities, a leadership group can be formed among community members. This group will be more successful if they develop an education program to inform citizens about the risk, gain support for expenditures, deal with hazard psychology, a nd involve local residents in the preparedness process. 2 Public education Public education can be used to inform residents about fire risk and change individual be havior for fire preparedness activities. This includes risk reduction, fuel mitigation, or structural mitigation activities on their property. Also, education can be used to te ll homeowners about evacuation procedures, emergency pla nning, and steps to smooth recovery. 3 Increase interaction, trust, networks, support between agency and residents Fire preparedness may be a function of the interaction between forestry and fire agencies and residents. The act of educating the public helps increase interaction within agencies and trust between agencies and residents. 4 Adopt new philosophy Reducing wildfire risk may i nvolve changing the focus of existing fire preparedness in itiatives. This could involve more aggressive programs that empower citizens and agencies to take action. 5 Recruit and recognize leadership Leadership is an important part of community preparedness. Education can be used to inform residents in an area about the various ways they can get involved in reducing their community’s risk of wildfires and thus attract interested citizens to take leadership role s in the preparedness process. Also, wildfire education meeti ngs can be used to identify community representatives that would better serve the community’s preparedness initiatives. 6 Develop codes/plans Community leaders a nd other fire and forestry agencies can use information on wildfire preparedness from successful communities to develop codes a nd plans that best suit their community preparedness activ ities. Community education activities can be used to inform residents of new policies and practices to gain their support for the activities, and involve them in action to reduce wildfire risk. 7 Build on local knowledge Local successful fire prepare dness activities can be used as examples to justify the eff ectiveness of the preparedness activities. Also, local knowledge on fire mitigation and risk reduction should be incorporat ed into public education planning and implementation.

PAGE 39

24 Community education should help improve community preparedness by information dissemination and behavior cha nge. Information dissemination may include providing information to suppor t activities, expenditures, and policies; providing information to support planning and policy; and understanding importance of wildfires at the community level. Behavior change may in clude engaging the public in reducing risk, making personal preparedness plans, and partic ipating in community activities to reduce risk of wildfires at an individual level. It also may include attr acting and involving local residents to take leadership roles in thei r community’s preparedness plans and improving interaction and trust among agencies and residents. Wildfire Education Programs As this study focuses on communities in Florida that are at risk of wildfires and that are involved in wildfire edu cation programs in Florida, this section provides an overview of some successful and widely used wild fire education programs in Florida. A wildfire education initiative for resident s of a community is one that provides information on history and effects of wildfire and informs them about actions taken by fire agencies to reduce risk of wildfire. In addition it may help gain acceptance and support for those actions, create a sense of responsibility for protecting th eir own homes during a wildfire, gain support for expenditure s, pass ordinances/dir ectives for reducing risk, educate about steps resi dents can take to reduce thei r vulnerability, and create a norm of reducing wildfire risk. In a comprehensive wildfire program, a variety of communication strategies are us ed to deliver these messages. These include electronic media sources (e.g., websites, televi sion news/advertisements, radio news/advertisements, videos); print medi a sources (e.g., brochures, door hangers, newsletters); and events (e.g., demonstra tions, meetings, community picnics).

PAGE 40

25 Numerous efforts have been established at the national and stat e level to educate citizens about the risk of wildfire and how to reduce the risk. Following are some wildfire educational initiatives that are helping to make a difference by increasing awareness, involving people at risk in the preparedness process, improving partnership among agencies and homeowners, and se tting up demonstration areas. FireWise The FireWise Communities program (FireWis e, 2001) helps people living in a fire prone environment to adopt strategies to be better prepared for a fire before it occurs. The program follows a three-step template that can be adapted to different communities. In the first step, the program leadership or the sponsoring agency invites residents of a community at risk of wildfire to attend a wo rkshop where staff from federal, state, and/or local forestry or fire agencies introduce th em to wildfire concep ts and their risk of wildfire. These natural res ource professionals provide people with information about coexisting with wildfire, along with mitigation information tailored to that specific location. In the second step, the community a ssesses its risk and cr eates its own network of cooperating homeowners, agencies, and organizations. The program generates avenues for networking among the community and agencies. The third step paves the way for people to work together to identify and implement local solutions. In 2005, 61 communities in 22 states across the United States received FireWise Communities USA recognition. Of those, the foll owing nine were in Florida: Briargate, Cypress Knoll, Lakewood, Muse, Pioneer Pl antation, Placid Lakes, River Camps on Crooked Creek, Verandah, and Wedgefield.

PAGE 41

26 Wildland Fire Education Toolkit This toolkit is an assembly of audiovisu al and print sources for extension agents and Florida Division of Forestry staff. It provides them with tools to conduct public programs on wildfire, improve public awareness on wildland fires, and develop demonstration areas on FireWise landscaping a nd uses of prescribed burning. It gives them information on history of fires in Florid a, associated risk to homeowners, ways to improve public participation, a nd steps that homeowners can take to protect their homes from wildfires (Monroe et al. , 2000). Within four months of the program’s inception, county agents and forestry staff reached more than 2000 residents through programs, 23,000 people through county fairs and exhibi ts, and more than 2.1 million people through the local media (Monroe, 2002). Florida Division of Forestry Wildfire Education Initiatives The Florida Division of Forestry (DOF) plays a significant role in educating homeowners about the risk of wildfires through various media and organizing local FireWise programs. The DOF conducts training for Prescr ibed Burner Certification as well as a number of Incident Command c ourses at the new Center for Wildland and Forest Resource Management Training. Four wi ldfire mitigation teams head a statewide effort to reduce hazardous fuel loads near communities. Regional wildfire mitigation specialists coordinate media and education programs. The DOF is also working to increase awareness and understanding about wildland fire and risk reduction practices (DOF, 2003). Some activities include speak ing at local civic groups and community meetings, mobilizing neighborhoods to c onduct hazard assessments, and rallying residents to take precautionary actions.

PAGE 42

27 Social Capital An important element of the study was to understand the relationship between participation in disaster education programs and perception of social capital. This section provides an overview of the literature on social capital—defi nitions, theories, advantages, criticisms, and role of education in building social capital. In 1920, Hanifan provided the first contemporary definition of the term social capital: “those tangible assets that count for most in the daily lives of people, goodwill, fellowship, sympathy, and social intercourse am ong families who make a social unit” (as cited in Putnam, 2000, p 19). Since then severa l researchers in the field of anthropology, sociology, political science, and economics have studied and defined social capital through different perspectives. The book The Well-being of Nations: The Role of Human and Social Capital (OECD, 2001) lists the appro aches that distinguish the c oncept of social capital by analyzing the views espoused by anthropology, sociology, economy, and political science. Literature in anthropology emphasizes that humans have a natural instinct to form groups and become involved in associa tions, while the sociol ogical literature explores the features of so cial organization and motiva tions of humans to form associations (OECD, 2001). The economic lit erature focuses on human strategies for maximizing interest and economic benefits by interacting with others and the political science literature emphasizes the role of inst itutions in shaping human behavior ( Portes, 1998; OECD, 2001). Broadly, social capital can be viewed at two levels: th e macro-level and the microlevel (Krishna and Shrader, 1999) . Macro level social capital refers to relationships and structures in formal organizat ions or institutions. It places emphasis on the vertical or

PAGE 43

28 hierarchical structures and networks. Micro level social capital focuses on horizontal organizations and social networ ks and is further sub-divided into two elements: structural and cognitive social capital. St ructural social capital is mo re tangible and objective. It refers to the social structures such as ne tworks, associations, a nd institutions that facilitate information sharing, collective action, and decision making. Cognitive social capital is subjective and refers to the in tangible aspects of attitudes and norms of behavior, values, trust, and reciprocity (Grootaert and Ba stelaer, 2001) that create conditions for communities to work together. Social capital is “social” because it exis ts in the relationship among individuals. It is called “capital” because it is a resource in which one can invest and obtain a stream of benefits (OECD, 2001). Social capital has char acteristics of a public good. It requires at least two individuals who are associated in some fashion. Thus, social capital exists not in the individuals but in the structure of relationship among individuals. It accumulates as a result of use and diminishes due to misuse, i.e., it increases with interaction among people in a group but can fade if interactions cease. It re sults in increased trust and strengthens the bond within members of the gr oup, resulting in more access to resources of individuals in that group. It is an asset that requires ma intenance, investment, and time. It also leads to the establishment of norms of trust and reciprocity. Theories of Social Capital Essentially there are three major defin itions of social cap ital, one proposed by Bourdieu (1987), another by Coleman ( 1988, 1990), and the third by Putnam (1993, 1995, 2000).

PAGE 44

29 Bourdieu and social capital Bourdieu (1987, p 248) defines social cap ital as “the aggregate of actual and potential resources which are linked to the possession of a durable network of more or less institutionalized relations hips of mutual acquaintan ce and recognition—or in other words, to membership in a group—which provides each of its members with the backing of the collectively-owned capital , a ‘credential’ which entitles them to credit, in the various senses of the word.” He focuses on be nefits accrued to indi viduals by virtue of participation in groups, and the deliberate c onstruction of sociability for the purpose of creating the resour ce” (Pope, 2003). Bourdieu mentions that profits (mater ial and symbolic) are the basis of group solidarity. He explains that although economic cap ital is the basis of all interaction, other forms of capital (social and cultural) produce th eir benefits when this fact is concealed. Social capital helps gain access to economi c resources and increase cultural capital. Transaction of social capital involves uncertain time, uns pecified obligation, and possible violation of reciproc ity (Bourdieu, 1987). Coleman and social capital Coleman (1988, p S98) defines social capital by its function, “not a single entity, but a variety of entities with two things in common. They a ll consist of some form of social structure, and they facilitate certain actions of actors—whether persons or corporate actors— within the structure.” When individuals interact w ith each other, they form social relations that they utilize to achieve their personal goa ls. Over a period of time these relations become a res ource possessed by the individuals. Coleman bases the social process formation on the Rational Choice Theory, which explains that individuals freel y choose to build networks to further their self-interest

PAGE 45

30 (Pope, 2003). Rational and purposeful individuals create social capital to maximize their individual profits and opport unities; these are contracts be tween individuals unrestricted by economic constraints. Social capital is pr oductive and helps achieve certain ends that may not be obtained in its absence. It is not completely fungible, which means that social capital that may be valuable for one activity may not be useful for another activity. Obligation is one form of social capital; it depends on level of trustworthiness of the social environment and extent of obliga tions held. An example would be the whole sale diamond market in New York (Coleman, 1988). Within this market, merchants have free exchange of precious stones for purpose of inspection; this transaction takes place without any formal insurance or documenta tion. Merchants hand stones to each other for inspection without the threat that the other might exchange it for inferior stones. This free exchange can be attributed to the social stru cture of the community that is predominantly Jewish. Typically this is a closed commun ity with strong family ties and religious affiliation; they live and work in the same co mmunity. In fact it is these ties that provide insurance for not defecting, as defecting could cause loss of status a nd wither ties in the society. This ethnic group exhibits high le vel of social capital in the form of trustworthiness, obligations, and formation of norms. Such obligations (in social relations) constitute useful capital for the individuals. Coleman explores the human motives for creating obligations and suggests that rational persons extend obligations (when they are sure of it being repaid at a later stage) only when they expect to obtain something gr eater in return. Social capital also has the potential as an information source and can be useful in setting effective norms and sanctions in the community that may act in th e interest of the collectivity. However, the

PAGE 46

31 Rational Choice is primarily an economically based model and it has been severely criticized in the literature for not being ve ry predictive of the choices people make. Putnam and social capital Putnam defines social capital at the level of the community. He defines social capital as those features of social organiza tions such as networks, norms, and trust that facilitate action and cooperation am ong people for mutual benefit (Putnam et al., 1993). He equates social capital with the level of community participati on and engagement for collective action (Putnam, 2000). He further ad ds that life may be easier in communities that have higher stocks of social capital (Putnam, 1995). Putnam analyzed the social organizations (e.g., the PTA, women’s club, and churches) as networks of civic engagement that foster norms of reciprocity, develop trust, and fac ilitate coordination and communication for mutual benefit. Putnam (2000) in his book Bowling Alone looks at trends in civic engagement and social capital in the U.S. Like Coleman, he t oo explicates externaliti es of social capital, such that all the benefits do not accrue to the lone individu al who makes the contact; the wider community may also benefit. He specifica lly looks at trends in people’s political, civic, and religious particip ation; trends in formal (work place); and informal connections, such as trust, reciprocity, altr uism, and honesty and explains several factors that have impacted participation and social capital in the U.S. Compare and Contrast Social Capital Perspectives The definition of social capital proposed fo r this research integrates these different perspectives as stated by Bourdieu, Cole man, and Putnam and provides a more holistic viewpoint . Many researchers believe these varying pe rspectives of social capital are not alternatives, but in fact complementary to one another (The World Bank, 1998; Wall et

PAGE 47

32 al. , 1998; Grooatert, 1998; Stone, 2001). Putn am (1993), Coleman (1990), and Bourdieu (1987) each view social cap ital as a resource for colle ctive action, however their approaches differ in the outcomes of social capital. Social capital outcomes are described by Bourdieu in terms of ec onomic well-being, by Coleman for educational achievement and acquisition of human capital, and by Putn am for prosperity of communities and nations. All three theorists agree that social capita l is a valuable resource that accumulates over a period of time and cannot exist outside the broader community context. However, they differ in their reference of size of the community: Bourdieu talks about class factions; Coleman about families and organizations; and Putnam stretches it to the national level (Wall et al. , 1998). For all three, social capit al is goal oriented and plays a role in helping individuals a nd groups achieve what they deem important. However, they vary on scale. While Bourdieu and Colema n focus on individuals and the resources appropriated to them, Putnam looks at the regi onal level. Bourdieu vi ews social capital as individual power; Coleman views its use to improve educational achievement and increase human capital; Putnam sees social cap ital as improving the prosperity of nations. All view social capital as social control; for Bourdieu social capital as social control is exclusion of some people, for Coleman a nd Putnam it is a way of setting norms and standard in the community to realize safety and smooth functioning (Wall et al., 1998). All three theorists include quantitative measures of social capital assessment (Wall et al. , 1998). However, Coleman specifies both qu alitative and quantitative analysis, or quantitative analysis that use qualitative indicators (Coleman, 1990). Coleman’s indicators of social capital include parent ’s presence and intera ction with the child,

PAGE 48

33 number of siblings, mother’s expectation of child’s education, and frequency of interaction and talking with parents. Some indicators us ed by Putnam include personal participation variables like vol unteering and leadership in asso ciations or voting, informal sociability, beliefs abou t trust and honesty, and a count of number of associations people belong to. Value of Social Capital Social capital provides a number of bene fits to the communities in which it occurs. Several studies have documented the role of social capital in alleviating poverty (Collier, 1998) , fostering economic growth (Putnam, 2000), increasing academic achievement (Coleman, 1988; Israel et al. , 2001), and attaining jobs (Gra novetter, 1973). High social capital is associated with cooperative probl em solving, effective governance, and rapid economic development (Putnam et al., 1993). Social capital plays an important role in fostering social networks that are needed to take collec tive action (Putnam, 2000). When a community has large stocks of social capital, pe ople in that community should be more willing to participate in community activiti es and solve common problems together. For example, residents in a subdivision vulnerable to wildfires with high social capital may be more successful at working together to c onduct a prescribed fire , install a dry hydrant, rent a chipper, or widen an access road. Although social capital accumula tion leads to a stream of benefits for individuals, it may also have less desirable consequences. A very tight and closed group structure may lead the community to excl ude certain individuals fr om the group. Excessive group closure may also lead to the problem of free riding (Portes, 1998), where individuals in a group might try to extract benefits from th e well-off individuals, and in addition may restrict the progress of these individuals by constantly tapping on the freely available

PAGE 49

34 resources. Another negative consequence of group solidarity is that it might create demands for conformity by setting restrict ions on personal freedom (Portes, 1998). Although group solidarity may, in most cases, be beneficial for indivi duals in that group, it might not always lead to desirable outcome s for the society. For example, it is the strong bonds of trust that exis t in groups like crime syndicates and gangs that have detrimental effects on the social fa bric of the society (Portes, 1998). Contrasting Views on Social Capital Several researchers have debated the use of the term “social capital” for numerous reasons. Some believe that the use of the word capital, which is traditionally used in economic terms, implies that community i nvolvement and political participation are forms of economic activities (Smith and Ku lynych, 2002). Also, some believe that the word “capital” implies traditional aspect of capitalism, such as individualism, competition, and pursuit for wealth that are not necessarily characteristics of social capital as used in literature (Coleman, 1988; Putnam, 2000) DeFilippis (2001) argues that the interest of community development theorists in the term social capital is misguided because the concept fails to encompass power issues in the development of communities. DeFilippis criticizes Putnam’s scale of measurement of social capital—taking individu al attributes and aggregati ng them at the level of the community. He further adds that commun ities are outcomes of social, economic, political, and power-laden relationships bot h internally within the community and externally with the rest of the world (DeF ilippis, 2001). He emphasizes that for social capital to have meaning, it needs to be reconnected with economic capital. For the purposes of this study the term “soc ial capital” will be used because of the belief that capital encompasses both ec onomic and non-economic benefits that

PAGE 50

35 individuals, groups, and communities can achie ve because of the structure of their relationships. In this study social capital is measured at the leve l of the individual and defined as an individual’s perception of th e quality of relationship he/she holds with his/her neighbors and other people in the co mmunity, and community characteristics like community participation. Indicators and Measures of Social Capital As there is no one definition of social cap ital across all disciplines, measures of social capital vary. However, in the last decade, researchers have worked to create tools that are a reliable and valid measure of social capital (The World Bank, 1998; Krishna and Shrader, 1999; Stone, 2001). Stone (2001) defined the core dimensions of social capital as two aspects: the structural aspect of social re lation (networks) and the quality of social relations (norms of trust and reciprocity arising from them). Clear ly stating the component s of social capital helped her define the measurable characteristics of each component. She emphasizes that the structural component can be m easured by looking at the following: Structural dimension of networks (formal/informal) Size of the network (number of individuals) Spatial arrangement (household/global) Type of the network (vertical/horizontal) She classified trust into social and civic trust; social trust defined in terms of personal trust, like trust in family members a nd relatives, and civic trust as generalized trust. Reciprocity includes ex changes or favors that individu al or groups might do for one another with or without expecting anything in return (Stone, 2001).

PAGE 51

36 Community participation is one important measure of social capital ( Scaff, 1952; Hagedorn and Labovitz, 1968; Kasarda and Janowitz, 1974; Buckland and Rahman, 1999; Putnam, 2000). It is measured by member ship in organizations, participation in community activities, membership in PTA, religious bodies, attend ing meetings, etc. Literature on social capital suggests that communities with higher levels of social capital function smoothly (Putnam, 1995). Putnam analyzed the social organizations as networks of civic engagement that foster norms of reciprocity, encourage the emergence of trust, and facilitate coordination and communication for mutual benefit. Neighborliness is one aspect of informal social capital (Bullen and Onyx, 2000). Research also indicates a close correlati on between social trust and associational membership (Bullen and Onyx, 2000). Broadly, researchers have measured social capital at the individual, community, and national level by looking at member ship and participation in community organizations, voter turn-out, attachment to pl ace, participation in religious organizations, informal relations like friendships, work-bas ed associations, nei ghborliness, norms of trust, and reciprocity. It is common to ascertain social capit al from surveys. For this study, mailed surveys were used, followed by quantitative an alysis to measure social capital using qualitative indicators, which was obtained from literature review and interviews with community leaders. Independent variables Social capital is also correlated with th e community demographic measures such as length of residence, age of residents, population size, soci o-economic status, occupation, and level of education ( Scaff, 1952; Hagedorn and Labovitz, 1968; Kasarda and

PAGE 52

37 Janowitz, 1974; Foskett, 1995). It is higher in places with l ong-term residents and higher levels of education, for example. Scaff (1952) studied the effects of co mmuting to work on participation in community organizations. Results of the st udy indicate that commuter families had the lowest participation score while retired people in the community had the highest participation. Length of residence in the community too was found to influence participation. Residents who lived in the community for more than 20 years had the highest rate of participation as compared to those who lived there for 3 to 5 years or less than 3 years. Educational status measured through the number of years of formal education indicated that participation wa s least among persons who had not completed high school, and highest among those with coll ege education. High part icipation rates for people who had additional schooling beyond co llege indicated a positive relationship between participation rates and college education. Participation rates varied with differences in occupational groups, which were highest among professionals and public service people and lowest among those employed in the industry. Kasarda and Janowitz (1986) examined th e sociological factors that influence community participation and attachment. They developed a model of community attachment that measured the impact of popul ation size, density, le ngth of residence, social class, and stage in life cycle (young family, middle-ag ed, retired) on friendship, kinship, associational bonds in the community, and on attitudes toward the community. Noteworthy in the study is that population size and density do not significantly effect local friendship, kinship, and associational bonds. Length of residence was shown to have positive effects on social bonds, but not on partic ipation in informal social activities in

PAGE 53

38 the community. Interest in community affairs is shown to incr ease with length of residence and decline with age. This study indicates that le ngth of residence is the key exogenous factor that influences attachment to the local community that in turn has an effect on participation in community affairs. Many of these demographic variables may al so influence the success of a disaster education program. Communities of long-term residents with greater income and education may have the commitment and ability to participate in disa ster preparation to a greater degree than others. They may also percei ve they have a lot to lose in a disaster and therefore work hard to prevent one. Fo skett’s (1955) study on pa rticipation in policy formation process at the community level deve loped a scale to measure factors associated with community participation. He found th at participation scores were positively correlated with the level of fo rmal education and income leve l of residents. Hagedorn and Labovitz (1968) also reported a direct positive relation betw een level of education and social capital. Studies suggest that social capital is linked to a number of different variables. However, things that correlate with social capital are not always stable. For example, some communities may have high social capital but insufficient funds. Other communities may have lower social capital ev en if people have lived in the community for a long time. Criticisms of social capital indicators Indicators used to measure social capital include trust and trustworthiness, reciprocity, networks, sense of community, and participation. There are four criticisms for indicators that measure social capital (P ope, 2003). The first criticism is that these indicators lack clear definition. Trust a nd reciprocity are vague and ambiguous. In

PAGE 54

39 addition, researchers emphasize the need of a clear distinctio n between sources (indicators) of social capital and outcomes of social capital. To counter this criticism in this research, interviews were conducted to better understand how people perceive social capital and used those perceptions in the de velopment of the social capital scale. The second criticism deals with the unit of measurement—individual or community—and using individual level data to describe collective social capital. Putnam has been criticized for making individual social capital the property of communities and nations (Portes and Landolt, 1996). Additionall y, the resources available to individuals and groups as a result of their so cial capital and factors that affect social capital at an individual level may be different than thos e at the community level. To counter this criticism, this research makes the clarificati on that it is measuring social capital at the individual level, more specifically measuring individual’s perception of social capital. In this study, no attempt has been made to gene ralize the findings from the individual level to the collective community level. The third criticism is the glorification of outcomes of social capital. Most research on social capital assumes positive outcomes. However, as mentione d earlier the use of social capital may have some undesirable cons equences. The fourth criticism is that the individualized notion of social capital may enforce inequality and in some cases may not even work (Leeder and Dominello, 1999). There are several criticisms of social capital measurement that helped define this study. Role of Education Institutions in Building Social Capital There is some evidence to suggest that educational institutio ns may facilitate actions for mutual benefit, thus potentially enhancing social capital formation. One example is the establishment of agricultura l extension programs and community colleges

PAGE 55

40 (Putnam, 1995). Extension programs can br ing various institutional actors—church groups, health services, senior groups, businesses, and banks—together to improve quality of life of people in these groups. Ex tension programs facilitate development of public educational materials to build indi vidual and community cap acities (from website http://www.cpn.org ) for problem solving and decision-making. Educational endeavors like Service Learning programs that are aimed at engaging youth in community projects and problem-solvi ng skills are activities that may increase social capital. Role of Social Capital in Disaster Preparedness As mentioned earlier, social capital is positively correlated with economic development and individual and community devel opment. It is also a factor in disaster preparedness. Community well-being is meas ured by economic and social indicators like income level, productivity, education, dens ity of civic engagement, and normative behavior in a social sphere (Bucklan d and Rahman, 1999). A comparison of three communities in the 1997 Red River flood in Canada indicates that economic status, human capital, and social capital play an impor tant role in effective disaster management. Such management requires social organizations that are partic ipatory in nature and that foster cooperation among various sectors in the communit y. These characteristics link disaster preparedness with social capital. Buckland and Rahman’s study (1999) emphasized both the benefit and cost of social capital. For example, although social capital can effectively mobilize people to work together and solve problem s, it may also lead to delay in decision making due to the need to reach consensus w ith all the stakeholders.

PAGE 56

41 Social capital can be built or rebuilt th rough an incremental process that helps individuals break out of thei r isolation, brings them into interaction with other individuals, improves connectedness among one another, and makes them responsible for their actions (Wilson, 1997). Thus , a disaster education program may be successful at improving disaster preparedness and also in bu ilding social capital when it introduces its participants to the threat; provides viable solutions for mitigating the threat; brings together individuals and agencies; involves participants in activities that require cooperation, coordination, trus t, and understanding; improve s or builds decision-making and problem-solving skills; and makes partic ipants responsible for their actions. This Research This research looks at social capital as an indi vidual’s perception of the quality of relationships he/she holds w ith his/her neighbors and other people in the community, and community characteristics like community par ticipation. Social capit al in this context means simple things that people might do for one another (e.g., keep an eye on a neighbor’s house when they are not in town and take the neighbor’s dog for a walk), or things they may do with one another (e.g., barbeques and fuel removal for reducing risk of wildfires), or comfort each other in times of grief and share in times of happiness. They may share tools and material resources (e.g., clippers or chain saws for creating defensible space), share information on j obs, websites on wildfire information or FireWise plants, childcare, and may support one another’s cause. They make these efforts without expecting a direct economic or non-eco nomic gain in return. Their actions and deeds with neighbors may be a result of longterm relationships; their actions in the community may be a result of their position in the community. In addition, their actions may have been stimulated from a perception of risk and the desire to avoid negative

PAGE 57

42 consequences or as a result of a situation that stimulates interaction among members of the community (e.g., a wildfire educational program). Social capital exists in the relationships among individuals. It helps individuals achieve a variety of things, (e.g., inform ation on jobs, safer communities because of social norms, and reduced risk of wildfire s), that might not ot herwise be achieved. Simply stated social capital is the combination of econo mic and non-economic benefits achieved by individuals, groups, and communities because of the structure of their relationships. Social capital can be built or fostered by participation, interaction, and communication and nurtured by trust, sense of trustworthiness, reciprocity, and mutual respect. Broadly, this is the theoretic al framework in which this research study was carried out. However, a connection must be made betw een social capital at the individual level and at the community level. A lthough benefits of social cap ital may be acquired as an individual, there can be spill over effects for those who do not participate. In case of wildfire education prog rams, although the ultimate motive of participation is protection of individual hom es and property (economic), there are social benefits of retaining an intact community. As wildfire is a risk to the community, it is the individual and community actions together that help reduce the community’s vulnerability to wildfire and make it better prepared for a wildfire (Jakes et al. , 2002). Thus creating norms of wildfire preparedness is helpful to reduce risk of wildfire at the level of the individual and the collective. When people in the community interact with each other as individuals for alleviating risk of wildfires, the networks formed (formal and informal) are resources

PAGE 58

43 possessed by the individuals in the community or by the community. These networks can be used to achieve a variety of objectiv es. Although success of a community-based educational program can be attributed to exis ting networks in the community, it is also possible that participatory educational programs may help develop or strengthen new relationships and networks—an added res ource for individuals in a community. Figure 2.1 summarizes the rese arch hypothesis through a mode l of relations hip diagram between participation in a wildfire educati on program and perception of social capital. The relationship diagram aids in the understand ing of the objectives of the study and the confounding variables that may pot entially effect perception of social capital and risk reduction activities. This study attempts to understand the relationship be tween wildfire education programs and people’s perception of social capital. It does not attempt to show whether one causes the other, just whether there is a relationship, which is symbolized by the double arrows in the diagram. There are thr ee main sources of information regarding wildfire education programs—electronic, print, and participation in ev ents. The nature of interaction in the education program may help individuals become aware of the risk of wildfires and actions they can take to reduce th at risk. It also help s them understand civic responsibilities and creates opportunities for them to work together. The educational programs can create platforms for people to ta lk about a concern, take responsibility, or improve communications and networks, whic h can help enhance social capital.

PAGE 59

44 Wildfire educational program Social capital Electronic Print Events Use information to reduce wildfire risk Become aware & knowledgeable about wildfire & risk reduction actions Understand civic responsibilities Enhance partnerships Increase volunteerism Improve communication Increase networks Renewed community interest Take responsibility Move out of isolation Gender Education Residency status Length of residence Lot characteristic Experience of wildfire Perception of wildfire risk Work together to reduce risk Nature and level of interaction in the education program Create a platform for people to talk about a concern Minimize negative impact of wildfire Networks Trustworthiness Employment status Ethnicity Income Age Confounding Variables Sources of information/ participation Reciprocity Partici p ation Sense of community Figure 2.1: Model of relationship between wild fire education programs and perception of social capital in communiti es at risk of wildfires

PAGE 60

45 The relationship diagram (figure 2.1) al so lists the confounding variables that literature suggests affect per ception of social cap ital: gender, age, education, income, length of residence, ethnic ity, residency status, and employment. Other confounding variables include those that impact action to reduce wildfire risk: lot characteristics, experience with wildfire, and perception of wild fire risk. Thus it would be expected that people who have higher incomes, are older, more educated, and retired; and have lived in the community for a longer time would perc eive higher social capital than people who have lower incomes, are younger, less educate d, and employed, and have not lived in the community very long. Also, it would be expected that people who have a greater risk of wildfire due to lot size or perception of risk to be more engaged in fire education programs. To develop the survey, interviews were conducted with program leaders in each neighborhood. A mail survey based on the relationship diagram was used to collect data from individuals on each of the demographic charact eristics. Also, analys is was conducted to determine a relationship between perception of social capital and these demographic characteristics. If the demographic variables show a relationship with perception of social capital, then it would suggest that it the demographic variables that are causing the differences in people’s perception of social capital. However, if it does not show a relationship then it could suggest that someth ing else is causing di fferences in people’s perception of social capital. Summary This review of the literature defines disa ster as a social ra ther than a natural phenomenon and a result of pre-existing social and ecological forces and human interaction (Quarantelli, 1978; Oliver-Smith and Hoffmann, 1999). Research on disaster

PAGE 61

46 preparedness focuses on the need for training and orientation on differe nt topics such as disaster management, activities for disa ster preparedness, community organizing, evacuation management, environmenta l education, and emergency response to prepare a community for disasters (Luna, 2001). It al so emphasizes the need of non-formal, community-specific, issue-specific, continui ng education that informs people about the risk and consequences of not taking preventive measures (Brown, 1979). A disaster education program can help pr oduce a citizenry that is aware of and motivated to take actions to solve an envir onmental problem. In the case of a wildfire education initiative, a direct intended outcome may be increas ed preparedness efforts that decrease the community’s vulnerability to wildfires. Based on the strategies employed in an initiative, an indirect outcome may be increased trus t, improved communication, and better relationship among residents and with agencies. Social capital plays an important role in fo stering social networks that are needed to take action. Thus, if a community has large “stocks” of social capital, then people in that community are more willing to participate in community activities and take actions to solve problems they face in common, such as taking steps to increase preparedness for wildfire. Also, existing strong social networ ks (formal and informal) and the norms of trust and reciprocity among individuals in a community might re sult in successful implementation of a wildfire educational initiative. For example, a community in which individuals communicate with one another, participate in community activities, are concerned about the well-being of people in their community, and have strong social networks, might be more successful in adap ting to and implementing a program targeted

PAGE 62

47 toward reducing the risk of w ildfires, as compared to a community where individuals do not trust or interact with each other. Thus both disaster education initiatives a nd social capital can assist in community preparedness for disasters. This study is aimed at exploring the relationship between disaster education programs and social capi tal. Putnam (1993) suggests that cooperation among individuals in a community is important for long-term mutual benefits and this depends on formation of social capital. If components of a disaster education init iative that may improve or build social capital can be extracted, then these initiativ es can play an important role not only in building a community that is prepared to pr otect itself from an impending disaster, but also in building networks and relationships that the community can use to solve other problems. This study aims to assess the value of aspects of wildfire education programs to determine which elements might improve social capital.

PAGE 63

CHAPTER 3 METHODOLOGY The objective of this stu dy was to explore the relationship between wildfire education initiatives and social capital in communities at risk of wildfires. The study used in-depth interviews, Conceptual Content Cognitive Mapping (Kearney and Kaplan, 1997), and mail surveys to collect data from i ndividuals in three communities in Florida. In-depth interviews were used to obtain information on wildfire education programs in each community, including goals, objectives, ta rget audience, anticipated outcomes, and unintended consequences. Personal intervie ws using Conceptual Content Cognitive Mapping (3CM) prompts provided an understanding of people’s perceptions of their neighborhoods and interaction in their ne ighborhoods. The 3CM interviews were a qualitative data collection tool that aided in the design of the quantitative survey and better understanding of individua ls’ perceptions of social capital. The mail surveys afforded a large sample size, which was useful in statistical analysis. The quantitative survey helped explore the relationship betw een participation in wildfire education programs and perception of social capital. The research used mixed methods to better understand perceptions and practices. The unit of analysis for this research study is the individual. This study did not attempt to show a causa l relationship betwee n participation in education programs and an indi vidual’s perception of social capital because most of the neighborhoods had already participated in some sort of wildfire educational program and pre-test was not possible. A control group could not be used because all neighborhoods in 48

PAGE 64

49 the study were exposed to some kind of w ildfire educational program. The dependent variable for the study was i ndividual perception of soci al capital. The independent variable for the study was compos ed of participation in wildfi re education programs, age, educational level, income, employment, ethnic ity, gender, and length of residence. Risk reduction activities were treated as both an independent and a dependent variable. Research Questions The study addresses the follo wing research questions: 3. What is the relationship between disaster education programs a nd social capital? 4. What elements of an educational program are more likely to be associated with enhanced social capital? Study Communities Selection of Study Communities Three communities in Florida were select ed based on their risk of wildfires and varying levels of participati on in wildfire educa tion programs. Selection of two of the communities (Wedgefield and Palm Coast) was facilitated by the national study “Community Partnerships: Landscape Level St rategies to Reduce the Risk and Loss from Catastrophic Fires” (Jakes et al., 2002). Identification and selection of the other community (Lake Wales Ridge) was based on cons ultation with forestry officials and The Nature Conservancy (TNC) in Florida. A total of seve n neighborhoods in the three communities were selected for the study. A ll seven neighborhoods were at risk of wildfire, had been exposed to some educatio nal program to educate residents, and were taking steps to increase wildfire prep aredness. The neighborhoods selected had populations between 100 and 2700. The following communities and neighborhoods were selected for the study:

PAGE 65

50 Wedgefield (in Orange County, east of Orlando) Two neighborhoods of Palm Coast (in Flagler County, north of Daytona)— Seminole Woods and Cypress Knoll Four neighborhoods in the ecosystem on the Lake Wales Ridge (in Polk and Highlands County)—Tiger Creek Forest, In dian Lakes Estates, Placid Lakes, and Leisure Lakes. For the purpose of this study, communities are Wedgefield, Palm Coast, and Lake Wales Ridge. They are defined as being geogr aphically based, having a local society, and taking locality-oriented actions to increase wildfire preparedness. A neighborhood refers to a subdivision within the larger community or geographic area. Cul-de-sacs and street patterns limit access to these subdivisions, making them a geographically defined neighborhood. They have homeowners associatio ns, newsletters, and varying degrees of infrastructure that helps create an identity. Wedgefield is identified as a community and a neighborhood because it is isolated; it is one huge neighborhood with no commercial zone. No baseline data were available on people’s perceptions of social capital before the implementation of the wildfire educational programs. Thus, in this study data were collected from participants a nd non-participants of wildfire education programs in each community. There are two basi c assumptions of the study: 1. People in each of these communities have been exposed to both a variety of and varying levels of educational program s and messages using one or more strategies. The sum of these programs and messages will result in people with greater and lesser exposure. As a result, they face similar risk but may not have similar levels of knowledge about wildfire issues. 2. Individuals perceive higher and lower levels of social coherence.

PAGE 66

51 Figure 3.1: Map of study communities Palm Coas t Wed g efiel d Lake Wales Rid g e Cypress Knoll Se m i nole W oods Tiger Creek Forest Indian Lake Estates Leisure Lakes Placid L akes Description of Study Communities Figure 3.1 provides a map of study co mmunities (Florida County Outline Map, 2006). Summarized below is a brief descri ption of the study communities: location, population, wildfire history, w ildfire threat, and educat ion/communicatio n activities related to wildfire preparedness (For detail ed summaries of wildfire education programs for each community see Appendix A, p 157.) Wedgefield Wedgefield is a large subdivision of 2700 people (2000 census) spread over 6500 acres. It is located 23 mile s east of downtown Orlando, well outside the city limits.

PAGE 67

52 Although development is quickly moving toward Wedgefield, the community is currently surrounded by undeveloped land, including a larg e private ranch and a county park. This area has a history of periodic wildfire s, as is typical in central Florida. When development began in Wedgefield, however, fire was suppressed. This led to an accumulation of heavy fuel loads and an incr eased risk of wildfire. In 1998 a large wildfire approached Wedgefield from the northeast and many residents were evacuated. If not for the change in wind direction at th e last moment, the comm unity might have lost homes. This fire was a wake-up call for people in Wedgefield, alerting th em to the risk of wildfire and setting the st age for community actions. In April 2001 a Wildfire Risk Assessment Team made up of DOF and Ora nge County Fire officials carried out a wildfire hazard assessment in Wedgefield. The assessment placed Wedgefield in the high hazard category. Based on this, Wedgefield was encouraged to become a pilot community in the National FireWise program, the first in Florida. The FireWise committee of Wedgefield used several strategies to make residents aware of the risk of fire. These strategies include FireWise meetings, new homeowner packets, setting up tables with information at yard sales and golf tournaments, FireWise articles in the Wedgefield newsletter, fire aw areness days, and modifications to codes and covenants. Palm Coast Palm Coast was designed as a model community in the early 1970s. A total of 42,000 acres were platted and sold in quarteracre lots. Many of the early inhabitants were retirees from New York, New Jersey, a nd Michigan, anxious to enjoy their later years with sunshine, ocean breeze, and lowmaintenance lots. Most of them had no idea they were moving into an area that was likely to experience wildfire.

PAGE 68

53 An early court agreement prevented developers from building Palm Coast in phases, so lots were sold w ith no restriction on when buildi ng would occur. This resulted in many vacant lots owned by absentee owners across the world sprinkled among homes with mowed lawns. The unbuilt lots contain you ng pine trees, saw palmettos, wax myrtle, and vines. Where the vegetati on is not managed, it is a wild fire hazard to neighbors, despite being a wildlife habita t and useful privacy screen. Palm Coast has experienced two major wildfire events. In 1985, 130 homes were lost when two wildfires swept through the co mmunity. In 1998 a complex of several fires became a major disaster. The city and the en tire county (100,000 people) were evacuated over the July 4 weekend. Seventy homes were lo st. No lives were lo st in either fire. After the 1998 fires, several actions were initiated at the st ate, county, and city levels to reduce the risk of wildfires. Thes e included purchasing additional equipment to fight fires (e.g., trucks, helicopter), improve d wildland fire suppression and incident command training, mitigation to reduce hazar dous vegetation, improved evacuation plans and communication procedures , and educational programs and communication activities to educate residents about actions they c ould take to reduce their vulnerability to wildfires. Palm Coast: Cypress Knoll: This is the first pilot FireWise community in Palm Coast. The local Neighborhood Watch program helped launch the FireWise program. A demonstration project to retrofit one home and mitigate the surrounding vegetation was conducted to show people how to comply with their fuel reduction or dinance; especially how much vegetation could be left around a home. The organizers took photographs before and after the activity and organized ne ws coverage of the demonstration project.

PAGE 69

54 Cypress Knoll received its FireWise Commun ity certification in 2004. This part of Palm Coast has not experienced a wildfire. Palm Coast: Seminole Woods: This is the second neighborhood in Palm Coast that is in the process of being cer tified as a FireWise community. In the 1998 fires, 20 homes were destroyed, and 17 others were damaged. Of all the neighborhoods in Palm Coast, it has the greatest number of unbuilt lots, making its residents quite vulnerable. Lake Wales Ridge The Lake Wales Ridge geological feature is primarily located in Highlands and Polk counties and includes remnant scrub ecosy stems. The Ridge has a large collection of rare plants and animals. The Ridge’s well-dr ained sand attracts both citrus farming and developments and has resulted in approximate ly 85 percent of the dry uplands being converted to cultivation and commercial and residential development. Many species of plants and animals at the Ridge are threatened to extincti on. Lightning fires are prevalent throughout the region and tend to create habita ts that require frequent, patch fires to maintain biodiversity. The area is extremely popular with reti rees and has contributed to the recent population in crease here. The region was severe ly impacted by hurricanes in 2004. Lake Wales Ridge North: Tiger Creek Forest (TCF): This is a community of about 100 people (in 2004) on 1000 acres. It is located 3 miles east of downtown Lake Wales on State Road 60. When the commun ity was developed on cattle grazing land in the 1980s, most homeowners were retirees. In the past 10 to 12 years, younger people have moved to the community. TNC owns so me land in Tiger Creek Forest and an adjacent 5000-acre preserve. It conducts pe riodic prescribed burns on these lands.

