<%BANNER%>

Predicting Beef Cattle Producer Wildlife Conservation Behavior and Participation in Cost-Share and Technical Assistance ...

Permanent Link: http://ufdc.ufl.edu/UFE0042087/00001

Material Information

Title: Predicting Beef Cattle Producer Wildlife Conservation Behavior and Participation in Cost-Share and Technical Assistance Programs
Physical Description: 1 online resource (134 p.)
Language: english
Creator: Willcox, Adam
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: agency, behavior, questionnaire, rancher, wildlife
Wildlife Ecology and Conservation -- Dissertations, Academic -- UF
Genre: Wildlife Ecology and Conservation thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Natural resource and wildlife agencies have been increasing efforts to involve private landowners in wildlife management with legislation that has created new or expanded existing programs that offer technical and financial assistance to landowners for approved wildlife management activities. One of the primary stakeholders, cattle ranchers, is well poised to benefit from these programs, especially if they can easily integrate wildlife management activities into daily cattle operations. One of the keys to a successful wildlife management program for ranchers is to understand how ranchers perceive wildlife management and assistance programs and explain why they choose to participate in them. It is also important that agency personnel working with these ranchers are able to accurately assess rancher perceptions of wildlife management to tailor programs that meet rancher needs and foster strong working and interpersonal relationships. I surveyed 1,634 ranchers and 52 agency employees in the southeastern states of AL, FL, MS, and GA, to understand their perceptions of wildlife benefits and problems, likability of certain species and groups of wildlife, and financial and technical assistance programs. Rancher and agency responses were similar for many wildlife benefits and likability of different types of wildlife, yet agency personnel overestimated rancher perceptions of wildlife problems and the economic benefits of wildlife enterprises. Agencies can improve current assistance programs for cattle ranchers based on both rancher and agency perceptions of wildlife management that include emphasizing programs that feature preferred species and integrating wildlife with routine cattle operations. The Theory of Planned Behavior proved a good model to explain why ranchers integrate wildlife management into cattle operations and participate in cost-share programs and technical workshops and field days. All three theory components, attitudes, subjective norms, and perceived behavioral control, were important to future participation in technical workshops and field days, but only attitudes and subjective norms predicted participation in cost-share programs and the integration of wildlife management into cattle operations. Past participation in programs, likeliness to work with different agencies, and several other demographic variables also proved important. By designing programs that utilize existing social networks and increasing positive attitudes about the programs, agencies should be well-suited to increase participation in wildlife habitat enhancement programs.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Adam Willcox.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Giuliano, William M.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2012-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042087:00001

Permanent Link: http://ufdc.ufl.edu/UFE0042087/00001

Material Information

Title: Predicting Beef Cattle Producer Wildlife Conservation Behavior and Participation in Cost-Share and Technical Assistance Programs
Physical Description: 1 online resource (134 p.)
Language: english
Creator: Willcox, Adam
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: agency, behavior, questionnaire, rancher, wildlife
Wildlife Ecology and Conservation -- Dissertations, Academic -- UF
Genre: Wildlife Ecology and Conservation thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Natural resource and wildlife agencies have been increasing efforts to involve private landowners in wildlife management with legislation that has created new or expanded existing programs that offer technical and financial assistance to landowners for approved wildlife management activities. One of the primary stakeholders, cattle ranchers, is well poised to benefit from these programs, especially if they can easily integrate wildlife management activities into daily cattle operations. One of the keys to a successful wildlife management program for ranchers is to understand how ranchers perceive wildlife management and assistance programs and explain why they choose to participate in them. It is also important that agency personnel working with these ranchers are able to accurately assess rancher perceptions of wildlife management to tailor programs that meet rancher needs and foster strong working and interpersonal relationships. I surveyed 1,634 ranchers and 52 agency employees in the southeastern states of AL, FL, MS, and GA, to understand their perceptions of wildlife benefits and problems, likability of certain species and groups of wildlife, and financial and technical assistance programs. Rancher and agency responses were similar for many wildlife benefits and likability of different types of wildlife, yet agency personnel overestimated rancher perceptions of wildlife problems and the economic benefits of wildlife enterprises. Agencies can improve current assistance programs for cattle ranchers based on both rancher and agency perceptions of wildlife management that include emphasizing programs that feature preferred species and integrating wildlife with routine cattle operations. The Theory of Planned Behavior proved a good model to explain why ranchers integrate wildlife management into cattle operations and participate in cost-share programs and technical workshops and field days. All three theory components, attitudes, subjective norms, and perceived behavioral control, were important to future participation in technical workshops and field days, but only attitudes and subjective norms predicted participation in cost-share programs and the integration of wildlife management into cattle operations. Past participation in programs, likeliness to work with different agencies, and several other demographic variables also proved important. By designing programs that utilize existing social networks and increasing positive attitudes about the programs, agencies should be well-suited to increase participation in wildlife habitat enhancement programs.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Adam Willcox.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Giuliano, William M.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2012-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042087:00001


This item has the following downloads:


Full Text

PAGE 1

1 PREDICTING BEEF CATTLE PRODUCER WILDLIFE CONSERVATION BEHAVIOR AND PARTICIPATION IN COST SHARE AND TECHNICAL ASSISTANCE PROGRAMS By ADAM S. WILLCOX A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORI DA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 Adam S. Willcox

PAGE 3

3 To ranchers, private landowners, and wildlifers who appreciate that the air we breathe, the water we drink and the nature we enjoy is largely provided free of charge by private lands May you continue supporting initiatives that foster Aldo Leopolds Land Ethic.

PAGE 4

4 ACKNOWLEDGMENTS This dissertation could not have been realized without the assistance of many people. One of my main intellectual drivers throughout this fiveyear process was my graduate committee. Dr. Martha Monroe, through her teachings and mentoring, inspired and encouraged me to focus my research on conservation behavior as an applied science grounded in social theory to present unbiased information useful to conservation practitioners. Mr. Scott Sanders, through his deep practical understanding of private lands wildlife conservation from both agency and landowner perspectives, helped form ulate my specific research objectives, to make my results directly applicable to agency staff and ranchers. Dr. Glenn Israel inspired an appreciation for the process and rigor involved in quantitative social science research, and the simple joys researchers can experience by uncovering novel data through experimentation and twisting the knobs on statistical tests. Dr. Perran Ross was always there to remind me of the bigger picture goals to conserve wildlife and ecosystems by integrating rigorous biologi cal research into programs and policy. Finally, I thank my advisor, Dr. Bill Giuliano, for all the help and support he provided over years. His assistance extended well beyond being an excellent research mentor by enrolling me in programs like the Florid a Natural Resource Leadership Institute, including me in meetings with agencies and other stakeholders, helping me procure grants and contracts beyond my PhD research, and encouraging me to present my findings at local, regional, national, and international conferences. His open door policy and genuine concern for his students success and well being were integral in achieving my goals. In addition to my committee, I thank Mr. Jim Selph who helped arrange most of my rancher interviews and had a careers wo rth of experience with cattle ranchers and

PAGE 5

5 wildlife he shared with me. This project could not have been realized without the 52 agency staff who carefully completed an online survey and the ranchers I interviewed, those who assisted with survey drafts, an d the 1,093 who took the time to complete the survey, and for this, I express sincere gratitude. I also thank my research assistant Dixie Cline for her meticulous data entry and all the people who helped me administer this project including: Laura Hayes, Monica Lindberg, Caprice Mcrae, Genn Vann, Elaine Culpepper, Claire Williams, and Carl Williams. Special thanks are due to my parents Joanne and Richard Washington. The summers spent driving across the country to see nearly all of the nations national parks and the weekend visits to the Smithsonians various museums in Washington D.C., inspired a lifetime love of nature and learning. I also thank Patti and Nigel House for their continued encouragement and support along this educational journey. Finally I acknowledge the greatest thing higher education has brought to my life, my wife Emma Willcox. We met during our masters degrees and never looked back. From the rainforests of Cameroon and the Congo, up the white cliffs of Dover, through the plains of the Serengeti and mountains of Mahale, around the glaciers and forests of Alaska, to the sandhills, rivers, springs, and saltwater flats of Florida, we have shared some of the most beautiful places on the planet. I look forward to every day with you and cant wait to see what the future brings. This research was realized with funds from the Renewable Resources Extension Act and the Department of Wildlife Ecology and Conservation at the University of Florida.

PAGE 6

6 TABLE OF CONTENTS ACKNOWLEDGMENTS .................................................................................................. 4 page LIST OF TABLES ............................................................................................................ 9 ABSTRACT ................................................................................................................... 11 CHAPTER 1 INTRODUCTION .................................................................................................... 13 Importance of Cattle Producing Lands .................................................................... 13 Objectives ............................................................................................................... 15 2 CATTLE RANCHER AND CONSERVATION AGENCY PERCEPTIONS OF WILDLIFE MANAGEMENT AND ASSISTANCE PROGRAMS IN THE SOUTHEASTERN UNITED STATES ..................................................................... 17 Introduction ............................................................................................................. 17 Study Area .............................................................................................................. 21 Methods .................................................................................................................. 22 Rancher Survey ................................................................................................ 22 Agency Survey ................................................................................................. 24 Data Analyses .................................................................................................. 25 Results .................................................................................................................... 25 Response ......................................................................................................... 25 Wildlife Management Practices ........................................................................ 26 Agency Program Recommen dations ................................................................ 30 Improving wildlife management on ranches ............................................... 30 Wildlife program participation ..................................................................... 33 Improving wildlife management programs .................................................. 35 Improvements to technical workshops and field days ................................ 36 Rancher Attitudes and Perceptions of Programs .............................................. 36 Discussion .............................................................................................................. 37 3 EXPLAINING THE INTEGRATION OF WILDLIFE MANAGEMENT AND CATTLE OPERATIONS IN THE SOUTHEASTERN UNITED STATES ................. 50 Introduction ............................................................................................................. 50 Theoretical F ramework ........................................................................................... 51 Methods .................................................................................................................. 53 Rancher Survey ................................................................................................ 53 Data Analyses .................................................................................................. 54 Results .................................................................................................................... 56 Response, Demographics, and Ranch Characteristics .................................... 56

PAGE 7

7 Theory of Planned Behavior Components Analyses ........................................ 58 Behavioral intent ........................................................................................ 58 Attitudes ..................................................................................................... 58 Subjective norms ....................................................................................... 59 Perceived behavioral control ...................................................................... 59 Main Analyses .................................................................................................. 59 Discussion .............................................................................................................. 60 4 EXPLAINING CATTLE RANCHER PARTICIPATION IN COST SHARE FINANCIAL AND TECHNICAL ASSISTANCE PROGRAMS IN THE SOUTHEASTERN UNITED STATES ..................................................................... 72 Introduction ............................................................................................................. 72 Methods .................................................................................................................. 74 Rancher Survey ................................................................................................ 74 Data Analyses .................................................................................................. 76 Results .................................................................................................................... 78 Response, Dem ographics, and Ranch Characteristics .................................... 78 Theory of Planned Behavior Components Analyses ........................................ 79 Behavioral intent ........................................................................................ 79 Attitud es ..................................................................................................... 79 Subjective norms ....................................................................................... 80 Perceived behavioral control ...................................................................... 81 Main Analyses .................................................................................................. 81 Discussion .............................................................................................................. 82 5 EFFECTS OF TOKEN FINANCIAL INCENTIVES ON RESPONSE RATES AND ITEM NONRESPONSE FOR MAIL SURVEYS ...................................................... 98 Introduction ............................................................................................................. 98 Methods .................................................................................................................. 99 Results .................................................................................................................. 100 Discussion ............................................................................................................ 102 6 CONCLUSIONS ................................................................................................... 109 APPENDIX A SOUTHEAST CATTLE RANCHER WILDLIFE MANAGEMENT SURVEY ........... 112 B BIVARIATE CORRELATIONS OF ALL VARIABLES INCLUDED IN SOUTHEASTERN UNITED STATES RANCHERS LIKELINESS TO CONSIDER WILDLIFE MANAGEMENT WHEN CONDUCTING RANCH ACTIVITIES MODEL, 2009. .................................................................................. 124 C BIVARIATE CORRELATIONS OF ALL VARIABLES INCLUDED IN SOUTHEASTERN UNITED STATES RANCHERS LIKELINESS TO PARTICIPATE IN COST SHARE PROGRAMS, 2009. ......................................... 125

PAGE 8

8 D BIVARIATE CORRELATIONS OF ALL VARIABLES INCLUDED IN SOUTHEASTERN UNITED STATES RANCHERS LIKELINESS TO PARTICIPATE IN A VARIETY OF WORKSHOP TOPICS, 2009. ......................... 126 LIST OF REFERENCES ............................................................................................. 127 BIOGRAPHICAL SKETCH .......................................................................................... 134

PAGE 9

9 LIST OF TABLES Table page 2 1 Perceptions of ranchers and agency personnel of wildlife benefits on ranches in the southeastern United States 2009 ............................................................. 44 2 2 Rancher and agency personnel perceptions of wildlife problems on ranches in the southeastern United States 2009 ............................................................. 45 2 3 Rancher and agency personnel perceptions of wildlife likability on ranches in the southeastern United States 2009 ................................................................ 46 2 4 Rancher and agency personnel perceptions of how likely ranchers were to consider wildlife management when ranching in the southeastern U nited.S tates 2009 ............................................................................................ 47 2 5 Rancher and agency personnel perceptions of how interested or likely cattle ranchers are to participate in financial assistance programs and selected workshop or field day topics in the southeastern United States 2009 ................ 48 2 6 Cattle rancher agreement with statements about cost share programs, wildlife workshops and field days, and wildlife on their ranch in the southeastern United States 2009 ...................................................................... 49 3 1. Cattle rancher ( n = 1,093) likeliness to work with organizations over the next five years in the southeastern United States 2009 ............................................ 66 3 2 Cattle rancher ( n = 1,093) agreement with outcome evaluation and attitudinal strength statements about combining wildlife management with beef cattle operations i n the southeastern United States 2009 ........................................... 67 3 3 Cattle rancher ( n = 1,093) agreement with normative belief and motivation to comply statements about combining wildlife management with beef cattl e operations in the southeastern United States 2009 ........................................... 68 3 4 Cattle rancher ( n = 1,093) agreement with control belief and control power statements about combining wildlife management with beef cattle operations in the southeastern United States 2009 ............................................................. 69 3 5 Forced hierarchical regression model explaining how likely cattle ranchers ( n = 1,071) are to consider wildlife management in b eef cattle operations in the southeastern United States 2009 ...................................................................... 70 3 6 Final stepwise regression model explaining how likely cattle ranchers ( n = 1,071) are to consider wildlife management in beef cattle operations in the southeastern United States 2009a ..................................................................... 71