PAGE 70

55 Tiger Creek Forest had a 200-acre wildfire on the southern portion of the property in the 1980s; no lives or homes were lost. The community also suffered another fire 8 to 10 years ago. The community is located in a wooded setting and is densely overgrown, which creates a high risk of wildfire. Lot size varies from 5 to 30 acres. Of the 194 sites, 100 have homes or trailers. In order to address the is sue of increased risk of wildfires, TNC and the DOF organized a community picnic. The Tiger Creek Forest homeowner association promoted the picnic, where members began to discuss th eir vulnerability and strategies for reducing their risk of wildfire. Lake Wales Ridge North: Indian Lake Estates: This is a community of more than 1000 people (in 2004) of which about 35 percent are seasonal residents. The community is spread over 8000 acres, most of which (7458 acres) are platted, and about 500 are developed with single-family residen ces. Most lots average about one-half acre. The initial residents were retirees, but now younger families are moving in. In 2004, approximately 15 percent of the community was built, and the vacant lots in the community contributed to risk of wildfires. Indian Lake Estates had a large fire in 1999. No homes were lost, but 10 homes were damaged. Education/communication activities in I ndian Lake Estates include community meetings and fuel reduction activities by the DOF to manage vegetation on vacant lots. Lake Wales Ridge South: Placid Lakes: This is a community of 2500 people (in 2004) spread over 3,333 acres. Many people who moved into Placid Lakes had careers in the armed forces. It has a strong and active homeowners asso ciation, with many

PAGE 71

56 volunteers. During the period of the study, about 80 percent of people participated in community events. In 1999, Placid Lakes had a wildfire in the wooded areas of the community. This prompted residents and the homeowner associ ation board to become more aware of the threat of wildfires. They initiated a di scussion with the DOF and learned about the FireWise program. As a result, five peopl e from Placid Lakes attended a FireWise workshop in Brooksville, Florida. The Fire Wise program was announced at a general membership meeting for Placid Lakes, and th e community had its first FireWise day in October 2004. Lake Wales Ridge South: Leisure Lakes: This area is more rural than Placid Lakes. Of the approximately 8000 lots, only 25 pe rcent are built. The neighborhood has an active homeowners association. The neighborhoo d experienced a fire in 2004. Only 300 acres burned; no homes were lost or damage d. The area is sparsely populated and this poses a major threat for wildfires. However, neither the homeowners association nor the DOF have made any efforts to educate the community about wildfire protection. The only activity of wildfire prevention in Leisure La kes is the prescribed burns conducted by the local fire department around the area. Organization of Study Communities Based on these elements, a 2 x 2 matrix with varying levels of wildfire education program participation and social capital was created (see Table 3.1). Neighborhoods for this study were selected to show diversity in involvement in education programs and perception of social capital as shown in Table 3.1. Based on the interviews conducted with program develope rs and leaders in each neighborhood, it is assumed that Cypress Knoll falls in Categor y A, Seminole Woods in Category B, Tiger

PAGE 72

57 Creek Forest in Category C, Placid Lakes in Category A or C a nd Leisure Lakes and Indian Lakes Estates in category B or D. Wedgefield residents have varying levels of social capital depending on where they live in the neighborh ood, but the entire community has been exposed to the same educational program; hence it would fall in category A and C. Table 3.1: Community organization criteria Involvement in wildfire education program Information Gathering: In-depth Interviews at Community Level In-depth interviews were conducted with 20 program developers and implementers in seven neighborhoods to obtain information on the wildfire education programs in each neighborhood, including goals, obj ectives, target audience, anticipated outcomes, and unintended consequences. Information gathered from the interviews was used to generate questions on the survey. From these intervie ws, wildfire education program descriptions for each community were generated. (For more information on the wildfire education program in each community, see Appendix A, p 157.) The interviewees reviewed their respective program descriptions. Social Capital High Low A B Cypress Knoll Seminole Woods C D Tiger Creek Forest High Low Placid Lakes Wedgefield Leisure Lakes Indian Lake Estates

PAGE 73

58 Qualitative Research: 3CM in Each Neighborhood 3CM is a cognitive mapping technique designed to provide “graphical representations of knowledge structures embod ying people's assumptions, beliefs, facts, and misconceptions about the world” (K earney and Kaplan, 1997, p 580). The method helps assess people’s mental models—factors that an individual perceives as being important and relevant for a particular issu e and the structure of relationship among those factors (Kearney et al. , 1999). According to Kearney and Kaplan (1997, p 583), “mental or cognitive models of both physic al and conceptual aspects of the world are necessary to enable people to think about things that are not present in the environment and to access information related to the problem at hand.” It helps participants differentiate between factors that they own and those they do not. In this study the participants were asked to respond to a scenario followed by a prompt that helped generate components that were relevant to them. This method has been used to assess pe ople’s cognitive maps to explore issues such as the values of stream health (Bou ma, 2000), forest management in the Pacific Northwest (Kearney et al. , 1999), promotion of carpooling (Kearney and De Young, 1995), perspectives of non-industrial priv ate forest owners (Irvine, 1997), and environmental policymaking (Austin, 1994). It can be used in education research to study changes in student’s cognitive models before and after an educational intervention or to design effective educational interventions. It can also be used in needs assessment, formative and summative evaluation, or in answering different research questions. 3CM is a card sorting technique that fo cuses only on issues that the respondent believes are relevant to th e topic. It has two implem entations: open-ended 3CM and structured 3CM. Austin and Kaplan de veloped open-ended 3CM (Austin, 1994). The

PAGE 74

59 method works well in exploratory studies with relatively small sample sizes. Structured 3CM can be used when dealing with larger sample sizes and when a more thorough statistical analysis is desired. A 3CM exercise provides a “map” of how the participant conceptualizes the topic. The map has essentially two parts: component s (participant’s res ponses in words or phrases) and categories (the groups into which these responses are placed). A cognitive map is a mental structure that enables peopl e to store and access information (Kaplan, 1982). Details on how a participant generates a cognitive map are explained in detail in the 3CM process and data collecti on section in this chapter. 3CM is a useful technique in collecting unbiased information because it minimizes the amount of guidance the participant receives from the researcher. The technique does not judge the participant; it do es not distinguish between tr uth or fallacy and facts or belief. The method helps understand the mental information an individual has related to a particular topic, as opposed to what the re searcher wishes to find. The technique also facilitates discovery. Generally individuals are not fully aware of their cognitive structures and express satisfacti on at the end of the process. Th is technique aids people in exploring their own knowledge structures in the proce ss of externalizing them. Knowledge is organized into cognitive maps in a coherent way and provides indications of how people might use this information (Kearney and Kaplan, 1997). In this study, an open-ended 3CM process was used to explore people’s perceptions of their neighborhoods and the social interactions within th ose neighborhoods. It helped articulate what items people felt were impor tant in describing their neighborhoods. As 3CM has been useful in measuring abstract i ssues, the technique was suitable to collect

PAGE 75

60 data from a purposeful sample of individua ls to understand how they perceive their neighborhood, people, and the related social interactions. Data obtained from the 3CM were used to understand people’s percep tion of their neighborhood and generate categories and items for the social capital scale to be used in the mail survey. Data Sampling Officials from DOF, TNC, local fire departments, and local homeowners associations identified leaders in each nei ghborhood who might be able to participate in the 3CM interview. Table 3.2 gives detailed information on the number of participants interviewed in each community and neighborhood. A total of 15 interviews were conducted in three communities. Table 3.2: 3CM participants in each neighborhood Community name Neighborhood Number of interviews conducted Lake Wales Ridge North Indian Lake Estates 2 Tiger Creek Forest 1 Lake Wales Ridge South Leisure Lakes 2 Placid Lakes 2 Palm Coast Cypress Knoll 3 Seminole Woods 1 Wedgefield Wedgefield 4 TOTAL 15 3CM Process and Data Collection The 3CM interview began with questions th at are easy to answer and that introduce the topic. Participants were asked questions about the number of year s they lived in that neighborhood, number of family members who lived in the neighborhood, the city from which they moved, the number of associati ons and organizations they belonged to, and their level of involvement with these groups. This information was helpful in interpreting the 3CM data.

PAGE 76

61 The essence of a 3CM interv iew is generating and categor izing a list of things. The question that drives the task asks about item s or things. Because the goal is for people to describe how they store information, it is ofte n easier for people to respond to a question that asks them to explain something to a peer or a neighbor. The act of explaining to another person reveals the important features or elements of their mental models. In this study two scenarios were use d. The first scenario was designed to understand people’s perception of their neighborhood. The second scenario helped understand social aspects of the neighbor hood. Probing questions were asked following each scenario to better understand participants’ perceptions of their neighborhoods and the quality of relationships they hold with people in the neighborhoods. Scenario 1: You are a resident of this commun ity. Imagine that a person is trying to make a decision about moving into your neighborhood and is interested in learning more about your neighborhood, particularly about how people get along. What would you tell them? Follow-up probing questions: What things or factors would you mention in describing whether there is a feeling of “closeness with nearby neighbors” in your neighborhood? What things would you talk about in telling/describing how people relate to one another? What things would you tell that pe ople in the neighborhood do together? What things would you tell that people in the neighborhood do for one another? Does your neighborhood have community-wid e events? What role do you play in those events? In November-December 2003, a 3CM interview guide was developed and pilot tested with a sample of five people who lived in wildland-urban in terface areas. The 3CM interview guide was revised and modified based on data collected from the pilot test and comments from experts in 3CM and environmen tal education. Specific changes made as

PAGE 77

62 a result of the pilot test include adding a second scenario that would help understand aspects that shaped the neighborhood and ma de things work in the neighborhood and asking additional questions: Scenario 2: This potential new resident is partic ularly interested in learning about social aspects of your neighborhood. If you were to tell about things that make this community work, what would that be? Follow-up probing questions: What characterizes the relationships pe ople in the neighborhood have with one another? How well do people in this neighborhood get along? What makes this possible? Additional questions: List five things that you lik e the most about your neighborhood. List five things that you like about people and their relationships in your neighborhood. List five things that you would like to change in your neighborhood. Why? The selected individuals were contacted by phone and asked to participate in the 3CM process. Some people were forthcomi ng and were happy to set up an interview while others were concerned that they mi ght not know enough. Some were hesitant to talk about their percep tions of their neighborhoods. Howeve r, after explaining to them the importance of the study in terms of wildfire preparedness, mo st were willing to proceed. Twenty people in three communities d eclined the interview, 15 agreed. The participants were introduced to the in terview process with an explanation about the interview questions and the 3CM task. Each step was explained as often as needed to make them comfortable. Th ese steps are outlined here:

PAGE 78

63 Generating components After the scenario was read to the partic ipant, their comments (components) were recorded on sticky notes and placed in front of the participant. When writing comments care was taken so that no phrases or se ntences were noted, just nouns. Once all the prompts were completed for the two scenarios, participants were asked if they wanted to add something to the list or if there was something important about the issue that they thought the interviewer should know. Organizing categories Participants were asked to arrange all of the items (components) on sticky notes into categories. They were told that they ar e free to generate as many categories as they wanted and that they could place the items in any category they chose. Participants categorized these items giving context to th e words and explaining why the words were placed as they were. The participants genera ted the concepts and structured them, which reveals how information was organized in their minds. People were generally comfortable when asked to respond to a prompt. But once asked to organize their thoughts, some people felt that the method was trying to judge them. Participants were told that there were no right and wrong answers and that they were not being tested. Sometimes it worked to tell them that they are doing really well. A few people benefited from a little help on organizing the compone nts, and quickly took over organizing the items into categories. Labeling categories Participants were then asked to label each category. They were also asked to explain reasons for grouping items together.

PAGE 79

64 In the final part of the interv iew participants were asked to list five things each that they liked about their neighborhoods and peopl e in their neighborhoods. They were also asked to list five things that they woul d like to change in the neighborhoods. Most participants repeated items that they ha d mentioned during the 3CM task. Very few people came up with something different a bout their neighborhoods. On an average, interviews lasted from 30 to 60 minutes. Table 3.3 is an example of the results of a 3CM map for one participant. Table 3.3: Sample of components and categories of a participant’s map Real America Best p eo p le here Neighborhood association Get alon g q uite Excellent community This is what America should All ethnic groups People get along Same economic level Independent people Mind about your business Wish list for an American dream Living your own life People look for a decent community Advice Need something physically they are there Pride in the community Highest land value in Palm Coast Realize who should not be here Who lives in what house Social functions Social life emanates through golf Social functions Church Calmness Not fast pace life Lends a hand Get with each other as nothing more to prove Miscellaneous attributes Watch out for one another Do not worry about anything Picnic Christmas Golf course Talk to one anothe r Screen them No challenge Property values escalated

PAGE 80

65 Quantitative Research: Mail Surveys In this study mail surveys were used to collect data to explore the relationship between wildfire education programs and social capital. A survey is a systematic tool for collecting information from a sample of the people and using these data to make generalizations about the la rger population (Penfield, 2002; Jacobson, 1999). They are used to collect data on people’s opinions, attitudes, and knowledge on societal and environmental issues and gather other demogr aphic information about the participants. The use of mail surveys can be justified as a data collection tool for this study as it helps cover a wider geographic area and enab les a larger sample size. The method also helps overcome interviewer bias. As compared to phone and personal surveys, mail surveys are relatively inexpensive to ad minister (Jacobson, 1999). In addition, mail surveys can be used to more easily reach larger p opulations. However, with mail surveys the researcher does not have control ove r misunderstood questions, cannot ask for clarification or ask follow-up questions, and usually has to work with missing data. Errors using surveys may include biases, content and construct validity, and reliability of the data. Low response rate can be a limitati on of survey research (Dillman, 2000). Each of these is addressed in the following sections. Data Sampling The names and addresses of property owne rs in each community were obtained from the respective county property apprai ser offices. Those whose mailing address did not match the property address were excluded from the population. The final mailing list contained 3,744 individuals in seven neighborhoods.

PAGE 81

66 Survey Development Process To develop the mail survey a relationship diagram was created th at illustrated the relationship between wildfire e ducation programs and perception of social capital. It also indicated confounding variables that might affect that re lationship (For relationship diagram, see Appendix B, p 170). The survey contained six sections with a total of 39 questions. The six sections in clude: housing, individual’s perception of wildfire risk, participation in wildfire education programs, steps taken to reduce risk of wildfires, involvement and perception of community, and personal information. Of the five questions on housing, three were categorical and two were open-ended. The section on perception of wildfire risk c ontained eight questions. Out of these, three were categorical questions and five were ordi nal questions about people’s pe rception of wildfire risk. Questions on participation in wildfire edu cation programs were divided into three sets: participation in electronic, print, and events related to wildfire preparedness. For each question, a list of activities on ways to obtain information about risk reduction and creating defensible space was provided. Pe ople were asked if they had received information from that source (dichotomous ques tion). If they did receive the information or attended an event, they were asked if they learned from the source (dichotomous question), the extent to which they interacted with the information (1 to 4 Likert scale), and the extent to which they used the informa tion to reduce wildfire risk (1 to 4 Likert scale). This section had a total of 13 questions. The section on wildfire risk reduction activities contained on e dichotomous and two continuous scale variable questions. The Likert scale questions focused on the extent to which people pa rticipated in various risk reduction activities and their pe rceptions of the extent to which people (e.g., leaders, educators, neighbors) and information influe nced their risk reduction activities. Three

PAGE 82

67 questions on people’s perception of their community contained both dichotomous and continuous variables. The section on persona l information contained seven categorical questions. The survey was pilot tested with nine people, which included wildland-urban interface residents in Florida and experts a nd students in the field of environmental education, wildfire preparedness, and social capital. Social capital scale development An important aspect of the survey was deve lopment of the social capital scale. The social capital scale was used to measur e people’s perception of their neighborhood’s social capital. For purposes of this study soci al capital is defined as an individual’s perception of the quality of relationship he/she holds with his/her neighbors and other people in the community and community charac teristics like community participation. The literature provided initial ideas in the development of a scale ( Scaff, 1952; Hagedorn and Labovitz, 1968; Kasarda and Janowitz, 1974; Coleman, 1988; Foskett, 1995; Brown and Ashman, 1996; Grooatert, 1998; Buck land and Rahman, 1999; Bullen and Onyx, 2000; Putnam, 2000; Purdue, 2001; Flora and Flora, 2004). Based on the review, five measures of social capital appeared most dominant: reciprocity, trustworthiness, networks, participation in co mmunity activities, and sense of community. Results of the 3CM interviews helped understand how resident s perceived these factors. Data obtained from 3CM interviews for each group were used to develop items within each category. By using the terms and phrases that arose from the 3CM process, it was anticipated that residents in the community would be able to re late to the items on the social capital scale. A 33-item social capital Likert scale, from strongly agre e to strongly disagree, was developed and pilot tested with the mail survey. The scale contained questions on

PAGE 83

68 people’s perception of their community and interaction in their community. The scale included the five categories previously men tioned, with specific items from both the literature and 3CM interviews. The literature suggests that respondents provide more accurate opinions when the “don’t know” response is placed at the far right and when the neutral option is eliminated (Tourangeau et al., 2004). Accordingly in this study when developing and administering the scale, the option “don’t know” (neutral) was placed at the end of the continuum. Thus, a four-point strongly disagree to str ongly agree Likert scale was developed with “don’t know” placed at the end of the con tinuum and given a value 5. However for coding and analysis purposes, the “don’t know” response was treated as neutral and assigned a value 3 (in the middle of the scale) . Thus, social capital scale was recoded on scale of 1 to 5 as 1 being st rongly disagree, 2 being disagr ee, 3 being don’t know (treated as neutral), 4 being agree, and 5 being st rongly agree. Negatively worded items were reverse coded. Content validity of the social capital scal e for measuring individual’s perception of social capital was established by comments from social science experts in the fields of scale development and measurement and social capital. Construct validity of the social capital scale was conducted using factor analysis with SPSS 10.0 for Windows. Exploratory factor analysis data indicated that the items load on two factors—neighborhood engagement (networks, participation in neighborhood activities) and affective attr ibutes (trust, reciprocity, a nd sense of neighborhood). These factors are similar to the two factors developed by Stone (200 1). Items that had a factor rating lower than .4 were removed from the scal e. The final social cap ital scale contained

PAGE 84

69 21 items. There is a considerable overlap a nd high inter-correlation among subconstructs within the scale. Due to the di fficulty in obtaining more than two strong distinct factors, it is recommended that this social capital scal e be used as a one-dimensional measure of social capital. Caution should be taken when using the subconstruct scales for making comparisons among groups. Tables 3.4 to 3.7 provide the communalitie s, total variance explained, component matrix, and the rotated component matrix for the 21 items social capital scale. A social capital summated scale score was used for an alysis. Reliability of the 21 items social capital scale (n=1243) as measured by Cr onbach’s alpha is 0.8711 (see Appendix F, p 235). Table 3.4: Factor analysis: Communalities Survey item no. Statement Initial Extraction F I trust my neighborhood associa tion to make decisions on my behalf. 1.00 .32 L Most people in my community do voluntary work for the community. 1.00 .45 O Most people in the community are involved in activities that benefit the community. 1.00 .53 P People in this community are easy to contact. 1.00 .43 T This community offers enough chance for a person to do volunteer work. 1.00 .35 U People in this community work together to solve problems. 1.00 .58 X I volunteer in my community. 1.00 .28 BB Some of my neighbors attend several community functions. 1.00 .41 DD People in the community share common interests. 1.00 .34 C People in the community are willing to help each other whenever they can. 1.00 .45 E I can call on my neighbor for help anytime. 1.00 .37 G People in this community ha ve mutual respect for one another. 1.00 .59 H For most part, people are w illing to make the community a better place to live. 1.00 .57 J I greet my neighbors. 1.00 .47 K This community is a safe place for children. 1.00 .47

PAGE 85

70 Table 3.4 continued Survey item no. Statement Initial Extraction M I think people in this community cannot be trusted. 1.00 .19 Q For the most part, people in this community are friendly. 1.00 .58 S Usually people in the community greet one another. 1.00 .51 W People in the community get along with each other. 1.00 .47 Y For the most part, people in the community obey community codes and covenants. 1.00 .33 CC Most people in my community can be trusted. 1.00 .53 Extraction Method: Principal Component Analysis. Table 3.5: Total variance explained Initial Eigenvalues Extraction Sums of Squared Loadings Rotation Sums of Squared Loadings Component Total % of Variance Cumulative % Total % of Variance Cumulative % Total % of Variance Cumulative % 1 7.24 34.47 34.47 7.24 34.47 34.47 5.21 24.82 24.82 2 1.97 9.36 43.83 1.97 9.36 43.83 3.99 19.02 43.83 3 1.07 5.07 48.90 4 .99 4.70 53.60 5 .96 4.57 58.17 6 .87 4.14 62.30 7 .81 3.87 66.17 8 .74 3.52 69.69 9 .67 3.16 72.85 10 .65 3.10 75.95 11 .61 2.91 78.87 12 .60 2.84 81.70 13 .55 2.60 84.30 14 .51 2.42 86.72 15 .47 2.25 88.97 16 .45 2.16 91.13 17 .42 2.00 93.13 18 .41 1.94 95.08 19 .38 1.82 96.90 20 .35 1.66 98.55 21 .30 1.45 100.00 Extraction Method: Principal Component Analysis.

PAGE 86

71 Table 3.6: Component matrix a Component Survey item no. Statement 1 2 F I trust my neighborhood association to make decisions on my behalf. .53 .20 L Most people in my community do voluntary work for the community. .46 .49 O Most people in the community are involved in activities that benefit the community. .44 .58 P People in this community are easy to contact. .63 .20 T This community offers enough chance for a person to do volunteer work. .48 .35 U People in this community work together to solve problems. .69 .33 X I volunteer in my community. .38 .37 BB Some of my neighbors at tend several community functions. .51 .38 DD People in the community share common interests. .55 .21 C People in the community are willing to help each othe r .67 whenever they can. -.08 E I can call on my neighbor for help anytime. .57 -.23 G People in this community have mutual respect for one another. .76 -.11 H For most part, people are willing to make the community a better place to live. .74 -.15 J I greet my neighbors. .48 -.48 K This community is a safe place for children. .61 -.31 M I think people in this community cannot be trusted. .36 -.24 Q For the most part, people in th is community are friendly. .70 -.30 S Usually people in the community greet one another. .65 -.30 W People in the community get along with each other. .65 -.22 Y For the most part, people in the community obey community codes and covenants. .57 .028 CC Most people in my community can be trusted. .70 -.18 Extraction Method: Principal Component Analysis. a 2 components extracted.

PAGE 87

72 Table 3.7: Rotated component matrix a Component Survey item no. Statement 1 2 F I trust my neighborhood associ ation to make decisions on my behalf. .29 .49 L Most people in my community do voluntary work for the community. .054 .67 O Most people in the community are involved in activities that benefit the community. -.014 .73 P People in this community are easy to contact. .37 .54 T This community offers enough chance for a person to do volunteer work. .16 .57 U People in this community work together to solve problems. .33 .69 X I volunteer in my community. .06 .53 BB Some of my neighbors attend several community functions. .16 .62 DD People in the community share common interests. .30 .51 C People in the community are willing to help each other whenever they can. .58 .35 E I can call on my neighbor for help anytime. .59 .17 G People in this community ha ve mutual respect for one another. .67 .38 H For most part, people are w illing to make the community a better place to live. .67 .34 J I greet my neighbors. .68 -.08 K This community is a safe place for children. .67 .13 M I think people in this community cannot be trusted. .43 .03 Q For the most part, people in this community are friendly. .74 .20 S Usually people in the community greet one another. .70 .16 W People in the community get along with each other. .65 .23 Y For the most part, people in the community obey community codes and covenants. .43 .38 CC Most people in my community can be trusted. .66 .29 Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. a Rotation converged in 3 iterations. Survey administration The survey was revised approximately 15 times before administering it. The Dillman method (Dillman, 2000) for administer ing mail surveys was used to improve response rate. Because Dillman suggests that adding an incentive may improve response

PAGE 88

73 rate, specially designed FireWise magnets were attached to each cover letter. The survey packet also contained a copy of the mail su rvey, and business reply envelope. The first survey mailing was followed by a reminder postcard two weeks later. After deleting the names of people who responded to the first mailing, a second mailing (without the magnet) was done two weeks after the fi rst reminder postcard. A second reminder postcard followed two weeks later. The campus mail service may not have gotten everything out exactly accordi ng to the plan and the interv al between first and second mailings and postcards varied from 2 to 3 weeks. Mail surveys were sent to 3,744 homeow ners in the seven neighborhoods in Florida. Usable data were obtained from 1,350 participants in three communities in Florida. Seventeen surveys were receive d without the community name and were excluded from the analysis. Twelve surveys we re returned blank. No envelopes with bad addresses were returned. The overall res ponse rate for all the communities was 36.83% (ranging from 43.16% in Tiger Creek Fore st to 26.16% in Seminole Woods). Table 3.8: Mail survey response rate Community name County name Neighborhood name Surveys sent Surveys received Response rate (%) Lake Wales Ridge Highlands Tiger Creek Forest 95 41 43.16 Indian Lake Estates 367 149 40.6 Polk Leisure Lakes 425 159 37.41 Placid Lakes 869 353 40.62 Palm Coast Flagler Seminole Woods 497 130 26.16 Cypress Knoll 746 317 42.49 Wedgefield Orange Wedgefield 745 201 26.97 Blank surveys 12 Surveys without community name 17 Total 3744 1379 36.83

PAGE 89

74 Table 3.8 provides detailed information on the response rate for each study neighborhood in Florida. All data were c oded and entered in Microsoft Excel. Two graduate students assisted with data entr y. Data were randomly checked for errors in coding. Also, frequency analysis tables were used to check for potential errors during data entry. Analysis was c onducted using SPSS 10.0 for Windows. Nonresponse Analysis A nonresponse analysis was conducted for five neighborhoods to check if there were any response biases among respondents. For this analysis, property values were obtained for five neighborhoods from their resp ective county property appraisers offices. Property value information was not availa ble for Polk County, which excluded the neighborhoods of Leisure Lakes and Pl acid Lakes from this analysis. For each neighborhood an independent sample t-test was conducted to determine a significant difference between property va lues among people who did and did not respond to the mail surveys. T-test results showed no significant difference at 95% confidence interval: Indian Lakes Estates (p< .05, n=359) Tiger Creek Forest (p< .05, n=94), Wedgefield (p< .05, n=745), Cypre ss Knoll (p< .05, n= 712), and Seminole Woods (p< .05, n=495). These results and the pr oximity of the four Lake Wales Ridge neighborhoods suggest there woul d not be a response bias on basis of property value found in Placid Lakes or Leisure Lakes. (F or non-response analysis, see Appendix F, p 231).

PAGE 90

75 CHAPTER 4 RESULTS AND DISCUSSION This chapter reports the results of the th ree data collection tools described in the previous chapter: in-depth interviews, 3CM (Conceptual Content Cognitive Mapping) interviews, and mail surveys. In-depth interviews with program lead ers and developers helped describe the various wildfire educational initiatives in each neighborhood. The analysis presents excerpts from these interviews that help e xplain the diversity of educational programs, level of involvement with the educationa l program in the neighborhoods, and outcomes of these educational programs. 3CM interviews helped articulate pe ople’s perceptions of their neighborhoods. Data were analyzed qualitatively and excerpts are included. Data from mail surveys helped explain relationships between participation in wildfire education programs, risk reduction, and people’s perceptions of social capital. Data were analyzed quantitatively using SPSS: t-tests, ANOVA (Analysis of Variance), multiple regression, correlation, and chi-square. Qualitative Data Analysis: In-depth Interviews Program developers and implementers were interviewed in each neighborhood. These interviews were used to obtain in-depth information on each wildfire education program—its goals, objectives, target audien ce, anticipated outcomes, and unintended consequences; data were used to generate program descriptions (Appendix A, p 157).

PAGE 91

76 Interviews revealed that programs we re designed to provi de a variety of opportunities for people to particip ate in the wildfire educatio nal programs. At one end of the continuum were education programs that involved little or no interaction with the agency or people delivering information. These included information dispensed through electronic media, such as websites, televisi on news, radio news, compact disc, or a video. People viewed the information at their conven ience with little or no interaction with others. At the other end of the continuum were wildfire educational programs that involved a significant interaction among agency staff and residents. These include events such as meetings, presentations, prescrib ed burn or equipment demonstrations, or community picnics. Somewhere in the middl e of the continuum fell those educational programs that involved some level of inte raction depending on how the information was delivered (mail or in person). These include d information through print media, such as handouts, brochures, door hangers, newspapers, etc. This idea of educational programs along a continuum depending on medium and delivery helped conceptualize three categories—electronic, event, and print. Th ese categories were used in the survey development and have been used through out the study. Specific comments from program leaders helped explain some of the innovative educational initiatives that worked in neighborhoods as well as the reasons these initiatives worked. In Wedgefield, after initia l low participation from the community, the Wedgefield FireWise committee became quite astute at identifying opportunities to contact their residents. The President of th e Wedgefield FireWise committee commented: “We [FireWise committee] try to put, we try to have a FireWise article in every month's newsletter, if it's onl y to brag that we got anot her award. We try to put a little tidbit in there... We tr y to have an article every month on some kind of ideas that people can think about... Golf tournament is a big moneymaker for us. Usually

PAGE 92

77 we used to have like a spring party or fall party where we could meet, meet neighbors type thing. And we try to alwa ys be involved wherever we can. Whenever there is going to be something, we try to be there participate in homeowners association” In Placid Lakes, a neighborhood in La ke Wales Ridge, participation in the FireWise initiative was a result of a wildfi re in a heavily wooded area near the house of the neighborhood homeowner association’s boa rd member. The board member met with some members of the neighborhood and the Di vision of Forestry (DOF) personnel to look at ways to minimize risk. The chair of the FireWise committee in Placid Lakes commented: “The DOF contacted us about a [FireWise] seminar in Brooksville, Florida. Five people from Placid Lakes went to the Fire Wise program. [After that] we announced about the program at the General Membership meeting. October 2 nd [2004] is the FireWise day for the community. DOF pe rsonnel will address the community for two hours. It will be a festive atmosphere and we will distribute literature.” The president of homeowners associ ation of Placid Lakes commented: “There are lots of vacant lots in Placid Lakes. When there is a dry spell these vacant lots is the biggest problem. Als o, many teenagers come to party on the weekend and that may increase the threat of fires [due to cigarette butts]. In October, we are going to host a FireWise meeting and we will invite all the five County Commissioners to the meeting.” Program leaders and developers in each community credit the success of the educational programs to networks and key people in their community. The extension agent of Orange County Cooperative Extension Service commented on the success of FireWise in Wedgefield: “I think [the program is successful because of] the agencies, I do not think the residents would have made it without the Fo restry Service, the Extension Service, and County Planning, and fire people. They just wouldn't have they would have run out of steamI think it is a model [FireWise] that could be studied, I really do think it could be studied exactly how this went. Some people say it was personalities that made it go. Well, of c ourse, they played a key role, but I won’t agree with that. I think in this case it’s the model and there are elements that could be identified—key leadership, people w ho really believe in what they're doing,

PAGE 93

78 resident involvement, Forestry Service, Extension Service. [These are] key elements that have to be there [for the model to be successful] and I think the model could be replicated.” In addition to education programs, so me neighborhood used mitigation, policies, and ordinances as ways to educate people a bout actions to reduce risk of wildfires and improve community preparedness. The direct or of Flagler County Extension commented on the attitude change with re gard to prescribed burning: “Before, the fires, people forget and so when they do control burns, people said, hey, ‘I want to get out here in the eveni ng on my back porch and there is smoke. I don't want that,’ and they would compla in.’ And DOF saw this throughout the state, and they started backing off becau se of the complaints and the liability, people want to sueSo, Forestry [staff] backed off until the fires got bad in and now they say ‘We are going to burn and this is the way it is.’ They have kind of drawn the line. They don’t let peop le who don't know the reason for control burns, control what is going on. So there is an attitude change I think with specifics about what they do.” Wedgefield invited companies to demonstr ate their mulching equipment to clear underbrush. This served two purposes—getting the community and its leaders together and understanding the usefulness of the equipment in reducing the understory. Wedgefield also conducted prescribed burns to create firebreaks around the community. The forest area supervisor of Division of Forestry in Orange County talked about prescribed burning in Wedgefield: “We have prescribed burns on the boundaries of Wedgefield. I am also responsible to come up with a strategy for mitigati on planning in Wedgefield. We have nice fire breaks in Wedgefield —some roadwa ys, and some waterways. The Ranger Drainage District owns a lot of these br eaks and with their permission we are trying to establish fire breaks on their boundaries . We worked with the Deseret Cattle and Citrus and burned their property on the bor der of Wedgefield with no cost to them. This helps the community, as it is not at the risk of getting burned because of the fire coming from outside that has started somewhere else. For the Deseret Cattle and Citrus it gives them the assurance that they can manage their land with prescribed burning, now that they alr eady have a border burnedWe plan on doing more prescribed burning. The Hawkins Law allows us to chemically, mechanically, and with fire [reduce fuel on the property ] that are considered a fire hazard by notifying the landowner through tax a ssessments and putting out a public

PAGE 94

79 announcement. That announcement needs to clearly mention that date and the location of the burn. If the landowner does not want a burn to take place then they need to notify us by a letter. Unless they send us a letter re jecting it, it is taken as acceptance.” In Palm Coast, an ordinance requiring ow ners of documented hazard lots to clear undergrowth has been an effective strategy to reduce the risk of wildfire. The ordinance went into effect at the prodding of one citi zen leader after a second major wildfire. The wildfire mitigation specialist from Bunnell Di strict (DOF) spoke a bout the ordinance: “In the city of Palm Coast there are numb er of lots that are not built on and the owners of these plots are all over the country. The city passed a bush-clearing ordinance with help from DOF. This ordinance applies to hazar d lots adjacent to houses or structures. DOF set up a fire mitigation board to hear appeals of community leaders.” A landscape architect from Palm Coast commented on the ordinance: “It took two years to get the ordinance passe d. The city council had the idea that the pine trees are bad. Later the DOF told th em that it was not the trees but the vegetation and the palmettos that grew on the ground that posed the biggest hazard. County commission was earlier defensive. [Bunnell District Manager, DOF] persuaded the city council to pass the or dinance. Initially th ere was little support from the people who were involved in pa ssing it, but there wa s a concern about property rights. No one was against it but there was resistance as they thought that the traditional civil rights were threatened.” Summary: Interviews with program developers and implementers were used to generate a list of all the wild fire education activities that people in each neighborhood participated in. This list wa s used to develop questions on the mail survey pertaining to participation in wildfire education programs. The interviews also revealed key components to the success of educational programs in each of the study communities. Interviews with program leaders revealed that communities do a lot of different things to improve preparedness, such as, implement po licies, put on demonstrations, and provide mitigation activities. Education is just one element of their comprehensive community preparedness program.

PAGE 95

80 Qualitative Data Analysis: 3CM The objective of the 3CM (Conceptual Cont ent Cognitive Mapping) activity was to understand people’s perception of their ne ighborhoods and ways in which social interaction occurred in those neighborhoods. Data obtai ned from 3CM interviews were analyzed in keeping with the objective s of the study. A 3CM task generates both components and categories. Components (als o called items) refer to a participant’s response (words or phrases); categories refer to the labels and groups in which participants place these responses. For the purpose of the analysis, the entir e data set was divided into four groups based on the participants’ places of reside nce. The four groups are—Lake Wales Ridge South (LWR South), Lake Wales Ridge North (LWR North), Palm Coast, and Wedgefield. This consolidation allowed sma ll samples from similar neighborhoods in the same county to be grouped together: Leis ure Lakes and Placid Lakes became LWR South; Tiger Creek Forest and Indian Lake Estates became LWR North; Seminole Woods and Cypress Knoll are Palm Coast. Also, wherever the participants have used the word “community” to refer to neighborhood, the wo rd “neighborhood” has been substituted. However, the transcripts’ in Appendix E (p 192) retains or iginal language. Component analysis was conducted to identify the common and unique components within each group. In the thema tic analysis components from each group were arranged into common themes and compared. Category analysis helped compare category themes within and among groups and understand how respondents categorize their perception of social capital. A qualitat ive analysis provided information on people’s perceptions of things they liked and disliked about their neighborhood.

PAGE 96

81 Component Analysis In this study, 15 participan ts generated 63 categories made up of 459 components. All participants indicated that they were year-round residents and the number of years they lived in the neighborhood va ried from 1 to 25 years. Component analysis can be used to identify components both common and unique to each group. The 459 components were reviewed to identify terms and concepts that were similar and different across the four groups. Components were grouped together if they had similar meaning or if different words were used to talk about a single con cept. For example, in Wedgefield although the participants talked about the homeowners’ a ssociation activiti es, they used several terms to describe it: “social,” “ac tivities,” and “it bri ngs people together.” After regrouping, the term association activity was used to represent all th ese aspects of the homeowners association. This data reduction activity result ed in a total of 219 co mponents in the four groups. Participants of at least two groups mentioned 40 components. Eight components were common across all groups. These include golf, association activity, Christmas (the social event), people get along well with everyone, fr iendly, something in common, help each other, and close-knit community. Respondents from two groups LWR North and LWR South together mentioned 10 unique components. The following were unique components generated by participants of LWR North participants: ‘the older population needs to socialize’ , ‘needs each other’, and ‘gets involved in the neighborhood’; the younger popula tion ‘does not find many facilities for activities’ and ‘does not get i nvolved.’ A comparison of LWR North to the other groups on the basis of age does not i ndicate any significant difference. Unique

PAGE 97

82 components generated by participants of LWR South are the ‘help cris is center at Lake Placid’, ‘arts and crafts festival at Lake Pl acid’, and ‘nursing home community project at Lake Placid.’ Although participants of ot her neighborhoods are involved in events outside their neighborhood, participants of onl y this group mentioned events outside their neighborhood. Summary: One of the objectives of the 3CM interviews was to identify common components across groups that could be used to generate items on the survey’s social capital scale. Of the eight components men tioned by participants of all groups, some were developed into items on the social cap ital scale. Because people in all groups mention these components, these phrases are probably understandable and common across the survey population. Identifying unique components in each group helped further understanding of the perception of social cap ital in each group. It is clear from the components that respondents appreciate the social opportuni ties in and outside their neighborhoods and they perceive supportiv e interaction among their neighbors. Thematic Analysis This analysis evaluates the content of people’s cognitive structures by organizing components into common themes and looking at differences and si milarities among and within groups. Of the 219 components generated through the data reduction step, those representing common themes within each group we re placed together. This resulted in 12 common themes across all groups. As can be seen from the following examples, the common themes are much broader concepts than the common components identified in the previous analysis. Social events: This theme includes or ganized celebrations or parties that involve the entire neighborhood.

PAGE 98

83 Characteristics of people : This theme includes people’s relationship with others in their neighborhood, reciprocal behavior, and camaraderie. Associations/clubs : This theme includes recognized or organized clubs or associations in the neighborhood. Neighborhood characteristics: This theme includes com ponents on nature of the neighborhood—retirement, young, peaceful, attr actions, etc., level of interaction, and related characteristics. Negative perceptions : This theme includes complaints and concerns people expressed about their ne ighbors and neighborhood. Nearby neighbors : This theme includes people’s per ceptions of their relationships with other people on their st reet and feelings about th em; and their perceptions about other people in the neighborhood. Recreation : This theme includes components that describe activities happening in small groups. It is different from the th eme “social events” as these recreational activities may not necessarily involve the entire neighborhood. Fire education : This theme includes components about activities to educate people on wildfires and reducing risk of wildfires. Location : This theme includes components that describe the physical location of the neighborhood with regard to geographical surrounding. Older people: This theme includes components that describe the older people in the neighborhood. This theme was unique to LWR North. Younger people: This theme includes components that describe the younger people in the neighborhood. This theme was unique to LWR North. Social events—Lake Placid : This theme includes components on social events outside the neighborhood in which people from more than one neighborhood may actively participate. This th eme was unique to LWR South. Small sample sizes make quantitative co mparison of common themes across groups difficult. This analysis gives an overview of themes that are mentioned by people in each group and also helps compare how all groups perceive these themes. A comparison revealed that all four groups mentioned the five themes, and more than one group mentioned 10 themes. The five common themes include: Social events, Characteristics of

PAGE 99

84 people, Associations/clubs, Neighborhood characteristics, and Negative perceptions . Tables 4.1 to 4.5 list a sampling of items men tioned by participants for the five common themes. Table 4.1: Items relating to theme Social events Wedgefield Lake Wales Ridge North Lake Wales Ridge South Palm Coast New year party July 4 celebrations Men’s breakfast October fest Friday night dinners Picnic Clean the lake Holiday dance Halloween party Christmas party for kids Ladies luncheon Garage sale Christmas competition Memorial day program Fireman's barbecue in Placid Lakes Block parties Fall/ Spring fling Christmas gathering at town hall Social functions Super bowl party October Garage sale Picnic Functions for children Yard sales to sponsor needy families Christmas party Table 4.2: Items relating to th eme Characteristics of people Wedgefield Lake Wales Ridge North Lake Wales Ridge South Palm Coast Get involved Get along well Everyone waves Most get along, as nothing to prove Family oriented people Friendly Take elderly or disabled to the hospital People want to be involved Get along with everyone Different (two) groups in the community Very friendly and welcome Friendly Friendly/nice, lot of people have many networks People live and let live People get along well Help one another Talk to each other Help—pick up relatives from airport /rescue Help each other Share thoughts about community Relate well regardless of status and income Willing to apply knowledge and skills Pull together Everyone waves Something in common

PAGE 100

85 Table 4.3: Items relating to theme Associations/clubs Wedgefield Lake Wales Ridge North Lake Wales Ridge South Palm Coast Garden club Internal neighborhood watch program Homeowners association Cypress Knoll neighborhood watch Homeowners association activity Computer club Aluminum recycling for homeowners association Homeowners association Parent Teachers association Homeowners association Welcoming committee Yacht club Crime watch program Garden club Table 4.4: Items relating to them e Neighborhood characteristics Wedgefield Lake Wales Ridge North Lake Wales Ridge South Palm Coast Similar economic class Deed restricted Retirement community Mix of people Different ethnic group Small community Quiet and serene Most retired 1-5 acre lots, Outpouring of love from all groups Deed restricted community Attractive Plenty of room Nice community No t many parties Not cold community Park Community is spread out Close knitted neighborhood Same economic level Homes close together Close knit community Many from military Young families Relaxed High middle class/educated Many ladies are volunteers Single-women families Safe/codes Groups center around interest Seasonal residents Library run by volunteers Referred by friends and relatives Condition of roads/pave main road Picnic area, boat ramp That is what America should be Wide range of ages Park Relief from rat race

PAGE 101

86 Table 4.5: Items relating to theme Negative perceptions Wedgefield Lake Wales Ridge North Lake Wales Ridge South Palm Coast Homeowners association not social thing Renewed life ministries broke the community Problems with ATVers Not all people likeable Do not say hi No feeling of neighborhood Visitors in the area Split Not cohesive Increase tax Nice feelings before fire People negative about stuff Increased maintenance fee Do not care enough Not many social organizations L o s t i n t e r e s t For all the five themes, some items ar e common across all four groups and some are not. The items that are common across all gr oups are the same eight items listed in the component analysis. The items that are uni que within a particul ar theme help us understand how people in these groups perc eive their neighborhoods differently. For example, a review of items under theme Social events suggests avenues for interaction at the neighborhood level for all four groups. Among the less common themes, Location of the neighborhood and Nearby neighbors seems important to the participants in LWR South, Wedgefield, and Palm Coast. However, they did not emerge as a component theme from participants of LWR North. Participants of LWR North a nd Palm Coast mentioned the theme Fire education . Although not mentioned during the 3CM intervie w process, residents in both Wedgefield and LWR South participated in fire education activities, as learned from the in-depth interviews with program leaders and developers.

PAGE 102

87 Category Analysis Category analysis helps identify and co mpare categories within and among groups. The importance of this analysis is that it helps understand what broad categories are important and how people structure th eir perception of social capital. As part of the 3CM process, 15 particip ants generated a total of 63 categories. Participants organized their components into tw o to six categories. On an average people organized their information in 4.2 categories. This matches the assumption of construct validity of 3CM that people tend to organize th eir information into five plus or minus two categories (Kearney and Kaplan, 1997). The assumption is that due to the brain’s limited capacity to hold elements in a working memory, individuals would express their knowledge in five plus or minus two categ ories (Kearney and Ka plan, 1997). However, this rule does not hold true for the total number of components an individual can create within each category. For the purpose of analysis, all the cate gories mentioned by the participants were listed and category labels that were similar to each other in terms of content, words, or phrases were grouped together. This process reduced the number of different categories from 63 to 23. The next step was to iden tify broad category themes across groups to eliminate the repetitive individual category la bels. For this, the 23 category labels were reviewed, and the ones that described a common theme were placed together. For example, category labels such as ‘advantage s of living in the neighborhood’, ‘things you can tell someone about the neighborhood’, ‘sense of neighborhood’, ‘social aspects of the neighborhood’, ‘security of th e neighborhood’, and ‘Real America’ were grouped into Neighborhood attributes . This resulted in eight category themes: Neighborhood attributes, Neighborhood infrastructure, People in the neighborhood, Social events,

PAGE 103

88 Challenges, Immediate nei ghborhood, Personal contribution in neighborhood activities, and Children. Category themes listed by all groups were Neighborhood attributes and Neighborhood infrastructure . These categories form the basi s of people’s perceptions of and social interactions in th eir neighborhoods. Participants in three out of the four groups mentioned Challenges in their neighborhoods. These include deficiencies about their neighborhoods and negative feelings a bout people in their neighborhoods. This information about challenges people face in th eir neighborhood was useful to develop an item on the social capital scale and helps as sess people’s perceptions of social capital (negative or positive) in their neighborhoods. The category theme Social events were important to participants in LWR North and Palm Coast but not in LWR South and Wedgefield. The category theme Immediate neighborhood was an important aspect in LWR South and Wedgefield. The category theme Children were important to respondents in only one group, Wedgefield. The category theme Personal contribution in neighborhood activities was important to respondents in two groups: LWR South and Pa lm Coast. This information would be useful when developing and implementing ed ucational programs in these neighborhoods. Summary: One of the objectives of the 3C M interviews was to understand respondents’ perceptions of their neighborhoods and interact ions in their neighborhoods. Category analysis revealed that people’s perceptions could be broadly grouped into two aspects: Neighborhood attitudes and Nei ghborhood features. Neighborhood attitudes include categories related to how people f eel about their neighbor hoods and the people who live there, for example, Neighborhood attributes , People in the neighborhood , and

PAGE 104

89 Immediate neighborhood . Neighborhood features include categories that allow for resident interaction and may influen ce neighborhood attitudes—for example, Social events , Challenges, and Neighborhood infrastructure . From the analysis, it could be suggested that for participants in Wedge field the neighborhood at titudes aspect was stronger than their neighborhood features aspe ct. However, participants of LWR North, LWR South, and Palm Coast placed equal emphasis on both the neighborhood attitudes and neighborhood features aspects. A second objective of the 3CM interviews was to understand how people structure information with regard to perception of social capital. This objective included determining if there was a difference in people’s cognitive structures, in neighborhoods with high and low social capital. Are the c ognitive maps of people in neighborhoods with higher social capital denser compared to t hose in neighborhoods with low social capital or are they the same? Comparing the number of components within each category for each respondent revealed that respondents had similar density of components within each category, which could be because only leaders were interviewed in each neighborhood. These were the people who were involved and engaged in the neighborhood, and could thus provide richness in their cognitive maps. 3CM interviews were also conducted to generate items to measure people’s perception of social capital using a mail surv ey. Based on this part of the analysis, the social capital scale was deve loped to contain items about both the neighborhood attitudes and neighborhood features aspect.