PAGE 10

10 4 1 Cattle rancher ( n = 1,093) agreement statements directly measuring attitudes towards cost share financial assistance programs and indirectly measuring attitudes towards technical workshops and field days through outcome evaluation and attitudinal strength statements in the southeastern United S tates ., 2009 ...................................................................................................... 88 4 2 Cattle rancher ( n = 1,093) agreement statements indirectly measuring subjective norms towards cost share financial assistance programs and workshops and field days through normative belief and motivation to comply statements in the southeastern United Stat es 2009 .......................................... 89 4 3 Cattle rancher ( n = 1,093) agreement statements indirectly measuring perceived behavioral control towards cost share financial assistance programs and workshops and field days through agreement with control belief and control power statements in the southeastern United States 2009 ... 91 4 4 Forced threestep hierarchical regression model explaining how likely cattle r anchers ( n = 1,093) are to participate in cost share financial assistance programs in the southeastern United States 2009 ............................................. 93 4 5 Forced threestep hierarchical explaining how likely cattle ranchers ( n = 1,093) are to participate in wildlife management workshops and field days in the southeastern United States 2009 ................................................................ 95 4 6 Final stepwise model explaining how likely cattle ranchers ( n = 1,093) are to participate in cost share financial assistance programs in the southeastern United States 2009a. .......................................................................................... 97 4 7 Final stepwise model explaining how likely cattle ranchers ( n = 1,09 3) are to participate in wildlife management workshops and field days in the southeastern United States 2009a ..................................................................... 97 5 1 Response rates and costs for wildlife questionnaires mailed to cattle ranc hers with and without a one dollar coin incentive ....................................... 104 5 2 Demographic and ranch size variables for wildlife questionnaires mailed to cattle ranchers with and without a one dollar coin incentiv e ............................. 105 5 3 Demographic and ranch size variables for the 2007 Census of Agriculture compared with wildlife questionnaires mailed to cattle ranc hers with and without a one dollar coin incentive .................................................................... 107 5 4 Demographic and threatened or endangered species question item nonresponse percentage comparisons for wildlife questionnaires mailed to cattle ran chers with and without a one dollar coi n incentive ............................. 108

PAGE 11

11 Abstract of Dissertation Presented to t he Graduate School of the University of Florida in Partial Fulfillment of the Requirements f or the Degree of Doctor o f Philosophy PREDICTING BEEF CATTLE PRODUCER WI LDLIFE CONSERVATION BEHAVIOR AND PARTICIPATION IN COST SHARE AND TECHNICAL ASSISTANCE PROGRAMS By Adam S. Willcox August 2010 Chair: William M. Giuliano Major: Wildlife Ecology and Conservation Natural resource and wildlife agencies have been increasing efforts to involve private landowners in wildlife management with legislation that has created new or expanded existing programs that offer technical and financial assistance to landowners for approved wildlife management activities. One of the primary stakeholders, cattle ranchers, is well poised to benefit from these programs, especially if they can easily integrate wildlife management activities into daily cattle operations. One of the keys to a successful wildlife management program for ranchers is to understand how ranchers perceive wildlife management and assistance programs and explain why they choose to participate in them. It is also important that agency personnel working with these ranchers are able to accurately assess rancher perceptions of wildlife management to tailor programs that meet rancher needs and foster strong working and interpersonal relationships. I surveyed 1, 634 ranchers and 52 agency employees in the southeastern states of AL, FL, MS, and GA, to understand their perceptions o f wildlife benefits and problems, likability of certain species and groups of wildlife, and financial and technical assistance

PAGE 12

12 programs. Rancher and agency responses were similar for many wildlife benefits and likability of different types of wildlife, yet agency personnel overestimated rancher perceptions of wildlife problems and the economic benefits of wildlife enterprises. Agencies can improve current assistance programs for cattle ranchers based on both rancher and agency perceptions of wildlife management that include emphasizing programs that feature preferred species and integrating wildlife with routine cattle operations. The Theory of Planned Behavior proved a good model to explain why ranchers integrate wildlife management into cattle operations and participate in cost share programs and technical workshops and field days. All three theory components, attitudes, subjective norms, and perceived behavioral control were important to future participation in technical workshops and field day s, but o nly attitudes and subjective norms predicted participation in cost share programs and the integration of wildlife management into cattle operations Past participation in programs, likeliness to work with different agencies, and several other demographic variables also proved important. By designing programs that utilize existing social networks and increasing positive attitudes about the programs, agencies should be well suited to increase participation in wildlife habitat enhancement programs

PAGE 13

13 CHAPTER 1 INTRODUCTION Importance of Cattle Producing Lands Starting with the Native Americans and gradually intensifying through European colonization, the most productive and accessible natural ecosystems in North America were the first to undergo agricultural t ransformation. When the government first reserved public lands for conservation, officials targeted productively marginal areas that were left unclaimed and unwanted by agriculturalists, meaning public lands were largely isolated fragments of low agricult ural value, with characteristics such as high elevation, little rainfall, or poor soil (Shands and Healy 1977, Scott et al. 2001). For example, early settlers found it difficult to cultivate most of the first areas set aside for conservation including the mountainous areas of Rocky Mountains National Park, the geothermal hotspots of Yellowstone, and the slopes of the Grand Canyon (National Parks Service, 2006). During the second wave of public lands acquisition, some areas containing human settlements wer e authorized as parks. Settlers in these areas were bought out or provided homesteads, but those who refused to leave were forcibly evicted and their buildings demolished (e.g. the mountaineers of Shenandoah National Park; Lambert 1979). As government a gencies no longer establish protected areas by forcibly removing landowners, they have turned to purchasing private lands to establish conservation reserves (e.g. Redwood National Park ; Bearss 1969). Unfortunately, as human populations increase, it becomes increasingly difficult for government agencies to purchase large continuous tracts for conservation due to high agricultural and development land values. In Florida, the state with the highest percentage of public lands in the Southeast, of 179 rare tax a, existing public lands do

PAGE 14

14 not adequately protect 56 species and it would take an estimated $8.2 billion to purchase the remaining 1.65 million ha needed to do so (Kautz and Cox 2001). Even if Florida realized this purchase, biodiversity management on t hese lands would cost an estimated $122 million/year (Kautz and Cox 2001), which is more than six times the $19 million/year currently expended by the Florida Fish and Wildlife Conservation Commission (FWC) to manage their conservation and recreational lands (FWC 2006). Similarly, national wildlife analyses indicate that for largescale wildlife conservation, current public conservation lands are too few, of low quality, or too fragmented to conserve both common and rare species (Scott et al. 2001). Purch ase and management of public lands will continue to play an important role in wildlife conservation, but a comprehensive wildlife reserve system is likely an unattainable goal. Therefore, it is integral that government wildlife agencies and nongovernment al organizations work with private landowners to improve wildlife management and conservation to fill gaps in reserved land coverage. The integration of agriculture and wildlife management on working lands is particularly challenging on the modern farm. T he mechanization of agriculture and large scale clean farming practices that promote monocultures and removal of all vegetation other than planted crops or forage do not support diverse or abundant wildlife populations (Vance 1976). Fortunately, some ag ricultural practices, specifically beef cattle production, maintain both native and improved grasslands for cattle, having the potential to integrate wildlife conservation and management practices with agricultural operations. Grassland habitats, including pastures, hay fields, and grazable woodlands, can support diverse communities of common and listed wildlife species due

PAGE 15

15 to the proliferation of graminoids and forbs (Willcox 2010) These types of vegetation are readily consumed by both livestock and wil dlife, attract diverse assemblages of insects and insectivores, and often provide other key habitat components These systems can be suitably managed for both wildlife and cattle though rotational grazing, fencing, stocking rate manipulation, and other forms of vegetation management (West 1993, Payne and Bryant 1994, Morrison and Humphrey 2001). Overwhelmingly, the majority of private land holdings in the United States are devoted to growing food and fiber for human consumption. Twenty percent of the land is used for crops, 26% for forage or range for livestock, and 29% for production forest with only 3% in urban development (Lubowski et al. 2002). Once areas such as grazable woodlands are included in the calculation, 326,000,000 ha (35%) of the United St ates is producing livestock forage, making it one of the most prominent land uses I f wildlife conservation agencies want to affect a large proportion of land by working with a specific stakeholder group beef cattle producers are a key target audience. A gencies should design, monitor, and adapt rancher wildlife conservation programs that meet the needs of both wildlife and ranchers to maximize effects. The social aspects of this necessary data can be procured through careful surveys of ranchers and private lands agency staff. As a human dimensions of wildlife researcher, the overall goal of my research was to explain rancher participation in wildlife management activities and the major rancher assistance programs offered by agencies to improve and increase wildlife management on cattle ranches. Objectives Since the initial research of cattle producer attitudes toward wildlife, little has been done to specifically address the wildlife conservation beliefs and practices of beef cattle

PAGE 16

16 producers in the Unite d States (Kellert 1979, 1980). Although some primary work on the biological wildlife effects of cost share and technical assistance programs has been realized ( Haufler and Ganguli 2007), aside from a descriptive economic analysis based on basic farm survey data (Lambert et al. 2006), little is known about the factors influencing participation in federal and state conservation programs. Additionally, aside from a study of agriculturalists and agency staff in Montana ( Saltiel and Irby 1998), little is known of the rancher and agency relationship and their perceptions of one another and wildlife management. Therefore, my objectives in this study were to: 1. Understand the perceptions and relationships of beef cattle producers and wildlife and natural resource a gencies have of one another and current programs (Chapter 2) 2. Explain why ranchers choose to integrate wildlife management into their cattle operations (Chapter 3). 3. Predict participation in financial and technical wildlife assistance programs for ranchers (Chapter 4) 4. Examine the effect small financial incentives have on the quality and quantity of data procured from ranchers on human dimensions of wildlife surveys (Chapter 5). Together, it is my hope that this exploratory research will be used by wildlife conservation and natural resource agencies to modify existing or design new targeted technical and financial assistance programs for integrating wildlife management and ranching while improve producer agency relations for increased impact and efficiency.

PAGE 17

17 CHAPTER 2 CATTLE RANCHER AND CONSERVATION AGENCY PERCEPTIONS OF WILDLIFE MANAGEMENT AND ASSISTANCE PROGRAMS IN THE SOUTHEASTERN UNITED STATES Introduction Many beef cattle producers have positive attitudes toward wildlife and frequently combine wildlife conservation and management objectives in their cattle operations. National surveys conducted in the late 1970s indicated that, in hypothetical situations on public rangelands, many cattle producers approved of reductions in cattle grazing allotments and w ild horse numbers to protect native wildlife ( Kellert 1980). They were also very knowledgeable about wildlife, were concerned with the ecological functions wildlife provide, had an affinity for wildlife and the outdoors, and did not have negative wildlife values. More recent studies indicate that 5155% of farmers and cattle producers report they manage for wildlife on their farms or ranches (Conover 1994, 1998). This management costs an estimated $2.5 billion and 120 million hours per year, with only 5% of this expenditure recouped through fee hunting and leasing. This suggests the majority of farmers and cattle producers value wildlife for personal enjoyment, aesthetics, or other unknown reasons over profit, and could indicate the development of land a nd wildlife stewardship ethics (Leopold 1959, Conover 1998). In 1998, the United States Department of Agriculture (USDA) Natural Resource Conservation Service (NRCS) extended technical and financial assistance conservation programs that historically focuse d on general environmental issues (e.g., Environmental Quality Incentives Program [EQIP]) to specifically address wildlife habitat (e.g., Wildlife Habitat Incentives Program [WHIP]) as part of the Agricultural Research, Extension, and Education Reform Act of 1998. These programs promote a mixture of financial and

PAGE 18

18 technical assistance for activities that improve environmental and wildlife management on agricultural lands. Financial assistance programs typically provide cost share assistance, whereby both t he farmer and government split the cost of an approved management activity such as planting native vegetation, fencing sensitive areas, restoring native habitat, etc. (approved activities are determined on a state or regional basis). Technical assistance programs vary, but typically include producing educational materials, conducting technical workshops, assisting with financial assistance program applications, conservation planning, individual technical advice, and program monitoring. To better facilitat e WHIP implementation and integrate wildlife management into EQIP at the state level, the NRCS hired additional biologists with wildlife expertise or funded private lands biologists within state fish and wildlife agencies (J. S. Sanders, Florida Fish and W ildlife Conservation Commission, personal communication). Although these programs are intended for all agriculturalists, beef cattle ranchers are well suited to utilize them as many cattle production activities, excluding the most intensive beef production operations, are compatible with wildlife management (Haufler and Ganguli 2007). Technical and financial assistance programs rely on agency personnel to actively recruit program applicants or assist self motivated landowners who apply independently. Thro ughout the application progress, applicants must work intensively with agency personnel to fulfill application criteria and complete forms. Therefore, the agency rancher relationship is key to the success of these assistance programs. Natural resource ag ency professionals have first hand experiences and perceptions of current programs, and thus serve as critical resources for program implementation,

PAGE 19

19 being well placed to suggest improvements to existing programs. Additionally, most ranchers have a deep understanding of their land and operation, and generally enjoy nature (Kellert 1980). Mutual respect, understanding, and knowledge transfer between agency professionals and ranchers should facilitate the design, modification, and implementation of programs that meet the needs of ranchers, conserve wildlife and the environment, and easily integrate into ranch management. When comparing agency perceptions of agricultural producers to actual producer perceptions, wildlife agency personnel sometimes misperceive producer attitudes and actions towards wildlife management (Saltiel and Irby 1998). These misperceptions could have wider repercussions if they become established as pervasive faulty assumptions that can negatively influence stakeholder interactions, wildlife decisions, and programming (Enck and Decker 1997). For example, a past pervasive assumption in wildlife agencies was a perceived dichotomous split between consumptive and nonconsumptive users of wildlife. Empirical data show this is not the case, as many primarily consumptive users participate in nonconsumptive activities and vice versa. These faulty assumptions influenced decisions by many wildlife agencies to segregate programs into game and nongame and manage certain public lands exclusively for game species, which could have adverse effects on nongame species. By recognizing this faulty assumption, most of the former species specific game management areas are now considered wildlife management areas, and the general habitat management of these areas benefits a variety of species and visitors with mixed and diversified interests (Enck and Decker 1997).

PAGE 20

20 By continually reexamining agency and stakeholder attitudes, values, and assumptions toward programs, activities, and each other assumptions of ag ency staff used in conservation program development and implementation can be examined and if necessary corrected. T hese assistance programs cannot be realized in absence of a strong agency rancher relationship. D ata can provide insight into program part icipation and draw attention to the state of interpersonal relationships between the stakeholder group and agency. It is therefore important to compare how agency employees perceive wildlife management on ranches with rancher wildlife attitudes, practices and needs to better understand similarities and differences in opinion. If differences exist, agencies should work to dispel faulty assumptions internally, improve their interpersonal and professional relationships with their clientele, and align themselves with rancher expectations and needs before designing and implementing conservation programs. Although much is known about rancher perceptions toward wildlife, studies in North America have focused on conflicts with wolves (Williams et al. 2002), cougars (Riley and Decker, 200 0 ), bears (Rice et al. 2007), threatened and endangered species (Conley et al. 2007), and wild ungulates (Irby et al. 1997). Rancher perceptions of wildlife other than those involved in conflicts have been studied in the past (Kel lert 1980), but in recent times, rancher perceptions have been imbedded in studies concerning agriculturalists in general (Conover 1994; 1998) or private landowners (Willcox et al. 2010b). The literature comparing agency and cattle producer perceptions is also scarce, and it has focused on monitoring western public rangelands (Fernandez Gimenez et al. 2005). To date, the only research published regarding natural resource agency personnel perceptions of cattle producers and ranchland wildlife management is