PAGE 105

90 People’s Perception of Neighborhood Attributes During the interviews, participants were as ked to list five things that they liked about their neighborhoods and the people in th eir neighborhoods. They were also asked to list five things that they would lik e to see change in their neighborhoods. Participants of LWR North mentioned that they liked the isolation and the solitude provided by their neighborhood, the location of their neighborhood, and people in their neighborhood, especially the fact th at they could call on neighbors for help anytime. The thing that they liked most about people in their neighborhood was that people were concerned about one another and would watch out for one another. “ I like that they [people] watch out for each other. They gather together and work together on different projects— they have a volunteer spir it. I appreciate the older people—the wisdom from them and their stories.” [Participant of Indian Lake Estates, LWR North] “Most people here are family oriented. Pe ople are willing to give so that everyone can be comfortable. People are concerned about other people. Th ey do not try to get into other people’s busine ss-basically live and let liv e. However, if you had a problem there would be someone to help.” [Participant of Tiger Creek Forest, LWR North] When asked about things that they woul d like to see change in their neighborhood, participants of Tiger Creek Forest mentione d they would like to see improved roads, as that would increase the property values. Partic ipants of Indian Lake Estates mentioned that they would like to see more activiti es and venues that involve children. In LWR South, participants mentioned th at they liked the friendliness of the people, weather, location, the neatness , the waterfront, and beautiful homes. “[I think] it is one of the best areas ar ound Lake Placid. People [here] are great [as compared to] people in other areas. There is willingness to help one another. There is friendliness—everyone will wave. Also, the weather is good and the values of houses have increased.” [Participan t of Leisure Lakes, LWR South]

PAGE 106

91 They mentioned that most people in thei r neighborhood were retirees. The things they liked most about people in their neighborhood were that th ey were active, interested in things, and willing to help whenever they can. “I like the people. Everyone [here] is different. You should take the time to ask— people have interesting things to contribu te. They [people] show interest when you are kind; it is a sign for them. Since I ha ve been active in the board [homeowners association] I think people seem to appreciate me more. They think I am fighting for them.” [Participant of Placid Lakes, LWR South] When asked about things that they woul d like to see change in their neighborhood, participants of LWR South me ntioned that kids on four wheelers speeding through the neighborhood and kids with radios were major nuisances to the neighborhood. They wanted to see activities organized to keep kids out of trouble. Participants of Palm Coast mentioned th at they liked their neighborhood because of its location—its proximity and accessibility to the beach, Orlando, and Jacksonville. They also liked their neighborhood for its attractiveness, friendline ss, and people. They liked the attitude of people in th eir neighborhood—the fact that people had mutual respect for one another; they were willing to help and work together to solve problems in the neighborhood. “[We have] large number of artistic people—people who are involved in acting, writing, music, classical arts, etc. People al ways socialize; they go out together and have each other at home for dinner. People [here] communicate. They are not cold to each other, for example, when you ride bikes, walk, etc. People in this neighborhood work together; if there is a concern they will not fight but work together.” [Participant of Cypress Knoll, Palm Coast] “People are friendly. [People he re have] mutual respect for one another. They share common interest and want to see one anothe r.” [Participant of Cypress Knoll, Palm Coast]

PAGE 107

92 Participants in Palm Coast wanted two th ings changed in their neighborhood: more sidewalks and improved safety through bette r lighting, and reducti on in the number of people moving into their neighborhood. In Wedgefield, participants mentioned th at they liked their neighborhood because of its low-crime rate, safet y, security, and people. They also liked it for its location— proximity to the beach and the green space around their neighborhood. “I like the neighborhood for its geog raphic location—how the neighborhood is away from Orlando, away from the hustle and bustle of Orlando. I like it for the people—they are pretty good and tight knit. If you need something you know whom to call. I also like how things ar e structured in this neighborhood; rules and regulations are not overbearing. I like how the neighborhood goes about solving problems. I like that the neighborhood is situated with green space around it.” [Participant of Wedgefield] Another participant commented on the things that he likes about people and their relationships: “People in this neighborhood look out for one another. People look out for children. They are willing to help whenever they ca n. I like the camaraderie in general. I like the relaxed atmosphere here.” [Participant of Wedgefield] Comments about what people would like to see change included the need for more stop signs and stricter and newl y established codes and covena nts. They also wanted to see more people get involved with the homeowners association. Participants of Wedgefield also wanted to find ways to deal with nearby growth and building issues. Summary: People’s actions are often base d on their knowledge, values, and beliefs, which are contained in their mental models. Understanding mental models can be useful when providing information to cha nge people’s behavior, as in wildfire preparedness. Key themes that were common across the components mentioned by all four groups are Social events, Characteristics of people, Associations/clubs, Neighborhood

PAGE 108

93 characteristics, and Negative perceptions . However, the study also helped identify component themes that were unique for each neighborhood. For example, participants in LWR North mentioned two groups: Younger and Older people. Although most communities had these groups, the distinction was probably more problematic for these respondents. However, in this group Children was not mentioned as a category. LWR North participants mentioned that most activ ities were organized and scheduled around a time that was appropriate for the elderl y. Thus the people in the younger group were rarely involved in these ac tivities. In LWR South when asked about neighborhood-wide events and activities that bring people together, particip ants mentioned that some activities that were held outside their neighborhood but within the county. The participants also had a strong sense of Location , which was mentioned by participants in Wedgefield and Palm Coast. Participants in Palm Coast emphasized the activities happening in their neighborhood. They had good neighborhood interaction and neighborhood committees were designed to benefit the residents of the neighborhood. 3CM technique has high construct validit y because the technique performs in accordance with three theoretica l expectations. One, by generating their own components participants were able to re port meaningful concepts. Second, participants organized their concepts in five plus or minus two categor ies. On an average people organized their information in 4.2 categories. Th ird, 3CM appeared to facilitate discovery. At the end of the 3CM interview process, most participants said that the process helped them clarify their own understanding of th eir perceptions of their ne ighborhood and relationships in their neighborhood. The unstructured version of the 3CM helped identify concepts and themes that were similar across groups as well as different among groups. 3CM can also

PAGE 109

94 be useful in generating categor ies and themes that can be used to develop items on scales for surveys, or better understand an issu e. By using the components and categories participants mentioned in the development of a survey, other residents should be better able to understand the questions. The 3CM anal ysis not only helped identify survey items but also provided information about what is important within each neighborhood— information that can help interpret the survey data. 3CM helps identify the intangible aspects of abstract issues a nd categorize themes associated with that concept, for example, people’s perception of so cial capital in their neighborhood. Analysis of 3CM is time-c onsuming, requiring a combination of qualitative and quantitative analytical methods. This study focused on qualitative methods for recognizing broader themes among individual’s cognitive maps. Comparative Neighborhood Analysis The research questions of this study requi red that neighborhoods represent diversity in both involvement in wildfire education programs and percepti on of social capital. Following is a summary of what was simila r and different about these neighborhoods in terms of wildfire risk an d education program, interaction and social capital. The interviews with program developers, 3CM inte rviews with program leaders, observations in each neighborhood provided insight into th e similarities and differences among these neighborhoods. This is not an attempt to speak for all the people in each neighborhood, but rather a way to provide clues about what might be occurring. All of the study neighborhoods are new—none existed before 1970. Only Cypress Knoll in Palm Coast is completely built, which reduces fire risk. Indian Lake Estates, Seminole Woods, and Wedgefield had devasta ting fires inside the neighborhoods. Three neighborhoods—Tiger Creek Forest, Placid Lakes, and Leisure Lakes had minor fires

PAGE 110

95 inside the neighborhoods; Cypress Knoll has expe rienced fire at the community level but not in the neighborhood. All neighborhoods had been exposed to some kind of wildfire education program— some more and some less. Wedgefield, Cypre ss Knoll, and Placid Lakes are part of the FireWise Communities program. Seminole Woods is attempting FireWise certification. Others had varying levels of involvement with the wildfire educational program, such as, using a community picnic to improve inter action among agencies a nd residents, inviting DOF personnel to the neighborhood meetings, and conducting prescribed burns around the neighborhood. Some neighborhoods have a predominance of retired residents; others have a greater percentage of working parents and ch ildren. The structural differences in the neighborhoods may also affect perception of social capital. Of the seven neighborhoods in the study, Tiger Creek Forest and Wedgefiel d are the only two that have lots larger than quarter of an acre. In Tiger Creek Fore st most homes are not visible from the road. In Wedgefield people live in four different types of reside ntial development: the City, consisting of quarter-acre plots; the Estate, consisting of one to five acre lots; the Village, consisting of condominiums; and the Reserve, a gated community. Each format may contribute to different levels of social interaction, expectations, and comfort with neighbors. In Cypress Knoll, a neighborhood of Palm Coast, all homes are built on quarter-acre lots and are clos e to each other; thus creati ng opportunities fo r interaction among residents in the neighborhood. In Se minole Woods unbuilt lots separate houses. This is also the case with the other three neighborhoods where less than 25 percent of the neighborhood is built.

PAGE 111

96 Some neighborhoods use social opportunitie s to connect people to each other. For example, the homeowners association and th e golf tournament in Wedgefield are widely attended and provide avenues for residents to interact. Other ne ighborhoods lean on the community to provide with networking assi stance. For example, in Palm Coast, the phone book lists people by the place from wh ich they came. Some neighborhoods are designed to facilitate interaction among people. Residents of Indian Lake Estates do not get their mail in their mailboxes; they have to collect it from the local post office. The post office acts as a meeting point for ma ny residents and a place for exchange of information, news, and views. In some neighborhoods daytime social opportunities draw retires but exclude working people. Some neighborhoods like Placid Lakes, Cypress Knoll, and Wedgefield enjoy greater volunteer involvement for activ ities with their homeowners association, while others do not. In Cypress Knoll, Wedgefield, and Indian Lake Estates the physical design and social opportunities help enhance social capi tal. In Tiger Creek Forest, Seminole Woods, and Leisure Lakes the physical design and social structure prov ide fewer opportunities for interaction. All of these elements provide the dive rsity of neighborhoods and programs needed in our study. Quantitative Data Analysis In the light of the objectives indicated in Chapter 1, this section will answer the following two research questions: What is the relationship between wildfire education programs and social capital?

PAGE 112

97 What elements of an educational program are more likely to be associated with enhanced social capital? In each community, survey data were collected from people who did and did not participate in wildfire education programs. Fo r the purpose of this research, “participation in wildfire education activities or programs” is defined as receiving, using, and learning from electronic media, print information, and local events. The varia tion in participation was between 15-74% (see Table 4.6). Of the total respondents, 300 participants did not view electronic media and 849 (74%) participants did. Of the total respondents, 771 reported that they did not receive printed information and 427 (36%) participants did. Overall, 1028 participants did not attend events on ways to reduce risk of wildfires and create defensible space while 175 (15%) pa rticipants did. Throughout the analysis, comparisons are made between two groups—participants and non-participants through each of the three media. Table 4.6: Variation in participati on in wildfire educational programs Viewing electronic media Receiving printed information Attending events Nonparticipants Participants Nonparticipants Participants Nonparticipants Participants 300 849 (74%) 771 427 (36%) 1028 175 (15%) The first section focuses on exploring the relationship between wildlife education and perception of social capital by comparing differences in perception of social capital among participants and non-participants in wildfire education programs at the neighborhood level. The second section of this analysis compares respondents’ perceptions of social capita l with participation in e ducation programs across all communities. It has four parts—wildfire education programs and perception of social capital, wildfire preparedness and percepti on of social capital, wildfire education

PAGE 113

98 programs and wildfire preparedness, and pa rticipation in community activities and perception of social capital. The third section attempts to explore characteristics of an educational program that are better associated with enhanced social capital; and therefore explore whether interaction in educationa l programs is associated with improved perception of social capital. Dealing with Missing Data During data entry, missing data and inconsis tent marking patterns were observed for data on participation in wildfire educati on programs and percepti on of social capital. This section discusses the method used to cons istently treat missing data on participation in wildfire activities and perception of so cial capital. The next section looks at demographic characteristics of respondent s. The Appendix F (p 236) discusses the method used to consistently treat missing data on participation in wildfire activities. Throughout the analysis, numb ers (n) reported are based on the number of respondents for each question on the survey. During data entry, missing values were c oded as 9. To maximize the usefulness of respondent information with missing social cap ital data values, th e following rules were created. Rules for deletion: Rule 1: If any respondent had more than 25% (more than 5 items) missing value (9) on the social capital scale, the respon dent was deleted from the analysis. Rule 2: If any respondent had 25% or less (5 items or less) missing value (9), the missing value was recoded as don’t know (neutral value = 3). Since these statements were on an agree/disagree s cale, and not answering probably means the person is relatively ambivalent, a mid-va lue seems to represent that person’s perception.

PAGE 114

99 Demographic Characteristics of Respondents Tables 4.7 through 4.12 provide a brief description of the demographic characteristics of the respondents for each neighborhood. Detailed tables on each variable for each community are presented separately (A ppendix F, p 239). In each of the tables 4.7 to 4.12 all missing data are ex cluded from the analysis. Table 4.7 suggests that with the exception of Seminol e Woods, more of the respondents were males than females. Indian Lake Estates and Cypress Knoll had a more balanced male-female ratio than Tiger Creek Forest and Wedgefield. It is possible that the initial introduction on the survey stating wildfire prep aredness activities and risk reduction skewed response toward males. Table 4.7: Gender distribution of respondents by neighborhood Gender Tiger Creek Forest Indian Lake Estates Wedgefield Cypress Knoll Seminole Woods Placid Lakes Leisure Lakes Male 27 (66%) 78 (54%) 127 (65%) 163 (52%) 60 (47%) 204 (60.0%) 116 (74%) Female 14 (34%) 66 (46%) 69 (35%) 148 (48%) 69 (53%) 135 (40.0%) 41 (26%) 2.00 1.00 Male (59.3%) Female (40.7%) Figure 4.1: Distribution of all respondents by gender Table 4.8 provides age distribution of th e respondents across each neighborhood. In this analysis, based on the fre quency distribution, age variable s were regrouped into three

PAGE 115

100 groups: 20-50 (20-35 and 36-50), 51-65, and 66-81 and above (66-80 and 81 and above). Approximately 61% of the respondents of Indian Lake Estates and Placid Lakes were above 65 years. Respondents of Wedge field and Seminole Woods were younger; approximately 47% and 53% were in the age group 25-50 years respectively. Approximately 40% and more of the responde nts of Tiger Creek Forest and Cypress Knoll fall in the age group 51-65 years. Clearly some communiti es appeal to, attract, and house younger residents with children, wh ile others have mostly retirees. Table 4.8: Age distribution of respondents by neighborhood Age Tiger Creek Forest Indian Lake Estates Wedgefield Cypress Knoll Seminole Woods Placid Lakes Leisure Lakes 20-50 10 (24%) 11 (8%) 92 (47%) 62 (20%) 68 (53%) 51 (15%) 26 (17%) 51-65 20 (49%) 45 (31%) 65 (34%) 131 (42%) 39 (31%) 76 (23%) 54 (34%) 66-81 and above 11 (27%) 89 (61%) 37 (19%) 116 (38%) 21 (16%) 212 (62%) 77 (49%) 5.00 4.00 3.00 2.00 1.00 51-65 years (34.3%) 20-35 years (6.8%) 36-50 years (18.5%) 81 and above ( 7.5% ) 66-80 years (32.9%) Figure 4.2: Distribution of all respondents by age group Table 4.9 describes the educational leve l of the respondents for each neighborhood. For the purposes of the descri ptive analysis, educ ation level variables were regrouped into three groups as classifi ed in the table. Approximat ely 50% of respondents of Wedgefield and Cypress Knoll had a college de gree. Respondents of Tiger Creek Forest, Placid Lakes, and Leisure Lakes had an al most equal education distribution among the

PAGE 116

101 three educational level groups. About 40% of the respondents of Seminole Woods had a college degree. So, respondents of Wedgefiel d and Palm Coast had higher education and lower age than respondents of LWR South and LWR North. Table 4.9: Educational level of respondents by neighborhood Educational Level Tiger Creek Forest Indian Lake Estates Wedgefield Cypress Knoll Seminole Woods Placid Lakes Leisure Lakes Some High school or less/ HS or GED 11 (27%) 48 (34%) 35 (18%) 54 (17%) 39 (30%) 112 (34%) 59 (39.3%) Some college coursework 14 (34%) 54 (38%) 53 (27%) 91 (30%) 36 (28%) 105 (31%) 39 (26%) College degree 16 (39%) 40 (28%) 106 (55%) 161 (53%) 54 (42%) 116 (35%) 42 (34.7%) 3.00 2.00 1.00 Some high school or less (3 .2 %) 6.00 Graduate or professional degree (18.4%) High school diploma or GED (22.2%) 5.00 Undergraduate degree ( 12.9% ) 4.00 Associates degree ( 12.1% ) Some college coursework (31.2%) Figure 4.3: Educational level of all respondents Table 4.10 provides the income level of respondents for each neighborhood. Approximately 64% of the respondents of Wedg efield fall in the in come bracket $ 60,000 and up, which can be considered as hi gh-income group. Approximately 53% of respondents of Tiger Creek Forest, 47% of resp ondents of Cypress K noll, and 35% of the respondents of Indian Lake Estates and Se minole Woods fall in their high-income group. Other neighborhoods reported as low as 25% in this high-income group. Approximately 50% of the respondents of Indian Lake Esta tes, Cypress Knoll, and Seminole Woods and

PAGE 117

102 about 61% of the respondents of Leisure Lake s fall into the income group $ 20,000 to $ 59,999, which can be considered as the low-income group. Table 4.10: Income level of respondents by neighborhood Income Tiger Creek Forest Indian Lake Estates Wedgefield Cypress Knoll Seminole Woods Placid Lakes Leisure Lakes Less than $ 20,000 3 (9%) 17 (15%) 6 (3%) 9 (4%) 11 (10%) 47 (17%) 17 (13%) $ 20,000 $59,999 12 (37.5%) 57 (50%) 55 (33%) 124 (50%) 59 (54%) 147 (55%) 78 (61%) $ 60,000 $99,999 12 (37.5%) 29 (25%) 62 (37%) 83 (33%) 35 (32%) 61 (23%) 21 (17%) $100,000$149,999 5 (16%) 7 (6%) 35 (21%) 20 (8%) 5 (4%) 12 (5%) 9 (7%) $150,000 and above 0 4 (4%) 10 (6%) 13 (5%) 0 0 2 (2%) 5.00 4.00 3.00 2.00 $150,000 and more (2 .8 %) 1.00 Less than $20,000 (8.7%) $100,000 to $149,999 ( 8.9% ) $60,000 to $99,999 (29.2%) $20,000 to $59,999 (50.3%) Figure 4.4: Income level of all respondents Table 4.11 explains the employment st atus of the respondents for each neighborhood. For the purpose of this anal ysis, respondents are grouped as employed (full-time or part-time), retired, and others (student/homemaker/unemployed). Approximately 60% or more of the respondents of Indian Lake Estates, Cypress Knoll, Placid Lakes, and Leisure Lakes were reti red. A large percentage of respondents of Wedgefield (68.5%) and Seminole Woods (64.1 %) were employed. Respondents of Tiger Creek Forest had an equal distributio n of retirees and employed people.

PAGE 118

103 Table 4.11: Employment status of respondents by neighborhood Employment Tiger Creek Forest Indian Lake Estates Wedgefield Cypress Knoll Seminole Woods Placid Lakes Leisure Lakes Employed 22 (54%) 31 (22%) 131 (68%) 97 (32%) 82 (64%) 98 (29%) 51 (35%) Retired 19 (46%) 107 (75%) 51 (27%) 191 (63%) 32 (25%) 223 (67%) 89 (60%) Others 0 5 (3%) 9 (5%) 14 (5%) 14 (11%) 12 (4%) 8 (5%) 6.00 5.00 4.00 3.00 2.00 1. Homemaker ( 3.8% ) Student (0.3%) Unemployed (0.6%) 00 Full-time ( 36% ) Retired (53.8%) Part-time (5.6%) Figure 4.5: Employment st atus of all respondents Table 4.12 describes the ethnicity of respondents of each neighborhood. For the purpose of the descriptive analysis, based on the low frequency distribution in each group, all ethnicities (Hispanic or Latino, Afri can American or Black, Asian or Pacific Islander, American Indian or Native Al askan, and Native Hawaiian) were grouped together as others. More than 80% of th e respondents of each neighborhood were White. Wedgefield had a larger ethni c mix: 17% of the responde nts of Wedgefield were nonCaucasian. This may likely to be a functi on of its location, between Orlando and the space industries near Titusville. The resi dents are younger and working and the two nearby cities draw a greater ethnic mix to this community. Approximately 7% of the respondents of Cypress Knoll, Seminole W oods, and Leisure Lakes, and 3% of the respondents of Indian Lake Es tates were non-Caucasians.

PAGE 119

104 Table 4.12: Ethnicity of respondents by neighborhood Ethnicity Tiger Creek Forest Indian Lake Estates Wedgefield Cypress Knoll Seminole Woods Placid Lakes Leisure Lakes White 40 (100%) 135 (97%) 159 (85%) 271 (91%) 118 (94%) 316 (96%) 144 (95%) Others 0 4 (3%) 31 (15%) 26 (9%) 8 (6%) 13 (4%) 8 (5%) 7.00 5.00 4.00 3.00 2.00 1. Hispanic or Latino (2.4%) African American or Black (2.6%) Others (1.1%) American Indian or Native Alaskan ( 0.4% ) Asian or Pacific Islander (0.8%) 00 White (92.8%) Figure 4.6: Ethnicity of all respondents Section 1: Neighborhood Data Analysis This section explores the relationship betw een participation in wildfire education programs and perception of social capital at the neighborhood level in an attempt to answer the first research question. On the survey, participation in wildfire education was divided in three activities— viewing electronic media, receiving printed information, and attending events. In all analysis, Type 1 error was se t at 0.05. Any missing value was excluded from the analysis. Social capital was measured with a Likert scale of 21 items that asked how respondents perceived their neighborhoods and their commun ities. A summated social capital scale was created for analysis. The value of answ ering this question at the neighborhood level is that each of these neighborhoods varies in their level of particip ation in educational programs and their perception of social capita l. Before we proceed with the analysis

PAGE 120

105 across communities, it will be us eful to understand the relati onship between participation and perception of social cap ital at this level. Tiger Creek Forest In Tiger Creek Forest, 41 people responded to the survey. Table 4.13 provides mean social capital values, standard deviat ion, and mean error for respondents of Tiger Creek Forest who did and did not particip ate in wildfire education activities. Table 4.13: Mean social capital values for respondents of Tiger Creek Forest Electronic media Printed information Events Nonparticipants Participants Nonparticipants Participants Nonparticipants Participants N 10 18 12 16 19 9 Mean Social Capital 71.70 71.72 68.08 74.44 69.21 77.00 Standard Deviation 12.74 9.63 12.38 8.47 10.68 8.78 Standard Error 4.028 2.27 3.58 2.12 2.45 2.94 For the purpose of analysis, responses were collapsed into a scale that enabled a designation for those who participated in any aspect of a program and those who did not. One-way ANOVA measured no sign ificant difference in per ception of social capital between respondents of Tiger Cr eek Forest who did and did not participate in any aspect of the wildfire education program. No significant difference in perception of social capital was obtained for respondents of Tiger Creek Forest who did or did not view electronic media, receive printed information, or attend events on ways to reduce risk of wildfires. Certainly the small sample size of respondents contribute s to these results.

PAGE 121

106 Indian Lake Estates In Indian Lake Estates, 149 people responded to the survey. Table 4.14 provides mean social capital values, standard devi ation, and standard e rror for respondents of Indian Lake Estates who did a nd did not participate in wild fire education activities. Table 4.14: Mean social capital values fo r respondents of Indian Lake Estates Electronic media Printed information Events Nonparticipants Participants Nonparticipants Participants Nonparticipants Participants N 42 77 93 34 102 23 Mean Social Capital 78.74 78.32 77.60 a 83.59 a 78.02 a 84.09 a Standard Deviation 11.18 11.32 11.03 10.71 11.02 10.89 Standard Error 1.72 1.29 1.14 1.8355 1.09 2.27 a = p<. 05 One-way ANOVA measured no significant difference in perception of social capital between respondents of Indian Lake Estates who did and did not participate in any aspect of the wildfire education program. No significant difference in perception of social capital was observed among people who viewed electronic media and those who did not. A one-way ANOVA measured a significant difference in perception of social capital for people who received printed in formation (p< .05, n=127) and those who did not. Among those who received information in print, significant difference in social capital was observed for people who did not receive a handout a nd those who received the handout in person (p< .05, n=38). A significa nt difference in perception of social capital was also observed for people who r eceived the newspaper in mail and those who received it in person (p< .05, n=38). A signi ficant difference in perception of social capital was observed for respondents of Indian Lake Estates who attended an event (p<

PAGE 122

107 .05, n=125) and those who did not . No significant difference in perception of social capital was found among other types of prin t information or specific events attended. Wedgefield In Wedgefield, 201 people responded to the survey. Table 4.15 provides mean social capital values, standard deviati on, and standard error for respondents of Wedgefield who did and did not participat e in wildfire education activities. One-way ANOVA measured a significant difference in perception of social capital among people who viewed electronic media and those who did not (p< .05, n=181). Similarly, a significant difference in pe rception of social capital was obtained for respondents of Wedgefield who received inform ation in print and those who did not (p< .05, n=184). No significant diffe rence in perception of soci al capital was observed for respondents who attended events on ways to re duce risk of wildfire s and those who did not. Table 4.15: Mean social capital valu es for respondents of Wedgefield Electronic media Printed information Events Nonparticipants Participants Nonparticipants Participants Nonparticipants Participants N 35 146 73 111 156 27 Mean Social Capital 67.23 a 73.53 a 69.59 a 74.09 a 71.96 74.22 Standard Deviation 12.91 9.70 12.18 9.00 10.83 9.17 Standard Error 2.18 .80 1.43 .85 .87 1.76 a = p< .05

PAGE 123

108 Cypress Knoll In Cypress Knoll, 317 people responded to the survey. Table 4.16 provides mean social capital values, standard deviation, a nd standard error for respondents of Cypress Knoll who did and did not participate in wildfire education activities. Table 4.16: Mean social capital values for respondents of Cypress Knoll Electronic media Printed information Events Nonparticipants Participants Nonparticipants Participants Nonparticipants Participants N 62 222 157 128 221 64 Mean Social Capital 74.16 76.71 73.82 a 79.14 a 75.04 a 80.23 a Standard Deviation 10.29 10.11 9.29 10.53 10.00 9.89 Standard Error 1.31 .68 .74 .93 .67 1.24 a = p< .05 A significant difference in perception of social capital was observed for respondents of Cypress Knoll w ho participated in any wildfire education programs and those who did not (p< .05, n=297). No signifi cant difference in perception of social capital was observed among respondents of C ypress Knoll who viewed electronic media and those who did not. One-wa y ANOVA measured a significan t difference in perception of social capital among respondents of Cypre ss Knoll who received information in print and who did not (p< .05, n=285), and who atte nded events and those who did not (p< .05, n=285).

PAGE 124

109 Seminole Woods In Seminole Woods, 130 people responded to the survey. Table 4.17 provides mean social capital values, standard deviation, a nd standard error for respondents of Seminole Woods who did and did not participate in wildfire education activities. No significant difference in perception of social capital was observed for respondents of Seminole Woods who participated in any aspect of the wildfire education programs and those who did not. Similarly, no significant difference in perception of social capital was observed for respondents of Seminole Woods who viewed electronic media, received printed information, or attended events and those who did not. Table 4.17: Mean social capital values for respondents of Seminole Woods Electronic media Printed information Events Nonparticipants Participants Nonparticipants Participants Nonparticipants Participants N 25 89 88 29 112 5 Mean Social Capital 67.08 69.53 69.35 69.55 69.05 77.20 Standard Deviation 7.45 8.61 8.17 10.56 8.64 9.23 Standard Error 1.49 .91 .87 1.96 .82 4.13 Placid Lakes In Placid Lakes, 353 people responded to the survey. Tabl e 4.18 provides mean social capital values, standard deviation, and standard error for respondents of Placid Lakes who did and did not participate in wildfire education activities. No significant difference in perception of social capital was observed for respondents of Placid Lakes who participated in any of the wildfire education programs and those who did not. One-wa y ANOVA measured a significan t difference in perception of social capital among responde nts of Placid Lakes who view ed electronic media and

PAGE 125

110 those who did not (p< .05, n=276). A significant difference in perception of social capital was also observed among respondents of Placid Lakes who received information in print (p< .05, n=270) and those who did not. No signi ficant difference in perception of social capital was observed for respondents who atte nded events on ways to reduce risk of wildfires and those who did not. Table 4.18: Mean social capital values for respondents of Placid Lakes Electronic media Printed information Events Nonparticipants Participants Nonparticipants Participants Nonparticipants Participants N 80 196 209 61 242 28 Mean Social Capital 71.55 a 74.94 a 73.07 a 76.34 a 73.43 77.07 Standard Deviation 13.76 10.45 12.20 8.67 11.65 10.46 Standard Error 1.54 .75 .84 1.11 .75 1.98 a = p< .05 Leisure Lakes In Leisure Lakes, 159 people responded to the survey. Table 4.19 provides mean social capital values, standard deviation, and standard error for respondents of Leisure Lakes who did and did not participate in wildfire education activities. No significant difference in perception of social capital was observed for respondents of Leisure Lakes who participated in any wildfire education program and those who did not. Similarly, no significant diffe rence in perception of social capital was observed for respondents of Leisure Lakes who viewed electronic media, received printed information, or attended events and those who did not.

PAGE 126

111 Table 4.19: Mean social capital values for respondents of Leisure Lakes Electronic media Printed information Events Nonparticipants Participants Nonparticipants Participants Nonparticipants Participants N 42 93 110 29 133 9 Mean Social Capital 73.12 76.81 75.20 78.14 75.92 74.33 Standard Deviation 10.43 10.75 9.64 12.37 9.94 16.76 Standard Error 1.61 1.11 .92 2.30 .87 5.59 Summary: Results of this analysis suggest no difference in perception of social capital among participants and non-participants in various educational programs in Tiger Creek Forest, Seminole Woods, and Leisure Lakes. In Tiger Creek Forest this could be a result of small sample size. From the interviews with community leaders, Leisure Lakes was identified as the one community wher e the homeowners association had limited educational programs to encourage its homeown ers to reduce risk of wildfires. However the DOF conducted prescribed burns around th e community to reduce wildfire risk. People in Leisure Lakes who did participate in the wildfire education programs were those who obtained information from electroni c media or received information in print. Electronic information can be obtained freely from the web, radio, or television news and may not require any initiation from the comm unity leaders and program developers. The same could be the case with the printe d information obtained through newspapers, magazines, etc. Thus, it validat es the result of no difference in perception of social capital among participants and non-participants . The third neighborhood that showed no significant difference in perception of social capital among participants and nonparticipants in wildfire educational pr ograms was Seminole Woods. From interviews

PAGE 127

112 with program leaders it was obs erved that this neighborhood wa s at the initial phases of its wildfire education initiative. However, as the community of Palm Coast was greatly involved in wildfire educational initiativ es, it was initially assumed that this neighborhood too might have had high partic ipation in educational programs on risk reduction activities and that differences in perception of social capital among participants and non-participants might be observed. Howe ver, results indicate otherwise. Further research could look at when these reside nts came to Seminole W oods. Was it after the 2000 fires? Neighborhoods that showed differences in perception of social capital among participants and non-participan ts in educational programs were Indian Lake Estates, Wedgefield, Cypress Knoll, and Placid Lakes. Of these four neighborhoods, participants of Wedgefield and Placid Lakes who viewed electronic media perceived higher social capital than those who did not view electronic media. Pa rticipants of these four neighborhoods who received printed information perceived higher social capital than those who did not. Thus, receiving printe d information is associated for these neighborhoods with higher perception of soci al capital. Participants of Indian Lake Estates and Cypress Knoll who attended events perceived highe r social capital than those who did not attend events. Although residents of both Wedgefield and Placid Lakes were exposed to several events, data do not show si gnificant differences in perception of social capital among participants and non-participants . This could be a result of small sample size. Section 2: Aggregate Data Analysis This section of the analysis explores the relationship between participation in wildfire education programs and percep tion of social capital across all study

PAGE 128

113 communities. It also explores the relative adva ntages and impacts of different media. For the purposes of this analysis, data from all the communities were collapsed together. The analysis is divided into four parts: partic ipation in wildfire education programs and perception of social capital; wildfire prep aredness and perception of social capital; participation in wildfire education programs and wildfire preparedne ss; and participation in community activities and perception of social capital. Part one: Wildfire education programs and perception of social capital Participation in wildfire education pr ograms was divided in three activities: viewing electronic media, receiving printed information, and attending events. A social capital scale was calculated for analysis. Responses were collapse d into a scale that allowed us to designate those who participated in any aspect of a program and those who did not. Results of a one-way ANOVA suggest that people who particip ated in any aspect of the education program (viewed electronic media, received printed information, or attended events) perceived a greater social cap ital than people who did not participate (p< .05, n=1230). In the neighborhood analysis, pa rticipation in sp ecific educational activities, electronic, print, or events showed significant differences in perception of social capital. Table 4.20 provides the mean social capital values for participants and non-participants in the diffe rent educational media. A one-way ANOVA measured a significant difference in perception of social capital for people who viewed electronic media and those who did not (p< .05, n=1149). Similar results were obtained for those who received printed information (p< .05, n=1198), or attended an even t (p< .05, n=1203), and those w ho did not (see Appendix F, p 250). Thus, people who viewed electronic media, received printed information, or attended events perceived greater soci al capital than people who did not.

PAGE 129

114 Table 4.20: Mean social capital values for pa rticipants and non-parti cipants in different educational media Electronic media Printed information Events Nonparticipants Participants Nonparticipants Participants Nonparticipants Participants N 300 849 771 427 1028 175 Mean Social Capital 74.26 a 77.00 a 75.24 a 78.77 a 75.82 a 80.73 a Standard Deviation 12.61 11.02 11.38 11.09 11.31 11.27 Standard Error .73 .38 .41 .53 .35 .85 a = p< .05 Participation in educational activities: Further, for each educa tional activity, one-way ANOVA was run to examine if there was a diffe rence in perception of social capital for people who did and did not enga ge in specific educational activities within each media— electronic, print, and events (Appendix F, p 250). This analysis helps explore the elements or components of an educational pr ogram that are associated with enhanced perception of social capital. Among respondents who received information from various electronic media sources like websites, television advertisements , television news, television shows, radio advertisements, radio news, and CD/videos, only those who listened to television news (p< .05, n= 1116) and radio news (p< .05, n= 1098) perceived greater so cial capital than those who did not. No significant difference was found in perception of social capital among participants and non-participants in other types of electronic media. Print information about reducing risk of wildfires included handouts, brochures, door hangers, newsletters, magazines, and newspapers. Respondents who received a handout (p< .05, n=452) or a newspaper (p< .05, n= 447) perceived greater social capital than those who did not. No significant difference was observed among those who

PAGE 130

115 received the handout in person or those w ho received it in the mail. A significant difference in perception of social capital was observed among respondents who did not receive the newspaper and those who received it in person (p< .05, n=447) and those who received it in the mail (p< .05, n=447). Thus , people who received the newspaper in mail or in person perceived great er social capital than pe ople who did not receive the newspaper. No significant difference was found in perception of social capital among participants and non-part icipants in other types of print information. Attending an event related to reducing risk of wildfires included presentations, meetings to discuss wildfires, pres cribed burn demonstrations, equipment demonstrations, landscape demonstrations, information booths, and community picnics. A significant difference in perception of social capital was obtaine d for respondents who said they attended a meeting (p< .05, n= 194) , a prescribed burn demonstration (p< .05, n= 190), or a picnic (p< .05, n= 187) and t hose who did not. No si gnificant difference was observed among people who attended the m eeting alone or with someone. However, a significant difference was observed among th ose who attended the meeting alone and who did not attend the meeting (p< .05, n= 194), and those who attended the meeting with someone and who did not attend the meeting (p< .05, n= 194). Thus, attendance in program does matter and may affect perception of social capital; it is equally effective if alone or with someone. A significant differe nce was observed among those who attended the prescribed burn demons tration with someone and those who did not attend a prescribed burn demonstration (p< .05, n= 190). No significant di fference was observed among people who attended the picnic and th ose who attended the pi cnic alone or with someone. Thus, some events appear to correlat e with social capital more than others.

PAGE 131

116 Tables 4.21, 4.22, 4.23, and 4.24 provide the mean social capital values, standard deviation, and standard error for electronic , print, and event ac tivities that showed significant difference in perception of social capital among participants and nonparticipants. Table 4.21: Mean social capital values for pa rticipants and non-participants for specific electronic media Television News Radio News Non-participants Participants Non-participants Participants N 392 725 854 245 Mean Social Capital 75.02 a 76.96 a 75.88 a 77.62 a Standard Deviation 12.51 10.90 11.48 11.63 Standard Error .63 .40 .39 .74 a= p< .05 Table 4.22: Mean social capital values for pa rticipants and non-participants for specific print information Handout Newspaper Nonparticipants Received in mail Received in person Nonparticipants Received in mail Received in person N 230 65 158 250 115 83 Mean Social Capital 76.81 a 79.95 a 79.47 a 76.58 a 79.58 a 80.80 a Standard Deviation 10.16 10.66 12.47 11.06 11.04 11.12 Standard Error .67 1.32 .99 .70 1.03 1.22 a= p< .05 Table 4.23: Mean social capital values for pa rticipants and non-participants for specific events Meeting Prescribed burn demonstration Nonparticipant Alone With other Nonparticipant Alone With other N 81 38 76 130 18 43 Mean Social Capital 74.52 a 81.45 a 83.98 a 77.90 a 80.95 a 82.93 a Standard Deviation 12.38 11.84 8.97 11.93 14.48 9.55 Standard Error 1.38 1.92 1.03 1.05 3.42 1.46 a = p< .05

PAGE 132

117 Table 4.24: Mean social capital values for pa rticipants and non-participants in specific events Picnic Non-participants Alone With other N 151 11 26 Mean Social Capital 78.36 a 83.18 a 84.00 a Standard Deviation 12.46 6.72 7.93 Standard Error 1.01 2.03 1.56 a = p< .05 Effect of learning, using, and interacting with information: Among respondents who engaged in specific educational activities w ithin each group, a correl ation analysis was run to see whether learning, using, or in teracting with the information had any relationship with people’s pe rception of social capital (Appendix F, p 251). The purpose of this analysis was to explore whether peopl e who said they learned, used, or interacted with the information perceived greater social capital as compared to those who did not. The assumption being activities that improve perception of social capital may also enhance social capital. Correlation analysis showed statistical significance for responde nts who interacted with information they received from the handout and perception of social capital. Statistical significance was also observed am ong people who interacted with the handout and used the information from the handout (p< .01, n= 172, r= .590). Statistical significance was also observed for people who interacted w ith the information they received from the newspaper and their per ception of social capital. A significant correlation was observed among people who in teracted with the information they received from the newspaper and used th at same information (p< .05, n= 137, r= .450). No correlation was observed between learning, using, and interacting with information from other media sources and people’s percep tion of social capital. Results of this

PAGE 133

118 analysis suggest that people who interacted with the information and used the information they received from handouts and newspapers pe rceived greater social capital than those who did not interact or use the information. Th us, interaction with and use of information is related with higher per ception of social capital. Table 4.25: Effect of learning, using and interacting with the information on perception of social capital Perception of Social Capital Pearson correlation Significance value (2-tailed) N Perception of social capital 1.00 137 Receiving a newspaper -.05 .60 137 Learning from a newspaper .02 .80 137 Interacting with information from a newspaper .20 a .02 137 Using the information from a newspaper .03 .75 137 Receiving a handout .01 .94 172 Learning from a handout .03 .70 172 Interacting with information from a handout .17 a .03 172 Using the information from a handout .10 .18 172 a Correlation is significant at the 0.05 level (2-tailed) Effect of interaction on p erception of social capital: A regression analysis was conducted to identify whether participation in events, which may invol ve interaction with neighbors, residents, and agency personnel, ha d an effect on percepti on of social capital over and above participation in electronic and print media that may not involve direct interaction among people. For the purpose of th e analysis three indi ces (electronic index, print index, and event index) were devel oped. A summated score was used to develop each index. The minimum value on the electron ic index for each individual was 7 and maximum was 14. The minimum value on a prin t and event index was 6 and 7 and the maximum was 18 and 21 respectively.

PAGE 134

119 A correlation analysis (Appendix F, p 252) revealed a significant relationship between print index and perception of soci al capital (p< .05, n=114, r= .208). A medium effect was observed between event index and perception of social capital (p< .01, n=114, r= .349). No statistical significance was observed between electronic index and perception of social capital. Thus, analys is shows a stronger correlation coefficient between event index and perception of soci al capital as compared to print index. A regression analysis was conducted to determine if there was any effect of participation in all three media sources on perception of social capital. The dependent variable for the analysis wa s perception of social capital . Independent variables were electronic index, print index, and event index. Regression analysis showed a significant effect of participation in el ectronic, print, and events on perception of social capital (n=113, r= .351, Adjusted R 2 = .099). Table 4.26 shows standa rdized and unstandardized coefficients with t and significance values for the analysis. Table 4.26: Coefficients a for regression analysis of pr ogram participation indices and perception of social capital Unstandardized Coefficients Standardized Coefficients t Sig. Model B Std. Error Beta 1 (Constant) 64.38 5.87 10.97 .00 Electronic Index .05 .64 .01 .07 .94 Print Index .20 .60 .04 .34 .74 Event Index 1.14 .37 .33 3.14 .01 a Dependent Variable: Social capital Regression analysis showed th at participation in events had a significant effect on perception of social capital ove r and above participation in electronic and print media, though it is very small (n=113, r= .349, Adjusted R 2 = .114). Table 4.27 shows standardized coefficients, unstandardized coefficients, t value, and significance values for regression analysis of event index with pe rception of social ca pital. Although the

PAGE 135

120 regression analysis predicts only 11.4% of the variance in perception of social capital, it is important as it shows that participation in ev ents predicts a larger variance as compared to participation in all other media together. T hus, participation in events has a statistically significant effect on perception of social capital, at least relative to other tested variables. Table 4.27: Coefficients a of regression analysis of event index and perception of social capital Unstandardized Coefficients Standardized Coefficients t Sig. Model B Std. Error Beta 1 (Constant) 65.80 3.59 18.34 .00 Event Index 1.22 .31 .35 3.94 .00 a Dependent Variable: Social capital When a regression analysis was conducte d with event index as the dependent variable and perception of social capital as independent variable, similar results were obtained. Thus, the study cannot show the direction of the relationship between participation in wildfire education programs a nd perception of social capital, but indicate that there is a relationship. Summary: Results indicate a significant differen ce in perception of social capital among people who did and did not engage in all three types of wildfire education programs when all the communities are analyzed together. Differences were observed in specific educational activities. These differen ces were apparent only in viewing television news; receiving handouts or newspapers; and attending meetings, prescribed burn demonstrations, or community picnics. This research suggests that these types of communication are associated with and indeed may be support the overall program and the development of increased social capital. There was a statistically significant differe nce between people who interacted with information they received from a handout and perception of social capital and those who

PAGE 136

121 did not. Statistical significan ce was also observed for people who interacted with the information they received from the newspape r and their perception of social capital. Regression analysis suggests that participa tion in events had a significant effect on perception of social capital ove r and above participation in electronic and print media. These results suggest that interaction in educational programs may be to some extent influencing perception of soci al capital and helps ascertain a relationship between the two. This finding is in keeping with the literature, which suggests social capital is built when people move out of isolation and interact with others (Wilson, 1997). Part two: Wildfire preparedness a nd perception of social capital This section explores the relationship be tween taking action to reduce risk from wildfires and respondents’ per ception of social capital. Th e question on the mail survey used to conduct this analysis was a yes/no dichotomous scale that asked people whether or not they have taken steps around th eir house to reduce risk of wildfire. There was a statistically significant differe nce in perception of social capital for people who did and did not take steps to re duce risk of wildfire around their homes (p< .05, n= 1223); people who took steps to reduce th eir risk of wildfire perceived greater social capital than those who did not (Appendix F, p 254). This is a very small difference, and may not be meaningful. This is not unexpe cted as the theories about social capital suggest that communities high in social capital are more likely to work together to solve problems. However, one could also hypothesi ze that people who activ ely care about their community and their neighbors would be more anxious to do what they could to protect them. Table 4.28 provides mean scores for social capital, standard de viation, and standard error for people who did and did not take steps to reduce risk of wildfires.