PAGE 21

21 a study of Montana agency personnel and farmers (Saltiel and Irby 1998). My study addresses a gap in the human dimensions of wildlife literature, being the first to directly compare rancher wildlife management perceptions with natural resource agency per sonnel perceptions of ranchers and ranchland wildlife management. I designed this exploratory study to address rancher and agency perceptions regarding 1) the benefits and problems associated with wildlife on ranches, 2) the types of wildlife ranchers like and disli ke, 3) wildlife management behaviors associated with integrating wildli fe management on active ranches, and 4) participation in cost share financial programs and technical workshops. I further present agency opinions on the strengths and weaknes ses of current programs and their suggestions on how to improve them as compared with actual rancher attitudes about general ranch wildlife management and the benefits of financial and technical assistance programs. Study Area I conducted my study in the S outhern Coastal Plain and Southeastern Plains of Omernicks Level III Ecoregions in the southeastern states of AL, GA, MS, and FL ( Commissio n for Environmental Cooperation 1997). These ecoregional divisions are suitable for human dimensions of wildlife research because they combine anthropogenic land uses with ecological and geologic features found in the landscape. They are also used by most federal and state wildlife and natural resource agencies and one of only two ecoregion classification systems in t he national atlas (USDI 2007; J. M. Tirpak, USFWS, personal communication). The Southern Coastal Plain and Southeastern Plains historically contained mixtures of pine and deciduous forests, woodlands, and native range. These flat, warm are as are conduciv e to agriculture; one

PAGE 22

22 of the major land uses is beef cattle production ( Commissio n for Environmental Cooperation 1997). Methods Rancher Survey I conducted a series of personal interviews with beef cattle producers and University of Florida (UF) Cooperative Extension Service faculty in spring 2008 to develop a self administered mail survey. The survey was designed based on these interviews and knowledge obtained from focus groups conducted as part of a concurrent study of private landowners in FL (Willcox et al. 2010b). I sent survey drafts to 10 human dimensions of agriculture, wildlife, and forestry experts, 5 wildlife agency personnel, and 5 ranchers in FL and GA for comment and testing. Drafts were revised and resent several times for comment before fi nalizing the survey for content validity The survey consisted of 35 numbered questions over 9 pages (Appendix A) An initial screening question was asked to confirm the respondent was an active cattle rancher. I asked questions related to wildlife bene fits and problems, types of wildlife, wildlife management behaviors, and participation in financial and technical assistance programs in batteries with a 5point scale using ordered responses (i.e., 1 = not a benefit to 5 = extreme benefit; 1 = not a probl em to 5 = extreme problem; 1 = extremely dislike to 5 = extremely like; 1 = never to 5 = all of the time; and 1 = strongly disagree to 5 = strongly agree; Dillman et al., 2009). Ranchers were also asked questions about land use, ranch characteristics, and demographics, which were compared with demographics from the 2007 Census of Agriculture to assess nonresponse error (USDA 2009).

PAGE 23

23 I purchased a sampling frame of 24,049 nondairy cattle producers from the Farm Index of CAS Inc. during September 2008 (www.c as online.com). The Florida sampling frame was augmented with lists from the UFs Cooperative Extension Service as the purchased frame was low (227 ranchers) compared with the 15,717 farms with beef cattle reported in the USDAs National Agricultural Stat istics Service 2002 Census of Agriculture (USDA, 2004). My sampling frame for Mississippi was also small, but I was unable to procure any additional lists to increase the sampling frame. I reduced the combined list by zip code to my study area in the Southern Coastal Plain Southeastern Plains of Omernicks Level III Ecoregions, resulting in 15,023 addresses (Commission for Environmental Cooperation 1997). I stratified Florida by north and south at approximately 29N as the agricultural systems and environment differ, the north having more pine timber and native forest and the south containing more planted and native range (Clouser et al. 2007; Willcox et al. 2010b). North and south Florida were later pooled as northern and southern ranchers were not diff erent on any variables unlike the previous studies involving landowners and agriculturalists. I drew random samples of 500 ranchers from each state and the two Florida strata, except for Mississippi where I included all 271 ranchers. An additional 500 ranchers were drawn from the Georgia sampling frame to conduct a financial incentive effect methodological experiment (Willcox et al. 2010a). My final sample resulted in 2,771 addresses. I conducted a 5wave mailing following the Tailored Design Method ( Dillman, 2009). Preletters, initial surveys, and reminder postcards were mailed in November 2008, replacement questionnaires in January 2009, and final replacement questionnaires in February 2009.

PAGE 24

24 Agency Survey To allow direct comparison of results, I des igned the natural resource agency survey to be nearly identical to the rancher survey. The only difference between the agency and rancher surveys for most questions was that I asked How do ranchers perceive the following. before the battery questi ons about wildlife benefits and problems, likability of different wildlife types, wildlife management activities, and participation in cost share financial assistance programs and technical workshops. Additionally, I asked agency personnel how important t hey felt wildlife are to ranchers and four openended questions on how to improve wildlife management on cattle ranches, why ranchers participate in cost share and technical assistance programs, to suggest ways to improve current wildlife management progra ms, and detail their experiences of successful and unsuccessful approaches used during wildlife technical assistance workshops and field days. I contacted individuals at the United States Fish and Wildlife Service (USFWS), NRCS, Alabama Department of Conservation and Natural Resources, Florida Fish and Wildlife Conservation Commission, Georgia Department of Natural Resources, and Mississippi Department of Wildlife, Fisheries, and Parks to procure an email sampling frame of agency employees working with beef cattle ranchers on private lands conservation programs. I crosschecked the sampling frame with contact lists found on agency websites, resulting in 71 email addresses. In October 2009, I surveyed agency personnel via the Internet using Survey Monkey (ww w.surveymonkey.com). I employed a 4wave emailing including an introductory letter with no survey link, an email with the survey link several days later, a second email with a survey link to nonrespondents one week later, and a final email with survey link to nonrespondents one week after that.

PAGE 25

25 Data Analyses Likelihood ratio ( LR ) was used to assess binary response (yes or no) wildlife management questions and nonresponse error by comparing my data with the 2007 Census of Agriculture (USDA 2009). Missing data were imputed for all other questions using the expectationmaximization function with 20 iterations and 0.001 convergence to reduce uncertainty from missing values and use completedata techniques ( Shafer and Graham 2002; SYSTAT 2007). I used principal components analysis with a Varimax rotation for exploratory factor analyses categorizing wildlife benefits, wildlife problems, likability of different wildlife, wildlife management behaviors, and topics for wildlife management workshops and field days. Extracted factor reliability was assessed using Chronbachs Analysis of variance with Bonferroni post hoc tests were used to compare factor scores among ranchers and agencies, blocking where appropriate to control interstate variation. Open ended questions were coded by the researcher and analyzed for prevailing themes reported using a combination of summaries and exemplary quotations (Bernard 2000). I used SYSTAT 12 for all analyses and concluded statistical significance at P All data were tested for normality and homogeneity of variance. Those violating test assumptions were rank transformed prior to analysis (Conover 1980). Results Response Of the 2,771 surveys mailed to ranchers 1,634 responded and my adjusted response rates were: overall = 59%, AL = 57%, FL = 72%, MS = 56%, and GA = 49% (discounting undeliverable surveys). This resulted in 1,093 usable surveys after discounting those returned blank or respondents answering no to the screening

PAGE 26

26 question indicating they were no longer active ranchers. By comparing these results with the 2007 Census of Agriculture on corresponding demographic (i.e., gender, age, and ethnic group) and ranch size variables, I found nonresponse error in al l my samples (USDA 2009). Across all states, I had fewer <35 yr old ranchers ( LR = 80.73, df = 8, n = 69,008, P < 0.001) and ranches <57 ha ( LR = 1443.69, df = 11, n = 68,566, P < 0.001) than the 2007 Census. Additionally, Florida had more women ( LR = 23 .67, df = 1, n = 17,281, P < 0.001) and Mississippi had fewer other (than white) ethnic group respondents ( LR = 21.24, df = 5, n = 15,139, P < 0.001) than the 2007 Census. There were no differences for any other comparisons. This nonresponse error could actually be coverage error if my purchased sampling frame differed from the 2007 Census mailing list, but I cannot calculate coverage error from available data (Dillman et al. 2009). Regardless, care should be taken when extending my results to the entire population as my data best reflects the perceptions of producers whose ranches were 57 ha and were 35 yr s old. For the natural resource agency survey, 52 of 71 natural resource agency professionals responded. Discounting 5 undeliverable email addresses, 1 out of office reply, and 1 agency employee who did not work with ranchers, my adj usted response rate was 81%. It was not possible to calculate nonresponse error for natural resource agency personnel as their population demographics were not available, I did not ask demographic questions of my sample, and there is no comparable published literature on natural resource agency perceptions of cattle rancher wildlife management. Wildlife Management Practices Alabama (77%, n = 176), FL (79%, n = 464), GA (81%, n = 300), and MS (86%, n = 118) respondents indicated that their regular land management practices promoted

PAGE 27

27 wildlife and wildlife habitat with no difference among states ( LR = 3.99, df = 3, n = 1058, P = 0.263). I also detected no differences among AL (46%, n = 178), FL (52%, n = 466), GA (55%, n = 300), and MS (54%, n = 115) ranchers responding that they actively controlled for unwanted wildlife on their property ( LR = 4.03, df = 3, n = 1,059, P = 0.258). Agency personnel indicated they thought wildlife was important to ranchers on a 5point scale with ordered response categories ranging from 1 = very unimportant to 5 = very important (3.60 0.10 [ SE], n = 52) with no differences among states ( F = 0.53, df = 3, P = 0.662) or agency type ( F = 0.85, df = 2, P = 0.433). Mean rancher and agency personnel responses to questions about the benefits of wildlife on ranches ranged from not a benefit to serious benefit (Table 2 1). The top 5 benefits perceived by ranchers and agency personnel, all of which fell between slight benefit and serious benefit, were: enjoyment for future generations, maintain healthy land and environment, conservation, family wildlife watching, and family hunting opportunities. They expressed two wildlife benefit factors: 1) intrinsic benefits related to wildlife recreation by rancher families and the ecological or conservation value of wildlife on ranches, and 2) economic benefits related to wildlife recreational enterprises or mitigation and species banking. Rancher factors were largely consistent among states, except for Mississippi rancher s who realized higher intrinsic benefits from wildlife ( F = 6.41, df = 3, P < 0.001) than Alabama and Florida and higher economic benefits ( F = 3.91, df = 3, P = 0.008) than Florida. Agency personnel factors were consistent among states ( F = 0.53, df = 3, P = 0.667; F = 0.71, df = 3, P = 0.552) and agency types ( F = 0.311, df = 2, P = 0.734; F = 0.03, df = 2, P = 0.971) for both intrinsic and economic benefits, respectively. Agency personnel were not different from ranchers for intrinsic

PAGE 28

28 wildlife benefit s ( F = 3.76, df = 1, P = 0.051) but overestimated perceptions of economic wildlife benefits ( F = 145.25, df = 1, P < 0.001). Mean rancher and agency personnel responses to questions about wildlife problems on ranches ranged from not a problem to moderate problem (Table 2 2). Crop damage, livestock predation, and burrows in tunnels or fields were in the top five problems for both agency personnel and ranchers, but ranchers also had fence damage and spread unwanted plants as problems, whereas agency per sonnel had feed damage and constraints from endangered species regulations. Regardless, wildlife problems overall were consistent as only one factor was extracted. Rancher problems did not differ among states ( F = 1.29, df = 3, P = 0.277) and agency per sonnel did not differ among states ( F = 0.25, df = 3, P = 0.862) or agency type ( F = 0.09, df = 2, P = 0.910). Compared to ranchers, agency personnel generally overestimated rancher perceptions of wildlife problems on ranches ( F = 12.40, df = 1, P < 0. 001). When asked about the likability of certain species of wildlife on ranches, mean agency personnel and rancher responses ranged widely from extremely dislike to extremely like (Table 2 3). I found 3 factors that I labeled: 1) enjoyed, as these ani mals are commonly enjoyed in both consumptive and nonconsumptive activities and were generally liked or extremely liked; 2) damaging, as these animals are known to damage livestock operations or humans and were generally disliked or extremely disliked; an d 3) mixed, as although these animals average score indicated they were neither disliked or liked by ranchers, there are dual benefits and problems associated with many of them (e.g., gopher tortoises can generate income from species mitigation banking yet their burrows can damage livestock and pastures). The enjoyed wildlife

PAGE 29

29 factor differed among states for ranchers ( F = 7.54, df = 3, P < 0.001), with Mississippi > Alabama, Florida, and Georgia, and Florida > Alabama and Georgia. The rancher damaging wi ldlife factor also differed among states ( F = 10.22, df = 3, P < 0.001), with Florida > Alabama, Georgia, and Mississippi. The mixed wildlife factor did not differ for ranchers among states ( F = 1.14, df = 3, P = 0.332). The enjoyed wildlife factor dif fered among states for agencies ( F = 4.23, df = 3, P = 0.010), with Georgia agency personnel being lower than Mississippi. Both damaging ( F = 1.33, df = 3, P = 0.276) and mixed ( F = 0.58, df = 3, P = 0.634) wildlife factors did not differ among states and agency type did not differ for enjoyed ( F = 2.07, df = 2, P = 0.137), problem ( F = 0.76, df = 2, P = 0.473), or mixed ( F = 1.43, df = 2, P = 0.250). Agencies and ranchers did not differ on the enjoyed species factor ( F = 0.02, df = 1, P = 0.901), but agencies overestimated rancher perceptions of the likability of damaging wildlife ( F = 30.10, df = 1, P < 0.001). Mixed wildlife factor likability was underestimated by agency personnel compared to ranchers ( F = 80.01, df = 1, P < 0.001). Mean ranc her and agency responses to how often they considered wildlife management when conducting various ranch activities ranged from seldom to some of the time for all variables measured with 1 factor extracted (Table 2 4). Rancher ( F = 0.71, df = 3, P = 0 .574) perceptions of the factor did not differ among states and agency perceptions did not differ among states ( F = 1.85, df = 3, P = 0.151) or agency type ( F = 0.44, df = 2, P = 0.866). Agencies underestimated how often ranchers consider wildlife manag ement when compared with ranchers ( F = 4.48, df = 1, P = 0.035). Interest in participating in cost share financial assistance programs ranged from neither uninterested or interested to interested for both rancher and agency