PAGE 137

122 Table 4.28: Mean social capital values for pa rticipants and non-participants of wildfires preparedness activities People who did not take steps to reduce wildfire risk People who took steps to reduce wildfire risk N 310 913 Mean Social Capital 75.06 a 77.11 a Standard Deviation 11.15 11.47 Standard Error .63 .38 a = p< .05 Of those who took steps around their homes to reduce their risk of wildfires, the next question asked was what influenced their decision to take these steps. This question is important to the research as it helps us understand what activitie s people believe were influential in their decision to reduce wild fire risk. The response categories allowed respondents to indicate if activities that involved interaction influenced them more than those that involved litt le interaction. A four-point Likert scale question was used for this analysis. People were asked whether viewi ng electronic information, receiving print information, or attending events influenced th em to take action. They were also asked whether talking to neighbors, experts, leaders, friends, or participa ting in workdays or a FireWise program influenced them. This was an effort to understand whether engagement in educational programs and with others influe nced perception of soci al capital and risk reduction activities. Of all the types of educati onal activities assessed, the two most strongly correlated to perception of social cap ital were talking to friends in the neighborhood (p< .01, n= 663, r= .287) and talking with a community leader (p< .01, n= 663, r= .272). A small correlation coefficient was also observed be tween perception of social capital and influence of attending an event on wildfire preparedness (p< .01, n= 664, r= .213), participating in a FireWise program (p < .01, n= 662, r= .210), and seeing neighbors do

PAGE 138

123 risk reduction around their homes (p< .01, n= 665, r= .203) on taking actions to reduce risk of wildfires around homes. All these ac tivities involve intera ction among people, be it a neighbor, community leader, or forest or fire protection agency personnel. Thus, it reiterates the fact that engagement in activities and interaction with people has a significant effect on percepti on of social capital. The correlation analysis showed statistical significance between pe rception of social capital a nd perception of influence of educational activity or interac tion on taking steps to reduce ri sk of wildfires (Appendix F, p 254). However, there can be other factors that affect or influence risk reduction behaviors, such as, perception of risk, res ources, tools, and the physical landscape. Effect of educational media on risk reduction: A regression analysis was conducted to see the effect of participati on for those who believed that participation influenced action to reduce risk of wildfires and perception of social capit al. For this analysis, three separate regression models were run for electronic media, print, and events. For those who viewed electronic media, bot h the independent vari able participation in electronic media and people’s perception of influence of electronic media were scale variables. Participation in electronic media was measured through an electronic media index as discussed in the prev ious section. The dependent va riable for the analysis was perception of social capital. Regression analys is showed a signifi cant relationship, though small, between viewing electronic media a nd people’s percepti on of influence of electronic media on taking action to reduce risk of wildfire s and perception of social capital (n= 627, r= .151, Adjusted R 2 = .020). Regression analysis also s howed a significant relationshi p between receiving print information and people’s perception of influence of written information on taking action

PAGE 139

124 to reduce risk of wildfires and perception of social capit al (n= 326, r= .196, Adjusted R 2 = .033). Both the independent variables for the analysis—receiving printed information and influence of written information were scale variables. Receiving printed information was measured through a print index as discus sed in the previous section. The dependent variable for the analysis was percepti on of social capital. Tables 4.29, 4.30, and 4.31 show unstandardized coefficients, standardized coefficients, t value, and significance value for three regression analyses for el ectronic, print, and event information. Table 4.29: Coefficients a for regression analysis of electronic media and perception of social capital Unstandardized Coefficients Standardized Coefficients t Sig. Model B Std. Error Beta (Constant) 72.6 2.52 28.82 .00 Perception of influence of electronic information 1.78 .51 .15 3.47 .00 Electronic Index .04 .30 .01 .13 .90 a Dependent Variable: Social capital Table 4.30: Coefficients b for regression analysis of print media and perception of social capital Unstandardized Coefficients Standardized Coefficients t Sig. Model B Std. Error Beta 1 (Constant) 69.63 2.78 25.04 .00 Print Index .36 .28 .07 1.32 .19 Perception of influence of written information 1.82 .63 .16 2.90 .01 b Dependent Variable: Social capital Regression analysis also s howed a significant relationshi p between attending events and people’s perception of influence of events on taking action to reduce risk of wildfires and perception of social capit al (n= 151, r= .390, Adjusted R 2 = .141). Both the independent variables for the analysis (attending events a nd influence of events) were scale variables. Attending events was measur ed through an event i ndex as discussed in

PAGE 140

125 the previous section. The depe ndent variable for the analys is was perception of social capital. Table 4.31: Coefficients b for regression analysis of event and perception of social capital Unstandardized Coefficients Standardized Coefficients t Sig. Model B Std. Error Beta (Constant) 65.27 3.40 19.21 .00 Perception of influence of events 3.34 .90 .32 3.70 .00 Event Index .46 .33 .12 1.39 .17 b Dependent Variable: Social capital Thus, results of these three regression m odels suggest that perception of social capital among those who attended events and credit events with influencing their risk reduction actions was greater than those who received electronic or printed information and credited them to risk reduction activities. Summary: There was a statistically significant difference in perception of social capital for people who did and did not take st eps to reduce risk of wildfires around their homes. People who took steps around their homes perceived greater social capital than people who did not. Results of this study also suggest that partic ipation in education programs is associated with higher percepti on of social capital fo r those who did take steps to reduce risk of wildfires. Howeve r, we do not know the direction of this relationship. Regression analysis showed that percep tion of social capital among those who attended events and credit events with influe ncing their risk reduction actions was greater than those who received elec tronic or printed information and credit them for risk reduction actions. Also, they credit risk reduction to participation in events and FireWise programs. Each of these involves interaction among people and agencies in the

PAGE 141

126 community and could either build social capit al or be an outcome of existing social capital in the community. All of these separa te analyses build a case for considering the benefits of participatory educational program s for both social capital and wildfire risk reduction. Part three: Wildfire education pr ograms and wildfire preparedness This section explores the relationship betw een participation in wildfire education programs and taking action to reduce risk of wildfires. People who did and did not participate in wildfire edu cation programs showed a signi ficant chi-square association (p< .05, n= 1215, Cramer’s V= .192) with those who did and did not take steps around their homes to reduce their ri sk of wildfires (Appendix F, p 256). Thus, people who did participate in the wildfire education program we re more likely to take steps to reduce risk of wildfire as compared to those who did not. The value of Cramer’s V suggests that the strength of the relationship between participation in wild fire educational programs and taking steps to reduce wildfire risk is relatively small. A chi-square test showed a significant association (p< .05, n= 1134, Cramer’s V= .154) between people who did and did not view electronic media and those who did or did not take steps around their homes to reduc e their risk of wildfires. As before, the relationship between viewing electronic media and taking steps to reduc e wildfire risk is very small. Chi-square test reveal a signi ficant relationship for people who did and did not receive printed information (p< .05, n= 1134, Cramer’s V= .171) and those who did or did not take steps around their homes to reduce their risk of wildfires. Chi-square test also showed a significant association for people who did and did not attend events (p< .05, n=1188, Cramer’s V= .157) and those who did or did not take steps around their homes to reduce their risk of wildfires.

PAGE 142

127 Effect of learning, using, and interacting with information: A correlation analysis was run to explore whether learning, using, or inte racting with the information received from educational programs had any effect on people taking action to reduce risk of wildfires above and beyond the value of just receivi ng information. Among those who viewed TV news, a significant correlation was observ ed among respondents who learned from TV news (p< .01, n= 522, r= .144), used the information from TV news (p< .01, n= 522, r= .180), and interacted with that informati on (p< .01, n= 522, r= .394) and taking steps to reduce wildfire risk. For respondents who recei ved the handout, a significant correlation was observed for people who inte racted with the informati on in the handout and those who took steps around their homes to reduce ri sk of wildfires (p< .01, n= 205, r= .285). A significant correlation coefficient was also obs erved for people who interacted with the information they received from newspaper a nd those who took steps to reduce risk of wildfire (p< .01, n= 166, r= .294). Among respondents who attended events, a significant correlation coefficient was observed among peop le who interacted with the information they received in the meeting and who took steps to reduce wildfire risk (p< .05, n= 101, r= .242); see Appendix F, p 260. Summary: This analysis showed a statisti cally significant relationship between people who did and did not participate in va rious wildfire education programs and those who did and did not take steps around their homes to reduce their risk of wildfires. Among people who participated in a variety of wildfire educati on programs and took steps around their homes to reduce risk of wildfires, a correlation analysis showed significant relationship between learning, using, and interacti ng with the information and taking steps to reduce risk of wildfires.

PAGE 143

128 Results of this analysis show that education programs are effective at getting people to take action. In addi tion, those who take action also perceive greater social capital. Thus, participation in education program has a positive relationship on both risk reduction activities and percep tion of social capital. Part four: Participation in community acti vities and perception of social capital In this section, other fact ors that may affect percep tion of social capital are explored. The literature suggest s that social capital is associated with participation in community activities and in gr oups and associations. For this analysis, questions on the survey that related to awareness and belonging to groups and associa tions were used to see whether knowing about or belonging to a community group influe nced perception of social capital. A one-way ANOVA was conducted to see if there is any relationship between perception of social capital and awarene ss of community groups and associations (Appendix F, p 262). Community groups and asso ciations listed on the survey include religious groups, homeowners associations, sports groups, garden clubs, youth organizations, ethnic organizations, soci al groups, neighborhood groups, and civic groups. Respondents who said they were aw are of religious groups (p< .05, n=983), homeowners associations (p< .05, n=983), s ports groups (p< .05, n=983), garden clubs (p< .05, n=983), social groups (p< .05, n=983), neighborhood committees (p< .05, n=983), and civic groups (p< .05, n=983) perceived social cap ital differently than those who were not aware of these groups. Similarly, a one-way ANOVA was conducted to measure whether there was a difference in perception of social cap ital among respondents who belonged to a community group or association and t hose who did not (Appendix F, p 263).

PAGE 144

129 Respondents who said they belonged to re ligious groups (p< .05, n=785), homeowners associations (p< .05, n=785), sports gr oups (p< .05, n=785), garden clubs (p< .05, n=785), social groups (p< .05, n=785), nei ghborhood committees (p< .05, n=785), and civic groups (p< .05, n=785) perceived social capital differently than those who did not belong the group. A chi-square test measured a significant association between be ing aware of groups and associations and belonging to that group or association. This in cludes being aware of a group and belonging to that group respectivel y (see Appendix F, p 264). The significant associations for the groups are as follows : religious groups (p< .05, n=786, Cramer’s V= .345), homeowners associations (p< .05, n=786, Cramer’s V= .409), sports groups (p< .05, n=786, Cramer’s V= .270), garden clubs (p< .05, n=786, Cramer’s V= .172), youth groups (p< .05, n=786, Cramer’s V= .233), et hnic groups (p< .05, n=786, Cramer’s V= .171), social groups (p< .05, n=786, Crame r’s V= .354), neighborhood committees (p< .05, n=983, Cramer’s V= .304), and civic gr oups (p< .05, n=983, Cramer’s V= .321). Participation in activities and perception of social capital: Descriptive statistics were calculated to see whether participation in community activities had any effect on perception of social capital . Tables 4.32, 4.33, and 4.34 show mean values for social capital, with standard deviation and standard error values for people who did or did not participate in activities in their communities. As expected, people who participated in various community activities showed greater percepti on of social capital as compared to non-participants.

PAGE 145

130 Table 4.32: Mean social capital values for pa rticipation in various community activities Community events Club events Sports events No Yes No Yes No Yes N 310 439 617 132 492 257 Mean Social Capital 71.29 a 79.26 a 74.69 a 81.89 a 74.54 a 78.66 a Standard Deviation 9.98 10.40 10.70 10.14 10.82 10.70 Standard Error .57 .50 .43 .88 .49 .67 a = p< .05 Table 4.33: Mean social capital values for pa rticipation in various community activities Meetings Educational meetings Work projects No Yes No Yes No Yes N 461 288 613 136 595 154 Mean Social Capital 72.82 a 80.98 a 75.09 a 79.88 a 74.47 a 81.70 a Standard Deviation 10.32 10.03 10.82 10.70 10.34 11.38 Standard Error .48 .59 .44 .92 .42 .92 a = p< .05 Table 4.34: Mean social capital values for pa rticipation in various community activities Meetings to resolve problems inside the community Meetings to resolve problems outside the community No Yes No Yes N 544 205 663 86 Mean Social Capital 73.80 a 81.70 a 75.10 a 82.56 a Standard Deviation 10.42 10.25 10.63 11.14 Standard Error .45 .72 .41 1.20 a = p< .05 Regression analysis indicated that partic ipating in meetings, such as homeowner associations and garden clubs was a predictor of individual’s percepti on of social capital (p< .05, n=748, Adjusted R 2 =. 144), Table 4.35. Participating in meetings to solve problems inside the community; participating in work projects; and community events such as picnics, yard sales and hol iday parties (p< .05, n=748, Adjusted R 2 =. 181) were also significant predic tors of individual’s perception of social capital, Table 4.36.

PAGE 146

131 Table: 4.35: Coefficients b for regression analysis of atte nding meetings and perception of social capital Unstandardized Coefficients Standardized Coefficients t Sig. Model B Std. Error Beta (Constant) 68.72 .74 92.52 .00 Participating in meeting 4.33 .39 .38 11.24 .00 b Dependent Variable: Social capital Table: 4.36: Coefficients b for regression analysis of atte nding meetings and perception of social capital Unstandardized Coefficients Standardized Coefficients t Sig. Model B Std. Error Beta (Constant) 63.60 1.05 60.65 .00 Participating in community events 2.64 .44 .22 5.96 .00 Participating in work projects 2.74 .62 .15 4.40 .00 Participating in meetings to resolve problems inside the community 2. 60 .49 .20 5.31 .00 b Dependent Variable: Social capital Summary: These analyses showed a significant difference in perception of social capital among people who were aware of or belonged to a community group or association and those who were not. A si gnificant correlation was obtained for participation in community events, part icipation in community meetings, and participating in meetings to resolve probl ems inside the community and individual’s perception of social capital. Results of regre ssion analysis suggested that participation in community meetings is a predictor of an indi vidual’s perception of social capital. Thus, results suggest that people who particip ate in community meetings have higher perception of social capital or that participation in the m eetings improves perception of social capital. In either case, participating in meetings is associated with enhanced

PAGE 147

132 perception of social capital. It suggests that this is an important indicator for measuring individual’s percepti on of social capital. Community demographic variables and perception of social capital The literature suggests that social cap ital is correlated w ith the community demographic measures, such as length of residence, age of residents, gender, socioeconomic status, occupation, and level of education (Scaff, 1952; Hagedorn and Labovitz, 1968; Kasarda and Janowitz, 1974; Foskett, 1995). People who have lived in a community longer, and/or are older, wealth ier, and more educated should perceive greater social capital than people who are new residents, younger, less educated, etc. It is suggested that the latter will be less enga ged in the community, and therefore will not contribute to or perceive social capital. The literature also suggests that social capital can be improved or enhanced through engagement and interaction with others. A one-way ANOVA (for continuous variables) and General Linear Model (GLM) univariate analysis (for categorical variables) we re conducted to examine if there was a relationship between community demographic variables and percep tion of social capital. If there is a relationship between community demographic variables and pe rception of social capital then maybe the demographic variables are causi ng the difference in perception of social capital. However, if there is no relationsh ip between community demographic variables and individual’s perception of social capital then, something else may be causing the difference. Age, education, gender, and employment were treated as discrete variables; income and length of residence were treated as cont inuous variables. A GLM univariate analysis shows a significant effect of all age categories on percepti on of social capital (p< .05, n=951, Adjusted R 2 = .046), (see Appendix F, p 273). A one-way ANOVA was run to see

PAGE 148

133 which specific age categories showed significance with perception of social capital. A significant difference in perception of social capital was observed for people in age group 20-35 (p< .05, n=951) and 36-50 years (p< .05, n=951). People who belonged to age group 20-35 and 36-50 perceived lower social ca pital than those who did not belong to these groups respectively. Thus, the younger pe ople perceived lower social capital when compared to people in age group 51 and older. A significant differen ce in perception of social capital was obtained for respondent s who belonged to age group 66-80 years (p< .05, n=951). No significant difference in pe rception of social capital was found for respondents who belonged to age groups 51 to 65 and 81 and above. This suggests that the elder people in the community, age 66-80 ye ars perceived greater social capital than respondents in other age groups. A GLM univariate analysis showed a si gnificant effect of all employment categories on perception of social capital (p< .05, n=951, Adjusted R 2 = .043). One-way ANOVA showed a significant difference in perc eption of social cap ital among those who were employed full-time and those who we re not (p< .05, n= 951). People who were employed full-time perceived lower social capital as compared to people who were not employed full-time. Respondents who were re tired perceived greater social capital compared to people who were not (p< .05, n=951). For people who did or did not participate in any wildfire education progr am, no significant relationship was observed between perception of social capital and in come, education, length of residence, and gender. Summary: One would expect that as demographi c factors such as education, income, length of residence, and age increase, so would perceptions of social capital. In this study

PAGE 149

134 most demographic variables did not show a ny significant relationshi p with perception of social capital. This is curious, because some individuals do perceive that their communities have social capit al, and one might wonder what variables make those people different from others, except for age and employment . Interestingly, participation in wildfire education programs does show a significant relationship with perception of social capital. Perhaps participation in wildfire education programs and activities helped to contribute to these differences in perception of social capital. Section 3: Interaction in Disaster Educ ation Program and Perception of Social Capital So far, the results of this study show that there is a significant relationship between participation in disaster education programs and perception of social capital. In addition, people who participated in the education progr ams and took actions to reduce their risk of wildfires perceived greater soci al capital than people who di d not participate and did not take action. The next question looked at how the different types of education program elements compared and whether a perception of social capital could be predicted from participation in different educational program s. Was it possible to predict social capital differently depending on the level of e ngagement in educational programs for participation in electronic, print, or events ? This was in anticipa tion of discovering which elements might play a greater ro le in building social capital. For the analysis, gender, age, education, and employment were treated as discrete variables; for each, (n-1) dummy variables we re created (n is the number of original variables for each demographic characteristic ). When conducting regression analysis if the nominal variable had more than two levels , dummy variables were created to take the place of the original nominal variable. In come, length of residence, perception of

PAGE 150

135 influence of educational activity on taking steps to reduce risk of wildfire, and participation in various educational media were treated as scale variables. These variables were chosen for the model, because the literatu re suggests that these affect perceptions of social capital. Participation in educationa l programs was measured through electronic, print, and event indices. Perception of social capital was treated as the dependent scale variable. Conditions of linearity, normality, and multi-collinearity were met before conducting regression analysis. Regression analysis helped identify the effect of confounding variables on relationship between participation in different wildfire education programs and per ception of social capital. Viewing electronic media This regression analysis showed that part icipation in electronic media, perception of influence of electronic media on taking step s to reduce risk of w ildfire, income, length of residence, age, education, and employmen t significantly predict individual perception of social capital (n=506, r= .361, Adjusted R 2 = .096). This model explained approximately 10% of the variance in per ception of social capita l. Table 4.37 shows unstandardized coefficients, B, standard error, standardized coefficients , t, and significance values for the model. In this model, electronic index came out to be nonsignificant. This may be a result of intera ction with the perception of influence of electronic media variable and not the fact that it is not significant. To validate this, the same regression model was run using the block method, where th e first block was demographic variables, the second was el ectronic index, and the third block was perception of influence of electronic media. Results indicate that if perception of influence of electronic media is not incl uded in the model, electronic index was

PAGE 151

136 significant with perception of social capital. This valid ates the earlier finding that participation in electronic media has a signifi cant relationship with perception of social capital. The electronic index variable is not significant in the regression model (Table 4.37) because of its interaction with the perception variable. Table 4.37: Coefficients b for regression analysis for elect ronic media participation and demographic variables with perception of social capital Unstandardized Coefficients Standardized Coefficients t Sig. Model B Std. Error Beta 1 (Constant) 69.53 4.11 16.90 .00 Electronic Index .42 .33 .06 1.28 .20 Perception of influence of electronic media 1.46 .57 .12 2.56 .01 Length of residence .04 .06 .03 .72 .47 Gender: Male .89 1.05 .04 .85 .40 Age: 20-35 years -1.64 3.18 -.04 -.52 .61 Age: 36-50 years .74 2.81 .03 .26 .79 Age: 51-65 years 2.12 2.55 .09 .83 .41 Age: 66-80 years 2.84 2.48 .11 1.15 .25 Age: 81 and above .00 .00 .00 .00 .00 Some high school or le ss -5.95 3.22 -.09 -1.85 .07 High school diploma or GED -1.25 1.57 -.04 -.80 .42 Some college coursework -3.20 1.37 -.13 -2.33 .02 Associates degree -2.64 1.79 -.07 -1.48 .14 Undergraduate degree -.43 1.64 -.01 -.26 .79 College degree .00 .00 .00 .00 .00 Income .59 .63 .05 .93 .35 Unemployed -8.13 6.47 -.05 -1.26 .21 Home maker -8.00 2.61 -.15 -3.07 .00 Student -6.38 6.52 -.04 -.98 .33 Full-time employee -4.10 1.53 -.18 -2.69 .01 Part-time employee -6.29 2.05 -.14 -3.07 .00 Retired .00 .00 .00 .00 .00 b Dependent Variable: Social capital

PAGE 152

137 Receiving print information Regression analysis showed that receiv ing printed information, perception of influence of written information on taking step s to reduce risk of w ildfire, income, length of residence, age, education, and employmen t significantly predict individual perception of social capital (n=249, r= .495, Adjusted R 2 = .182). This model explained approximately 18% of the variance in per ception of social capita l. Table 4.38 shows unstandardized coefficients, B, standard error, standardized coefficients , t, and significance values for the model. Table 4.38: Coefficients b for regression analysis for prin t participation and demographic variables with perceptio n of social capital Unstandardized Coefficients Standardized Coefficients t Sig. Model B Std. Error Beta 1 (Constant) 70.85 4.96 14.29 .00 Print Index .30 .28 .07 1.08 .28 Perception of influence of written information 1.92 .66 .18 2.90 .00 Length of residence .08 .09 .06 .96 .34 Gender: Male -1.96 1.37 -.09 -1.43 .15 Age: 20-35 years -.85 4.46 -.02 -.19 .85 Age: 36-50 years 1.60 3.72 .06 .43 .67 Age: 51-65 years 2.86 3.31 .13 .86 .39 Age: 66-80 years 2.80 3.21 .12 .87 .39 Age: 81 and above .00 .00 .00 .00 .00 Some high school or less .76 5.58 .01 .14 .89 High school diploma or GED -3.14 2.09 -.11 -1.50 .14 Some college coursework -3.69 1.80 -.16 -2.06 .04 Associates degree -2.96 2.31 -.09 -1.28 .20 Undergraduate degree -3.42 2.20 -.11 -1.56 .12 College degree .00 .00 .00 .00 .00 Income 2.0 .78 .18 2.56 .01 Unemployed -20.18 7.59 -.17 -2.66 .01 Table 4.38 continued

PAGE 153

138 Unstandardized Coefficients Standardized Coefficients t Sig. Model B Std. Error Beta Full-time employee -6.64 1.93 -.30 -3.45 .00 Part-time employee -3.68 2.74 -.09 -1.34 .18 Home maker -7.22 3.82 -.13 -1.90 .06 Student -4.62 10.39 -.03 -.45 .66 Retired .00 .00 .00 .00 .00 b Dependent Variable: Social capital In this model, print index was not significant. This may be a result of interaction with the perception of influe nce of written information va riable. To validate this, the model was run using the block method, where th e first block was demographic variables, the second was print index, and the third bl ock was perception of influence of written information. Results indicate that if perception of influence of written information is not included in the model, print index was significa nt with perception of social capital. This validates the earlier finding th at participation in print medi a has a signific ant relationship with perception of social capital. The prin t index variable is not significant in the regression model (Table 4.38) b ecause of its interaction with the perception variable. Attending events Regression analysis showed that attending events, perception of influence of attending events on taking steps to reduce risk of wildfire, income, length of residence, age, education, and employment significantly predict individual perception of social capital (n=109, r= .693, Adjusted R 2 = .378). This model explained approximately 38% of the variance in perception of social capital. Table 4.39 shows unstandardized coefficients, B, standard erro r, standardized coefficients , t, and significance values for the model.

PAGE 154

139 Table 4.39: Coefficients b for regression analysis for pa rticipation in events and demographic variables with perception of social capital Unstandardized Coefficients Standardized Coefficients t Sig. Model B Std. Error Beta 1 (Constant) 61.00 8.09 7.55 .00 Event Index .66 .36 .18 1.84 .07 Perception of influence of events 1.97 1.02 .20 1.93 .06 Length of residence -.03 .12 -.03 -.33 .74 Gender: Male .53 2.12 .02 .25 .80 Age: 20-35 years 5.94 7.92 .10 .75 .46 Age: 36-50 years -1.79 7.03 -.05 -.26 .80 Age: 51-65 years .43 6.17 .02 .07 .95 Age: 66-80 years 5.07 5.95 .20 .85 .40 Age: 81 and above .00 .00 .00 .00 .00 Some high school or less -13.80 7.93 -.19 -1.74 .09 High school diploma or GED -3.05 3.33 -.09 -.92 .36 Some college coursework -5.51 2.71 -.21 -2.03 .05 Associates degree .69 3.26 .02 .21 .83 Undergraduate degree -1.44 3.26 -.04 -.44 .66 College degree .00 .00 .00 .00 .00 Income 2.96 1.18 .23 2.51 .01 Full-time employee -6.97 2.96 -.28 -2.36 .02 Part-time employee 4.40 4.49 .08 .98 .33 Home maker 7.89 11.71 .06 .67 .50 Unemployed -20.34 12.40 -.16 -1.64 .10 Retired .00 .00 .00 .00 .00 b Dependent Variable: Social capital In this model, event index was not significan t. This may be a result of interaction with the perception of influence of events variable. To vali date this, the same regression model was run using he block method, where th e first block was demographic variables, the second was event index, and the third bloc k was perception of influence of events. Results indicate that if percep tion of influence of events was not included in the model, event index was significant with perception of social capital. This validates the earlier

PAGE 155

140 finding that participation in events has a signi ficant relationship with perception of social capital. The event index variable is not si gnificant in the regre ssion model (Table 4.38) because of its interaction with the perception variable. Summary: Results of this analysis suggest th at keeping all other variables in the model the same, attending events and percep tion of influence of attending events on taking steps to reduce risk of wildfire was a be tter predictor of an individual’s perception of social capital. The model predicted 37.8% of the variance in perception of social capital as compared to the other two analyses that predict 18.2% (printed information) and 9.6% (electronic media) of variance in pe rception of social capital respectively. Thus, these results suggest that engagement in events makes a greater cont ribution to perception of social capital than viewing electronic media and receiving printed information. Although the study does not show a causal relationship between participation in education programs and perception of social capital, it suggests that participation in events offers something different than the other two media. Of course, the people who choose to participate in events could be ones that perceive greater social capital than those who view electronic media or receive printed information. To account for the variability in neigh borhood, n-1 dummy variables were created for each neighborhood. Regression analysis wa s conducted to explore if neighborhood had any effect on the relationship between participation in edu cation programs and perception of social capital, when all other independent variables in the regression model were the same. Regression an alysis revealed no effect of neighborhood on perception of social capital for the two models—receiving printed information and attending events (Appendix F, p 276). This suggests that existi ng levels of social capital in neighborhoods

PAGE 156

141 did not have an influence in these two mode ls. In the model of viewing electronic media on perception of social capita l, a significant effect of the neighborhood Seminole Woods was observed. Conclusion This chapter presents the results of thr ee data collection tools used in the study— interviews with program developers and implementers, 3CM interviews with active residents, and mail surveys with homeowners in the study communities. Interviews with program developers and implementers of the wildfire education programs revealed that people’s participati on could be better unde rstood on a continuum that involved no engagement to a lot of enga gement in the education program. Wildfire education program summaries developed fo r each community elaborated on the goals, objectives, target audience, anticipated outco mes, and unintended consequences of the programs. The 3CM interviews with key residents pr ovided a better understanding of people’s perception of their neigh borhoods and interaction in their neighborhoods. Component analysis revealed eight components that were mentioned by participants of all groups. These were golf, association activity, Christmas (as a social event), people get along well with everyone, friendly, something in common, help each other, and close-knit community . Category analysis revealed that peop le’s perceptions of their neighborhoods could be categorized into two as pects—Neighborhood attitudes and Neighborhood features. 3CM data also helped understand th ings that participants liked about their neighborhood, things that they liked about people in their neighborhood, and things that they would like to see ch ange in their neighborhood.

PAGE 157

142 The mail surveys were analyzed at tw o levels: neighborhood level and across communities. Results of analysis conducted at the neighborhood level reveal no difference in perception of social capital among participan ts and non-participants in various educational programs in Tiger Creek Forest, Seminole Woods, and Leisure Lakes. A significant difference in percep tion of social capit al was observed among participants and non-participan ts of educational programs in Indian Lake Estates, Wedgefield, Cypress Knoll, and Placid Lakes. People in these communities who participated in specific educational programs perceived greater social capital than people who did not participate in those programs. Aggregate level analysis revealed a signi ficant difference in perception of social capital among people who did and did not e ngage in wildfire education programs. Specific activities that showed greater perc eption of social capital among participants as compared to non-participants were viewi ng television news; receiving a handout or a newspaper; and attending meetings, prescr ibed burn demonstration, or a community picnic. These types of communication are associated with and indeed may be supporting the overall program and the development of social capital. The study also indicated that participation in events had a greater effect on perception of social capital than other educational activities. Ther efore, including events a nd opportunities to increase engagement in education programs may effect perception of social capital. Results of the study also showed a signifi cant difference in perception of social capital for people who did and did not take st eps to reduce risk of wildfires around their homes. People who took action around their homes perceived greater so cial capital than people who did not take action. Of those who took action, a strong correlation was

PAGE 158

143 observed between perception of social capital and influence of talking to friends in the neighborhood or influence of talking to a co mmunity leader. According to the Diffusion of Innovations theory these people may be the opinion lead ers of a community who may informally influence people’s attitudes (Rogers, 2003). Data showed a significant relationship between particip ation in education programs and taking steps to reduce wildfire risk. Learning, using, and interacti ng with information obtained through various media showed a significant effect on perception of social capital. In this study, most of the demographic vari ables that the litera ture suggests have a positive relationship with social capital did not show the predicted relationship with perception of social capital. However, participation in wildfire education program did show a significant effect on perception of social capital. Ba sed on this result, the study suggests that perhaps participation in wildfire education programs and activities helped to contribute to these differences in perception of social capital. Finally, the study showed that engagement in events makes a greater contribution in predicting percepti on of social capital than engagement in electronic or print media. Also, engagement in events was associated with higher risk reduction activities. Overall, the three analyses used in th e study helped improve understanding of the education programs in each community, people’s perception of social capital, and suggest a relationship between participation in wild fire education programs and perception of social capital.

PAGE 159

CHAPTER 5 IMPLICATIONS AND CONCLUSIONS We are faced with many environmental and social problems—increased pollution, increasing population, and rele ntless environmental degr adation. All of these are compounded by an increased disconnectedne ss among individuals and reduced civic engagement. This trend results in decreas ed social and economic well-being and a deterioration of quality of life. One type of environmental problem, seemingly unrelated to pollution and resource degrad ation, is natural disaster. Interestingly, preparedness to reduce vulnerability to disaster can incr ease community engagement and connections among residents. The current reduced civic engagement is relatively new to the United States. “For the first two-thirds of the twentieth century a powerful tide bore Americans into deeper engagement in the life of their communities, but a few decades ago—silently, without warning—that tide reversed and we were ove rtaken by a treacherous rip current. Without at first noticing, we have been pulled apar t from one another and from our communities over the last third of the cen tury” (Putnam, 1995, p 27). Putnam goes on to say that the deteriorating fabric of American society thr eatens to shrink access to social capital and cause civic and personal strife. Loss in social capital is asso ciated with increased crime rates, lower educational performance, child suicide, teen pregnancy, low birth weight, and prenatal mortality (Putnam, 1995). Social capital is thus a vital ingredient that strengthens and enhances a community, a resource that stabilizes and sustains a community. Environmenta l educators, program 144

PAGE 160

145 developers, and researchers shoul d be exploring the concept of social capital. Some types of environmental education may have the poten tial to also build community relations and motivate people to work together to solv e other community problems, resulting in enhanced social capital. The literature suggests that in creased social capital is as sociated with alleviating poverty (Collier, 1998) , fostering economic growth (P utnam, 2000), increasing child academic achievement ( Coleman, 1988; Israel et al. , 2001), and attaining jobs (Granovetter, 1973). High social cap ital is associated with cooperative problem solving, effective governance and rapid economic development (Putnam et al., 1993). Social capital plays an important role in fostering social networks that are needed to take collective action (Putnam, 2000). So cial capital is also associ ated with improved disaster preparedness (Jakes et al. , 2002). The literature suggests that communities with higher social capital are better able to reduce their risk of wildfires as compared to communities with lower social capital (Jakes et al. , 2002). The objective of this stu dy was to explore the relationship between wildfire education programs, risk reduction, and per ception of social cap ital. Specifically, it explored where there is a difference in per ception of social cap ital among people who did and did not participate in vari ous aspects of local wildfire education programs. The study was also designed to improve the understandi ng of the elements of educational programs that are more likely to be associated with enha nced social capital, i. e., if interaction and level of involvement in education programs are associated with enhanced social capital. It examined characteristics of w ildfire education programs that helped change risk reducing behaviors and improve social capital, which were impor tant outcomes for achieving

PAGE 161

146 environmental, economic, and social sustai nability. The results suggest that adult environmental education programs can be de signed to build community relations and motivate people to work together to solve other community problems. The study used in-depth interviews, Conceptual Content Cognitive Mapping (3CM), and mailed surveys to collect data from individuals in seven neighborhoods in three communities in Florida. Study nei ghborhoods included Wedgefield (in Orange County, east of Orlando), two neighborhoods of Palm Coast (in Flagler County, north of Daytona)—Seminole Woods and Cypress Kno ll, and four neighborhoods on the Lake Wales Ridge (in Polk and Highl ands Counties)—Tiger Creek Fo rest, Placid Lake Estates, Indian Lakes Estates, and Leisure Lakes. In-depth interviews were conducted with 20 program developers and implementers in all seven neighborhoods to obtain informa tion on the wildfire education programs in each neighborhood. This interviews included information about their goals, objectives, target audiences, anticipated outcomes, and unintended consequences. Fifteen people in seven neighborhoods were inte rviewed with an open-ended 3CM process to explore people’s perceptions of their neighborhoods and the social interaction within their neighborhoods. This interview process helped articulate items people felt were important in describing their neighborhoods; these were used in the development of the social capital scale. Mail surveys were sent to 3,744 homeowners in the seven neighborhoods in Florida. Usable data were obtained from 1,350 respondents. The overall response rate for all the communities was 36.83%. No response bias was observed among people who did and did not respond to the mail surveys, ba sed on a comparison of respondent and nonrespondent property values.

PAGE 162

147 Results in Brief Results of the mail survey indicate a significant relationship be tween participation in wildfire education programs and perception of social capital; people who participated in the wildfire education program perceived higher social capital in their communities than people who did not participate. The study di d not definitely find that participation in educational programs builds social capital becau se perceiving greater social capital might be the reason people particip ate in wildfire awareness a nd action programs. Additional analysis was conducted to try to determine whet her perception of social capital could be an outcome of participation in wildfire education programs. 1. The study reveals a significant difference in perception of so cial capital among participants and non-particip ants of specific wildfire preparedness educational activities. These activities include view ing television news, listening to radio news, receiving a handout or a newspa per, attending meetings, attending a prescribed burn demonstration, and atte nding a community picnic. One may argue that people who participated in these activities are the ones who already had a greater perception of social capital or, conversely, that participation changed how people perceived their neighbors and thei r community to the extent that it influenced their perception of social capital. In either case, participation in these activities is associated with enhan ced perception of social capital. 2. Participation in events had a significant e ffect on perception of social capital, and it was larger than the effect of partic ipation in electronic and print media. Participation in events by definition i nvolves interaction among participants and program implementers. The literature suggest s that social capital can be built or improved by participation, interaction and communication, trust, sense of trustworthiness, reciprocity, and mutual respect (Wilson, 1997) . Participation in an event increases the chances of inte raction and communication through talking, sharing and comparing information a nd experiences, sharing concerns, and catching up on local events and may thus pr ovide a platform to build or improve social capital. Therefore, including events and opportunities to increase engagement in education programs may enha nce the perception of social capital. 3. The data show a significant difference in perception of social capital for people who did and did not take steps to reduce ri sk of wildfires aro und their homes. This is not unexpected as the theories about social capital suggest that communities high in social capital are mo re likely to work together to solve problems. Also, the data show that most of the people who took risk reduction efforts and who

PAGE 163

148 engaged in education programs perceived greater social capital than those who did not take actions and partic ipate in education programs. 4. Another approach to sorting out the imp act of various aspects of a wildfire education program and perception of social capital is to ask the participants which elements they believe were most helpful at prompting them to take risk reduction action. In this study, data show that that there is high perception of social capital among those who credit taking risk reduction actions to talking to friends in the neighborhood and talking to a community leader for their action to reduce wildfire risk. The respondents also cred it risk reduction activities to attending events and being a part of FireWise co mmunities more than receiving information through electronic or print media. This s uggests that people perc eive that talking to neighbors, which involv es interaction and communication among residents is more successful at influencing them to ta ke actions to reduce wildfire risk. Also, events can influence interaction with one another, engagement in risk reduction activities and planning, discussion on ways to improve preparedness, and improve communication—all aspects that the liter ature suggests help improve social capital. To successfully involve resident s in taking steps to reduce risk of wildfires, it may be useful to consid er designing educational programs that encourage action by strengthening social capital. 5. The data show a significant relationshi p between participation in education programs and taking steps to reduce risk of wildfire. Thus, participation in education programs is related to risk re duction. This result reinforces that the goals of the education programs are succe ssful; they improve risk reduction and disaster preparedness. 6. The data also show a significant relati onship between level of engagement with educational materials and risk reduction act ivities. Interaction with information received through specific e ducational activities like television news, handout, newspaper, and attending a meeting is asso ciated with taking st eps to reduce risk of wildfires. So, people who interacted with the inform ation they received from electronic, print, or events took actions to reduce wildfire risk as compared to people who did not interact with the information. As we know, interaction is an important element to build social cap ital. Thus, building engagement and interaction with informati on and people in educational programs can help involve people in risk reduction and enhance social capital. 7. Data reveal that engagement in events makes a greater contribution to predicting perception of social capital than engageme nt in electronic or print media. Also, engagement in events is associated with higher risk reducti on activities. Program developers may want to improve engagement in activities and with others as this can help improve perception of social capital and disaster preparedness.

PAGE 164

149 Implications Implications for Environmental Educators This study has applications for environm ental educators in the formal and nonformal arena. The study shows the usefulne ss of program elements that provide for interaction among neighbors for the reduction of wildfire risk and for enhanced perception of social capital. Environmental educators working with youth and adults in various formal and non-formal areas could gain from these results. Environmental educators create opportunitie s for people to learn, gain skills, and take action to solve environmental problems. This is often done through engaging people in problem solving. Certain education me thodologies that involve interaction and communication may be more able to improve social capital. For example, many service learning programs are aimed at educating you th about local environmental issues and providing them with the attit ude and skills they need to undertake a project to help resolve an environmental problem (Billi ng, 2000). These programs provide youth with the opportunity to make decisions, help th eir community, and use their talent and knowledge to implement an action project. In a service learning program when youth interact with each other, they might help each other, do favors and develop trust in each other, all of which enhances social capital. When youth intera ct with community partners they learn about and get exposed to existi ng networks community functioning, which also enhances social capital. When people in the community learn about actions taken by children, such as rebuilding a city park or initiating a cleanup campaign, they might start taking pride in children’s action. A more positive attitude towards one’s community and an example of collective action may encourag e the development of social capital. Thus, service learning programs that result in community action have the potential to strengthen

PAGE 165

150 social capital. One could hypothesize, howev er, that such a program may be more successful in communities with existing soci al capital. But since service learning programs are successful with at-risk youth, dr opouts, and minority groups and since these populations are typically disenfranchised (B illing, 2000), service learning is probably responsible for enhanced social capital. A recent study showed significant correlations between the number of service learning program s attended by the students and increased social capital, measured by the decreased number of hours spent watching television (Howard, 2006). Implications for Resource Managers This study also has implications for natu ral resource managers. Results of the study indicate that specific educatio nal activities are associated with enhanced perception of social capital. Public affairs officers could take advantage of fire stories in newspapers and other appropriate channels to place other fire-related stories useful to mitigation and preparedness and to encourage residents to talk to neighbors and work together. Most importantly, resource managers could use dem onstrations, meetings, and get-togethers as ways to disseminate wildfire information to residents. Greater success would be achieved if these opportunities were designed to have people meet each other or work together. The study suggests that people who perceive greater social capital are more likely to take action around their homes to reduce ri sk of wildfires. If people who are more connected in their community take action around their homes to reduce their risk of wildfires, others will notice them. Resource managers can influence these people to volunteer their homes as demonstration sites for ways to do mitigation and create defensible space. Homeowners can provide tour s to educate their peer s and others in the community about risk reduction methods.

PAGE 166

151 Results of the study suggest th at people credit their risk reduction actions to talking to friends in the neighborhood or talking to a community lead er. If community leaders are more influential at motivating people to take action to reduce risk of wildfires, resource managers should focus attention on incorpora ting them in the outreach education. Also, they should design the programs to involve in teraction with friends, neighbors, and other people in the community. For example, a Neighborhood Work Day or a Neighbor Appreciation Day would bring together frie nds and others members of the neighborhood. At these functions, people who have taken act ions around their homes to reduce wildfire risk can talk to peers about things that they have done around their homes to reduce wildfire risk. They can talk about what motivated them and why others in the neighborhood should also get involved. Recommendations for Future Research Future research can be focused on specifi c elements of the educational activities that showed significant correla tions with perception of soci al capital. An experimental study could be done to isolate the engageme nt variables more precisely. Qualitative studies can be conducted with pe ople who participated in spec ific activities that showed higher perception of social capital to understand what aspe cts of the program they believed influenced interaction, perception of social capital, and risk reduction. This could help define specific elements of th e educational program, such as type of interaction, platforms for comm unication, learning activities, and hands-on activities that provided opportunities for people to in teract and network with others. It would be useful if future research were conducted to show a causal relationship between participation in non-fo rmal adult education program s and an increase in the perception of social capital. The data obtained from this study can be used as a baseline

PAGE 167

152 social capital marker in the study communities th at continue activity in wildfire education programs. However, such a relationship may be difficult to ascertain, as communities are dynamic; they are changing and so are re lationships within the community. These changes might be independent of participa tion in an education program. Additionally, when retired adults move to Florida, they probably bring skills in building networks and relationships with them and begin to put them into effect. This might also affect social capital. A causal relationship between participati on in education programs and social capital, however, may be achieved through future research focused on youth in formal and non-formal education systems. The cu rrent emphasis on stude nt achievement may necessitate that social capital be linked to test scores. The litera ture suggests that increased family social capital is associat ed with improved educational achievement at the school level (Coleman, 1988). It would be interesting to see if opportunities to interact with peers within the education sy stem affect perception of social capital, educational achievement, and performance at school levels. Litera ture suggests that participation in environment education or environment-based programs shows support for improved student academic achievements (Lieberman and Hoody, 1998; Randall, 2001; Ernst and Monroe, 2004). A study by Ernst and Monroe (2004) reports that students who participated in environment-based programs had greater critic al thinking skills than their peers. Students and teachers in the study suggested that interdisciplinary nature of the program, issuebased instruction, learner-cente red instruction, and construc tivist approaches in the environment-based programs may have influe nced student’s critical thinking skills and

PAGE 168

153 disposition towards critical thinking (Ern st and Monroe, 2004). The nature of an environment-based program involves interact ion among peers, teachers, stakeholders, and members of the community. It integrates multiple disciplines and empowers students to be responsible for their own learning. T hus, it could provide opportunities to improve or build social capital. Research focused on these programs may help establish a causal relationship between participat ion in environment education programs and social capital. A study by Lieberman and Hoody (1998) s uggests that students learn more effectively within an environment-based c ontext rather than traditional education framework. The study also suggests that using Environment as an integrating concept for learning (EIC) appears to signi ficantly improve student performance, which is measured through standardized tests. By definition EIC programs involve interdisciplinary, studentcentered projects that use collaborative, hands-on learning (Lieberman and Hoody, 1998). The literature suggests that social ca pital can be built through communication, participation, and interaction (Wilson, 1997). In addition to benefits, such as educational achievement, advanced thinking skills, appr eciation of the divers e viewpoints of the society, EIC approaches may also help enhance social capital. This study did not attempt to measure social capital but instead tried to get a handle on how people perceived social capital in their communities. The 3CM interviews found that people did not conceptualize social capital in the five categories that the literature suggests: trust and trustworthiness, reciprocity, networking, partic ipation in community activities, and sense of community. The study found evidence of these concepts within the components and categories for each inte rview. A category analysis suggested two broad categories: Neighborhood attitudes a nd Neighborhood features. Due to the small

PAGE 169

154 sample size of the 3CM intervie ws (15 participants), there can be no conclusion about the power of the 3CM analysis. So, the social cap ital scale was created to include items on these five categories. Factor analysis of the 1,350 respondents revealed that responses did not group into the five categories found in th e literature. People basically classified perception of social capital in two categories: neighbor hood engagement and affective attributes. Thus, this study suggests that pe ople’s organization of perception of social capital is different from what literature suggests; it is more nebulous. Further research is needed to better conceptualize and measure social capital. Conclusions In a community setting, some adult envi ronmental education programs are focused on changing particular behavi or (e.g., recycling, reducing wi ldfire risk). The success of such an environmental education program may be affected by community characteristics, such as the relationships between homeowne rs and agency staff delivering the program, trustworthiness, people’s sense of place, a nd participation in community activities. These characteristics are together refe rred to as social capital. In addition, an education program that i nvolves interaction among its participants may help improve communication and enhance or build relationships. Research indicates that the more people in a community interact, sh are, participate, and trust each other, the better is the community’s ability to solve common problems (Putnam, 2000). Social interaction and ne tworking not only benefit the individuals making the connections, but also the wider neighborhood in which they live. For example, at the neighborhood level people may interact with each other through civic, religious, social, and recreational groups. In each case, although the interaction among people is shortlived, and sometimes impersonal, the inte raction may provide its members with

PAGE 170

155 friendships and business contacts that can be useful in case of pot ential need to solve problems in the neighborhood. A well-netw orked and involved neighborhood where people do favors for each other might be more successful in adapting or implementing programs that improve quality of life and reduc e threat from natural calamities, such as wildfires, as compared to a neighborhood where individuals do not trust or interact with each other. Results of this study show that wildfire education programs are successful and they achieve their developmental goals. More people who participated in the wildfire education programs took action to reduce their risk of wildfire than those who did not participate in the educational programs. Data show that social capital and risk reduction behaviors are associated with certain aspects of wildfire e ducation programs. This study, however, did not ascertain which causes the ot her, and there are theoretical grounds for explaining the relationship in both directions. Although a direct intended out come of a disaster education program is reducing risk for the disaster, this study also contributes to the knowledge of how education can provide additional and important benefits. For in stance, the nature and level of interaction within the educational program may he lp improve neighborhood networks and connectedness. A neighborhood can increase its wildfire preparedness by using education as a tool. Certain types of education can bu ild social capital, which in turn lays the foundation for community support for prepare dness initiatives. Disa ster preparedness managers and neighborhood leaders who wish to improve community preparedness could focus on delivering messages through existi ng community groups (where social capital

PAGE 171

156 might already exist) and through formats that encourage community engagement (where social capital might be nurtured). Presence or lack of social capital also ha s implications in our daily lives. Social capital is the non-economic value of the networks and relationships that people hold with others around them. People who are better c onnected have an improved chance of coping with traumas. Loneliness and depression may sometimes stem from being unconnected in a society. Social capital ma y provide a support from various sources including friends, family, and groups in times of distress, failu res, or bad health. Social capital may help access resources not othe rwise available. As well as being an important tool for disaster preparedness (reducing the risk of wildfires in this study), social capital is al so important for quality of life and community connectedness. Education programs can be one way for communities to enhance social capital and reducing vulnerability. In doing s o, education may have greater value at the community level than just information dissemination.

PAGE 172

APPENDIX A DESCRIPTION OF WILDFIRE ED UCATION PROGRAM IN STUDY COMMUNITIES (The wildfire education program summaries for each community was reviewed by program developers in the respective communities) Wildfire Education Initiatives on Lake Wales Ridge This document summarizes the wildfire education efforts in Tiger Creek Forest, Indian Lake Estates, Placid Lakes, and Leis ure Lakes in Polk and Highlands counties in Florida. It includes a description of the co mmunities, their wildfi re history, and their wildfire education efforts—objectives, target audience, activities, collaborations, and outcomes—to reduce the risk of wildfires. Polk County Tiger Creek Forest Tiger Creek Forest (TCF) is a commun ity of about 100 people spread over 1000 acres. It is located 3 miles east from dow ntown Lake Wales on SR 60. The community was developed on cattle grazing land in the 1980 ’s; most homeowners then were retirees. In the past 10 to 12 years younger people have started moving in the community. The community is located in a natural wooded setting. Lot size varies from 5 to 30 acres. The area is largely w ooded and densely overgrown. Wildfire history The Nature Conservancy and the Florida Division of Forestry rate Tiger Creek Forest at high risk of wildfires. TCF had a 200-acre wildfire on the southern portion of the property in the 1980’s; no lives or homes were lost. The community also suffered another fire 8-10 years ago. The community has three entrances—one from Dude ranch Road on to 60; second, on rattlesnake Road on to 60, and third off Glenn St Mary Road. At most times the exit of 60 is used for commute. The risk of wildfires comes from the overgrown vegetation and the community’s natural dense wooded setting. This not only increases the risk of the fire spreading quickly but in some places restricts the fire fi ghters’ ability to protect homes. Access is an important issue as many driveways and roads ar e very narrow. To address this issue, the volunteer fire department surveyed TCF and th en sent out letters to people whose houses had poor access. 157

PAGE 173

158 Goals To develop an ongoing fire awareness progr am to encourage residents to adopt FireWise practices to reduce the risk of wildfires, and increase the chance of their home surviving in a wildfire To increase awareness of the risk of wildfires To improve forest health in TCF Objectives To involve the community in taking step s to reduce risk of wildfires—creating defensible space To improve access to homes—widen driveways and roads To create a demonstration home based on FireWise principles To build community support for prescribed burning Target audience Homeowners and property owners of TCF Agencies involved The Nature Conservancy (TNC) Florida Division of Forestry (DOF) Volunteer fire department TCF Homeowners Association Education/communication activities 1. Survey of homes: Recognizing the increased risk of wildfires due to overgrown vegetation and dense woods, fire fighters from the volunteer fire department for Tiger Creek Forest surveyed access routes to all homes in TCF and sent letters to people whose homes presented a high risk. 2. Community picnic: To reach community members and educate them about the risk of wildfires, TNC and DOF participated in the annual community picnic promoted by the Tiger Creek Forest Ho meowner Association. About 50 people (30 families) attended the picnic in Ma y 2004. TNC brought displays at the picnic to educate homeowners about wildfires, defensible space, and native plants. 3. Community meetings: Residents of TCF have i nvited officials from DOF to speak to homeowners about FireWise prin ciples, explain their risk of fire, and describe the steps they can take to reduce the risk of wildfires. 4. Demonstration home: TNC has written a grant proposal to create a demonstration home in TCF. Around one of the trailers TNC wants to do prescribed burning, put a driveway, and retr ofit it using FireWi se principles, and use it as a demonstration home to advise people on ways they can make their homes fire safe.

PAGE 174

159 5. Fuel reduction: TNC owns lots in TCF and the adjacent Tiger Creek Preserve. They conduct prescribed burning on their land on a regular basis. Outcomes 50 people attended the picnic and gather ed information on how to make their homes FireWise Some people have started doing de fensible space around their homes TNC has applied for a grant to create the demonstration home based on FireWise principles Indian Lake Estates Indian Lake Estates (ILE) is a commun ity of over 1000 people of which 35% are seasonal residents. The community covers 8000 acres; of the 7458 plat ted residential lots, 500 have single family residences on them . Most lots are acre lots with some exceptions. The community has a lot of vacant lots. Most people who moved in initially were retirees, now many younge r people are also moving in. ILE has a newsletter that goes out every 6-7 weeks. There are two groups in ILE—the older group typically gets involved in the community, volunteer and uses the community as a social place to in teract with each other. The younger residents are less involve d in the community and do not volunteer. Wildfire history Indian Lake Estates had a large fire in 1999. No homes were lost , but 10 homes were damaged. Approximately only 15% of the commun ity is built; the vacant lots in the community contribute to risk of wildfires. To address the fire issue, the Division of Forestry and board members of Indian Lake Estates are increasing awareness about ri sk of fires and edu cating residents about steps they can take to reduce their vulnerability to wildfire s. Details of the education/ communication activities are listed. Earlier ILE had a volunteer fire department with three volunteer firemen, all in their 80’s. Some years ago ILE gave land to the county fire department and now it has a station. Also, the new fire department is a paid fire department. People in ILE share a special relationship with the fire department. Goals To increase awareness of the risk of wildfires To involve the community in taking step s to reduce risk of wildfires—creating defensible space To manage vegetation in ILE Objectives To increase support for prescribed burning To involve people to create defensible space to improve th e chances of their homes surviving a wildfire

PAGE 175

160 Target audience Residents of Indian Lake Estates Agencies involved Florida Division of Forestry County Fire department Indian Lake Estates homeowner association Education/communication activities 1. Community meetings: Members of the Polk County Fire Department spoke to homeowners about prescribed burns at the community’s general meetings, Snowbird meetings (welcome new resi dents in the community) and the Yacht club meetings. In the past 2-3 years, DOF has attended 3-4 general meetings to talk to people about risk of wildfires. Approximately 85-100 people attend general meetings, 100 people attend the Yacht club meetings and 120 people come to the Snowbird. The ILE board also talks about fire at the general meetings. For example, they inform people that they need a permit to burn in their backyard and that the fire department i ssues fines to anyone who does not comply with the burn permit. 2. Fuel reduction: The County Fire Department and the DOF conduct prescribed burns in ILE to manage vegetation. One way to advertise prescribed burns is posting a notice at the post office. In ILE, people do not get mail at their homes; everyone goes to the post office to collect mail. The fire department has also asked people in ILE to have at least 100 feet clearing around their homes; this will facilitate vehicle access in case of a wildfire. Outcome More homeowners are aware of the risk of wildfires More support for prescribed burn in the area Increased interaction between residents and fire department and forestry officials Highlands County Placid Lakes Placid Lakes (PL) is a community of 2500 people covering 3,333 acres. Of the 8250 lots, only 3500 have homes on them. The community of Placid Lakes is incorporated. Many people who moved into Pl acid Lakes had careers in the armed forces. Many people in Placid Lakes vol unteer for activities with th e homeowners association. Wildfire history In 1999, PL had a wildfire in the wooded areas of the community. This helped residents and the PL homeowner association bo ard become more aware of the threat of wildfires. In fact this fire took place near the house of a board member who lives on a street that is heavily wooded. After the fire, he took the initiative to talk with the DOF

PAGE 176

161 and learn about the FireWise program. As a result five people from PL attended the FireWise workshop in Brooksville, FL. Since then, the announcement about the FireWise program has been made at the general me mbership meeting for Placid Lakes. Vacant lots pose major threat for wildfires, especially when there is drought. There is also an added threat from teenagers as they are not careful about cigarette buds. Target audience Residents of Placid Lakes Agencies involved The Nature Conservancy Florida Division of Forestry Placid Lakes Homeowner Association Fire department Goals To reduce risk of wildfires To involve the community in taking step s to reduce risk of wildfires—creating defensible space Objectives To get involved in the FireWise program with the end result of an community ordinance to comply with FireWise recommendations Education/communication activity 1. Attending FireWise workshop: After the wildfires in 1999 some members of the homeowners association became involved in taking steps to reduce the risk of wildfires. They contacted the DOF for a dvice on ways to reduce the risk. Five people attended a FireWise workshop in Brooksville, FL. They reported on this workshop at the annual general meeting. 2. FireWise Day: Placid Lakes had their fi rst FireWise Day in October 2004. At the event, Gerry LaCavera (DOF) made a two-hour presentation on wildfires and FireWise to the community. He also dist ributed literature on FireWise and ways to reduce risk of wild fire at the event. Outcomes Five members of Placid Lakes attended th e FireWise workshop in Brooksville, FL First FireWise Day in October 2004 Leisure Lakes Leisure Lakes (LL) is more rural than compared to Placid Lakes. The community has an active Homeowners asso ciation. However, they have made no efforts to educate the community about wildfire protection. Of the approximate 8000lots, only 25% are built.

PAGE 177

162 Wildfire history Leisure Lakes had a fire in 2004 that an ar sonist started. 300 acres were burned in the fire; no homes were lost or damaged. The area is not completely populated and this poses a major threat for wildfires. However, the community has not taken any steps to create awareness or reduce risk of wildfires. The only fire activity comes from the fire department that conducts cont rol burns around the area. Wildfire Education Initiatives in Palm Coast Palm Coast was designed as a model co mmunity in the early 1970’s. 42,000 acres in the community were platted and sold in acre lots. Many of the early inhabitants were retirees from New York, New Jersey, and Michigan, anxious to enjoy their later years with sunshine, ocean breeze, and low-maintenance lots. Most of them had no idea they were moving into an area that wa s likely to experience wildfire. An early court agreement prevented the de velopers from building Palm Coast in phases. Instead, the entire road system was paved, providing water and sewer services throughout the developing region. Lots were so ld with no restriction on when building would occur. Even today some homes are isolated on wooded stre ets, the only house on the block. These vacant lots, owned by absent ee owners around the world, contain young pine trees, saw palmettos, wax myrtle, and vines. Where the vegetation has not been managed, it is a wildfire hazard to neighbors, despite its useful privacy screen and wildlife habitat. Wildfire history Palm Coast has experienced two major wildfires in the past. In 1985, 130 homes were lost when two wildfires swept th rough the community. In 1998 two different wildfires angled across Palm Coast, burning over 70 homes in their wake. The second of these fires was part of complex of several fires that became a major disaster. Flagler county and nearby communities (100,000 peopl e) were evacuated over the July 4 th weekend. Fortunately, no lives were lost in either fire. After the 1998 fires, several actions were initiated at the state, county and city level to reduce the risk of wildfires. These include: increased equipment to fight fires, improved wildland fire suppression and incide nt command training, mitigation to reduce hazardous vegetation, and education/communicati on activities to educate residents about actions they could take to reduce th eir vulnerability to wildfires. This document has two sections. The first se ction provides a summary of the wildfire education program in Palm Coast— its objectives, target audience, activities, and outcomes. The second section gives an overview of the education activ ities specific to the two Palm Coast communities—Cypress Kno ll and Seminole Woods, who are attempting to be recognized as FireWise communities.

PAGE 178

163 Section 1: Wildfire Education Programs in Palm Coast Goals Increased awareness about risk of wildfires Increased individual and community action to reduce vulnerability to wildfires Reduced fuel load through implementati on of the vegetation reduction ordinance Objectives To make homeowners aware of the risk of wildfires To reduce the vulnerability of homes during a wildfire To reduce fuel load in and around the properties To initiate action from homeowners a nd landowners to reduce vulnerability to wildfires Target audience Residents of Palm Coast Landowners in Palm Coast Participating agencies Florida Division of Forestry City of Palm Coast Fire Rescue City of Palm Coast Flagler County Extension Service Flagler County Emergency Services Education/communication activities It has two components: Aware ness and action component. Awareness component: Providing information to resident s that fire protection is the responsibility of many different agencies and offices in the ar ea. Various strategies have been used to make residents of Palm Coast more aware of the fire danger in the area. 1. Public information system: A county website provides information about weather, current fires, and prevention tips. A 24-hour public information hotline is updated daily. An Eagle Scout erected signs “Fire Dange r Today Is” at major intersections. A red flag flies at DOF, county, and city offices on extreme fire hazard days. The Flagler County Emergenc y Service has set up a Reverse 911 system to inform people of any given area about approaching wildfires or about a nearbyprescribed burn. The Flagler County Extension office pr ovides information and programs about designing a landscape with plants that reduce th e need for water and the risk of fire. The City of Palm Coast has information on it’s website with links to other sources of information to inform residents of FireWise Practices. FireWise Community update information if e-mailed to the Cypress Knoll residents who sign up to receive this information.

PAGE 179

164 A FireWise Communities Workshop was he ld for Cypress Knoll and Seminole Woods Residents at the county ex tension office this past spring. 2. Communication: Two-way radios were put in all sc hool buses as a re sult of the 1998 fires. Weather alert radios are set up in all schools and government buildings. 3. Guest speakers: The Division of Forestry (DOF) st aff speaks at club meetings to raise awareness about wildland fire and risk reduction practices on a regular basis. Representatives from the Flagler County Extension Service an d the City Fire department are also invited to speak at civi c club meetings. The City of Palm Coast’s landscape architect has made presentations to the City Council and Homebuilders Association on FireWise. 4. DOF initiatives: The DOF has assigned a Wildfire Mi tigation Specialist to the region who is devoted to increasing public awareness and community response by mobilizing neighborhoods, conducting ho meowner assessments, and rallying residents to take precautionary actions. DOF has also set up a Fire Prevention Strike Team that speaks to residents in targeted areas by knocking on doors. Action Component: This includes: 1. Vegetation Reduction Ordinance: In December 1998, Flagler County passed a vegetation reduction ordinance for Palm Coast. When Palm Coast became an incorporated city in December 1999, the task of reducing vegetati on (contributes to wildfire hazard) fell to the new city depa rtments. A slightly di fferent ordinance was approved in 2001 that directs the city code enforcement office to send letters to identified hazard lot owners informing them of their options for mitigation on their property. The options are—one, lot owners can reduce underbrush themselves, or two, lot owners can ask the city to perform the serv ice for them and pay a cost that usually averages between $200 and $400 plus a $100 ad ministration fee. According to the ordinance, if the lot owner does neither, th e city will clear the underbrush and charge the lot owner the cost. If no payment is made, the city will place a lien on the property that must be paid before the lot can be sold or built upon. The first round of letters were sent out Spring 2002 and received a fairly positive response. Some landowners were pleased th at the answer to reducing the risk of wildfire was so easy. Others were hesitant to invest in a lot they plan to sell, but eventually realize they had no other choi ce. From May to July 2003, 315 notices of violation were sent out to hazard lo t owners and 137 lots were mowed. Section 2: FireWise Program Goals Increased awareness about risk of wildfire s in areas not recently exposed to fire: Cypress Knoll and Seminole Woods Increased individual and community action to reduce vulnerability to wildfires

PAGE 180

165 Reduced fuel load through implem entation of the Vegetation Reduction Ordinance Objectives To make selected homeowners aware of the risk of wildfires To reduce the vulnerability of homes during a wildfire To initiate action from homeowners to take steps to reduce their vulnerability to wildfires To encourage homeowners to take responsibility for their property and spread the word to their neighbors as well Target audience Residents of Cypress Knoll and Seminole Woods Participating agencies Florida Division of Forestry City of Palm Coast Fire Rescue City of Palm Coast Flagler County Extension Service Flagler County Emergency Services Cypress Knoll Homeowners Association Seminole Woods Homeowners Association Education/communication activities 1. Cypress Knoll FireWise Community: Cypress Knoll is the first pilot FireWise community in Palm Coast. Cypress Kno ll has a Neighborhood Watch program; which helped launch the FireWise program. A dem onstration project to retrofit and mitigate one home was conducted to show people wh at the ordinance me ant and how much vegetation could be left around a home. The or ganizers took before and after pictures of the demonstration and organized news c overage of the demons tration project. The pilot community is in the process of ge tting FireWise Community certification. Cypress Knoll has not experienced a wildfire in recent years. 2. Seminole Woods FireWise program: Seminole Woods will be the second community to attempt to be certified as a FireWise community in Palm Coast. In 1998 fires 20 homes were destroyed he re, and 17 others were damaged. Outcomes Passed and implemented Vegetation Reduction Ordinance in Palm Coast Positive response from lot owners Cypress Knoll becomes the pilot Fi reWise community of Palm Coast Retrofitted a home to demonstrate FireWise principles Seminole Woods is planned to be the s econd FireWise community of Palm Coast

PAGE 181

166 Wildfire Education Initiatives in Wedgefield Wedgefield, FL is a large subdivision of 2700 people (2000 census) and spread over 6500 acres. It is located 23 miles east of downtown Orlando, well outside the city limits. Although development is quickly moving toward Wedgefield, the community is currently surrounded by undeveloped land, including a large private ranch and a county park. Wildfire history This area has a history of periodic wildfires, as is typical in central Florida. When development began in Wedgefield, however, fire was suppressed. This led to an accumulation of heavy fuel loads and an in creased risk of wildfire. In 1998 a large wildfire approached Wedgefield from the northeast and many residents were evacuated. If not for the change in wind direction at the last moment, the community might have lost homes. This fire was a wake-up call for people in Wedgefield, alerting them to the risk of wildfire and setting the st age for community actions. After the 1998 fires, Florida state and count y agencies initiated steps to reduce the risk of wildfires in several communities. In April 2001, a Wildfire Risk Assessment Team composed of Division of Forestry (DOF), Orange County Fire Rescue (OCFR), Division of Emergency Management (DEM), and local homeowners carried out a wildfire hazard assessment in Wedgefield. The assessment placed Wedgefield in the high hazard category. Based on this, Wedgefield was invited to become a pilot community in the National FireWise program. A FireWise Mitigation Board was formed; it comprised of homeowners and at least one member from the DOF, DEM, OCFR, Orange County Planning Division, Orange County Cooperativ e Extension, Orange County Emergency Management, Ranger Drainage District (RDD), and the St. Johns River Water Management District. Although people in Wedgefield live in f our different types of residential development, the FireWise program is targeted at the people living in the City (consisting of acre plots) and the Estate s (consisting of 1 to 5 acre lots). While wildfire is not likely to break out in the City, with its trim grass lawns, the City does back up to a public natural area. The risk of fire is much highe r in the Estates area because many homes are built close to flammable saw palmettos and pine trees, or next to unmanaged unbuilt lots. This document summarizes the Wedgefiel d wildfire education program evolved through the FireWise activities and include s objectives, target audience, activities, collaborations, and outcomes. Many regard Wedgefield as the most active and the most successful FireWise community in the nation. The FireWise Boar d meets regularly and has received several awards and recognition for their efforts to make Wedgefield better prepared for wildfire. Goals Increased awareness about risk of wildfires Increased individual and community action for reducing risk of wildfires More support and resident partic ipation in the FireWise program

PAGE 182

167 Objectives Make homeowners aware of the existence and activities of the FireWise committee Educate new homeowners about risk a nd history of wildfires in the area Educate residents on ways to reduce risk of wildfire Initiate homeowner support a nd commitment for the program Target audience Wedgefield residents New homeowners in Wedgefield Participating agencies (FireWise board) Every agency involved with FireWise in Wedg efield is also a member of the FireWise board. St. Johns River Water Management District Florida Division of Forestry Orange County Fire Rescue Ranger Drainage District Orange County Planning and Zoning UF-IFAS Orange County Extension Service Deseret Cattle and Citrus Wedgefield Homeowners Association Wedgefield Garden Club Homeowners and landowners in Wedgefield Education/communication activities The FireWise committee has used severa l strategies to reach out to the community, make residents aware of risk of fire , and initiate actions that will help reduce their vulnerability to wildfire s. These strategies include: Events 1. FireWise meetings: These meetings have two main objectives: to make homeowners aware of the existence and ac tivities of the FireWise committee, and to discuss development and implementa tion of the FireWise program. Other activities in the meeting include: pla nning for forthcoming events in the community, taking stock and assigning re sponsibilities to members of the committee, announcing awards and grants, discussing future collaborations, and applying for grants. 2. New homeowner packet: The FireWise committee has developed informational packets to educate new homeowners about th e history and risk of wildfires in the area. These packets also provide the newc omers with information about ways to reduce their risk of wildfires, and a list of people to contact for assistance.

PAGE 183

168 3. Piggybacking on community events: In view of the low homeowner participation in FireWise meetings, th e FireWise committee decided to piggyback on community events to reach the residents of Wedgefield. Members of the committee participate in community gatherings like yard sales, golf tournaments and picnics that are attended by many residents of Wedgefield. The committee sets up an informational booth, hands out brochures, displays fire fighting equipment, and members are available to talk with residents on actions to reduce fire risk, FireWise landscaping, and FireWise plants. 4. FireWise articles: The Homeowner’s Associati on in Wedgefield is active throughout the entire community, and FireWi se committee contributes articles to their monthly newsletters. Topics covered in the newsletter are targeted to make residents aware of upcoming FireWise events and committee meetings, provide information on the progress of the FireWise committee, talk about awards (state and national) received by the FireWise committee, and provide information on FireWise landscaping and ways to reduce risk of fire. 5. Homeowner Association meetings: Members of the FireWise committee are invited to attend the HOA meetings a nd speak to homeowners about FireWise events, introduce them about ri sk of fire, and steps that they can take to reduce the risk. 6. Fire Awareness Day: In 2002 the FireWise committee hosted a Fire Awareness Day at a Lowes, where they displayed a cabin landscaped with FireWise plants and handed out FireWise lite rature. This event is now hosted annually for the past two years. Last year (2003) it wa s hosted at Home Depot, another home improvement store. 7. Demonstration days: The general manager of the Ranger Drainage District has held several Demonstration Days, wher e equipment vendors come to Wedgefield with giant mulching machines to demons trate the usefulness and efficiency of their machines in clearing the underbrush. 8. Codes and Covenants: The chair of the FireWise committee is a member of many clubs and committees in Wedgefield and provides a useful link between FireWise and the community. She has been instrumental in bringing FireWise concepts to the Codes and Covenants Co mmittee and successful in changing the codes to meet requirements for mini mum gate width, etc. New codes in Wedgefield to meet FireWise guidelines include: The access gate should be minimum 15 feet wide (for easier access to emergency vehicles) and have a mini mum vertical clearance of 14 feet. This code has been rewritten and was signed in 2002. Collaborations 1. Collaborating with Garden Club: In February 2003 the FireWise committee collaborated with Wedgefield’s popular Ga rden Club to promote FireWise approved

PAGE 184

169 landscaping plants. During the same time, a representative of the FireWise committee (Extension Specialist) was invited at the Garden Club meeting to talk to people about FireWise plants and landscaping. Outcomes Increased interaction among agencies and residents More homeowners aware of the risk of wildfires More residents are participating in Fi reWise events, and many homeowners are employing FireWise principles on their property Wedgefield is a designated “FireWise Community USA” Wedgefield community and the FireWise Board has received several awards o Two at the national level from the National FireWise Communities/USA o One from Orange County for thei r tireless efforts to reduce the community’s risk to wildfires o One from FLASH (Florida Alliance for Safe Homes), recognizing their commitment to wildfire education o Two members of the FireWise board received plaques from the DOF in recognition of their efforts o Board has received donations from supportive contractors and builders in Wedgefield

PAGE 185

APPENDIX B SOCIAL CAPITAL AND WILDFIRE ED UCATION PROGRAM RELATIONSHIP DIAGRAM Electronic Print Events Use information to reduce wildfire risk Become aware & knowledgeable about wildfire & risk reduction actions Understand civic responsibilities Enhance partnerships Increase volunteerism Improve communication Increase networks Renewed community interest Take responsibility Move out of isolation Gender Education Residency status Length of residence Lot characteristic Experience of wildfire Perception of wildfire risk Work together to reduce risk Create a platform for people to talk about a concern Minimize negative impact of wildfire Trustworthiness Employment status Ethnicity Income Age Sense of community Confounding Variables Networks Reciprocity Partici p ation Nature and level of interaction in the education program Social capital 38 36 37 39 3 2 9 7 6 33 35 26 27-29 Linking hypothesis to survey questions: Q1-13: Information on confounding variables Q 14-26: Exploring sources of information leading up to building social capital Q27-29: Risk reduction activities Q 30-32: Measuring social capi tal based on what builds it Q 33-39: Information on confounding variables 30 31 32 20 21 22 16 17 18 14 15 Wildfire educational program Sources of information/ participation 170

PAGE 186

APPENDIX C 3CM INTERVIEW PROTOCOL AND INTERVIEW GUIDE Script for Setting up an A ppointment for 3CM Interview Good morning/afternoon. My name is Shruti Agrawal and I'm calling from the University of Florida. I understand you ar e a resident of _______ (community name). I got your name from ----. I would like to inv ite you to participate in a research study funded by the USDA Forest Service. This study is exploring the relationship be tween wildfire educ ation programs and social capital in communities at risk of wildfires. To be tter understand the communities, I am conducting interviews with a few selected participants. I would like to set up a time to interview you. Participating in the study means that one or two people will meet you at your home to discuss your perceptions of social interaction in you community. During the interview, the interviewers w ill want you to respond to few questions and respond to two or more prompts. As you answer, the interv iewer will write your responses on sticky notes and place them in front of you. After that he/she will ask you to cluster your responses into categories, and th en name each category. Interviews generally range from 30 to 60 minutes. We will explain the process of interview to you before and during the process. With your permission, we would like to tape record the conversation. That will allow us to faithfully interpret wh at you are saying. Only the research team will have access to those tapes, and when we ha ve concluded our study, they will be erased. Your name will never be associated with your responses—you will remain anonymous. Your identity will be kept anonymous to the extent provided by law. You do not have to answer any question you do not ch oose to answer, even though we hope that you will. In fact, you can stop the interview at any time. There is no benefit or risk to you if you participate in this study. We cannot offer any compensation to you, but please know that we gratefully appreciate your tim e and your thoughts. The results will lead to a PhD dissertation thesis and recomme ndations for improving our communities’ preparations for reducing the risk of wildland fires. If you have any questions about this rese arch project, feel free to call Shruti Agrawal at 352-846-2374. Shruti is a graduate student at the University of Florida carrying out this research for her PhD disse rtation. If you have any questions about participants' rights, please call the UFI RB at 352-392-0433. We will give you a card with these numbers when we visit with you. Do you voluntarily give your consent to participate in this study? Participant’s Name(s): YES: [ ] No: [ ] Date for Interview: Time: Address: 171

PAGE 187

172 3 CM Interview Guide Initial Questions Number of years he/she ha s lived in the community Number of family members who live in the community Where did you come from (move here from) How many organizations a nd clubs do you belong to Prompt 1. You are a resident of community -----. A person is trying to make a decision about moving into your community. He/she is in terested in learni ng more about your community, particularly about how peopl e get along. What w ould you tell them? What things or factors would you mention in describing whether there is a feeling of neighborhood in your community? What things would you talk about in telling/describing how people relate to one another? What things would you tell that peop le in the community do together? What things would you tell that people in the community do for one another? Does your community have community-wide events? What role do you play in those events? o Do you organize those? o Do you participate in those? 2. He/she is interested in l earning about social structure of the community. If you were to tell someone about things that make th is community work, what would that be? What makes people in this community get along? What shapes the relationship people in the community hold with each other? How efficient are people in this co mmunity in getting their work done? Please categorize these things and label each category. Indicate why you clustered the items in that fashion. Additional Questions Also, we would like you to: List five things that you lik e the most about your community. List five things that you like about people and their relationships in your community. List five things that you would lik e to change in your community. Why?

PAGE 188

173 Script for 3CM Process The 3CM task was explained to the part icipants as follows (Kearney, 1997): a. First, you will be given a scenario and asked to respond to it by generating concepts that you perceive as being impor tant (related to the question). I will record those components on sticky notes and place it in front of you. Once you are finished answering all the questions, you will have all the sticky notes in front of you. You are always free to add/remove any concept that you have said. b. Then, you will be asked to categorize the sticky notes in ways that depicts perceived relationship among the components. c. Finally, you will be asked to label the categories and indicate why you clustered those concepts together. Participant Consent Form Greetings! My name is Shruti Agrawal and I'm a graduate student at the University of Florida. Thank you for willi ng to participate in this research study. As mentioned earlier, this study is expl oring the relationship between wildfire education programs and social capital in comm unities at risk of wildfires. To better understand the communities, I am conducti ng interviews with a few selected participants. During the interview, we will want you to respond to few questions and respond to two or more prompts. As you answer, the interviewer will write your responses on sticky notes and place them in front of you. Af ter that the interviewer will ask you to cluster your responses into categories, and th en name each category. Interviews generally range from 30 to 60 minutes. We will explain the process of interview to you before and during the process. With your permission, we would like to tape record the conversation. That will allow us to faithfully interpret wh at you are saying. Only the research team will have access to those tapes, and when we ha ve concluded our study, they will be erased. Your name will never be associated with your responses—you will remain anonymous. Your identity will be kept anonymous to the extent provided by law. You do not have to answer any questi on you do not choose to answer, even though we hope that you will. In fact, you can stop the interview at any time. There is no benefit or risk to you if you participate in this study. We cannot offer any compensation to you, but please know that we gratefully appreciate your time and your thoughts. The results will lead to a PhD dissertation th esis and recommendations for improving our communities’ preparations for reducing the risk of wildland fires. Do you voluntarily give your consent to participate in this study? Participant’s Name(s): Date for Interview: Time:

PAGE 189

174 Card To Leave With Participants Thank you for agreeing to be a part of th e study “Exploring the relationship between wildfire education programs and social capital in communities at risk of wildfires.” Your responses are very important to us. No information about you (your name or your address) will be mentioned in any of our reports; your responses will be reflected in a general way to develop questions for the community survey. If you have any questions about this intervie w or the study, please contact Shruti Agrawal (Phone: 352-846-2374, Email: sagrawal@ufl.edu ) or Martha Monroe (Phone: 352-8460878, Email: mcmonroe@ufl.edu ) at the School of Forest Re sources and Conservation at the University of Florida. You can also discuss research participants’ rights with the UFIRB Office, University of Florida, PO Box 112250, Gainesville FL, 32611-2250, (352) 392-0433.

PAGE 190

APPENDIX D SURVEY PROTOCOL AND MAIL SURVEY Mail Survey Exploring Wildfire Education In itiatives and Your Community This survey is designed to learn more about your participation in the wildfire educational activities and your perception of your community. We will use the information to help other communities prepare for and recover from disasters like wildfires and hurricanes. We thank you for your participation in the research study. Please return the survey in the enclosed envelope in the next few days. School of Forest Resources and Conservation, University of Florida 175

PAGE 191

176 GETTING TO KNOW YOU In this section we would like to learn more about your housing situation. 1. Do you or your family own the home at this address or do you rent? (Check one) Own Rent 2. How would you best explain your housing situation in this community? (Check one) Full time resident Seasonal resident Weekend resident Other, please specify_____________ 3. How many year(s) have you lived in this community? _____________ Year(s) 4. Prior to living here, where did you live? (Check one) In the U.S. Which state? _____________ Outside of the U.S. 5. What is the approximate size of your property at this address? _____________ Acres YOUR PEREPTION OF WILDFIRE RISK In this section we would like to understand your perception about the risk of wildfire in your community. 6. In all the places you have lived, have you ever: (For each, check one) No Yes a. Experienced house fire b. Been in or close to a wildfire c. Been evacuated because of wildfire d. Been in or close to a prescribed burn A prescribed burn is fire set by land managers in a controlled fashion under selected weather conditions to control vegetation and acco mplish other land management objectives. 7. Do you think you and your home are at risk of a wildfire? Yes No 1

PAGE 192

177 8. What best describes the type and density of vegetation on the property adjacent to yours ? (Check all that apply) Cannot see through Can see through a. Mostly trees b. Mostly shrubs c. Mostly grass 9. What best describes the vegetation on your lot ? (Check all that apply) Dense vegetation (you can not see through it) Managed vegetation—trimmed trees and shrubs, reduced undergrowth Mowed lawn 10. How close is the dense vegetation to your home ? (Check one) There is none Less than 5 feet 6-30 feet 31-100 feet More than 100 feet 11. How concerned are you about a wildfire threatening your home and property ? (Check one) Not at all concerned Slightly concerned Concerned Extremely concerned 12. In your opinion, how likely is it that a wildfire in the next five years will: (For each, check the best response) Very unlikely Unlikely Likely Very likely Don’t know a. Burn a natural area near your house? b. Damage your property? c. Change your quality of life? 13. How aware are you about the things that can be done to reduce your risk of wildfire? (Check one) Not at all aware Somewhat aware Aware Extremely aware 2

PAGE 193

178 YOUR PARTICIPATION IN WILD FIRE EDUCATION INITIATIVES In this section we would like to explore your exposure to and level of participation in educational activities related to wildfire preparedness. There are several avenues to get information on wildfire, for example, attending a meeting, picking up a brochure at an event, or talking to a friend. When answering the next set of questions, please consider the information sources separately to the best of your ability. 14. Listed below are some electronic media sources that may provide information on various aspects of wildfires. Answer the following questions for each source that you are aware of for information to reduce wildfire risk and create defensible space (not for updates on nearby wildfires). If YES, answer the following questions A. Did you learn from these sources ? (Check one) B. To what extent did you interact with anyone with that information? (Circle one) C . To what extent have you used the information to reduce wildfire risk? (Circle one) Have you seen or heard these electronic sources for information on ways to reduce risk of wildfires and create defensible space? No Yes No Yes A lot Some A little Not at all A lot Some A little a. Websites 1 2 3 4 1 2 3 4 b. Television advertisement 1 2 3 4 1 2 3 4 c. Television news 1 2 3 4 1 2 3 4 d. Television shows 1 2 3 4 1 2 3 4 e. Radio advertisement 1 2 3 4 1 2 3 4 f. Radio news 1 2 3 4 1 2 3 4 g. CD/video 1 2 3 4 1 2 3 4 15. From your experiences with electronic media sources above, list up to three that were instrumental in promoting neighborhood interaction . Electronic media sources: What is the approximate number of times you heard/saw/visited it? (Circle one) a. ___________________ 0 1 2 3 4 5 6 7 >7 b. ___________________ 0 1 2 3 4 5 6 7 >7 c. ___________________ 0 1 2 3 4 5 6 7 >7 3

PAGE 194

179 16. Have you ever received information in print on ways to reduce risk of wildfires? Yes continue with Q 17, Q 18 and Q 19 No skip to Q 20 17. For all information that you have received in print on ways to reduce risk of wildfires and create defensible space; answer the following questions. If you received PRINTED MATERIAL, answer the following questions What printed material did you receive with information on reducing wildfire risk and creating defensible space and how did it arrive? A. Did you learn from these materials? (Check one) B. To what extent did you interact with anyone with that information? (Circle one) C. To what extent did you use the information to reduce wildfire risk? (Circle one) Mail In person Did not receive No Yes A lot Some A little Not at all A lot Some A little Not at all a. Handout 1 2 3 4 1 2 3 4 b. Brochure 1 2 3 4 1 2 3 4 c. Door hanger 1 2 3 4 1 2 3 4 e. Newsletter 1 2 3 4 1 2 3 4 g. Magazine 1 2 3 4 1 2 3 4 h. Newspaper 1 2 3 4 1 2 3 4 18. From your experience with printed material above, list up to three that were instrumental in promoting neighborhood interaction . Print sources: What is the approximate number of times you received the information? a. ___________________ 0 1 2 3 4 5 6 7 >7 b. ___________________ 0 1 2 3 4 5 6 7 >7 c. ___________________ 0 1 2 3 4 5 6 7 >7 19. You may have seen other printed it ems that remind you about wildfire risk and landscaping changes, such as magnets, billboards, and signs. Which ones have you seen that were effective at promoting community activity? _____________________________________________________________________ _____________________________________________________________________ 4

PAGE 195

180 20. Have you ever attended any event related to reducing risk of wildfires and creating defensible space? Yes continue with Q 21 and 22 No skip to Q 23 21. For each event that you attended, answer the following questions. If you ATTENDED THE EVENT, answer the following questions With whom did you attend the events related to reducing risk of wildfire? A. Did you learn from the information ? (Check one) B. To what extent did you interact with anyone with that information? (Circle one) C. To what extent did you use the information to reduce wildfire risk? (Circle one) With someone Alone Did not Attend No s Yes A lot Some A little Not at all A lot Some A little Not at all a. Presentation 1 2 3 4 1 2 3 4 b. Meeting to discuss wildfire 1 2 3 4 1 2 3 4 c. Prescribed fire demonstration 1 2 3 4 1 2 3 4 d. Equipment demonstration 1 2 3 4 1 2 3 4 e. Landscape demonstration 1 2 3 4 1 2 3 4 f. Information booth 1 2 3 4 1 2 3 4 g. Community picnic 1 2 3 4 1 2 3 4 22. From your experience with the events above, list up to three that were instrumental in promoting neighborhood interaction . Events: What is the approximate number of times you attended the event? a. ___________________ 0 1 2 3 4 5 6 7 >7 b. ___________________ 0 1 2 3 4 5 6 7 >7 c. ___________________ 0 1 2 3 4 5 6 7 >7 5

PAGE 196

181 23. Has anyone ever talked to you about getting involved in activities to reduce risk of wildfires? Yes continue with Q 24 No skip to Q 25 24. If yes, what was the effect of the conversation? (For each, check one) No Yes a. I was encouraged to take actions on my property b. I encouraged my neighbors to take actions on their property c. My neighbors and I helped each other take actions d. My neighbors and I decided not to take actions e. Other, please specify___________ 25. How active are you in community efforts to reduce your community’s vulnerability to wildfires? (Check one) Not at all active Slightly active Moderately active Very active 26. Do you think your community’s work in fire preparedness has helped you: (For each, circle one) Not at all Small extent Some extent Large extent a. Feel good about where you live 1 2 3 4 b. Talk about issues in your community 1 2 3 4 c. Get together to solve community problems 1 2 3 4 d. Volunteer in the community 1 2 3 4 e. Reach out to others 1 2 3 4 f. Improve communication with neighbors 1 2 3 4 g. Become less isolated 1 2 3 4 h. Take responsibility in community matters 1 2 3 4 i. Have a common goal to work towards 1 2 3 4 j. Others, please specify_____________ 1 2 3 4 6

PAGE 197

182 REDUCING YOUR RISK OF WILDFIRES In this section we would like to know what steps have you taken to reduce your wildfire risk. 27. Have you taken any steps around your house to reduce risk of wildfire? Yes go to Q 28 and 29 No skip to Q 30 28. What are some of the activities that you have done to protect your home from wildfires ? For each statement give below answer both questions. A. How important is the activity at your house? (Circle one) Home Protection Activity Large extent Some extent Small extent Not at all B. Have you done the activity? (Check one) No Yes a. Clean roof/gutters to avoid accumulation of leaves and twigs 1 2 3 4 b. Remove dead/overhanging branches from within 10 feet of your roof 1 2 3 4 c. Keep wood/lumber at least 30 feet from house 1 2 3 4 d. Plant new trees and shrubs at least 15 feet apart 1 2 3 4 e. Maintain a healthy green area around your home using grass, flower gardens or shrubs 1 2 3 4 f. Reduce the density of trees and shrubs within 100 feet of your home 1 2 3 4 g. Clear dead vegetation/leaves from near the house 1 2 3 4 h. Plant fire resistant vegetation 1 2 3 4 k. Improving emergency vehicle access 1 2 3 4 7

PAGE 198

183 29. How much did each of these influence you to take action to reduce risk of wildfires and create defensible space around your home? (For each, circle one) No influence Small influence Moderate influence Large influence a. Receiving electronic information on ways to reduce risk of wildfires 1 2 3 4 b. Viewing a television story 1 2 3 4 c. Receiving written information on ways to reduce risk of wildfires 1 2 3 4 d. Attending an event on wildfire preparedness 1 2 3 4 e. Seeing your neighbor doing it 1 2 3 4 f. Talking with an expert 1 2 3 4 g. Talking with friends in the neighborhood 1 2 3 4 h. Talking to a community leader 1 2 3 4 i. Participating in community work day 1 2 3 4 j. Participating in FireWise program 1 2 3 4 8

PAGE 199

184 YOUR INVOLVEMENT AND PERCEPTIONS OF YOUR COMMUNITY In this section we are exploring your involvement with and perception of your community. 30. Of the groups/associations/organizations listed below (Check one for each A and B) : a. Are you aware of this group’s existence in your community? b. Do you belong to this group? Groups/Associations/Organizations A. Aware of? No Yes B. Belong to? No Yes a. Religious group b. Homeowners association c. Sports groups, like golf league d. Garden club e. Youth organization f. Ethnic organization g. Social groups, like luncheon group h. Neighborhood committee, like crime watch group i. Civic group j. Other, please specify__________ 31. In the past year what are some of the activities that you have participated in with your neighbors or other people in the community? For each activity indicate how often you performed the activity (For each, circle one) . Never Once/ Year Few times/ year Once/ month Few times/ month a. Community events like picnics, yard sales, holiday party 1 2 3 4 5 b. Club events like Ladies luncheon, Red Hatters outing 1 2 3 4 5 c. Sports events like tournaments or games 1 2 3 4 5 d. Meetings like homeowner association annual meetings, garden club meetings 1 2 3 4 5 e. Special educational meetings like FireWise, hurricane preparedness 1 2 3 4 5 f. Work projects like neighborhood clean up 1 2 3 4 5 g. Meetings to resolve problems inside the community 1 2 3 4 5 h. Meetings to resolve problems outside the neighborhood 1 2 3 4 5 i. Other, please specify____________ 1 2 3 4 5 9

PAGE 200

185 32. Please tell us how you feel about the following statements using the scale of 1 to 5, 1 being Strongly Disagree (SD), 2 being Disagree (D), 3 being Agree (A), 4 being Strongly Agree (SA) and 5 being Don’t know. Circle one appropriate number for every statement. SD D A SA Don’t know a. I know most people in my neighborhood. 1 2 3 4 5 b. People in the community look out for one another. 1 2 3 4 5 c. People in the community are willing to help each other whenever they can. 1 2 3 4 5 d. Most people in the community are concerned about their own welfare. 1 2 3 4 5 e. I can call on my neighbor for help anytime. 1 2 3 4 5 f. I trust my neighborhood association to make decisions on my behalf. 1 2 3 4 5 g. People in this community have mutual respect for one another. 1 2 3 4 5 h. For most part, people are willing to make the community a better place to live. 1 2 3 4 5 i. People in this community do not get involved in community activities. 1 2 3 4 5 j. I greet my neighbors. 1 2 3 4 5 k. This community is a safe place for children. 1 2 3 4 5 l. Most people in my community do voluntary work for the community. 1 2 3 4 5 m. I think people in this community can not be trusted. 1 2 3 4 5 n. Most people in the neighborhood are connected through the association. 1 2 3 4 5 o. Most people in the community are involved in activities that benefit the community. 1 2 3 4 5 p. People in this community are easy to contact. 1 2 3 4 5 q. For the most part, people in this community are friendly. 1 2 3 4 5 r. I know a few people in the neighborhood, but most are strangers. 1 2 3 4 5 s. Usually people in the community greet one another. 1 2 3 4 5 t. This community offers enough chance for a person to do volunteer work. 1 2 3 4 5 u. People in this community work together to solve problems. 1 2 3 4 5 10

PAGE 201

186 SD D A SA Don’t know v. Most people in the community do not feel they are a part of this community. 1 2 3 4 5 w. People in the community get along with each other. 1 2 3 4 5 x. I volunteer in my community. 1 2 3 4 5 y. For the most part, people in the community obey community codes and covenants. 1 2 3 4 5 z. Few people socialize in the community 1 2 3 4 5 aa. People in the community show support for a cause that may not directly benefit them but benefits the community as a whole. 1 2 3 4 5 bb. Some of my neighbors attend several community functions. 1 2 3 4 5 cc. Most people in my community can be trusted. 1 2 3 4 5 dd. People in the community share common interests. 1 2 3 4 5 ee. One needs to be careful when dealing with people in this community. 1 2 3 4 5 ff. My actions have an impact making this community a better place to live in. 1 2 3 4 5 gg. The community is a melting pot of different ethnic groups. 1 2 3 4 5 PERSONAL INFORMATION In this section we would like to know more about you. This information is purely for statistical analysis and all anonymity will be preserved. 33. What is your gender? (Check one) Male Female 34. Besides yourself, who lives in your house? (Check all that apply) No one Partner/spouse Children Other________ 35. To what age group do you belong? (Check one) 20-35 36-50 51-65 66-80 81 and above 11

PAGE 202

187 36. What is your highest level of education? (Check one) Some high school or less High school diploma or GED Some college coursework Associates degree Undergraduate degree Graduate or professional degree 37. What is your current (approximate) annual household income from all members of the household, before taxes? ( Check one) Less than $ 20,000 $ 20,000 to $ 59,999 $ 60,000 to $ 99,999 $ 100,000 to $ 149,999 $ 150,000 or more 38. What is your employment status? (Check one) Full time employee Part time employee Retired Homemaker Student Unemployed 39. What is your ethnicity? (Check all that apply) White Hispanic or Latino African American or Black Asian or Pacific Islander American Indian or Native Alaskan Native Hawaiian Other, please specify__________ Thank You. COMMENTS Thank you once again for taking th e time to fill the survey for us . Please use this space to tell anything else that you think we should know about your community. 12

PAGE 203

188 If you have any questions about the questionnaire or the study, please free to contact Shruti Agrawal, Graduate Student at the School of Forest Resources and Conservation, Universi ty of Florida at 352-846-2374, sagrawal@ufl.edu , or Dr. Martha Monroe, Associate Professor, University of Florida at 352-846-0878, mcmonroe@ufl.edu . You can also discuss research parti cipants’ rights with the UFIRB Office, University of Florida, PO Box 112250, Gainesville FL, 32611-2250, 352-392-0433.