PAGE 30

30 personnel responses (Table 2 5). Rancher perceptions of cost share program interest differed among states ( F = 8.44, df = 3, P < 0.001), with Florida interest being lower than Georgia and Mississippi. Agency perceptions of rancher interest in cost share programs did not differ am ong states ( F = 0.20, df = 3, P = 0.895) or agency type ( F = 0.09, df = 2, P = 0.916). Agencies overestimated rancher interest in cost share programs when compared to rancher responses ( F = 10.31, df = 1, P = 0.001). Over the last 10 years, 20% of AL, 24% of FL, 29% of GA, and 33% of MS ranchers attended at least one technical workshop or field day that included wildlife management. Agency and rancher perceptions of how likely ranchers were to participate in several workshop or field day topics ranged from unlikely to likely with 1 factor extracted (Table 2 5). Rancher responses did not vary among states ( F = 1.73, df = 3, P = 0.160), and agency perceptions did not differ among states ( F = 0.32, df = 2, P = 0.814) or agency type ( F = 2.10, df = 2, P = 0.133). Agency personnel overestimated rancher likeliness to participate in workshops and field days compared to ranchers ( F = 20.93, df = 1, P < 0.001). Agency Program Recommendations Improving wildlife management on ranches When asked about how to improve wildlife management on cattle ranches, agency personnel generally responded to the question in two ways, with 57% suggesting specific management practices and 54% describing ways to improve agency wildlife programs. The management practices suggested included: reducing stocking rates, controlling cattle access to sensitive areas like riparian zones, managing forest patches within pastures, replacing improved exotic with native grasses, increasing the amount of prescribed fire, and managing fenc erow vegetation. While most agency employees

PAGE 31

31 suggested one or several practices, one respondent encapsulated most of the different practices, stating: The manner in which cattle ranches have been and still are managed is highly intensive for the sole purpose of cattle production. Most lands are degraded because of this. However, if there is someway that the market system or supply demand for cattle production could consistently allow for moderate or even low intensity management, the difference could be spent on conservation benefits per unit of land. Focus should subsequently be placed on high priority conservation practices like: 1) limiting animal units per area using carrying capacity calculations specific for the lands placement within ecoregions and habitats (e.g., native plant communities) can be restored; 2) eliminating stream access, associated erosion, and associated fecal urinary contamination; 3) designing conservation quality shade sites for livestock using fencing by selecting small, ample sized interior hardwood or pine forest areas that have flat terrain and stable soils; 4) converting persistent or invasive exotic grasses to indigenous native grasses and forb plant mixtures for grazing and habitat for wildlife followed by consistent implementation of maintenance practices (note: in general this is a critical hurdle for realizing conservation benefits at a landscape scale); 5) rotational grazing efforts keeping successful avian recruitment in mind; and 6) managing fence rows and field bor ders with fallow field management techniques to create and maintain quality early successional habitat for wildlife.

PAGE 32

32 The suggested improvements to wildlife programs for cattle ranches included providing and increasing financial incentives, improving the ag ency rancher interpersonal relationship, redesigning programs based on cattle rancher inputs, promoting wildlife enterprises as an alternative ranch income, a rancher mentoring program whereby ranchers interested in wildlife management receive advice from peers with established wildlife programs, and increasing the amount of outreach and education. Two respondents criticized current programs, specifically cost share financial assistance programs. For example, one agency employee responded: There is plenty of potential to improve wildlife management on cattle ranches. The problem is getting the ranch managers to buy into the programs and not tie their hands with limitations of what they can and cannot do. If I were serious about improving management on cattle ranches then I would not limit practices based on poor cost sharing percentages. I should evaluate the actual cost for some of these practices and be realistic. Most cattle operations in Florida are losing money in strict cow calf operation. They are looking to diversify and find other ways to use their property to bring in some income to the farm. Now that could be sod, crops, citrus, hunting leases, selling guided hunts, and real estate. If I come to them with a cost share just to incorporate wi ldlife practices, and it costs more money out of pocket than they are getting back, it is unlikely that they are going to exploit this service. With the exception of a very few large landowner tracts that primary operation is something other than cow calf operation, few participate in wildlife management programs as far as cattle ranchers go.

PAGE 33

33 Wildlife program participation When asked why some ranchers are reluctant to participate in wildlife conservation and management programs, agency personnel responses included rules and bureaucracy (47%), funding amounts (27%), rancher characteristics (33%), and the agency rancher relationship (29%). With respect to rules and bureaucracy, these were related to landuse restrictions of certain programs, the lengthy appl ication process, and the fear that threatened or endangered species would be found on properties, further restricting land use. For example, one agency employee wrote: I believe ranchers are reluctant to participate in agency programs because often the agencies goals and focus (endangered species management in particular) are not the same as the ranchers goals. Most ranchers do not like being involved with threatened and endangered species because they find limits and constraints are imposed on what they can do on their land and how they can do it. There is a history of almost punishing people for having a listed species present on their property. Most ranchers are game species oriented, and there are not many conservation programs focused on these species. Responses regarding the amount of money offered for cost share programs were diverse. Several agency employees thought the money offered for cost share programs was too low for many ranchers to expend time and money to apply. Others thought rancher s did not have enough disposable capital to meet cost share requirements, whereas some believed that ranchers were only interested in agency programs when financial incentives are offered.

PAGE 34

34 With respect to rancher characteristics, many agency employees thought that ranchers were private and proud people who did not like to take financial assistance. For example, one agency employee wrote: I believe ranchers are generally independent, self reliant individuals, and therefore are not looking for a handout. Another common characteristic was that they thought although ranchers valued wildlife, wildlife management was not a personal priority. This is exemplified by one respondent who wrote: They [ ranchers ] respect wildlife and enjoy it when they have time, but it is not a priority to them. The agency rancher relationship was also frequently mentioned by agency personnel as a reason why ranchers were reluctant to participate in wildlife management programs. The relationship had three components, a distrust of agency staff by ranchers, the lack of agency personnel s knowledge of ranch management and ranchers, and personal or anecdotal examples of ranchers historically having bad experiences with agency employees. The former was exemplified by one agency employ ee stating: The old statement hi, I am from the government and am here to help you does not set well with most landowners. Agency personnel need to build a relationship with the landowners and earn their trust to be effective in wildlife management. The second reason, lack of agency knowledge, was stated by one respondent as: I think there are very few rancher agency interactions for wildlife conservation due to the fact that most agency personnel have limited knowledge of cattle operations in their area. Finally, the personal or anecdotal bad experiences are exemplified by Some [ranchers] have heard many wives tales about how these government based programs will limit their current management objectives and are not

PAGE 35

35 cost effective. Basically, m any of these folks need to have someone to speak directly with them on their property, which would allow for the true facts to be relayed directly to them. Improvi ng wildlife management programs When asked how agencies could improve their current wildlife management and conservation programs, I was able to group agency personnel responses into three major themes: education (33%), simplification (28%), and cooperation (20%). Education was threefold, including educating agency personnel about cattle operat ions, promoting available programs to landowners, and educating landowners on ranchwildlife integration techniques. Simplification involved streamlining the application processes to be less cumbersome. One respondent wrote: An example: a beef producer who wanted to come to an NRCS office and get financial assistance to install or implement practices to benefit his cattle, his forest land, and manage for wildlife would likely have to develop three plans and three contracts under three different programs (EQIP, EQIP Forestry, and WHIP). Its not convenient or practical for the office or the producer to keep up with. Develop one program that can apply all the necessary practices to fit an interdisciplinary approach. Respondents remarking about collaborat ion focused on inter agency cooperation and the cooperation among ranchers themselves. Regarding inter agency cooperation, an employee responded: Agencies need to work more cooperatively selling each others programs to landowners, all the while, being able to maintain each programs separate identities. Inter rancher cooperation statements focused on rancher leader approaches, and as stated by one employee: My opinion is that cattle folks listen and look up to other cattle folks. [If an agency can] implement wildlife conservation with a