PAGE 204

189 School of Forest Resources & Conservation 347 Newins Ziegler Hall, PO Box 110410 Gainesville FL 32611 0410 www.sfrc.ufl.edu Dear, This survey is a part of a research study funded by the USDA Forest Service and the University of Florida. It is aimed at learning more about your participation in the w ildfire educational activities and your perception of your community. Your community is Cape Orlando (Wedgefield) . The survey should take about 20 minutes of your time. The adult in your home whose birthday is cl osest to July 1 should complete the survey. You do not have to answer any question you do not choose to answer, even though we hope that you will. In fact, you can stop filling the questionnaire at any time. There is no benefit or risk to you if you participate in this study. The results from this study will lead to a PhD disser tation and recommendations for strategies to improve community preparation for wildland fire. No information about you (your name or your address) will be mentioned in any of our reports—you will remain anonymous to the extent provided by la w. The number on your survey identifies your community and will help with our statistical efforts to use these re sults. The personal information you prov ide us will help with both statistical analysis and to remove any age, gender, education, and income bias from the analysis. Returning this survey will be understood as your consent to be a part of the research study. Please return the survey in the addressed, stamped envelope. If you have any questions about the questionnaire or the st udy, please contact Shruti Agrawal at the School of Forest Resources and Conservation, Un iversity of Florida at 352-846-2374, sagrawal@ufl.edu , or Dr. Martha Monroe, Associate Profes sor, University of Florida at 352-846-0878, mcmonroe@ufl.edu . You can also discuss research participants’ rights with the UFIRB Office, Un iversity of Florida, PO Box 112250, Gainesville FL, 326112250, 352-392-0433. We greatly appreciate your efforts to improve the way Fl orida communities prepare for disasters like hurricanes and fires. Thank you for your time and support. Warm regards, Martha Monroe and Shruti Agrawal Associate Professor Graduate Student Direct responses to: Martha C. Monroe , Assoc. Professor & Extension Specialist mcmonroe@ufl.edu 352-846-0878 An Equal Opportunity / Affirmative Action Institution

PAGE 205

190 School of Forest Resources & Conservation 347 Newins Ziegler Hall, PO Box 110410 Gainesville FL 32611 0410 www.sfrc.ufl.edu Dear, Several weeks ago we sent you the enclosed survey. If y ou have already returned it, thank you very much for your assistance with our study. If you have not, we would very much appreciate your response. Here is why: As you know Florida has its share of disasters. Co mmunity education programs can help people reduce damage and make recovery easier. Experience tells us this is true for hurricanes and wildfires. With this is mind, we developed a research study that uses y our expertise and understanding of fire and your community interactions to help others better prepare for disasters like hurricanes and wildfires. We would greatly appreciate your time in completing and returning the enclosed survey to us. This survey is a part of a research study funded by the USDA Forest Service and the University of Florida. It is aimed at learning more about your participation in the w ildfire educational activities and your perception of your community. Your community is Cape Orlando (Wedgefield) . The survey should take about 20 minutes of your time. The adult in your home whose birthday is cl osest to July 1 should complete the survey. You do not have to answer any question you do not choose to answer, even though we hope that you will. In fact, you can stop filling the questionnaire at any time. There is no benefit or risk to you if you participate in this study and we cannot offer you any compensation. The results from this study will lead to a PhD dissertation and recommendations for strategies to improve community preparation for wildland fire. No information about you (your name or your address) will be mentioned in any of our reports—you will remain anonymous to the extent provided by la w. The number on your survey identifies your community and will help with our statistical efforts to use these re sults. The personal information you prov ide us will help with both statistical analysis and to remove any age, gender, education, and income bias from the analysis. We want to make sure our results will help all people. Returning this survey will be understood as your consent to be a part of the research study. Please return the surv ey in the addressed, stamped envelope. If you have any questions about the questionnaire or the st udy, please contact Shruti Agrawal at the School of Forest Resources and Conservation, Un iversity of Florida at 352-846-2374, sagrawal@ufl.edu , or Dr. Martha Monroe, Associate Profes sor, University of Florida at 352-846-0878, mcmonroe@ufl.edu . You can also discuss research participants’ rights with the UFIRB Office, Un iversity of Florida, PO Box 112250, Gainesville FL, 326112250, 352-392-0433. We greatly appreciate your efforts to improve the way Fl orida communities prepare for disasters like hurricanes and fires. Thank you for your time and support. Warm regards, Martha Monroe and Shruti Agrawal Associate Professor Graduate Student

PAGE 206

191 Sample Post card February 2005 Dear Homeowner, About two weeks ago we sent you a survey about your participation in wildfire education activities and your perception of your community. If you have returned the survey we would like to say thank you! If you have not retuned the survey we would request you to do so as soon as possible. Your input is very valuable to us; we will use it to help other communities prepare for and recover from disasters like hurricanes and wildfires. If you have any questions or concerns, please free to contact us by phone at 352-846-2374 or by email at sagrawal@ufl.edu Regards, Martha C. Monroe and Shruti Agrawal March 2005 Dear Homeowner, About two weeks ago we sent you a survey about your participation in wildfire education activities and your perception of your community. If you have returned the survey we would like to say thank you! If you have not retuned the survey we would request you to do so as soon as possible. Your input is very valuable to us; we will use it to help other communities prepare for and recover from disasters like hurricanes and wildfires. If you have any questions or concerns, please free to contact us by phone at 352-846-2374 or by email at sagrawal@ufl.edu Regards, Martha C. Monroe and Shruti Agrawal

PAGE 207

APPENDIX E 3CM INTERVIEWS AND ANALYSIS Conceptual Content Cognitiv e Mapping (3CM) Interviews Polk County Tiger Creek Forest, Lake Wales Ri dge Communities: 3 CM Interview Participant 1 Date of interview: June 16, 2004 Initial questions Number of years he/she has lived in the community: owned property here since 1989, built a mobile home on the property in 1990 (used as a vacation place). In 2000 we moved into the trailer, and in 2002 we build this house Number of family members who live in the community: none Where did you come from (move here from): Miami, Dade county How many organizations a nd clubs do you belong to: o Tiger Creek owners association (eve ryone who owns property becomes a member). It is active organization and looks into maintenance of main gate, roads, trimming roadsides , electricity, garbage pickup Additional questions Also, we would like you to: List five things that you like the most about your community. 1. Like woods—trees, birds, animal/wildlife 2. Quietness 3. Solitude 4. Do what I want to 5. Security-can call on neighbor for help List five things that you like about people and their relationships in your community. 1. Most people are family oriented 2. People are willing to give so that everyone can be comfortable 3. People are concerned about other people 4. Do not try to get into other people’s business—live and let live 5. If you had a problem there w ould be someone to help 192

PAGE 208

193 List five things that you would lik e to change in your community. Why? 1. No more rental properties-they do not care about construction and keeping the place 2. No more mobile homes-they affect property values 3. Roads paved-increases property values How do you feel about Indian Lake Estates? How would you compare your community with Indian Lake Estates? Is a similar/Different? Indian Lake Estates was designed to sell homes (they may not have mobile homes). However it did not material ize to the extent they wanted. Tiger Creek Forest is in the woods, natural setting, and for different clientele. Advantage of living in Tiger Creek Forest is that it is three miles from downtown Lake Wales. When I moved here in 2000, I had no idea of FireWise. This area is different from Indian Lake Estates. They have sized lots and paved streets. TCF was designed in the 1980’s with cattle grazing l and. It has three entrances—one, from Dude Ranch Road on to 60; second, on Rattlesnake Road on to 60, and third off Glenn St Mary Road. Facts about Tiger Creek Forest: 194 home sites-either a tr ail or house on 100 of them 11 miles of dedicated roads (all dirt or shell rock) 2 miles we do not maintain (TNC owns 14 lots in our area and they maintain it, that saves us money) Minimum lots size in Tiger Creek Forest is five acres In TCF we have people with attitudes about wildfires ranging from we will do anything to save it to we want to [t hose who want to] be here and don't care. We have a Volunteer Fire Department: It also has couple of the firemen (the fire department is 2 miles on Dude ranch Road from 60).The firemen surveyed the entire area and then send out letters to people whose houses they could not get to or they thought they would get stuck (if they got there). I have created defensible spaces around my house on three sides, I'm working on the fourth). TNC has worked hand in hand to educate people about risk of fire. We had a community picnic promoted by the homeowne rs association to meet with neighbors (six weeks ago—approximately May 2004). TNC came in with displays to give information on wildfires-general knowledge, defensible space, native plants. About 50 people (30 families) came to the picnic. We had a similar thing last fall where TNC and DOF came in. In TCF, 80 people live here year-around (full timers), the rest are part timers. We have had community meetings with the Forest Service. TNC has written a grant to create a demonstration house for wildfires. On one of the mobile home trailer in Tiger Creek Forest, TNC wants to do prescribed burning, put a driveway, and advice people to ma ke their house fireproof. This will be

PAGE 209

194 used as a demonstration house in Tiger Creek Forest .We are also working closely with Bill Delph, Division of Forestry. Before joining DOF, Bill was a land developer for Marco Brothers; he was also a lumber dealer in the area. Not many people in Tiger Creek Forest realize that something needs to be done, not many are willing to do something. History of Tiger Creek Forest: It was developed in the early 80’s with retired people. Earlier things here were done by volunteerism, now not many people are willing to do anything. In the past 10 to 12 years we have many younger people move in. History of wildfires: We have had one in the mid 80s. A man who li ves in the area started it, the Forest Service contained it. No lives or homes were lost. Groupings of people Association Limits the neighborhood feeling Groups center around interest Association looks for general wellbeing Horseback riding groups Dues for association Condition of roads People willing to volunteer Elected board (5 people) to take care of it One paid person to clean near dumpster Gates close for security Annual meeting in November (50-60 people attend) Speeding Enforce restrictions No real feeling of neighborhood Wide range of a g es Airboats Gaps in ages Dinner with friends To volunteer work

PAGE 210

195 Call on for advice/suggestions Advantages/positive about TCF Unique (minimum lot size five acres People live and let live 1 inhabitable structure (on each property) Do not see houses next to you Never had crime problem Call on anyone for help Willing to apply knowledge and skills Most people get along well People contact board if they do not like somethin g Had time TCF in timeframe Challenges for association Relate pretty well No new mobile homes All volunteers Paving main road Help with cogen grass eradication Come to picnic Young not involved Category 5: TCF in timeframe Category 6: Challenges for association Explanation of categories: Category 1: Groupings of people: Revolves around people doing stuff together Category 2: Association Category 3: Limits the neighborhood feeling: What keeps it from being more of a neighborhood Category 4: Advantages/positives about Tiger Creek Forest

PAGE 211

196 Indian Lake Estates, Lake Wales Ridge Communities: 3 CM Interview Participant 2 Date of interview: June 16, 2004 Initial questions Number of years he/she ha s lived in the community: 13 years Number of family members who live in the community: two sisters Where did you come from (move here from): River range, originally from Kentucky How many organizations a nd clubs do you belong to: o Vice President, Indian Lake Estates board o Homeowners golf association-has b een president for many years, has been helping the secretary for 3-4 years o Key club o Director, security at Indian Lake Estates o Head, social committee (to get people of Indian Lake Estates together)— arrange new years dance, St. Patrick’s dance, welcome back party, Halloween dance o Art club (help Shirley) Additional questions Also, we would like you to: List five things that you like the most about your community. 1. People 2. Location-love living with animals coming to your backyard 3. Golf (one of the reasons we came here) 4. Fishing 5. Weather (do not want to go back north) List five things that you like about people and their relationships in your community. 1. Friendly 2. Be able to discuss things with them, say politics 3. Enjoy learning from them 4. I'm a people watcher 5. Enjoy hearing their stories-like World War II veterans (listen to history) List five things that you would lik e to change in your community. Why? 1. Get rid of renewed life ministries as splitting the community-it robs people, it is taking advantage of under educated people 2. Security-no police protection 3. Like to have something for the kids-to not want to give till they (parents) give

PAGE 212

197 4. Need more money to run the community How do you feel about Tiger Creek Forest? How would you compare your community with Tiger Creek Forest? Is a similar/Different? I think here (Indian Lake Estates) we help each other more. Notes: The community has 97 miles of paved road, 32 miles of unpaved road. Of the 5000 lots, 500 have homes on them. There are approximately 50 children in the community. When I came in here (ILE) I was 49 years old and the youngest person. Now I am 62 and still among the youngest in the community. We do not have anything for younger people with children in this community, although it was not designed to be this way. Fire education: The Fire department come and talk to us at the Yacht club (100 people) meetings We have had firemen come at the general meetings. ILE gave land to fire depar tment and now we have a st ation (Fire department) here. Fire department people told us that if we have 100 yards of clearing around our houses they could save our houses in a wildfire, and they did so (we had our last fire five years ago). We always had a volunteer fire de partment, have three people there and all of them are in their 80’s. The new fire department is a paid Fire department. We have had controlled burns (they adver tise about it at the Post Office). We do not get mail at our homes; everyone needs to go into the Post Office. And so everyone knows anything that is advertised at the Post Office. We also bring up issues about fire at the general meeting (generally 85-100 people attend the meeting), te ll people that they need perm it to burn. The fire department issues fines to anyone who does not comply by it. General meeting: The Division of Forestry has come 3-4 times in the past 2-3 years to talk to people about wildfires (50 to 75 people attended the general meeting). Snowbird meetings: The fire department co mes to these meeting and talks to the newcomers in the community; approxim ately 120 people attend the meeting. Approximately 25% of the people in ILE ar e aware of fire (total people in Indian Lake Estates= 500 homes x 3 (average) = 1500 people Cannot live in the island Community as a whole Younger group does not volunteer Watch each others house Pickup relatives from airport Go to town (buy things for others) Outpouring of love from all groups People treat each other kind Free food to increase participation Nice community Come to rescue from all walks of lif e Community is spread out Whole community jumps in Afford to live here

PAGE 213

198 Clubs and things people to July for tournament Booklet Administration Memorial Day program Junior golf program Information and clubs Office-handouts on ILE Renewed life ministries (religion based on Realty company Main problems in ILE Explanation of categories: Elderly Do not shun widows We were taught to give since childhood Volunteer-need not be asked Close-knit community Get along well High middle-class More educated Help each other Give to community Volunteerism Do things slower Enjoy each other's company Older and need each other Key club outings-red hatters (lunch) Eaters club (twice a month)to support the restaurant (24 people Golf cart parade Christmas party Golf Fishing Different groups Increasing tax, maintenance, fee Older community in ILE Broke the community Category 5 : O lder co mm u n it y ( peo p le in ILE): w e h ave o lder peop le co m i ng here even now Category 1: Community as a whole Category 2: Clubs and things people do together: things all people (anyone—young, old, rich, poor) participate in at one time Category 3: Administration: things/information people need Category 4: Main problems in ILE: All communiti es have problems; we have not had it before

PAGE 214

199 Indian Lake Estates, Lake Wales Ridge Communities: 3 CM Interview Participant 3 Date of interview: June 10, 2004 Initial questions Number of years he/she ha s lived in the community: 11 years (full-time) Number of family members who live in the community: none Where did you come from (move here from): Kentucky How many organizations a nd clubs do you belong to: o President, Indian Lake Estates homeowners association o Help with security Additional questions Also, we would like you to: List five things that you like the most about your community. 1. Isolation 2. People here 3. Things we have to offer-golf, fishing 4. Way we administer ourse lves, self-sufficient 5. Private community-we on everythi ng here (either owned by individual or Corporation) List five things that you like about people and their relationships in your community. 1. They watch out for each other 2. Gather together and what together on different projects -volunteers spirit 3. I appreciate the older people-the wi sdom from them, their stories List five things that you would lik e to change in your community. Why? 1. Community more user-friendly -friendly for younger people 2. Improved roads 3. Improve clubhouse facilities—upgrade golf course 4. Come up with a to increase revenue that Corporation has to work with How do you feel about Tiger Creek Forest? How would you compare your community with Tiger Creek Forest? Is a similar/Different? I am unfamiliar with Tiger Creek Forest. Notes: We have a community Newsletter—publish 8 in a year (one every six to seven weeks). The next newsletter will go out in three weeks.

PAGE 215

200 30-40% residents are seasonal (out for three to five months , will be back in September). A total of 600 homes in ILE, m any single dwelling. ILE has a total of about 1000 people (consider 2 per home)—retirees, many younger people. ILE has a total: 8200 lots, of which 7500 vacant lots. The Count y conducts control burns, people who live here are not contributing to the wildfire problem The Corporation has 23 paid employees. We need more money to keep the place functioning. Community events: Sheriff department came and talked to people Fish and Wildlife came and talked about game species Fire department talked about control burns One portion of the community does not get together Administration Younger—live here and come to sleep Not many facilities for younger people Deficiencies in ILE Younger-do not get together Two groups Some want to be left alone Older-they want to be involved Older-use it is a social place Older-watch out for each other Older people in ILE Forestry-talk about doing control burns Administration all volunteer Community governance Activities (events) Different activities Golf Friendly Everyone will come to participate/not participate Garden club (trips to gardens) Yatch club (sponsor get-together

PAGE 216

201 Exercise classes Activities (parties/groups) Do things to educate community (CPR) Fishing, boating History who lives here Internal neighborhood watch Homeowners association Everyone knows everybody Deed-restricted Small community July 4 celebrations Computer club Christmas party with kids Demographics Category 5: Older people in ILE: Their thoughts, why they like ILE Category 4: Deficiencies in ILE Category 1: Administra tion: Things that Board of Directors do Category 2: Activities (parties, groups): Activities that people do in ILE Category 3: Activities (events) Explanation of categories:

PAGE 217

202 Highlands County Placid Lakes, Lake Wales Ridge Communities: 3 CM Interview Participant 4 Date of interview: June 09, 2004 Initial questions Number of years he/she ha s lived in the community: 5 years Number of family members who live in the community: wife, younger son and his wife Where did you come from (move here from): Germany How many organizations a nd clubs do you belong to: o Homeowners Association (President of the HOA) o Masonry group Additional questions Also, we would like you to: List five things that you like the most about your community. 1. Low crime rate 2. Friendliness of the people in the community 3. Having our own park (for fishing, boat ramp) at the community 4. Neatly kept and beautiful community 5. Beautiful homes that attract many people List five things that you like about people and their relationships in your community. 1. Older generation who want to retire and be left alone List five things that you would lik e to change in your community. Why? 1. In this community we have a problem with everyone that rents the place, we do not have a problem with those who own land here 2. Old wrecked cars, cars without license pl ates—we are working with the county code to clear that up 3. Clean big trucks that people get 4. Junk in people’s yard 5. Need to maintain vacant lots for fi re—have an ordinance. We are planning the FireWise program for vacant lots 6. Realtors not cleaning th e land after building

PAGE 218

203 Views about Leisure Lakes: Placid Lakes is a leading community as compared to Leisure Lakes. 2 nd Monday of each month the Presidents of all Homeow ner associations in the county have a meeting to discuss about our problems. Difference between Leisure lakes and Placid Lakes: I do not think there is a big differen ce. PL has got more beautiful homes; we maintain grounds and medians better. The other better places are Sun and Lakes and Sebring, of course Sebring is a city. Also every person on the HOA board has been appointed to do something. For example, I have appointed a person for code enforcement, this is not the case in LL. Notes: HOA certification for Placid Lakes was signe d on August 20, 1969. PL is not a city (we are incorporated as the Corporation). We can get the grants, and PL has got everything that Lake Placid (city) has. Since 1999, we have put 52 more streetlight s in PL. We pay for majority of the streetlights. People here do vo lunteer work, which saves a lot of money for Homeowner association. In Placid Lakes, 80% people participate in the community by events. One reason the people here get along well is that many of them are from the military and the military probably teaches them to get along. Threat of fire: Of the 8250 lots, only 3500 have homes. Ther e are a lot of vacant lots in PL. When there is a dry spell thes e vacant lots is the biggest problem. Also, many teenagers come to party over the weekend and that may incr ease the threat of fires (cigarette buds). In October, we are going to host a FireWi se meeting, and we will invite all the five county commissioners to the meeting. Dumping is another big problem in PL . We are working with the county and Sheriff department—they get inmates to clean dump on vacant lots (generally dumping is done by people from outside). We’re tr ying to clean up and anyone who is caught dumping will be fined.

PAGE 219

204 Advantages of living in Placid Lakes Elderly/incapable to the hospital or shopping History of class of PL Welcome p acket Very friendly Picnic area, boat ramp Activities in the community Deed restricted community Retirement community Keep to themselves Move here to enjoy retirement Peace and privacy Not many parties Close knitted neighborhood People get along we ll Many from military Unique—as for the older people Park (in the community) Golf course Free tickets to Restaurants Quarterly newsletter October Garage sale December Annual Christmas party (to thank the community for help Fire department— and will barbecue (February) Dues for joining association ($ 20 initiation fee, $10 renewal fee Quarterly meeting Certificate of incorporation Explanation of categories: Category 1: Advantages of living in Placid Lakes Category 2: Activities in the community: things/activities that he would tell someone who is due to the community Category 3: Certificate of incorporation: people need to know about it because it gives rights to charge dues

PAGE 220

205 Placid Lakes, Lake Wales Ridge Communities: 3 CM Interview Participant 5 Date of interview: June 09, 2004 Initial questions Number of years he/she ha s lived in the community: 9 years Number of family members who live in the community: brother lives in Lake Placid Where did you come from (move here from): Miami How many organizations a nd clubs do you belong to: o Director, Placid Lakes Homeowner and Property owners Association o Chairman, FireWise committee o Takes care of correspondence between FireWise and Divisi on of Forestry Additional questions Also, we would like you to: List five things that you like the most about your community. 1. Quietness 2. Pure fresh air 3. Wildlife that visit your backyard 4. Weather 5. Feel comfortable with neighbors—it took time to get establish and get involved— but it is nice to know nice people List five things that you like about people and their relationships in your community. 1. Like people 2. Everyone is different 3. Take time to ask—people have inte resting things to contribute 4. They show interest when you ar e kind, it is a sign to them 5. Since I have been active in the board I th ink people seem to appreciate me more. They think I am fighting for them List five things that you would lik e to change in your community. Why? 1. I am happy with the community 2. One problem is the young kids—with radi os, speeding, ans they sometimes mash the Christmas tree FireWise: Eugene is a chair of the FireWise co mmittee. A board member had a fire near his house (he lives in an area with only 1-2 hom es). The area was heavily wooded and the fire department could not handle the fire, an airp lane was brought in to put off the fire.

PAGE 221

206 This person then introduced Eugene to the FireWise program for the end result of an ordinance to comply with FireWise recommendations. Since then, the DOF came to speak with this person and Eugene became the chair of the FireWise program. After that th e DOF contacted Eugene for a seminar in Brooksville, Florida. Five peopl e from Placid Lakes went to the FireWise program. The FireWise board announced about the program at the general membership meeting. Oct 2, 2004 is the FireWise day for the community. Gerry LaCavera will address the community for two hours. It will be a festival atmosphere and we will distribute literature. Things around (in) my neighborhood How people meet people Where to find information People pray for others illness Similar needs Information on board members and what drew them here Men's breakfast (1 st Tuesday of every month) Store by and introduce ourselves to new neighbors Everyone waves here is Clean the lake Send an ill lady flowers every month) online for PLHOA People know no one when they move in Look for lasting neighborly companionship Ladies out for dinner (1 a month) Things in newsletter Got camaraderie (morning walks helped develop or increase Go out to each together Arts and crafts festival in Lake Placid All here for nice retirement Associated with different people in different wa y s Christmas parade in Lake Placid Something to be cheerful Fireman's barbecue in Placid Lakes If open with people they are receptive Every morning groups walk around lake Category 4: Things in newsletter Explanation of categories: Category 1: Things around (i n) my neighborhood: activities happening in the community Category 2: How people meet people: we have similar needs, we meet and talk to newcomers because we remember what we needed when moved here Category 3: Where to find information Christmas gathering at town Hall

PAGE 222

207 Leisure Lakes, Lake Wales Ridge Communities: 3 CM Interview Guide Participant 6 Date of interview: June 09, 2004 Initial questions Number of years he/she ha s lived in the community: 21 years Number of family members who live in the community: none, family owns property here Where did you come from (move here from): Naples, FL How many organizations a nd clubs do you belong to: o Leisure Lakes homeowners association o In charge, Leisure Lakes ladi es luncheon group (for 20 years) o Head, Crime watch (homeowner association project) Additional questions Also, we would like you to: List five things that you like the most about your community. 1. Be able to live on water and drive water sports from home setting 2. Friendliness of people 3. Homeowner association-voice of the community 4. Feel good about fire department 5. Glad to see state interested in the are a, community is bordered by state park List five things that you like about people and their relationships in your community. 1. Happy people 2. Active 3. Interested 4. Willingness to help one another (supportive) List five things that you would lik e to change in your community. Why? 1. See more things for younger kids (more projects to entertain them) 2. Small town situation is a problem 3. Children on four wheelers is a problem (we have no designated area for children on four wheelers), children may sometimes also be responsible for wildfires and are destroying property How do you feel about Placid Lakes? How would you compare your community with Placid Lakes? Is a similar/Different? I know both the developers for Leisur e Lakes and Placid Lakes. They were friends. The developer of Placid Lakes had a choice between whether to develop Leisure

PAGE 223

208 Lakes or Placid Lakes. As he decided to d evelop PL he told his friend about LL and his friend developed LL. Leisure Lakes: It is more country. Here we do not need sophisticati on of streetlights, curbing, etc. The homeowner association of Placid Lakes is as active as ours. PL has good volunteer fire department's as ours. However both volunteer fire departments help each other. Crime watch program: This program is for an emergency; in case of an emergency the Sheriff department can activate our calling numbers. This can be useful for people with Alzheimer’s disease and can also be useful in wildfires. Many ladies here are volunteers 35 five lakes within 17 miles Things I would tell someone about the area Invite to LL ladies luncheon (meet neighbors) Two hours from each coast Cookouts Water sports Play cards Picnicing Go to theatres Golf courses Things going on the community that I am involved in Active owners association LL represented in County by HOA Welcome packet Information (copy) of people who do services Christmas parade Needy family project (for 13 yrs), help 8 families Help crisis center at Lake Placid (for teenage girls) Hospital volunteer Nursing home community project (Lake Placid) Active volunteer fire department Garage sale-ties to needy family project (FD gives their facility for use

PAGE 224

209 Feelings about people in the area Welcome and friendly People are very busy If anyone asks for something do it High retirement community Do things for one another Excited about being here Make sure people get to appointments (help needy) Enjoy one another’s company Very active Help with breakdown car People enjoy beach area at golf club Feel friendliness in the area People here are happy People friendly and helpful Category 1: Things going on in the community that I am involved in Category 2: Feelings abou t people in the area: My observation/judgment/ opinion Category 3: Things I would tell some one about the area: things about what people to, location (geographic location) Explanation of categories:

PAGE 225

210 Leisure Lakes, Lake Wales Ridge Communities: 3 CM Interview Participant 7 Date of interview: June 09, 2004 Initial questions Number of years he/she ha s lived in the community: 6 years Number of family members who live in the community: daughter in Lake Placid Where did you come from (move here from): North Carolina How many organizations a nd clubs do you belong to: o President, Leisure Lakes homeowners association Additional questions Also, we would like you to: List five things that you like the most about your community. 1. One of the best areas around Lake Placid 2. People are great than people in other areas 3. Willingness to help one another 4. Friendliness-everyone will waive 5. Weather is good, value of houses has increased List five things that you like about people and their relationships in your community. 1. 80% retirees in the area-so we all can relate very easily to one another. We agree with people's problems, a ilments and grieve for them. 2. Do not have a younger population (they ar e disruptive to the older generation), that helps to be closer together List five things that you would lik e to change in your community. Why? 1. More people from out-of-state involved with the HOA. Some of them had bad experiences in their earlier places and now shy away from us 2. State is buying up a lot of lots, which will make the taxes go up. The state should not do so as it hurts us more. How do you feel about Placid Lakes? How would you compare your community with Placid Lakes? Is a similar/Different? Not heard that PL has the re pertoire that we have here. There are more problems in PL then we have here (more problem with crime). This may be the case because they have teenagers there. This is a better community than PL.

PAGE 226

211 Notes: Notes: Many people move here from the East or the West Coast. There they are more problems like traffic, hurricanes, etc. This area is growing fast. In LL, a 2BD/BA in 1998 was available for $ 80,000. Now, a 2BD/BA is for $ 180,000 Many people move here from the East or the West Coast. There they are more problems like traffic, hurricanes, etc. This area is growing fast. In LL, a 2BD/BA in 1998 was available for $ 80,000. Now, a 2BD/BA is for $ 180,000 Wildfires: Wildfires: This year (2004) we had wildfires due to arson attack. It burned 300 acres. LL is a big area, not totally populated and rural. We also have many ATVers. They have beer parties and campfires and that may start of fire. This year (2004) we had wildfires due to arson attack. It burned 300 acres. LL is a big area, not totally populated and rural. We also have many ATVers. They have beer parties and campfires and that may start of fire. From the Homeowners association standpoint , we have not here talked about fire. We have many vacant lots that have growing brush and increase the risk of fire. The fire department sometimes does control burns. But they too cannot do to many control burns. From the Homeowners association standpoint , we have not here talked about fire. We have many vacant lots that have growing brush and increase the risk of fire. The fire department sometimes does control burns. But they too cannot do to many control burns. Sense of community Security of community Interaction of people Quite and serene Problems with Atvers-fighting with Sheriff and HOA Personal feeling or pride in the Sense of security Crime watch program visitors in the area Retired and are closer 7 days a week Intermingle through golf Everybody interacts Package for newcomers Close-knit community Golf course-several leagues Very friendly Joining golf association Fallbacks No major gettogether The The Help each other HOA picnic (March), 100120 attend Aluminum recycling for HOA Yard sales—FD, HOA (sponsored 8 needy families) Friendlier group than other areas Friendships in the community helps pull together Welcoming committee Some takers not givers 70% live year around, 30% live here 6 months People will come and help Category 4: Fallbacks: Bad items in the communit y Category 1: Sense of community: How will work together, explain the community Category 2: Security of community: Secure, want it safe and quiet Category 3: Interaction of people Explanation of categories: Some people stay here six m o nth s Pull together Transportation Interaction between next two neighbors

PAGE 227

212 Orange County Wedgefield: 3CM Interview Participant 8 Date of interview: January 29, 2004 Initial questions Number of years he/she ha s lived in the community: 11 years (in November) Number of family members who live in the community: 2 Where did you come from (move here from): Maryland How many organizations and clubs do you belong to: New Comers Club of Central Florida (play bridge for them), play bridge for church, Master Gardener (UF extension), golf club, Garden club, HOA Additional questions Also, we would like you to: List five things that you like the most about your community. 1. Like my house a lot 2. Like my friends 3. Golf course 4. People (for the most part) List five things that you like about people and their relationships in your community. 1. Nice (for the most part) 2. Pleasant (for the most part) 3. Friendly (for the most part) 4. Having young people, it helps List five things that you would lik e to change in your community. Why? 1. People to attend more functions (i.e. dinner at club house) 2. Come to HOA

PAGE 228

213 Something in common with neighbors Something in common Friendly neighbors Neighborliness Things for kids Halloween HOA big thing Luncheon Things for kids Love it Bowling Go out and find people Great p lace to be Gettin g to g ethe r People could be together Warmly received at HOA meetings Very friendly People negative about stuff Friday night dinners People you have Common with Know neighbors Get along with everyone Many friends Not cohesive Helping neighbors Make pickles Taking care of birds Walking neighbors dogs Explanation of categories: Category 1:Something in common with neighbors Category 2: Friendly neighbors: why she likes it here Category 3: Neighborliness Category 4: Things for kids Category 5: Helping neighbors Blue Collar community Pockets in the community Play golf Very nice

PAGE 229

214 Wedgefield: 3CM Interview Participant 9 Date of interview: January 29, 2004 Initial questions Number of years he/she ha s lived in the community: 8 years Number of family members who live in the community: none Where did you come from (move here from): Union Park, east of Orlando How many organizations and clubs do you belong to: Country Club (rides bike here on Sunday) Additional questions Also, we would like you to: List five things that you like the most about your community. 1. Beauty: beautiful community 2. People 3. Security/safeness 4. Well-kept, maintained well 5. Social activities List five things that you like about people and their relationships in your community. 1. “Watchdog”, people look out for each other 2. People look out for children 3. People willing to help whenever they can 4. Camaraderie in general 5. Relaxed atmosphere List five things that you would lik e to change in your community. Why? 1. Would like building to slow dow n; we are overbuilding now. 2. Need more codes and restrictions. People are not getting what they pay for in terms of materials, quality—lack of build er ethic (customer is being neglected) 3. Streetlights in acreage, for safety 4. Need to maintain what's going to be left after the builder leaves 5. More speed limit signs (people do not obey signs/limits not enforced), concern for children especially in the Reserve.

PAGE 230

215 Environment Open spaces Meet neighbors while biking Great for biking Camaraderie of neighbors Temperament of Community Everyone looks out for everybody Lend tractor Spring Fling Clubs in Wedgefield (Neighbors) Behind us Depend on neighbor Meet your neighbors Tennis courts Watchdog for community Closed community Environmental protection Wetland protection Hal-Scot Preserve Relaxed Homes close together $ of homes Social aspects Fall Fling Social activities centered around club Relate well regardless of status and income Super bowl party Socialize Category 1: Environment: places where you can go to relax Category 2: Temperament of the community: feeling/sense of community Category 3: Camaraderie of neighbors Category 4: social aspects Explanation of categories:

PAGE 231

216 Advantages of Living here Good school New High school in Avalon Different community Carpool-school Area Mix of people— different ethnic Space Center Beeline expressway good Disney Safe Planned activities for children Everyone similar Sense of Feeling for community Go to Orlando Not far from Coast Airport 20 minutes Children friendly-Village Subdivision life Codes to protect community/participation Open meetings Wedgefield Homeowners Association Wedgefield newsletter Codes Category 5:Advantages of living here: amenities and convenience Category 6: Sense of feeli ng for community: feeling of people here Category 7: Codes that protec t the community: Covenants, clubs, HOA Explanation of categories:

PAGE 232

217 Wedgefield: 3CM Interview Participant 10 Date of interview: January 29, 2004 Initial questions Number of years he/she ha s lived in the community: 9 years Number of family members who live in the community: 2 Where did you come from (move here from): Cocoa How many organizations a nd clubs do you belong to: none Additional questions Also, we would like you to: List five things that you like the most about your community. 1. Lot size. It is very import ant as it gives people room. 2. Low crime area, a quiet community 3. Friendly people 4. Nature (nature reserve, bigger lots have room for wildlife) 5. Location: between Orlando and Cocoa List five things that you like about people and their relationships in your community. 1. Friendly people 2. Middle-class (not dealing with an educated people , people have enough money for a nice home) 3. People tend to look out for each ot her, for any suspicious activity 4. People have a competitive spirit, golf, Christmas lights List five things that you would lik e to change in your community. Why? 1. Stop signs, need to add a few more I am pretty happy with the community, would like to change anything about the people here.

PAGE 233

218 Community Neighborhood Large lots 1-acre lots Similar economic class Friendly Feeling of neighborhood 1300 homes, 200 people in HOA Diversity in race Walking park Golf course makes contacts Lot of people have networks Referred by friends and rela tive s Not many social organizations Park for kids (in village) Close in geographic Like any community Plenty room Network No social club Geographic entrances Small groups celebrate different holida y s Earlier judging on Christmas lights Talk to each other Get along very well Explanation of Categories: Category 1: Community: This category l ooks at Geographics. How people discover the area, contacts lead to people m oving here. What is unique about lots/community? Category 2: Neighborhood: Things that develop in a neighborhood Neighbors Knowing your neighbors HOA Suspicious activ ity Look out for one another Watch super bowl Get along well HOA not a social thing

PAGE 234

219 Wedgefield: 3CM Interview Participant 11 Date of interview: January 29, 2004 Initial questions Number of years he/she ha s lived in the community: 15 years Number of family members who live in the community: zero family members besides wife and kids Where did you come from (move here from): Orlando How many organizations a nd clubs do you belong to: HOA Additional questions Also, we would like you to: List five things that you like the most about your community. 1. Geographic location: How the community is away from Orlando, away from the hustle and bustle of Orlando 2. People: pretty good/tight kn it “If you need something you know who to call” 3. How things are structured (rules and regulations not overbearing) 4. How community goes about solving problems 5. Community situated with the green space around it List five things that you like about people and their relationships in your community. 1. People are friendly 2. For most part, people are willing to ma ke the community a better place to live 3. For most part, people are concerned about each other 4. Diverse community (ethnicity). Everyone is involved in the community, no ethnic separation 5. Giving community List five things that you would lik e to change in your community. Why? 1. People get more involved in HOA and the community. Need more of a “What can I do for the community” attitude 2. Community should come together to d ecide on how to handle the growth in Wedgefield 3. See the community have its own community center 4. Tighter and more restrictions on ge neral appearances of properties 5. More people should attend functions and e ducate themselves about the community

PAGE 235

220 Function Home Owners Association Assets of the community Reasons for it is a good community Golf with associations Unique community Get involved Glue: HOA for the community Involved in the PTA Different ethnic group Functions for children Bring people together Family oriented people Very diverse community Convenient Garden club HOA Social Golf club community Know your nei g hbo r Activities by HOA Great place to live Relationship b etter now Once fancy Many attractions Turmoil brings people together Things people do with neighbors/each othe r HOA: center of community People do not sa y hi Monthly meetings Horses, Events in park People get along good Dispel misconceptions Dance Watch each others houses Watch animals Category 1: Function: things people in the community Category 2: HOA: what they provide to the community Category 3: assets of the community Category 4: reasons why it is the community Category 5: things people do for neighbors/with each other Explanation of categories: New year party Hay shipments in the bulk New year party

PAGE 236

221 Flagler County Cypress Knoll 3CM Interview Participant 12 Date of interview: March 10, 2004 Initial questions Number of years he/she has lived in the community: full time since 1994, house built in 1990 (less than 100 at that time) Number of family members who live in the community: daughter in West section Where did you come from (move here from): Woodbridge, NJ How many organizations a nd clubs do you belong to: Neighborhood watch block captain, plight of Palm Coast Am ateur Radio Club (President) Additional questions Also, we would like you to: List five things that you like the most about your community. 1. People—immediate neighbors, my block capt ain responsibility. Earlier I was incharge of all block captains, and good way to getting to know other block captains 2. Attractiveness of the area 3. Way people maintain (appearance) of the home 4. Having a golf course—great activity for people here, lot of people enjoy it 5. People helping people “necessary” List five things that you like about people and their relationships in your community. 1. Friendly 2. Mutual respect for one another 3. Sharing on common interest 4. Wanting to see one another List five things that you would lik e to change in your community. Why? 1. There is only road in/out of Cypress Knoll— need to change it for safety of people, emergency response to help evacuation 2. See side walk put in on the main road into the community—at some point the community center will be built across Cypress Knoll—it will be needed for people to be able to walk there 3. Street lighting at major intersect ions, not all need street lights 4. Street signs you can see at night 5. Small community park here—bring grand ki ds to swing, a little picnic area

PAGE 237

222 Grassroots of looking at a community is knowing your immediate neighbors My involvement/con tribution to the community Social aspects Most get along Mix of people— US, Canada Mutual common ground Cypress Knoll neighborhood watch Social events Most retired Picnic—fall and spring All walks of life Watch people’s home Helping one another Closer relationship Christmas party FireWise program— learning en j o y able Talk about family Go out together to eat outside the community Attrac tive Share thoughts about community Say hello What would you tell someone who came to Cypress Knoll Learning experience Get together with one another Nature of people—not all likeable Amateur radio People want to be involved Done lot surveys Interest in maintaining safety a pp earance Listen to one another Explanation of categories: 30 block captains Responsibility Category 1:Social aspects: What we do socially? Sets the community tone Category 2: What you would tell someone who came to Cypress Knoll: About people and the area Neighbors friendly Mutual respect Enjoy Category 3: Grassroot s of looking at the community is knowing your immediate neighbors: that grows to the community Can conserve Congeniality Category 4: My involvement/contribution to the community: Some of my activities not in the community but benefits the community as a w hole Unusual activity

PAGE 238

223 Cypress Knoll 3CM Interview Participant 13 Date of interview: March 10, 2004 Initial questions Number of years he/she ha s lived in the community: 1 year in Cypress Knoll Number of family members who live in the community: no Where did you come from (move here from): other side of Palm Coast (for2 years), before that I lived in New York o Moved in Cypress Knoll from the other side as this is an upcoming area, has a golf club, newer and cleaner appearance, property values will increase, and has less tra ffic/business down here. How many organizations a nd clubs do you belong to: o He is the chair of Cypress Knoll organization. A year ago he instituted the web (in December). Attended the town meeting, and talked with long term residents o His approach to people: we are a team and can make the changes that everyone wants o The web came from the social to real organization. If you have a concern get on the web and it will be addressed immediately o His wife is a member of many writing clubs here, and they both are involved with adult education. Additional questions Also, we would like you to: List five things that you like the most about your community. 1. Proximity to the beach 2. Accessibility to the beach 3. Scenic view—a pretty community 4. A new community with new construction going on 5. Friendliness of the community List five things that you like about people and their relationships in your community. 1. Large number of artistic people—acting, writing, music, classis arts, etc. 2. People always socialize—go out together , have each other at home for dinner 3. People communicate—not cold to each other when you drive bikes, walk, 4. People work together—if there is a concer n they will not fight but work together List five things that you would lik e to change in your community. Why? 1. Enjoy growth, but rather not see many young families as it increases taxes, schools

PAGE 239

224 2. People be more courteous as far as th eir pets are concerned—it bothers other people 3. Not enough parking areas here 4. Few people who do not maintain their house, it gives a bad look to the neighborhood, spoils others homes beauty. Views about Seminole Woods: Do not know anyone there, and do not care for the accessibility to it. There is a tremendous difference between Cypress Knoll and Seminole Woods. Now the town center is being build right across from Cypress Knoll. I n Seminole Woods they do not have a golf c ourse, as it is more expensive. In Seminole Woods the construction is cheaper. The lots are cheap th ere and homes are not as nice. As they are cheaper, th ey may get more young families. They have all problems that we have multiplied by 10. A sense of community Organizational Commonality People address litter problem HOA—no dues People can identify HOA website interactive Nice mix of families People clean streets Newsletter: 2-3 a yr Lot of retirees People do beautification Block captains deliver newsletter Young families Everyone waives Have something in common (come from north) People from NY, FL, Relief from rat race

PAGE 240

225 Social aspects of the community Caring community October f est Share information with new people Block parties Holiday dan c e New social Director Walk someone’s dog Spring picnic Socials Will have community wide garage sale Take in mail and papers Not cold community Friendly Watch each other’s home Help women Go to plant sale together Explanation of categories: Recreation Category 1:Social aspects of the community Play golf Category 2: Organizational: What helps us to do some of these things Several clubs— swimming and racket ball, gym Category 3: A sense of community: Cooperative effort between people (aware that it benefits them too) Category 4: Commona lity: commonality of people here, not too different Library—volunteers from within community use it Category 5: Caring community: caring community here Category 6 : R ecreation: things people do

PAGE 241

226 Cypress Knoll 3CM Interview Participant 14 Date of interview: March 10, 2004 Initial questions Number of years he/she ha s lived in the community: 4 years, since 1999 Number of family members who live in the community: none Where did you come from (move here from): Atlanta, GA How many organizations a nd clubs do you belong to: Additional questions Also, we would like you to: List five things that you like the most about your community. 1. People 2. Location—excellent to get here and there— easy to get to Jacksonville, Orlando 3. Weather 4. Golf 5. Chance to expand yourself—do volunteer work List five things that you like about people and their relationships in your community. 1. I have created/developed friendships with people of similar interest 2. Their attitude—get along 3. Willingness to help 4. Mind their own business 5. Lack of obvious prejudice in this co mmunity (changing of old guard, new ones) List five things that you would lik e to change in your community. Why? 1. Need side walks 2. Need street lights 3. Want to see few/less people coming in 4. Publix supermarket at the corner 5. If we can maintain assembly of people with similar attitudes More about his views on Cypress Knoll: We have a committee for mosquito control (spray everyday). I know every neighbors home; I know what they do, where they came from. Property values of Cypress Knoll: 4 years a go it was an average of $ 7000— today there is no lot less than $ 50,000. “This community is made by people.” “It is not a retiree community.” “It is a mix of age and ethnic groups.”

PAGE 242

227 Views about Seminole Woods: When I go to the church I see people from Seminole Woods, do not know many people from there. I would not choose to li ve there. The power line runs through the community and I do not like it. In fact the power line may create some controversy in Seminole Woods. Wish list for an American dream Real America Advice Best people here People get along Need something physically they are there Same economic level Neighborhood association Inde p endent p eo p le Pride in the community Get along quite well Mind about your business Highest land value in Palm Coast Excellent community Living your own life Property values escalated People look for a decent community This is what America should be All ethnic groups

PAGE 243

228 Miscellaneous attributes Social functions Realize who should not be here Picnic Christmas Calmness Golf course Who lives in what house Not fast pace life Talk to one another Social functions Lends a han d Screen them Social life emanates through golf Get with each other as nothing more to prove No challenge church Watch out for one another Do not worry about anything Explanation of categories: Category 1:Real America: American dream community Category 2: Wish list for an American dream: What is an American dream? More specifics about it –all this in this category Category 3: Social functions C at ego r y 4 : Mi sce llan eo us attri bu t es

PAGE 244

229 Seminole Woods 3CM Interview Participant 15 Date of interview: March 09, 2004 Initial questions Number of years he/she has lived in the community: 25 years, was the commissioner for mosquito control in 1996 Number of family members who live in the community: none Where did you come from (move here from): NY How many organizations a nd clubs do you belong to: Additional questions Also, we would like you to: List five things that you like the most about your community. None List five things that you like about people and their relationships in your community. None List five things that you would lik e to change in your community. Why? None How do you feel about Cypress Knoll as compared to Seminole Woods? None

PAGE 245

230 Feel about people and me Chronological changes in the community When I came here Happy family Split Nice Go to commissioners meetings On our own Things changed Call me for help Water Volunteering Mayor Nice feeling before fire Bicycle riding Became a city I-95 US1 Mistreated Do not backup Near to the beach Christmas get together Do not want to be bothered Country Play cards Alone Bitter Help people Lost interest Do not care enough Single women parents Houses separated

PAGE 246

APPENDIX F QUANTITATIVE DATA ANALYSIS TABLES NON-RESPONSE ANALYSIS Tiger Creek Forest Group1: Residents who returned surveys=respondents Group 2: Residents who did not return surveys=non-respondents Group statistics Group N Mean Std. Deviation Std. Error Mean Property Value 1.00 40 94478.00 46760.70 7393.51 2.00 54 81467.04 50303.87 6845.49 Independent samples test Levene's Test for Equality of Variances t-test for Equality of Means F Sig. t df Sig. (2tailed) Mean Difference Std. Error Difference 95% Confidence Interval of the Difference Lower Upper Property value Equal variances assumed .19 .662 1.28 92 .21 13010.96 10187.15 -7221.61 33243.53 Equal variances not assumed 1.29 87.31 .20 13010.96 10075.95 -7015.08 33037.00 231

PAGE 247

232 Indian Lake Estates Group1: Residents who returned surveys=respondents Group 2: Residents who did not return surveys=non-respondents Group statistics Group N Mean Std. Deviation Std. Error Mean Property value 1.00 146 81278.42 40637.05 3363.15 2.00 213 73584.69 40644.11 2784.89 Independent samples test Levene's Test for Equality of Variances t-test for Equality of Means F Sig. t df Sig. (2tailed) Mean Difference Std. Erro r 95% Confidence Interval o Difference f the Difference Lower Upper Property value Equal variances assumed .104 .75 1.76 357 .08 7693.73 4366.65 -893.85 16281.31 Equal variances not assumed 1.76 311.77 .08 7693.73 4366.50 -897.81 16285.27 Wedgefield Group1: Residents who returned surveys=respondents Group 2: Residents who did not return surveys=non-respondents Group statistics Group N Mean Std. Deviation Std. Error Mean Property value 1.00 160 148467.03 37328.82 2951.10 2.00 585 142360.11 38699.88 1600.04

PAGE 248

233 Independent samples test Levene's Test for Equality of Variances t-test for Equality of Means F Sig. t df Sig. (2tailed) Mean Difference Std. Error Difference 95% Confidence Interval of the Difference Lower Upper Property values Equal variances assumed .357 .55 1.78 743 .08 6106.91 3426.82 -620.49 12834.31 Equal variances not assumed 1.819 260.10 .07 6106.91 3356.95 -503.36 12717.18 Cypress Knoll Group1: Residents who returned surveys=respondents Group 2: Residents who did not return surveys=non-respondents Group statistics Group N Mean Std. Deviation Std. Error Mean Property value 1.00 298 137967.47 41738.19 2417.83 2.00 414 134458.33 36658.27 1801.66 Independent samples test Levene's Test for Equality of Variances t-test for Equality of Means F Sig. t df Sig. (2tailed) Mean Difference Std. Error Difference 95% Confidence Interval of the Difference Lower Upper Property values Equal variances assumed 1.70 .19 1.19 710 .24 3509.14 2952.44 -2287.41 9305.70 Equal variances not assumed 1.16 588.02 .25 3509.14 3015.27 -2412.87 9431.16

PAGE 249

234 Seminole Woods Group1: Residents who returned surveys=respondents Group 2: Residents who did not return surveys=non-respondents Group statistics Group N Mean Std. Deviation Std. Error Mean Property value 1.00 143 86190.28 39523.91 3305.16 2.00 352 78915.24 39472.40 2103.89 Independent samples test Levene's Test for Equality of Variances t-test for Equality of Means F Sig. t df Sig. (2tailed) Mean Difference Std. Error Difference 95% Confidence Interval of the Difference Lower Upper Property values Equal variances assumed .94 .33 1.86 493 .06 7275.04 3915.80 -418.67 14968.75 Equal variances not assumed 1.86 262.9 .07 7275.04 3917.96 -439.52 14989.61

PAGE 250

235 SOCIAL CAPITAL SCALE: RELIABILITY ANALYSIS Reliability analysis Reliability coefficients N of Cases = 1243.0 N of Items = 21 Alpha = .8711

PAGE 251

236 RULES FOR MODIFYI NG MARKINGS (CODING) OF QUESTIONS ON PARTICIPATION IN WILDFIRE EDUCATION INITIATIVES 1. For Q 14 (Have you seen or heard these electronic sources for information on ways to reduce risk of wildfire s and create defensible space?) a. If for options a-g either Yes or No is marked, no blanks, then Yes = Y = 2 No = N = 1 b. If for options a-g only Yes’s are marked, others are left blank Y = 2 Blank = 9 = N = 1 c. If for options a-g both Yes and No are marked, and there some are left blank, Y = 2 N = 1 Blank = 9 d. If options a-g all are left blank Blank = 9 (unless something is marked in the corresponding Q 14 a/b/c e. If for options a-g only No’s are marked, and other left blank N =1 Blank = 9 2. If Q 14 (Have you seen or heard these elec tronic sources for information on ways to reduce risk of wildfires and create defensible space?) is left blank, and a. Corresponding Q 14 a/b/c are either or all marked, then Mark corresponding Q 14 option as Y = 2 (irrespective whether the answer to Q 14 a is No or Q 14 b/c is 1) 3. If Q 16 (Have you ever received informati on in print on ways to reduce risk of wildfires?) is marked as Yes, and a. In Q 17 for options a-h, either/or bot h mail and in person are marked, Blank = 9 = Did not receive = 1 b. In Q 17 for options a-h, did not receive is marked for either option, Blank = 9 c. If Q 17 mail/in person/did not receive marked, Blank = 9 Exception to the rule 3b and 3c: If either of Q17 a/b/c is marked, and Q17 is blank, then mark corresponding Q 17 according to 17 a/b/c. 4. If Q 16 (Have you ever received informati on in print on ways to reduce risk of wildfires?) is marked as No, a. For Q 17 for options a-h, either mail or in person is marked then change Q16 to Y = 2 b. Q 17 for options a-h is blank, but Q 17a is marked as Yes, or Q 17 b/c is marked as <1, then change

PAGE 252

237 Q 16 to Y = 2 5. If Q 16 (Have you ever received informati on in print on ways to reduce risk of wildfires?) is marked as Yes, and Q 17 a/b/c is blank a. Q 17 a/b/c options a-h marked either/or all, then mark Q 17 correspondingly. b. If Q 17 not marked, as a rule if Q 17 a is Yes, then mark options for Q 17 as follows: Handout (in person = 2) Brochure (in person = 2) Door hanger (mail = 3) Newsletter (mail = 3) Magazine (mail = 3) Newspaper (mail = 3) c. If Q17 and Q 17a is blank, and Q 17b and Q 17c are marked not at all, then i. For corresponding print source option mark Q 17 according to rule 5b ii. Mark print options for 17 a blank = 9 d. If Q 17 is blank, Q 17 b and Q 17 c are blank, and only Q 17a is marked, then i. For corresponding print option in Q 17 mark according to rule 5b ii. Mark Q 17b and Q 17cas blank = 9 6. If Q 20 (Have you ever attended any event related to reducing risk of wildfires and creating defensible space?) is marked Yes, and a. Q 21 either/or all marked alone or with someone, consider Blank = 9 = did not attend = 1 b. Q 21 marked either did not attend/alone/with someone, consider Blank = 9 c. Q 21 did not attend is marked, ir respective alone/with someone not marked, consider Blank = 9 Exception to the rule 6b and 6 c: If either of Q21 a/b/c is marked, and Q21 is blank, then mark corresponding Q 21 according to 21 a/b/c. 7. If Q 20 (Have you ever attended any event related to reducing risk of wildfires and creating defensible space?) is marked No, and a. Q 21 either alone/with some one is marked, change Q 20 to Y = 2 b. Q 21 main is blank, but either of Q 21 a/b/c is marked, change Q 21 to Y = 2

PAGE 253

238 8. If Q 20 (Have you ever attended any event related to reducing risk of wildfires and creating defensible space?) is marked Yes, Q 21 is blank, and either of Q 21 a/b/c is marked, then, a. Mark corresponding Q 21 according to the following rule: Presentation (alone = 2) Meeting (alone = 2) Prescribed burn demonstration (alone = 2) Equipment demonstration (alone = 2) Landscape demonstration (alone = 2) Information booth (alone = 2) Community picnic (alone = 2) b. Q 21 and Q 21a is blank, Q 21b and Q21 care marked not at all, i. For corresponding Q 21 mark according to rule 8a ii. Mark 21a as blank = 9 c. Q 21 is blank, Q 21 b and/or Q 21c is blank, Q 21a marked, then, i. Mark corresponding Q 21 according to rule 8a ii. Mark Q 21 b and Q 21 c as blank = 9

PAGE 254

239 CHARACTERISTICS OF RESPONDENTS Tiger Creek Forest Surveys returned : 41 out of 95 Descriptive statistics Variables N Minimum Maximum Mean Std. Deviation Interpretation of mean values Age 41 1.00 5.00 3.02 .91 51-65 years Education 41 1.00 6.00 3.44 1.37 Some college coursework Income 32 1.00 4.00 2.60 .88 20,000-59,999 Education Frequency Percent Valid Percent Cumulative Percent Some high school or less 1 2.4 2.4 2.4 High school diploma or GED 10 24.4 24.4 26.8 Some college coursework 14 34.1 34.1 61.0 Associates degree 7 17.1 17.1 78.0 Undergraduate degree 4 9.8 9.8 87.8 Graduate or professional degree 5 12.2 12.2 100.0 Total 41 100.0 100.0 Employment Frequency Percent Valid Percent Cumulative Percent Full time employee 21 51.2 51.2 51.2 Part time employee 1 2.4 2.4 53.7 Retired 19 46.3 46.3 100.0 Total 41 100.0 100.0 Gender Frequency Percent Valid Percent Cumulative Percent Male 27 65.9 65.9 65.9 Female 14 34.1 34.1 100.0 Total 41 100.0 100.0 Ethnicity Race Frequency Percent White 40 97.6 Hispanic 0 0 African American 0 0 Asian 0 0 American Indian 0 0 Native Hawaiian 0 0 Other 0 0

PAGE 255

240 Income Frequency Percent Valid Percent Cumulative Percent Less than $ 20,000 3 9.4 9.4 9.4 $ 20,000-$ 59,999 12 37.5 37.5 46.9 $ 60,000-$ 99,999 12 37.5 37.5 84.4 $ 100,000-$ 149,999 5 15.6 15.6 100.0 Total 32 100.0 100.0 Age Years Frequency Percent Valid Percent Cumulative Percent 20-35 2 4.9 4.9 4.9 36-50 8 19.5 19.5 24.4 51-65 20 48.8 48.8 73.2 66-80 9 22.0 22.0 95.1 81 and above 2 4.9 4.9 100.0 Total 41 100.0 100.0 Indian Lake Estates Surveys returned: 153 out of 367 Descriptive statistics N Minimum Maximum Mean Std. Deviation Interpretation of mean values Age 145 1.00 6.00 3.70 .93 51-65 years Education 142 1.00 6.00 3.33 1.47 Some college coursework Income 114 1.00 5.00 2.33 .93 $ 20,000-59,999 Education Frequency Percent Valid Percent Cumulative Percent Some high school or less 3 2.1 2.1 2.1 High school diploma or GED 45 31.7 31.7 33.8 Some college coursework 54 38.0 38.0 71.8 Associates degree 6 4.2 4.2 76.1 Undergraduate degree 10 7.0 7.0 83.1 Graduate or professional degree 24 16.9 16.9 100.0 Total 142 100.0 100.0 Employment Frequency Percent Valid Percent Cumulative Percent Full time employee 28 19.6 19.6 19.6 Part time employee 3 2.1 2.1 21.7 Retired 107 74.8 74.8 96.5 Homemaker 5 3.5 3.5 100.0 Total 143 100.0 100.0

PAGE 256

241 Gender Frequency Percent Valid Percent Cumulative Percent Male 78 54.2 54.2 54.2 Female 66 45.8 45.8 100.0 Total 144 100.0 100.0 Ethnicity Frequency Percent Valid Percent Cumulative Percent White 135 97.1 97.1 97.1 African American 2 1.4 1.4 98.6 American Indian 1 .7 .7 99.3 Other 1 .7 .7 100.0 Total 139 100.0 100.0 Age Years Frequency Percent Valid Percent Cumulative Percent 20-35 4 2.8 2.8 2.8 36-50 7 4.8 4.8 7.6 51-65 45 31.0 31.0 38.6 66-80 63 43.4 43.4 82.1 81 and above 26 17.9 17.9 100.0 Total 145 100.0 100.0 Income Frequency Percent Valid Percent Cumulative Percent Less than $ 20,000 17 14.9 14.9 14.9 $ 20,000-$ 59,999 57 50.0 50.0 64.9 $ 60,000-$ 99,999 29 25.4 25.4 90.4 $ 100,000-$ 149,999 7 6.1 6.1 96.5 $ 150,000 and above 4 3.5 3.5 100.0 Total 114 100.0 100.0 Cape Orlando (Wedgefield) Surveys returned: 201 out of 745 Descriptive statistics N Minimum Maximum Mean Std. Deviation Interpretation of mean values Age 194 1.00 5.00 2.61 .94 36-50 years Education 194 1.00 6.00 3.92 1.41 Some college courseworkAssociates degree Income 168 1.00 5.00 2.93 .96 $ 60,000-99,999

PAGE 257

242 Education Frequency Percent Valid Percent Cumulative Percent Some high school or less 2 1.0 1.0 1.0 High school diploma or GED 33 16.4 17.0 18.0 Some college coursework 53 26.4 27.3 45.4 Associates degree 34 16.9 17.5 62.9 Undergraduate degree 35 17.4 18.0 80.9 Graduate or profe ssional degree 37 18.4 19.1 100.0 Total 194 96.5 100.0 Employment Frequency Percent Valid Percent Cumulative Percent Full time employee 125 65.4 65.4 65.4 Part time employee 6 3.1 3.1 68.6 Retired 51 26.7 26.7 95.3 Homemaker 7 3.7 3.7 99.0 Student 1 .5 .5 99.5 Unemployed 1 .5 .5 100.0 Total 191 100.0 100.0 Gender Frequency Percent Valid Percent Cumulative Percent Valid Male 127 64.8 64.8 64.8 Female 69 35.2 35.2 100.0 Total 196 100.0 100.0 Ethnicity Frequency Percent Valid Percent Cumulative Percent White 159 85.0 85.0 85.0 Hispanic 11 5.9 5.9 90.9 African American 4 2.1 2.1 93.0 Asian 7 3.7 3.7 96.8 American Indian 2 1.1 1.1 97.9 Others 4 2.1 2.1 100.0 Total 187 100.0 100.0 Age Years Frequency Percent Valid Percent Cumulative Percent 20-35 22 11.3 11.3 11.3 36-50 70 36.1 36.1 47.4 51-65 65 33.5 33.5 80.9 66-80 35 18.0 18.0 99.0 81 and above 2 1.0 1.0 100.0

PAGE 258

243 Total 194 100.0 100.0 Income Frequency Percent Valid Percent Cumulative Percent Less than $ 20,000 6 3.6 3.6 3.6 $ 20,000-$ 59,999 55 32.7 32.7 36.3 $ 60,000-$ 99,999 62 36.9 36.9 73.2 $ 100,000-$ 149,999 35 20.8 20.8 94.0 $ 150,000 and above 10 6.0 6.0 100.0 Total 168 100.0 100.0 Cypress Knoll Surveys returned: 317 out of 746 Descriptive statistics N Minimum Maximum Mean Std. Deviation Interpretation of mean values Age 309 1.00 5.00 3.15 .93 51-65 years Education 306 1.00 6.00 3.99 1.49 Some college courseworkAssociates degree Income 249 1.00 5.00 2.62 .89 $ 20,000-59,999 Education Frequency Percent Valid Percent Cumulative Percent Some high school or less 5 1.6 1.6 1.6 High school diploma or GED 49 16.0 16.0 17.6 Some college coursework 91 29.7 29.7 47.4 Associates degree 35 11.4 11.4 58.8 Undergraduate degree 53 17.3 17.3 76.1 Graduate or professional degree 73 23.9 23.9 100.0 Total 306 100.0 100.0 Employment Frequency Percent Valid Percent Cumulative Percent Full time employee 76 25.2 25.2 25.2 Part time employee 21 7.0 7.0 32.1 Retired 191 63.2 63.2 95.4 Homemaker 10 3.3 3.3 98.7 Student 1 .3 .3 99.0 Unemployed 3 1.0 1.0 100.0 Total 302 100.0 100.0

PAGE 259

244 Gender Frequency Percent Valid Percent Cumulative Percent Male 163 52.4 52.4 52.4 Female 148 47.6 47.6 100.0 Total 311 100.0 100.0 Ethnicity Frequency Percent Valid Percent Cumulative Percent White 271 90.9 90.9 90.9 Hispanic 7 2.3 2.3 93.3 African American 17 5.7 5.7 99.0 Asian 1 .3 .3 99.3 Other 2 .7 .7 100.0 298 100.0 100.0 Age Years Frequency Percent Valid Percent Cumulative Percent 20-35 20 6.3 6.5 6.5 36-50 42 13.2 13.6 20.1 51-65 131 41.3 42.4 62.5 66-80 105 33.1 34.0 96.4 81 and above 11 3.5 3.6 100.0 Total 309 97.5 100.0 Income Frequency Percent Valid Percent Cumulative Percent Less than $ 20,000 9 3.6 3.6 3.6 $ 20,000-$ 59,999 124 49.8 49.8 53.4 $ 60,000-$ 99,999 83 33.3 33.3 86.7 $ 100,000-$ 149,999 20 8.0 8.0 94.8 $ 150,000 and above 13 5.2 5.2 100.0 Total 249 100.0 100.0 Seminole Woods Surveys returned: 129 out of 497 Descriptive statistics N Minimum Maximum Mean Std. Deviation Interpretation of mean values Age 128 1.00 5.00 2.46 .99 36-50 years Education 129 1.00 6.00 3.36 1.41 Some college coursework Income 110 1.00 4.00 2.31 .71 20,000-59,999

PAGE 260

245 Education Frequency Percent Valid Percent Cumulative Percent Some high school or less 10 7.8 7.8 7.8 High school diploma or GED 29 22.5 22.5 30.2 Some college coursework 36 27.9 27.9 58.1 Associates degree 26 20.2 20.2 78.3 Undergraduate degree 15 11.6 11.6 89.9 Graduate or professional degree 13 10.1 10.1 100.0 Total 129 100.0 100.0 Employment Frequency Percent Valid P ercent Cumulative Percent Full time employee 69 53.9 53.9 53.9 Part time employee 13 10.2 10.2 64.1 Retired 32 25.0 25.0 89.1 Homemaker 12 9.4 9.4 98.4 Student 1 .8 .8 99.2 Unemployed 1 .8 .8 100.0 Total 128 100.0 100.0 Gender Frequency Percent Valid Percent Cumulative Percent Male 60 46.5 46.5 46.5 Female 69 53.5 53.5 100.0 Total 129 100.0 100.0 Ethnicity Frequency Percent Valid P ercent Cumulative Percent White 118 93.7 93.7 93.7 Hispanic 2 1.6 1.6 95.2 African American 6 4.8 4.8 100.0 Total 126 100.0 100.0 Age Years Frequency Percent Valid Percent Cumulative Percent 20-35 23 18.0 18.0 18.0 36-50 45 35.2 35.2 53.1 51-65 39 30.5 30.5 83.6 66-80 20 15.6 15.6 99.2 81 and above 1 .8 .8 100.0 Total 128 100.0 100.0

PAGE 261

246 Income Frequency Percent Valid Percent Cumulative Percent Less than $ 20,000 11 10.0 10.0 10.0 $ 20,000-$ 59,999 59 53.6 53.6 63.6 $ 60,000-$ 99,999 35 31.8 31.8 95.5 $ 100,000-$ 149,999 5 4.5 4.5 100.0 Total 110 100.0 100.0 Placid Lakes Surveys returned: 356 out of 869 Descriptive statistics N Minimum Maximum Mean Std. Deviation Interpretation of mean values Age 339 1.00 5.00 3.62 .99 51-65 years Education 333 1.00 6.00 3.37 1.49 Some college coursework Income 267 1.00 5.00 2.15 .77 20,000-59,999 Education Frequency Percent Valid Percent Cumulative Percent Some high school or less 17 5.1 5.1 5.1 High school diploma/GED 95 28.5 28.5 33.6 Some college coursework 105 31.5 31.5 65.2 Associates degree 32 9.6 9.6 74.8 Undergraduate degree 33 9.9 9.9 84.7 Graduate or profe ssional degree 51 15.3 15.3 100.0 Total 333 100.0 100.0 Employment Frequency Percent Valid P ercent Cumulative Percent Full time employee 83 24.9 24.9 24.9 Part time employee 15 4.5 4.5 29.4 Retired 223 67.0 67.0 96.4 Homemaker 10 3.0 3.0 99.4 Unemployed 1 .3 .3 99.7 Gender Frequency Percent Valid Percent Cumulative Percent Male 204 60.0 60.0 60.0 Female 135 39.7 39.7 99.7 Total 340 100.0 100.0

PAGE 262

247 Ethnicity Frequency Percent Valid Percent Cumulative Percent White 317 96.1 96.1 96.1 Hispanic 6 1.8 1.8 97.9 African American 2 .6 .6 98.5 Asian 2 .6 .6 99.1 American Indian 1 .3 .3 99.4 Other 2 .6 .6 100.0 Total 330 100.0 100.0 Age Years Frequency Percent Valid Percent Cumulative Percent 20-35 9 2.7 2.7 2.7 36-50 42 12.4 12.4 15.0 51-65 76 22.4 22.4 37.5 66-80 154 45.4 45.4 82.9 81 and above 58 17.1 17.1 100.0 Total 339 100.0 100.0 Income Frequency Percent Valid Percent Cumulative Percent Less than $ 20,000 47 17.6 17.6 17.6 $ 20,000-$ 59,999 147 55.1 55.1 72.7 $ 60,000-$ 99,999 61 22.8 22.8 95.5 $ 100,000-$ 149,999 11 4.1 4.1 99.6 $ 150,000 and above 1 .4 .4 100.0 Total 267 100.0 100.0 Leisure Lakes Surveys returned: 158 out of 425 Descriptive statistics N Minimum Maximum Mean Std. Deviation Interpretation of mean values Age 157 1.00 5.00 3.36 .91 51-65 years Education 150 1.00 6.00 3.36 1.66 Some college coursework Income 127 1.00 5.00 2.22 .83 $ 20,000-59,999 Education Frequency Percent Valid Percent Cumulative Percent Some high school or less 12 8.0 8.0 8.0 High school diploma or GED 47 31.3 31.3 39.3

PAGE 263

248 Education continued Frequency Percent Valid P ercent Cumulative Percent Some college coursework 39 26.0 26.0 65.3 Associates degree 10 6.7 6.7 72.0 Undergraduate degree 11 7.3 7.3 79.3 Graduate or profe ssional degree 31 20.7 20.7 100.0 Total 150 100.0 100.0 Employment Frequency Percent Valid Percent Cumulative Percent Full time employee 41 27.7 27.7 27.7 Part time employee 10 6.8 6.8 34.5 Retired 89 60.1 60.1 94.6 Homemaker 7 4.7 4.7 99.3 Unemployed 1 .7 .7 100.0 Total 148 100.0 100.0 Gender Frequency Percent Valid Percent Cumulative Percent Male 116 73.9 73.9 73.9 Female 41 26.1 26.1 100.0 Total 157 100.0 100.0 Ethnicity Frequency Percent Valid Percent Cumulative Percent White 144 94.7 94.7 94.7 Hispanic 2 1.3 1.3 96.1 African American 1 .7 .7 96.7 American Indian 1 .7 .7 97.4 Other 4 2.6 2.6 100.0 Total 152 100.0 100.0 Age Years Frequency Percent Valid Percent Cumulative Percent 20-35 5 3.2 3.2 3.2 36-50 21 13.3 13.4 16.6 51-65 54 34.2 34.4 51.0 66-80 67 42.4 42.7 93.6 81 and above 10 6.3 6.4 100.0 Total 157 99.4 100.0

PAGE 264

249 Income Frequency Percent Valid Percent Cumulative Percent Less than $ 20,000 17 10.8 13.4 13.4 $ 20,000-$ 59,999 78 49.4 61.4 74.8 $ 60,000-$ 99,999 21 13.3 16.5 91.3 $ 100,000-$ 149,999 9 5.7 7.1 98.4 $ 150,000 and above 2 1.3 1.6 100.0 Total 127 80.4 100.0

PAGE 265

250 WILDFIRE EDUCATION PROGRAMS AND PERCEPTION OF SOCIAL CAPITAL Participation in any wildfire education program Sum of Squares df Mean Square F Sig. Between Groups 545.46 1 545.46 4.19 .04 Within Groups 159879.04 1229 130.09 Total 160424.50 1230 Viewing electronic media Sum of Squares df Mean Square F Sig. Between Groups 1661.36 1 1661.36 12.66 .00 Within Groups 150565.72 1147 131.27 Total 152227.07 1148 Receiving printed information Sum of Squares df Mean Square F Sig. Between Groups 3428.39 1 3428.39 26.96 .00 Within Groups 152115.09 1196 127.19 Total 155543.47 1197 Attending events Sum of Squares df Mean Square F Sig. Between Groups 3591.71 1 3591.71 28.11 .00 Within Groups 153463.32 1201 127.78 Total 157055.03 1202 Participation in Educational Activities Television news: One-way ANOVA Sum of Squares df Mean Square F Sig. Between Groups 959.70 1 959.70 7.27 .01 Within Groups 147264.72 1115 132.08 Total 148224.41 1116 Radio news: One-way ANOVA Sum of Squares df Mean Square F Sig. Between Groups 573.97 1 573.97 4.33 .04 Within Groups 145473.22 1097 132.61 Total 146047.20 1098

PAGE 266

251 Handout: One-way ANOVA Sum of Squares df Mean Square F Sig. Between Groups 901.83 2 450.92 3.67 .03 Within Groups 55309.84 450 122.91 Total 56211.68 452 Newspaper: One-way ANOVA Sum of Squares df Mean Square F Sig. Between Groups 1433.05 2 716.53 5.85 .00 Within Groups 54502.70 445 122.48 Total 55935.75 447 Meeting: One-way ANOVA Sum of Squares df Mean Square F Sig. Between Groups 3674.62 2 1837.31 15.03 .00 Within Groups 23469.57 192 122.24 Total 27144.19 194 Prescribed burn demonstr ation: One-way ANOVA Sum of Squares df Mean Square F Sig. Between Groups 872.92 2 436.46 3.19 .044 Within Groups 25766.23 188 137.05 Total 26639.15 190 Picnic: One-way ANOVA Sum of Squares df Mean Square F Sig. Between Groups 869.35 2 434.68 3.18 .04 Within Groups 25308.60 185 136.80 Total 26177.96 187 Effect of Learning, Using, and Interacting with the Information Receiving handout: Pearson correlation coefficient Social capital Handout Learn Interact Use Social capital Pearson Correlation 1.00 .01 .03 .17* .10 Sig. (2-tailed) . .94 .70 .03 .18 N 172 172 172 172 172 Handout Pearson Correlation .01 1.000 .07 -.27** -.13 Sig. (2-tailed) .94 . .38 .000 .10 N 172 172 172 172 172 Learn Pearson Correlation .03 .07 1.000 .02 .04 Sig. (2-tailed) .70 .38 . .84 .61 N 172 172 172 172 172

PAGE 267

252 Receiving handout continued Social capital Handout Learn Interact Use Interac t Pearson Correlation .17* -.27** .02 1.000 .59** Sig. (2-tailed) .03 .00 .84 . .00 N 172 172 172 172 172 Use Pearson Correlation .10 -.13 .04 .59** 1.00 Sig. (2-tailed) .18 .10 .61 .00 . N 172 172 172 172 172 * Correlation is significant at the 0.05 level (2-tailed). ** Correlation is significant at the 0.01 level (2-tailed). Receiving newspaper: Pearso n correlation coefficient Social capital Newspaper Learn Interact Use Social capital Pearson Correlation 1.000 -.05 .02 .20* .03 Sig. (2-tailed) . .605 .80 .02 .75 N 137 137 137 137 137 Newspaper Pearson Correlation -.05 1.00 .12 -.12 -.15 Sig. (2-tailed) .60 . .16 .17 .09 N 137 137 137 137 137 Learn Pearson Correlation .02 .12 1.00 .15 .20* Sig. (2-tailed) .80 .16 . .07 .02 N 137 137 137 137 137 Interact Pearson Correlation .20* -.12 .15 1.00 .45** Sig. (2-tailed) .02 .17 .07 . .00 N 137 137 137 137 137 Use Pearson Correlation .03 -.15 .20* .45** 1.00 Sig. (2-tailed) .75 .09 .02 .00 . N 137 137 137 137 137 * Correlation is significant at the 0.05 level (2-tailed). ** Correlation is significant at the 0.01 level (2-tailed). Effect of Interaction on Perception of Social Capital Electronic index Print index Event index Social capital Electronic index Pearson Correlation 1.00 .46** .25** .11 Sig. (2-tailed) . .00 .01 .26 N 114 114 114 114 Print index Pearson Correlation .46** 1.00 .51** .21* Sig. (2-tailed) .00 . .00 .03 N 114 114 114 114 Event index Pearson Correlation .25** .51** 1.00 .35** Sig. (2-tailed) .01 .00 . .00 N 114 114 114 114

PAGE 268

253 Effect of interaction on perception of social capital continued Electronic Print index index Event index Social capital Social capital Pearson Correlation .11 .21* .35** 1.00 Sig. (2-tailed) .26 .03 .00 . N 114 114 114 114 ** Correlation is significant at the 0.01 level (2-tailed). * Correlation is significant at the 0.05 level (2-tailed).

PAGE 269

254 WILDFIRE PREPAREDNESS AND PE RCEPTION OF SOCIAL CAPITAL One-way ANOVA: Steps to reduce risk of wildfire and perception of social capital Sum of Squares df Mean Square F Sig. Between Groups 970.83 1 970.83 7.48 .01 Within Groups 158425.65 1221 129.75 Total 159396.48 1222 Influence of activities to reduce risk of wildfires and perception of social capital Electronic information Television story Written information Attending event Social capital Electronic information Pearson Correlation 1.000 .42** .40** .37** .15** Sig. (2-tailed) . .00 .00 .00 .00 N 665 664 664 664 665 Television story Pearson Correlation .42** 1.00 .31** .18** .09* Sig. (2-tailed) .00 . .00 .00 .02 N 664 664 663 663 664 Written information Pearson Correlation .40** .31** 1.00 .51** .15** Sig. (2-tailed) .00 .00 . .00 .00 N 664 663 664 663 664 Attending event Pearson Correlation .37** .18** .51** 1.00 .21** Sig. (2-tailed) .00 .00 .00 . .00 N 664 663 663 664 664 Social capital Pearson Correlation .15** .09* .15** .21** 1.00 Sig. (2-tailed) .00 .02 .00 .00 . N 665 664 664 664 665 ** Correlation is significant at the 0.01 level (2-tailed). * Correlation is significant at the 0.05 level (2-tailed). Influence of activities to reduce risk of wildfires and perception of social capital Social capital Neighbor Expert Friends Social capital Pearson Correlation 1.00 .20** .15** .29** Sig. (2-tailed) . .00 .00 .00 N 665 665 661 663 Neighbor Pearson Correlation .20** 1.00 .47** .57** Sig. (2-tailed) .00 . .00 .00 N 665 665 661 663 Expert Pearson Correlation .15** .47** 1.00 .50** Sig. (2-tailed) .00 .00 . .00

PAGE 270

255 Influence of activities to reduce risk of wildfires and perception of social capital continued Social capital Neighbor Expert Friends N 661 661 661 659 Friends Pearson Correlation .29** .57** .50** 1.00 Sig. (2-tailed) .00 .00 .00 . N 663 663 659 663 ** Correlation is significant at the 0.01 level (2-tailed). Influence of activities to reduce risk of wildfires and perception of social capital Social capital Community leader Workday FireWise program Social capital Pearson Correlation 1.00 .27** .19** .21** Sig. (2-tailed) . .00 .00 .00 N 665 663 665 662 Community leader Pearson Correlation .27** 1.00 .68** .63** Sig. (2-tailed) .00 . .00 .00 N 663 663 663 660 Workday Pearson Correlation .19** .68** 1.00 .63** Sig. (2-tailed) .00 .00 . .00 N 665 663 665 662 FireWise program Pearson Correlation .21** .63** .63** 1.00 Sig. (2-tailed) .00 .00 .00 . N 662 660 662 662 ** Correlation is significant at the 0.01 level (2-tailed).

PAGE 271

256 WILDFIRE EDUCATION PROGRAMS AND WILDFIRE PREPAREDNESS Participation in wildfire education program s and taking steps to reduce risk of wildfires: Chi square analysis Case processing summary Cases Valid Missing Total N Percent N Percen t N Percent Participation in education programs * steps to reduce risk of wildfire 1215 100.0% 0 .0% 1215 100.0% Crosstabulation Steps to reduce wildfire ris k Total 1.00 2.00 Participation in educational programs 1.00 112 163 275 2.00 195 745 940 Total 307 908 1215 Chi-Square tests Value df Asymp. Sig. (2sided) Exact Sig. (2sided) Exact Sig. (1sided) Pearson Chi-Square 45.00 b 1 .00 Continuity Correction a 43.94 1 .00 Likelihood Ratio 42.01 1 .00 Fisher's Exact Test .00 .00 Linear-by-Linear Association 44.95 1 .00 N of Valid Cases 1215 a Computed only for a 2x2 table b 0 cells (.0%) have expect ed count less than 5. The mini mum expected count is 69.49. Symmetric measures Value Approx. Sig. Nominal by Nominal Phi .19 .00 Cramer's V .19 .00 N of Valid Cases 1215 a Not assuming the null hypothesis. b Using the asymptotic standard error assuming the null hypothesis.

PAGE 272

257 Viewing electronic media and taking steps to reduce risk of wildfires: Chi square analysis Case processing summary Cases Valid Missing Total N Percen t N Percen t N Percent Viewing electronic media *taking steps to reduce wildfire risk 1134 93.3% 81 6.7% 1215 100.0% Crosstabulation Steps to reduce wildfire risk Total 1.00 2.00 Viewing electronic media 1.00 108 187 295 2.00 179 660 839 Total 287 847 1134 Chi-Square tests Value df Asymp. Sig. (2-sided) Exact Sig. (2sided) Exact Sig. (1sided) Pearson Chi-Square 26.94 b 1 .00 Continuity Correction a 26.14 1 .00 Likelihood Ratio 25.66 1 .00 Fisher's Exact Test .00 .00 Linear-by-Linear Association 26.92 1 .00 N of Valid Cases 1134 a Computed only for a 2x2 table b 0 cells (.0%) have expect ed count less than 5. The mini mum expected count is 74.66. Symmetric measures Value Approx. Sig. Nominal by Nominal Phi .15 .00 Cramer's V .15 .00 N of Valid Cases 1134 a Not assuming the null hypothesis. b Using the asymptotic standard error assuming the null hypothesis.

PAGE 273

258 Receiving Printed Information and taking steps to reduce risk of wildfires: Chi square analysis Case processing summary Cases Valid Missing Total N Percen t N Percen t N Percent Receiving printed information * taking steps to reduce wildfire risk 1183 97.4% 32 2.6% 1215 100.0% Crosstabulation Steps to reduce wildfire ris k Total 1.00 2.00 Receiving printed information 1.00 233 525 758 2.00 65 360 425 Total 298 885 1183 Chi-Square tests Value df Asymp. Sig. (2sided) Exact Sig. (2-sided) Exact Sig. (1sided) Pearson Chi-Square 34.47 b 1 .00 Continuity Correction a 33.66 1 .00 Likelihood Ratio 36.43 1 .00 Fisher's Exact Test .00 .00 Linear-by-Linear Association 34.44 1 .00 N of Valid Cases 1183 a Computed only for a 2x2 table b 0 cells (.0%) have expected count less than 5. The mini mum expected count is 107.06. Symmetric measures Value Approx. Sig. Nominal by Nominal Phi .17 .00 Cramer's V .17 .00 N of Valid Cases 1183 a Not assuming the null hypothesis. b Using the asymptotic standard error assuming the null hypothesis.

PAGE 274

259 Attending events and taking steps to redu ce risk of wildfires: Chi square analysis Case processing summary Cases Valid Missing Total N Percen t N Percen t N Percent Attending events * taking steps to reduce wildfire risk 1188 97.8% 27 2.2% 1215 100.0% Crosstabulation Steps to reduce wildfire ris k Total 1.00 2.00 Attending events 1.00 283 731 1014 2.00 15 159 174 Total 298 890 1188 Chi-Square tests Value df Asymp. Sig. (2-sided) Exact Sig. (2sided) Exact Sig. (1sided) Pearson Chi-Square 29.40 b 1 .00 Continuity Correction a 28.39 1 .00 Likelihood Ratio 35.34 1 .00 Fisher's Exact Test .00 .00 Linear-by-Linear Association 29.38 1 .00 N of Valid Cases 1188 a Computed only for a 2x2 table b 0 cells (.0%) have expect ed count less than 5. The mini mum expected count is 43.65. Symmetric measures Value Approx. Sig. Nominal by Nominal Phi .16 .00 Cramer's V .16 .00 N of Valid Cases 1188 a Not assuming the null hypothesis. b Using the asymptotic standard error assuming the null hypothesis.

PAGE 275

260 Effect of learning, using and interacting with the information Viewing TV news and taking steps to reduce risk of wildfire Learn: TV news Use: TV news Interact: TV news Steps to reduce wildfire risk Learn: TV news Pearson Correlation 1.00 .16** .23** .14** Sig. (2-tailed) . .00 .00 .00 N 522 522 522 522 Use: TV news Pearson Correlation .16** 1.00 .48** .18** Sig. (2-tailed) .00 . .00 .00 N 522 522 522 522 Interact: TV news Pearson Correlation .23** .48** 1.00 .40** Sig. (2-tailed) .00 .00 . .00 N 522 522 522 522 Steps to reduce wildfire risk Pearson Correlation .14** .18** .40** 1.00 Sig. (2-tailed) .00 .00 .00 . N 522 522 522 522 ** Correlation is significant at the 0.01 level (2-tailed). Receiving handout and taking step s to reduce risk of wildfire Learn: Handout Use: Handout Interact: Handout Steps to reduce wildfire risk Learn: Handout Pearson Correlation 1.00 .38** .38** .10 Sig. (2-tailed) . .00 .00 .18 N 205 205 205 205 Use: Handout Pearson Correlation .38** 1.00 .63** .05 Sig. (2-tailed) .00 . .00 .47 N 205 205 205 205 Interact: Handout Pearson Correlation .37** .63** 1.00 .29** Sig. (2-tailed) .00 .00 . .00 N 205 205 205 205 Steps to reduce wildfire risk Pearson Correlation .10 .05 .29** 1.00 Sig. (2-tailed) .18 .47 .00 . N 205 205 205 205 ** Correlation is significant at the 0.01 level (2-tailed).

PAGE 276

261 Receiving newspaper and taking steps to reduce risk of wildfire Learn: Newspaper Use: Newspaper Interact: Newspaper Steps to reduce wildfire risk Learn: Newspaper Pearson Correlation 1.00 .29** .37** .16* Sig. (2-tailed) . .00 .00 .04 N 166 166 166 166 Use: Newspaper Pearson Correlation .29** 1.00 .51** .09 Sig. (2-tailed) .00 . .00 .24 N 166 166 166 166 Interact: Newspaper Pearson Correlation .37** .51** 1.00 .30** Sig. (2-tailed) .00 .00 . .00 N 166 166 166 166 Steps to reduce wildfire risk Pearson Correlation .16* .09 .29** 1.00 Sig. (2-tailed) .04 .24 .00 . N 166 166 166 166 ** Correlation is significant at the 0.01 level (2-tailed). * Correlation is significant at the 0.05 level (2-tailed). Attending a meeting and taking steps to reduce risk of wildfire Learn: Meeting Use: Meeting Interact: Meeting Steps to reduce wildfire risk Learn: Meeting Pearson Correlation 1.00 .41** .41** .10 Sig. (2-tailed) . .00 .00 .35 N 101 101 101 101 Use: Meeting Pearson Correlation .41** 1.00 .60** .04 Sig. (2-tailed) .00 . .00 .73 N 101 101 101 101 Interact: Meeting Pearson Correlation .41** .60** 1.00 .24* Sig. (2-tailed) .00 .00 . .02 N 101 101 101 101 Steps to reduce wildfire risk Pearson Correlation .10 .04 .24* 1.00 Sig. (2-tailed) .35 .73 .02 . N 101 101 101 101

PAGE 277

262 ** Correlation is significant at the 0.01 level (2-tailed). * Correlation is significant at the 0.05 level (2-tailed).

PAGE 278

263 PARTICIPATION IN COMMUNITY ACTIVITIES AND PERCEPTION OF SOCIAL CAPITAL Awareness of groups and perception of social capital Religious groups Sum of Squares df Mean Square F Sig. Between Groups 750.23 1 750.23 5.65 .02 Within Groups 130293.51 982 132.68 Total 131043.74 983 Homeowners Associations Sum of Squares df Mean Square F Sig. Between Groups 1339.02 1 1339.02 10.14 .00 Within Groups 129704.72 982 132.08 Total 131043.74 983 Sports groups Sum of Squares df Mean Square F Sig. Between Groups 1777.04 1 1777.04 13.50 .00 Within Groups 129266.70 982 131.64 Total 131043.74 983 Garden clubs Sum of Squares df Mean Square F Sig. Between Groups 3369.48 1 3369.48 25.92 .00 Within Groups 127674.26 982 130.02 Total 131043.74 983 Social groups Sum of Squares df Mean Square F Sig. Between Groups 5484.46 1 5484.46 42.89 .00 Within Groups 125559.28 982 127.86 Total 131043.74 983 Neighborhood committees Sum of Squares df Mean Square F Sig. Between Groups 6100.25 1 6100.25 47.95 .00

PAGE 279

264 Within Groups 124943.49 982 127.23 Total 131043.74 983 Civic groups Sum of Squares df Mean Square F Sig. Between Groups 4304.19 1 4304.19 33.35 .00 Within Groups 126739.55 982 129.06 Total 131043.74 983 Belonging to groups and perception of social capital Religious groups Sum of Squares df Mean Square F Sig. Between Groups 1547.56 1 1547.56 12.78 .00 Within Groups 94921.55 784 121.07 Total 96469.11 785 Homeowners associations Sum of Squares df Mean Square F Sig. Between Groups 6757.32 1 6757.32 59.05 .00 Within Groups 89711.79 784 114.43 Total 96469.11 785 Sports groups Sum of Squares df Mean Square F Sig. Between Groups 1653.38 1 1653.38 13.67 .00 Within Groups 94815.73 784 120.94 Total 96469.11 785 Garden clubs Sum of Squares df Mean Square F Sig. Between Groups 1125.22 1 1125.22 9.25 .00 Within Groups 95343.89 784 121.61 Total 96469.11 785 Social groups Sum of Squares df Mean Square F Sig.

PAGE 280

265 Between Groups 5667.57 1 5667.57 48.94 .00 Within Groups 90801.54 784 115.82 Total 96469.11 785

PAGE 281

266 Neighborhood committees Sum of Squares d f Mean Square F Sig. Between Groups 4437.67 1 4437.67 37.80 .00 Within Groups 92031.44 784 117.39 Total 96469.11 785 Civic groups Sum of Squares df Mean Square F Sig. Between Groups 3032.32 1 3032.32 25.44 .00 Within Groups 93436.79 784 119.18 Total 96469.11 785 Relationship between awareness of groups and belonging to groups Religious groups Case processing summary Cases Valid Missing Total N Percent N Percen t N Percent Awareness of religious groups* 786 Belong to religious groups 100.0% 0 .0% 786 100.0% Crosstabulation Belong to religious groups Total .00 1.00 Awareness of religious groups .00 177 7 184 1.00 348 254 602 Total 525 261 786 Chi-Square tests Value df Asymp. Sig. (2sided) Exact Sig. (2sided) Exact Sig. (1sided) Pearson Chi-Square 93.63 b 1 .00 Continuity Correction a 91.91 1 .00 Likelihood Ratio 119.90 1 .00 Fisher's Exact Test .00 .00 Linear-by-Linear Association 93.52 1 .00 N of Valid Cases 786

PAGE 282

267 a Computed only for a 2x2 table b 0 cells (.0%) have expect ed count less than 5. The mini mum expected count is 61.10. Symmetric measures Value Approx. Sig. Nominal by Nominal Phi .35 .00 Cramer's V .35 .00 N of Valid Cases 786 a Not assuming the null hypothesis. b Using the asymptotic standard error assuming the null hypothesis. Homeowners associations Case processing summary Cases Valid Missing Total N Percent N Percen t N Percent Awareness of homeowners association* Belong to homeowners association 786 100.0% 0 .0% 786 100.0% Crosstabulation Belong to homeowners association Total .00 1.00 Awareness of homeowners association .00 145 2 147 1.00 298 341 639 Total 443 343 786 Chi-Square tests Value df Asymp. Sig. (2-sided) Exact Sig. (2sided) Exact Sig. (1sided) Pearson Chi-Square 131.41 b 1 .00 Continuity Correction a 129.30 1 .00 Likelihood Ratio 172.76 1 .00 Fisher's Exact Test .00 .00 Linear-by-Linear Association 131.24 1 .00 N of Valid Cases 786 a Computed only for a 2x2 table b 0 cells (.0%) have expect ed count less than 5. The mini mum expected count is 64.15.

PAGE 283

268 Symmetric measures Value Approx. Sig. Nominal by Nominal Phi .41 .00 Cramer's V .41 .00 N of Valid Cases 786 a Not assuming the null hypothesis. b Using the asymptotic standard error assuming the null hypothesis. Sports groups Case processing summary Cases Valid Missing Total N Percent N Percen t N Percent Awareness of sports groups * 786 Belong to sports groups 100.0% 0 .0% 786 100.0% Crosstabulation Belong to sports groups Total .00 1.00 Awareness of sports groups .00 248 2 250 1.00 421 115 536 Total 669 117 786 Chi-Square tests Value df Asymp. Sig. (2-sided) Exact Sig. (2sided) Exact Sig. (1sided) Pearson Chi-Square 57.41 b 1 .00 Continuity Correction a 55.79 1 .00 Likelihood Ratio 80.71 1 .00 Fisher's Exact Test .00 .00 Linear-by-Linear Association 57.34 1 .00 N of Valid Cases 786 a Computed only for a 2x2 table b 0 cells (.0%) have expect ed count less than 5. The mini mum expected count is 37.21. Symmetric measures Value Approx. Sig. Nominal by Nominal Phi .27 .00 Cramer's V .27 .00

PAGE 284

269 N of Valid Cases 786 a Not assuming the null hypothesis. b Using the asymptotic standard error assuming the null hypothesis. Garden clubs Case processing summary Cases Valid Missing Total N Percent N Percen t N Percent Awareness of garden club* 786 Belong to garden club 100.0% 0 .0% 786 100.0% Crosstabulation Belong to garden club Total .00 1.00 Awareness of garden club .00 337 1 338 1.00 415 33 448 Total 752 34 786 Chi-Square tests Value df Asymp. Sig. (2-sided) Exact Sig. (2sided) Exact Sig. (1sided) Pearson Chi-Square 23.27 b 1 .00 Continuity Correction a 21.59 1 .00 Likelihood Ratio 30.77 1 .00 Fisher's Exact Test .00 .00 Linear-by-Linear Association 23.24 1 .00 N of Valid Cases 786 a Computed only for a 2x2 table b 0 cells (.0%) have expect ed count less than 5. The mini mum expected count is 14.62. Symmetric measures Value Approx. Sig. Nominal by Nominal Phi .17 .00 Cramer's V .17 .00 N of Valid Cases 786 a Not assuming the null hypothesis. b Using the asymptotic standard error assuming the null hypothesis.

PAGE 285

270 Youth groups Case processing summary Cases Valid Missing Total N Percent N Percen t N Percent Awareness of youth group * 786 Belong to youth group 100.0% 0 .0% 786 100.0% Crosstabulation Belong to youth group Total .00 1.00 Awareness of youth group .00 440 4 444 1.00 302 40 342 Total 742 44 786 Chi-Square tests Value df Asymp. Sig. (2sided) Exact Sig. (2-sided) Exact Sig. (1sided) Pearson Chi-Square 42.60 b 1 .00 Continuity Correction a 40.58 1 .00 Likelihood Ratio 46.73 1 .00 Fisher's Exact Test .00 .00 Linear-by-Linear Association 42.55 1 .00 N of Valid Cases 786 a Computed only for a 2x2 table b 0 cells (.0%) have expect ed count less than 5. The mini mum expected count is 19.15. Symmetric measures Value Approx. Sig. Nominal by Nominal Phi .23 .00 Cramer's V .23 .00 N of Valid Cases 786 a Not assuming the null hypothesis. b Using the asymptotic standard error assuming the null hypothesis. Ethnic groups Case processing summary Cases Valid Missing Total N Percent N Percen t N Percent

PAGE 286

271 Awareness of ethnic group* 786 Belong to ethnic group 100.0% 0 .0% 786 100.0% Crosstabulation Belong to ethnic group Total .00 1.00 Awareness of ethnic group .00 496 5 501 1.00 264 21 285 Total 760 26 786 Chi-Square tests Value df Asymp. Sig. (2-sided) Exact Sig. (2sided) Exact Sig. (1sided) Pearson Chi-Square 23.05 b 1 .00 Continuity Correction a 21.10 1 .00 Likelihood Ratio 22.42 1 .00 Chi-square tests continued Value df Asymp. Sig. (2-sided) Exact Sig. (2sided) Exact Sig. (1sided) Fisher's Exact Test .00 .00 Linear-by-Linear Association 23.02 1 .00 N of Valid Cases 786 a Computed only for a 2x2 table b 0 cells (.0%) have expect ed count less than 5. The minimum expected count is 9.43. Symmetric measures Value Approx. Sig. Nominal by Nominal Phi .17 .00 Cramer's V .17 .00 N of Valid Cases 786 a Not assuming the null hypothesis. b Using the asymptotic standard error assuming the null hypothesis. Social groups Case processing summary Cases Valid Missing Total N Percent N Percen t N Percent Awareness of social group* 786 100.0% 0 .0% 786 100.0%

PAGE 287

272 Belong to social group Crosstabulation Belong to social group Total .00 1.00 Awareness of social group .00 371 6 377 1.00 300 109 409 Total 671 115 786 Chi-Square tests Value df Asymp. Sig. (2-sided) Exact Sig. (2sided) Exact Sig. (1sided) Pearson Chi-Square 98.63 b 1 .00 Continuity Correction a 96.63 1 .00 Likelihood Ratio 118.53 1 .00 Fisher's Exact Test .00 .00 Linear-by-Linear Association 98.50 1 .00 N of Valid Cases 786 a Computed only for a 2x2 table b 0 cells (.0%) have expect ed count less than 5. The mini mum expected count is 55.16. Symmetric measures Value Approx. Sig. Nominal by Nominal Phi .35 .00 Cramer's V .35 .00 N of Valid Cases 786 a Not assuming the null hypothesis. b Using the asymptotic standard error assuming the null hypothesis. Neighborhood committees Case Processing Summary Cases Valid Missing Total N Percent N Percen t N Percent Awareness of neighborhood committee* Belong to neighborhood committee 786 100.0% 0 .0% 786 100.0% Crosstabulation Belong to neighborhood committee Total

PAGE 288

273 .00 1.00 Awareness of neighborhood committee .00 279 1 280 1.00 390 116 506 Total 669 117 786 Chi-Square tests Value df Asymp. Sig. (2-sided) Exact Sig. (2sided) Exact Sig. (1sided) Pearson Chi-Square 72.46 b 1 .00 Continuity Correction a 70.69 1 .00 Likelihood Ratio 103.28 1 .00 Fisher's Exact Test .00 .00 Linear-by-Linear Association 72.37 1 .00 N of Valid Cases 786 a Computed only for a 2x2 table b 0 cells (.0%) have expect ed count less than 5. The mini mum expected count is 41.68. Symmetric measures Value Approx. Sig. Nominal by Nominal Phi .30 .00 Cramer's V .30 .00 N of Valid Cases 786 a Not assuming the null hypothesis. b Using the asymptotic standard error assuming the null hypothesis. Civic groups Case processing summary Cases Valid Missing Total N Percent N Percen t N Percent Awareness of civic group * 786 Belong to civic group 100.0% 0 .0% 786 100.0% Crosstabulation Belong to civic group Total .00 1.00 Awareness of civic group .00 409 5 414 1.00 294 78 372

PAGE 289

274 Total 703 83 786 Chi-Square tests Value df Asymp. Sig. (2-sided) Exact Sig. (2sided) Exact Sig. (1sided) Pearson Chi-Square 81.00 b 1 .00 Continuity Correction a 78.93 1 .00 Likelihood Ratio 93.93 1 .00 Fisher's Exact Test .00 .00 Linear-by-Linear Association 80.90 1 .00 N of Valid Cases 786 a Computed only for a 2x2 table b 0 cells (.0%) have expect ed count less than 5. The mini mum expected count is 39.28. Symmetric measures Value Approx. Sig. Nominal by Nominal Phi .32 .00 Cramer's V .32 .00 N of Valid Cases 786 a Not assuming the null hypothesis. b Using the asymptotic standard error assuming the null hypothesis.

PAGE 290

275 Attending meetings and perception of so cial capital: Regression analysis Model summary Model R R Square Adjusted R Square Std. Error of the Estimate 1 .38 .15 .14 10.13 a Predictors: (Constant), meeting ANOVA Model Sum of Squares df Mean Square F Sig. 1 Regression 12974.11 1 12974.11 126.41 .00 Residual 76671.44 747 102.64 Total 89645.55 748 a Predictors: (Constant), meeting b Dependent Variab le: social capital Coefficients Unstandardized Coefficients Standardized Coefficients t Sig. Model B Std. Error Beta 1 (Constant) 68.72 .74 92.52 .00 Meeting 4.33 .39 .38 11.24 .00 a Dependent Variable: social capital Model predicting individual’s perception of social capital: Regression analysis Model summary Model R R Square Adjusted R Square Std. Error of the Estimate 1 .429 .18 .18 9.91 a Predictors: (Constant), inside community problems, work projec ts, community events ANOVA Model Sum of Squares df Mean Square F Sig. 1 Regression 16494.16 3 5498.05 55.99 .00 Residual 73151.38 745 98.19 Total 89645.55 748 a Predictors: (Constant), inside community problems, work projec ts, community events b Dependent Variable : social capital

PAGE 291

276

PAGE 292

277 Coefficients Unstandardized Coefficients Standardized Coefficients t Sig. Model B Std. Error Beta 1 (Constant) 63.61 1.05 60.65 .00 Community events 2.64 .44 .22 5.95 .00 Work projects 2.74 .62 .15 4.40 .00 Inside community problems 2. 60 .49 .20 5.31 .00 a Dependent Variable: social capital Community demographic variables and perception of social capital Univariate Analysis of variance: Ag e and perception of social capital Between-subjects factors N Age 20-35 years 76 36-50 years 198 51-65 years 336 66-80 years 283 81 and above 58 Tests of between-subjects effects Dependent Variable: Social capital Source Type III Sum of Squares df Mean Square F Sig. Corrected Model 5768.56 a 4 1442.14 12.36 .00 Intercept 3437848.81 1 3437848.81 29459.34 .00 Age 5768.56 4 1442.14 12.36 .00 Error 110396.40 946 116.70 Total 5685863.00 951 Corrected Total 116164.96 950 a R Squared = .050 (Adjusted R Squared = .046) Age group 20-35 years and perception of social capital Sum of Squares df Mean Square F Sig. Between Groups 811.40 1 811.40 6.68 .01 Within Groups 115353.56 949 121.55 Total 116164.96 950

PAGE 293

278 Age group 36-50 years and perception of social capital Sum of Squares df Mean Square F Sig. Between Groups 3877.97 1 3877.97 32.78 .00 Within Groups 112287.00 949 118.32 Total 116164.96 950 Age group 66-80 years and perception of social capital Sum of Squares df Mean Square F Sig. Between Groups 2383.38 1 2383.38 19.88 .00 Within Groups 113781.57 949 119.90 Total 116164.96 950 Univariate Analysis of variance: Empl oyment and perception of social capital Between-subjects factors N Employment Full-time employee 386 Part-time employee 61 Retired 458 Homemaker 36 Student 3 Unemployed 7 Tests of between-subjects effects Dependent Variable: Social capital Source Type III Sum of Squares df Mean Square F Sig. Corrected Model 5540.84 a 5 1108.17 9.47 .00 Intercept 375944.07 1 375944.07 3211.48 .00 Employment 5540.84 5 1108.17 9.47 .00 Error 110624.12 945 117.07 Total 5685863.00 951 Corrected Total 116164.96 950 a R Squared = .048 (Adjusted R Squared = .043) Full-time employee percepti on of social capital Sum of Squares df Mean Square F Sig. Between Groups 3797.16 1 3797.16 32.07 .00 Within Groups 112367.80 949 118.41 Total 116164.96 950

PAGE 294

279 Retired and perception of social capital Sum of Squares df Mean Square F Sig. Between Groups 5047.81 1 5047.81 43.11 .00 Within Groups 111117.15 949 117.09 Total 116164.96 950

PAGE 295

280 INTERACTION IN DISASTER EDUCATIO N PROGRAMS AND PERCEPTION OF SOCIAL CAPITAL Effect of variation in communities: Viewing electronic media Variables Entered/Removed Model Variables Entered Variables Removed Method 1 Student, some high school or less, homemaker, community 2, community 1, Part-time employee, undergraduate degree, electronic index, 51-65 years, as sociates degree, community 3, years, gender, income, 20-35 y ears, high school diploma or GED, community 5, influence of electronic information, community 6, unemployed, some college coursework, 66-80 years, community 4, 36-50 years, fulltime a . Enter a All requested variables entered. b Dependent variable: social capital Model Summary Model R R Square Adjusted R Square Std. Error of the Estimate 1 .40 a .16 .11 10.68 a Predictors: (Constant), St udent, some high school or less, homemaker, community 2, community 1, Part-time employee, undergraduat e degree, electronic index, 51-65 years, associates degree, community 3, years, gende r, income, 20-35 years, high school diploma or GED, community 5, influence of electronic information, community 6, unemployed, some college coursework, 66-80 years, community 4, 36-50 years, fulltime ANOVA Model Sum of Squares df Mean Square F Sig. 1 Regression 10238.59 25 409.54 3.59 .00 a Residual 54829.38 481 113.99 Total 65067.97 506 a Predictors: (Constant), St udent, some high school or less, homemaker, community 2, community 1, Part-time employee, undergraduat e degree, electronic index, 51-65 years, associates degree, community 3, years, gende r, income, 20-35 years, high school diploma or GED, community 5, influence of electronic information, community 6, unemployed, some college coursework, 66-80 years, community 4, 36-50 years, fulltime b Dependent Variable: Social capital

PAGE 296

281 Coefficients a Unstandardized Coefficients Standardized Coefficients t Sig. Model B Std. Error Beta 1 (Constant) 71.27 4.42 16.14 .00 Electronic Index .44 .33 .07 1.34 .18 Perception of influence of electronic media 1.47 .57 .12 2.57 .01 Community 1 -3.90 2.95 -.07 -1.32 .19 Community 2 -.41 2.17 -.01 -.19 .85 Community 3 -.95 2.06 -.03 -.46 .65 Community 4 .44 1.90 .02 .23 .82 Community 5 -5.72 2.19 -.17 -2.61 .01 Community 6 -2.32 1.90 -.08 -1.22 .22 Length of residence .02 .07 .02 .34 .74 Gender: Male 1.01 1.05 .04 .96 .34 Age: 20-35 years -1.39 3.20 -.04 -.44 .66 Age: 36-50 years 1.15 2.83 .04 .41 .69 Age: 51-65 years 1.99 2.56 .08 .78 .44 Age: 66-80 years 2.87 2.48 .11 1.16 .25 Some high school or less -5.09 3.22 -.07 -1.58 .12 High school diploma or GED -1.05 1.57 -.04 -.67 .51 Some college coursework -2.89 1.37 -.12 -2.11 .04 Associates degree -1.67 1.80 -.05 -.93 .36 Undergraduate degree -.23 1.63 -.01 -.14 .89 Income .10 .66 .01 .16 .88 Unemployed -9.85 6.45 -.07 -1.53 .12 Homemaker -7.07 2.62 -.14 -2.70 .01 Student -5.93 6.47 -.04 -.92 .36 Full-time employee -3.36 1.53 -.15 -2.20 .03 Part-time employee -5.42 2.05 -.12 -2.64 .01 a Dependent Variable: social capital Effect of variation in communities: Receiving printed information Variables Entered/Removed Model Variables Entered Variables Removed Method 1 Student, some high school or less, homemaker, print index, associates degree, 51-65 years, community 1, community 2, part-time, years, high school diploma or GED, gender, community 5, income, undergra duate degree, influence of . Enter

PAGE 297

282 written information, community 3, 20-35 years, community 6, unemployed, some college coursework, 66-80 years, community 4, 36-50 years, fulltime a a All requested variables entered. b Dependent Variable: social capital Model Summary Model R R Square Adjusted R Square Std. Error of the Estimate 1 .40 a .16 .114 10.68 a Predictors: (Constant), St udent, some high school or le ss, homemaker, print index, associates degree, 51-65 years, community 1, community 2, part-time, years, high school diploma or GED, gender, community 5, income, undergraduate degree, influence of written information, community 3, 20-35 y ears, community 6, unemployed, some college coursework, 66-80 years, community 4, 36-50 years, fulltime ANOVA Model Sum of Squares df Mean Square F Sig. 1 Regression 8030.23 25 321.21 3.42 .00 Residual 21024.85 224 93.86 Total 29055.08 249 a Predictors: (Constant), St udent, some high school or le ss, homemaker, print index, associates degree, 51-65 years, community 1, community 2, part-time, years, high school diploma or GED, gender, community 5, income, undergraduate degree, influence of written information, community 3, 20-35 y ears, community 6, unemployed, some college coursework, 66-80 years, community 4, 36-50 years, fulltime b Dependent Variable: social capital Coefficients a Unstandardized Coefficients Standardized Coefficients t Sig. Model B Std. Error Beta 1 (Constant) 70.95 5.64 12.57 .00 Print Index .36 .28 .08 1.28 .20 Influence of writte n information 1.92 .69 .18 2.80 .01 Community 1 -5.33 3.57 -.12 -1.49 .14 Community 2 2.06 3.39 .05 .61 .54 Community 3 -1.48 2.87 -.06 -.52 .61 Community 4 -.46 2.88 -.02 -.16 .88 Community 5 -5.94 3.25 -.16 -1.83 .07 Community 6 -.26 2.96 -.01 -.09 .93 Length of residence .06 .09 .05 .73 .47 Gender -2.15 1.38 -.10 -1.55 .12

PAGE 298

283 Age: 20-35 years 1.43 4.60 .03 .31 .76 Coefficients continued Unstandardized Coefficients Standardized Coefficients t Sig. Model B Std. Error Beta Age: 51-65 years 3.79 3.45 .17 1.10 .27 Age: 66-80 years 3.26 3.29 .14 .99 .32 Some high school or less 3.43 5.76 .04 .60 .55 High school diploma or GED -2.80 2.10 -.10 -1.33 .18 Some college coursework -3.34 1.79 -.14 -1.86 .06 Associates degree -1.35 2.36 -.04 -.57 .57 Undergraduate degree -3.13 2.20 -.10 -1.42 .16 Income 1.80 .81 .16 2.21 .03 Homemaker -7.09 3.84 -.12 -1.85 .07 Student -5.74 10.50 -.03 -.55 .59 Unemployed -21.90 7.63 -.18 -2.87 .01 Full-time employee -6.01 1.96 -.28 -3.06 .00 Part-time employee -2.81 2.77 -.06 -1.02 .31 a Dependent Variable: Social capital Effect of variation in communities: Attending events Variables Entered/Removed Model Variables Entered Variables Removed Method 1 Homemaker, some high school or less, 20-35 years, part-time employee, community 6, community 1, undergraduate degree, community 5, 66-80 years, gender, some college coursework, community 3, influence of events, income, years, 36-50 years, associates degree, community 2, high school diploma or GED event index, fulltime, community 4, 51-65 years, unemployed a . Enter a All requested variables entered. b Dependent Variable: social capital Model Summary Model R R Square Adjusted R Square Std. Error of the Estimate 1 .72 a .52 .38 9.50

PAGE 299

284 a Predictors: (Constant), Homemaker, some high school or less, 20-35 years, part-time employee, community 6, community 1, undergra duate degree, community 5, 66-80 years, gender, some college coursework, com3, influence of events, income, years, 36-50 years, associates degree, community 2, high school diploma or GED event index, fulltime, community 4, 51-65 years, unemployed ANOVA Model Sum of Squares df Mean Square F Sig. 1 Regression 8263.45 24 344.31 3.81 .00 a Residual 7672.74 85 90.27 Total 15936.19 109 a Predictors: (Constant), Homemaker, some high school or less, 20-35 years, part-time employee, community 6, community 1, undergra duate degree, community 5, 66-80 years, gender, some college coursework, community 3, influence of events, income, years, 3650 years, associates degree, community 2, high school diploma or GED event index, fulltime, community 4, 51-65 years, unemployed b Dependent Variable: social capital Coefficients a Unstandardized Coefficients Standardized Coefficients t Sig. Model B Std. Error Beta 1 (Constant) 43.03 14.67 2.93 .00 Community 1 -7.96 5.60 -.16 -1.42 .16 Community 2 .12 5.19 .00 .023 .98 Community 3 -7.65 4.81 -.22 -1.59 .12 Community 4 -5.45 4.59 -.22 -1.19 .24 Community 5 -4.14 5.94 -.08 -.70 .49 Community 6 -2.11 4.59 -.07 -.46 .65 Length of residence -.10 .13 -.07 -.80 .43 Event Index .65 .36 .18 1.78 .08 Influence of events 2.23 1.03 .22 2.16 .03 Gender .75 2.20 .03 .34 .73 Age: 20-35 years 4.45 8.14 .08 .55 .59 Age: 36-50 years -1.65 7.27 -.05 -.23 .82 Age: 51-65 years .64 6.54 .03 .10 .92 Age: 66-80 years 5.65 6.22 .23 .91 .37 Some high school or less -16.59 8.73 -.23 -1.90 .06 High school diploma or GED -2.48 3.43 -.07 -.72 .47

PAGE 300

285 Some college coursework -6.22 2.87 -.23 -2.17 .03 Associates degree 2.19 3.43 .06 .64 .53 Coefficients continued Unstandardized Coefficients Standardized Coefficients t Sig. Model B Std. Error Beta Income 3.50 1.24 .27 2.82 .01 Full-time employee 13.34 12.17 .53 1.10 .28 Part-time employee 25.09 13.03 .47 1.93 .06 Retired 19.96 12.40 .82 1.61 .11 Home maker 24.77 17.60 .20 1.41 .16 a Dependent Variable: social capital

PAGE 301

LIST OF REFERENCES Agrawal, A. and C.C. Gibson. 1999. Enchantment and disenchantment: The role of community in natural resource conservation. World Development 27 (4):629-649. Austin, D. 1994. Incorporating cognitive th eory into environm ental policymaking. The Environmental Professional 16 (3):262-274. Billing, S.H. 2000. Research on K-12 schoolbased Service Learning: The evidence builds. Phi Delta Kappan 82:658-664. Bolin, R., and L. Standford. 1998. The Northridge earthquake: Community-based approaches to unmet recovery needs. Disasters 22:21-38. Bouma, J.K. 2000. Values and the concepts of stre am "health": Implications for stewardship in a Seattle watershed : Master’s thesis: University of Washington. Bourdieu, P. 1987. The forms of capital. In Handbook of theory and research for the sociology of education , edited by John G. Richardson. Pp. 241-260.Westport, CT: Greenwood Press, Inc. Brown, B.J. 1979. Disaster preparedness. In Disaster preparedness and the United Nations: Advanced planni ng for disaster relief. Pp. 32-48. Geneva: Unitar. Brown, D.L., and D. Ashman. 1996. Participatio n, social capital, and inter sectoral problem-solving: African and Asian cases. World Development 24 (9):1467-1479. Buckland, J., and M. Rahman. 1999. Community-based disaster management during 1997 Red River Flood in Canada. Disasters 23 (2):174-191. Bullen, P., and J. Onyx. 2000. Measuring social capital in five communities in NSW. Journal of Applied Behavioral Science 36 (1):23-42. Coleman, J.S. 1988. Social capital in the creation of human capital. American Journal of Sociology 94:S95-S120. Coleman, J.S. 1990. Social capital. In Foundations of social theory , edited by James S. Coleman. Pp. 300-321. Harvard, MA: The Belknap Press of Harvard University Press. 286

PAGE 302

287 Collier, P. 2002. Social capital and poverty . Social Capital Initia tive Working Paper No. 4, The World Bank 1998 [accessed July, 05 2002]. Available from http://www.worldbank.org/poverty/scapital . DeFilippis, J. 2001. The myth of social capital in community development. Housing Policy Debate 12 (4):781-806. Dillman, D. A. 2000. Mail and telephone surveys: The tailored design method . New York, NY: John Wiley and Sons, Inc. Division of Forestry (DOF). 2003. Wildland Fire and Fire Prevention in Florida [Electronic]. Division of Forestry [a ccessed August 21, 2003]. Available from http://www.fl-dof.com/wildfire/index.html . Doran, D.K., C.K. Randall, and A.J. Long. 2004. Fire in the wildland-urban interface: Selecting and maintaining FireWise plants for landscaping: Circ ular 1445: One in the series Fire in the Wildland-Urban Inte rface of the School of Forest Resources and Conservation, Florida Cooperative Extension Service, Institute of Food and Agricultural Sciences, University of Florida. Dyer, C.L. 2002. Punctuated entropy as cu lture-induced change: The case of Exxon Valdez oil spill. In Catastrophe and culture: Th e anthropology of disaster , edited by A. Oliver-Smith. Pp 159-186. Santa Fe, NM: School of American Research Press. Dynes, R.R. 1993. Disaster reduction: The im portance of adequate assumptions about social organization. Sociological Spectrum 13:175-192. Ernst, J. and M. Monroe. 2004. The effects of environment-based education on student's critical thinking skills and dispos ition toward critical thinking. Environmental Education Research 10 (4):507-522. FireFree. 2004. FireFree program to support wildfire safety [Electronic] FireFree [accessed March 11, 2004]. Available from http://firefree.or g/ffreenew/index2a.htm . FireWise. 2001. FireWise communities workshop: Participant workbook . Quincy, MA: FireWise Communities. Flora, C.B., and J.L. Flora. 2004. Measuring and interpreting social capital on the community level: The difference and sim ilarities between social capital and entrepreneurial social infrastructure [Electronic] The World Bank Group [accessed February 04, 2004]. Available from http://www.worldbank.org/povert y/scapital/lib rary/flora2.pdf .

PAGE 303

288 Flora, C.B., J.L. Flora, and S. Fey. 2004. Rural communities: Legacy and change . 2nd ed. Boulder, CO: Westview press. Foskett, J.M. 1995. Social struct ure and social participation. American Sociological Review 20:431-438. Foth, M. 2003. Connectivity doesnot ensure comm unity: On social capital, networks, and communities of place. In ITIRA 2003 Conference track: Community informatics . Pp. 1-24. Fritz, C.E. 1961. Disasters. In Social problems , edited by R. Nisbet R. Merton. Pp. 651694. Beverly Hills, CA: Sage. Granovetter, M.S. 1973. The strength of weak ties. American Journal of Sociology 78 (6):1360-1380. Grooatert, C. Social capital: The missing link. Soci al Capital Initiative Working Paper No. 3. The World Bank 1998 [accessed July 05, 2002. Available from http://www.worldbank.org/poverty/scapital/ . Grootaert, C., and T. Bastelaer. Understanding and measuring soci al capital: A synthesis of findings and recommendations from the so cial capital initiative. Social Capital Initiative Working Paper No. 24 . The World Bank 2001 [accessed July 28, 2002. Available from http://lnweb18.worldbank.org/ESSD/sdvext.nsf/09ByDocName/SocialCapitalIniti ativeWorkingPaperSeries . Hagedorn, R., and S. Labovitz. 1968. Particip ation in community associations by occupation: A test of three theories. American Sociological Review 33 (2):272283. Hermansen, L.A., and E.A. Macie. 2002. Introduction. In Human influences on forest ecosystems: The Southern wild land-urban interface assessment , edited by E. A. Macie and L.A. Hermansen. Pp.1-10. Asheville, NC: U.S. Department of Agriculture, Forest Service, Southern Research Station, General Technical Report SRS-55. Howard, R. 2006. Bending towards justice: Serv ice-learning and social capital as means to the tipping point. Mentoring and Tutoring: Partnership in Learning 14 (1):515. Florida County Outline Map. Cartography. Indiana State University 2006 [accessed January 02, 2006. Available from website http://baby.indstate.edu/gga/gga_cart/basefl.gif

PAGE 304

289 Irvine, K.N. 1997. Stewardship in the management of pr ivate forests: Some psychological dimensions : Unpublished Master’s thesis : University of Michigan. Israel, G.D., L.J. Beaulieu, and G. Ha rtless. 2001. The influence of family and community social capital on educational achievement. Rural Sociology 66 (1):4368. Jacobson, S.K. 1999. Communication skills for conservation professionals. Washington D.C.: Island Press. Jakes, P., M. Monroe, and S. Agrawal. 2003. Waldo, Florida: Steps to improve community preparedness for wildfire. St. Paul, MN: U.S. Department of Agriculture, Forest Service, North Central Research Station. Jakes, P.J., K. Nelson, E. Lang, M. Monroe, S. Agrawal, L. Kruger, and V. Sturtevant. 2002. A model for improving community preparedness for wildfire. Homeowners, communities, and wildfire: Science findings from the National Fire Plan, Ninth International Symposium on Society and Resource Management. Pp 4-9. Minnesota: North Central Research Station, USDA Forest Service. Kaplan, S. 1982. On knowing the environment. In Humanscape . Pp. 54-58. Ann Arbor, MI: Ulrich's Books. Kasarda, J.D., and M. Janowitz. 1974. Co mmunity attachment in mass society. American Sociological Review 39 (3):328-339. Kearney, A.R., and R. De Young. 1995. A knowle dge-based intervention for promoting carpooling. Environment and Behavior 27 (5):650-678. Kearney, A.R., and S. Kaplan. 1997. Toward a methodology for the measurement of knowledge structures of ordinary people : The Conceptual Content Cognitive Map (3CM). Environment and Behavior 29 (5):579-617. Kearney, A.R., G. Bradley, R. Kaplan, and S. Kaplan. 1999. Stakeholders perspectives on appropriate forest management in the Pacific Northwest. Forest Science 45 (1):62-73. Kreps, G.A. 1984. Sociological i nquiry and disaster research. Annual Review of Sociology 10:309-330. Krishna, A., and E. Shrader. Social Capital Assessment Tool . The World Bank 1999 [accessed December 05, 2001. Available from http://wbln0018.worldbank.org/LAC/ LACInfoClient.nsf/1d7cc1630221ca478525 6cc5007a88fb/30d4d766e3e24bd485256ab900770dd6/$FILE/SocCap%20Assess %20Tool.pdf

PAGE 305

290 Leeder, S., and A. Dominello. 1999. Social capital and its rele vance to health and family policy. Australian and New Zealand Journal of Public Health 23 (4):424-429. Lieberman, G., and L. Hoody. 1998. Closi ng the achievement gap: Using the environment as an integrating contex t for learning. San Diego, CA: State Education and Environmental Roundtable. Long, A.J. 1999. Benefits of prescribed burning. Gainesville, FL: Univer sity of Florida. Long, A.J., D.D. Wade, and F.C. Bealls. 2005. Managing for fire in the interface: Challenges and opportunities. In Forests at the wildland-urban interface: Conservation and management , edited by Vince, S.W., M.L. Duryea, E.A. Macie and L.A. Hermansen. Pp201-223. Boca Raton, FL: CRC Press. Luna, E.M. 2001. Disaster mitigation and preparedness: The case of NGOs in the Philippines. Disasters 25 (3):216-226. Mobley, Hugh, E., and W.E. Balmer. 1981. Curre nt purposes, extent, and environmental effects of prescribed fire in the south. In Prescribed fire and wild life in southern forests, edited by Gene W. Wood. Pp. 1528. Georgetown, SC: The Belle W. Baruch Forest Science Institute of Clemson University. Monroe, M. 2002. Chapter 8: Fire. In Human influences on forest ecosystems: The southern wildland-urban interface assessment , edited by E.A. Macie and L.A. Hermansen. Pp. 133-150. Asheville, NC: USDA Forest Serv ice, Southern Research Station. Monroe, M., and A.J. Long. 1999. Protecting you r home from forest fires: Document FOR 59. One of a series of the School of Forest Resources and Conservation, Florida Cooperative Extension Service, Institute of Food and Agricultural Sciences, University of Florida . Monroe, M.C., S. Marynowski, and A.W. Bowers. 2000. Wildland Fire Education Toolkit. Gainesville, FL: University of Florida. National Climatic Data Center (NCDC). 2006. Billion dollar U.S. weather disasters [accessed Jan 20, 2006. Available from http://lwf.ncdc.noaa.gov/oa/reports/billionz.html . Nelson, K., M. Monroe, and J.F. Johnson. 2005. The Look of the land: Homeowner landscape management and wildfire prep aredness in Minnesota and Florida. Society and Natural Resources 18 (4):321-336. Organization of Economic Co-operation and Development (OECD). 2001. The well being of nations: The role of human and social capital. Paris: OECD.

PAGE 306

291 Oliver-Smith, A., and S.M. Hoffmann. 1999. The angry earth: disaster in anthropological perspective . New York, NY: Routledge. Penfield, R.D. 2002. The fundamentals of survey-based research. Gainesville, FL: University of Florida. Pope, J. 2003. Social capital and social capital indicators: A reading list. Working Paper Series No.1:10. Portes, A., and P. Landolt. 1996. The downside of social capital. The American Prospect 26 (May-June):18-21, 94. Portes, A. 1998. Social capital: Its origins and applications in modern sociology. Annual Review of Sociology 24:1-24. Purdue, D. 2001. Neighborhood governance: Lead ership, trust, and social capital. Urban Studies 38:2211-2224. Putnam, R.D. 1995. Bowling alone: America's declining social capital. Journal of Democracy 6 (1):65-78. Putnam, R.D. 2000. Bowling alone: The collapse and revival of American society . New York, NY: Simon and Schuster. Putnam, R.D., R. Leonar di, and R.Y. Nanetti. 1993. Making democracy work: Civic traditions in modern Italy . Princeton, NJ: Princeton University Press. Pyne, S.J. 1984. Introduction to wildl and fire: Fire management in the United States . New York, NY: John Wiley and Sons, Inc. Quarantelli, E.L. 1978. Disasters: Theory and research . London: Sage Publications. Quarantelli, E.L. 1997. Ten criteria for ev aluating the management of community disasters. Disasters 21 (1):39-56. Randall, J. 2001. Enhancing high school student writing skills with Florida biodiversity education. Master’s thesis: School of Fore st Resources and Conservation, University of Florida, Gainesville, FL. Rogers, E.M. 2003. Diffusion of Innovations . Fifth ed. New York, NY: Free Press. Roy, W.G., and P. Bonacich. 1998. Interlocking di rectorates and commun ities of interest among American railroad companies. American Sociological Review 53 (3):368379.

PAGE 307

292 Scaff, A.H. 1952. The effect of commuting on participation in community organizations. American Sociological Review 17 (2):215-220. Smith, S.S., and J. Kulynych. 2002. It may be social, but is it Ca pital? The social construction of social capital and politics of language. Politics and Society 30 (1):149-186. Stone, W. Measuring social capital: Towards a theoretically informed measurement framework for researching social ca pital in family and community life [Internet]. Australian Institute of Family Studies 2001 [accessed June 23, 2002. Available from www.aifs.org.au/ . Tanner, G.W., W.R. Marion, and J.J. Mullahe y. 2002. Understanding fire: Nature's land management tool. Wildlife Ecology a nd Conservation Department, Florida Cooperative Extension Service, Institute of Food and Agricultural Sciences, University of Florida. CIR 1018. Tourangeau, R., M.P. Couper, and F. Conard. 2004. Spacing, position, and order: Interpretive heuristics for visual features of survey questions. Public Opinion Quarterly 68 (3):368-393. United States Department of Agriculture Forest Service (USDA Forest Service). 1995. Fire ecology resource management education unit . Atlanta, GA: USDA Forest Service. Wade, D. D. 1989. A guide for prescribed fire in southern forests: Technical Publication R8-TP 11. Atlanta, GA: Forest Service Southern Region. Wall, E., G. Ferrazzi, and F. Schreyer. 1998. Getting the goods on social capital. Rural Sociology 63 (2):300-322. Wilkinson, K.P. 1991. The community in rural America . 1st ed. Westport, CT: Greenwood Press. Wilson, P.A. 1997. Building social capital: A learning agendas for the twenty-first century. Urban Studies 34 (5-6):745-760. Wood, H., D. Applegate, D. Akerman, and G. Whitney. 2005. Grand challenges for disaster reduction. Natural Hazards Observer 30 (2):1-3. The World Bank. 1998. In The initiative on defining, monitoring and measuring social capital: Overview and program descrip tion. Social Capital Initiative Working Paper No. 1 . Washington, D.C.: The World Bank.

PAGE 308

BIOGRAPHICAL SKETCH Shruti Agrawal was born on December 14, 1975 . She spent most of her youth in India, where she studied at Prakash Highe r Secondary School and Kendriya Vidyalaya (Space Applications Center), Ahmedabad. Her university experience includes an undergraduate degree in chemistry from St. Xavier’s College, Ahmedabad. She continued her education at the University School of Sciences, Ahmedabad, where she obtained a MS in organic chemistry. Following this sh e worked at the Center for Environmental Education, India, a non-governmental or ganization, where she was involved in developing K-12 educational materials on ener gy and Asian elephant conservation. She then moved to the University of Florida to pursue her PhD studies in 2001. She is married to Devang Parikh and currently resides in New Jersey. 293