PAGE 36

36 respected cattle rancher, the rest will follow for no other reason than fear of falling behind. Improvements to technical workshops and field days Of the agency responses to a question about the most and least successful field day or wildlife workshops, 70% of respondents indicated that outdoor field days with hands on activities were more effective than indoor workshops with a digital slide show lecture. Several thought a combination of classroom and field work could be effective as long as the topics were practical and used examples from the area. Forty eight percent of agency personnel also thought utilizing rancher leaders who were already effectively combining ranching and wildlife management as speakers along with a multi agency team was a particularly successful approach. One respondent wrote: The best workshops are generally ones where there is multiagency participation presenting real world situations, problems, and solutions where ranchers share their suc cesses and failures. Walking and hay ride tours are a big success as they are very interactive and lessons/themes can easily be illustrated. Too many classroom type presentations can get dry. Twenty percent also mentioned logistics and thought key components were: providing food; keeping presentations short, practical, and to the point; scheduling workshops and field days on Saturdays or times of the year when ranchers are less busy; and limiting the duration to a half day. Some agency personnel wrote about specific technical topics and thought those that focused on game species or integrating cattle and wildlife management were the most interesting to ranchers. Rancher Attitudes and Perceptions of Programs I asked several closed questions of ranchers r elated to attitudes and perceptions of programs, workshops, and field days that allow for some indirect comparisons of

PAGE 37

37 ranchers to agency personnel. I asked ranchers their level of agreement with statements about cost share programs, wildlife workshops and field days, and wildlife management (Table 2 6). All but two responses were above the middle response category, indicating they generally agreed with statements that cost share financial assistance programs are beneficial, but the application process can be complicated and the final cost share contracts restrictive. Similar to the questions about wildlife benefits, ranchers agreed to strongly agreed with statements about the importance of wildlife to ranchers with respect to recreation and other int rinsic wildlife values. The two statements that ranged from disagree to neither disagree or agree were that wildlife enterprises were important to ranchers and ranchers like having government personnel on their property. Discussion The results indicat e agency personnel and rancher perceptions of wildlife management agree on some, but not all, topics. Agency personnel overestimated the importance of economic wildlife enterprises on ranches. Additionally, ranchers agree with a general statement that wi ldlife enterprises can contribute to the ranch but disagree or are indifferent to the statement about wildlife enterprises being important to them. With respect to wildlife economic enterprises, leasing private lands to hunters has existed for more than 70 y ears in the United States and been well documented, especially in the literature regarding wildlife privatization and access (Brown and Messmer 2009). Respondents generally considered leasing and most other enterprises as not a benefit, whereas agency personnel, rated leases as a moderate benefit to ranchers. Similar findings were recorded in Montana, where approximately 90% of farmers and ranchers considered money from hunting fees of low importance, and 50%

PAGE 38

38 of agency personnel considered it of m oderate importance (Saltiel and Irby 1998). In recent times, hunters have been spending increasing amounts to lease private lands (Aiken 2005), but the percentage of landowners realizing income from fee hunting opportunities has remained low, from 4% in I llinois (Miller et al. 2002) to 6% in Florida (Willcox et al. 2010b) and New York (Siemer and Brown 1993). Conversely, wildlife professionals promote hunting leases as a viable means to increase incomes and believe they can increase wildlife conservation in general (Benson et al. 1999). These ideas are supported by case studies from the western United States (Benson 1998) and economic analyses in Texas that indicate net returns from hunting leases can exceed net income from typical cattle operations (Livengood 1983). My data reaffirms that agency personnel overestimate the importance of economic wildlife enterprises to ranchers, but it is notable that these ranchers understand the potential of wildlife enterprises, even if they are not conducting them. T he relationship between recognizing the importance of wildlife enterprises to ranchers and ranchers actually initiating them is not well understood and warrants further investigation. The overestimation of wildlife problems and underestimation of the likability of damaging wildlife by agency personnel may be because agency personnel are likely the first people ranchers contact when they encounter substantial wildlife problems Conversely, it is doubtful that ranchers would contact agency personnel if they had positive wildlife experiences on their ranch. It is also possible that agency personnel have extrapolated wildlife crop depredation problems of row crop, small livestock, and fruit crop farmers to ranchers that do not face as many crop depredation pro blems, but this has yet to be investigated. The unfortunate consequence of agencies receiving the

PAGE 39

39 majority of wildlife complaints is that they can project these negative interactions to the entire rancher population, a phenomenon observed in Montana game wardens (Saltiel and Irby 1998). In Montana, these negative interactions resulted in wardens underestimating landowners aesthetic wildlife values Conversely, the agency respondents in the southeast recognized and accurately predicted rancher intrinsic, many of which were aesthetic, wildlife benefits Agency personnel and ranchers were in general agreement about the intrinsic benefits of wildlife on ranches and the likability of beneficial animals on ranches. With half the intrinsic benefits ranging fro m moderate to serious, I suggest that ranchers appear to foster Leopolds land ethic as they do not generally recognize wildlife benefits as an economic commodity, yet realize wildlife benefits from intangible concepts including wildlife enjoyment by future generations and the health of the land, and tangible activities for personal recreation such as wildlife viewing, hunting, and fishing (Leopold 1959). As agency personnel seem to understand rancher appreciation of wildlife intrinsic benefits, and likely value these aspects of wildlife themselves, this shared appreciation should strengthen the agency rancher relationship. Of the wildlife that factored into the enjoyed category, rancher responses ranged from like to extremely like for all (i.e ., turkey, quail, deer, fish, songbirds, and waterfowl), giving clear direction to agency personnel as to which types of animals ranchers appreciate and would potentially manage sustainable populations of on their ranches. Agency personnel are already knowledgeable about rancher preferred species, most (but not all) of which were game species, and are thus well tuned to rancher wildlife priorities.

PAGE 40

40 Ranchers are considering wildlife management in general cattle operations some of the time for all variables measured, which was underestimated by agency personnel. This misperception should be encouraging to agencies, as many ranchers are already considering wildlife management in their daily activities and could be amenable to assistance that helps them bett er integrate wildlife and cattle management. In the openended questions, this sentiment was echoed by many agency personnel who acknowledged that although wildlife were important to ranchers, their management would always be secondary to ranch activities Therefore, the key is designing programs that integrate wildlife management into cattle operations without negatively affecting ranch income. Agency personnel seem to have accurately assessed rancher perceptions of some wildlife management aspects, lending support to the idea that they are well suited to assist ranchers with preferred species wildlife management. Additionally, in the statements of agreement by ranchers about the importance of seeing wildlife, enjoying wildlife recreational activities, and having both cattle and wildlife on the ranch should persuade agency personnel to consider designing programs that holistically manage habitat for both cattle and wildlife simultaneously, rather than separately. Although agency personnel overestimated r ancher interest in cost share financial assistance programs, ranchers generally agreed that the programs were beneficial to ranchers, were good for wildlife and conservation, and a worthwhile use of government funds. In openended questions, agency personnel had many ideas on how to make these programs more attractive and useful for cattle ranchers. There appears to be some frustration by agency personnel who feel the application processes are

PAGE 41

41 complicated and similar programs administered within or among agencies are not coordinated. This was marginally supported by ranchers who neither disagreed or agreed to agreed that the cost share application process is complicated. However, nationwide, cost share financial assistance programs for wildlife manag ement on private lands are a relatively new phenomenon (e.g., WHIP in 1998 and Partners for Fish and Wildlife national expansions in the 1990s), and it is understandable there will be resistance and a forced organizational change as agency employees must enact new programs based on new federal legislation (Burke 2008). Superficially, this organizational change appears to be a topdown benevolent autocracy, but agency personnel who implement these programs at the state or regional level have sufficient aut onomy to develop and change the operating procedures to administer cost share programs effectively, making it more of a consultative or participative management change (Likert 1967). In addition to agency surveys such as ours that present agency personnel opinions on how to improve assistance programs, some agencies in the southeast have formed teams and working groups to improve coordination among agencies and improve programs based directly on agency personnel recommendations similar to those reported here (J. S. Sanders, Florida Fish and Wildlife Conservation Commission, personal communication). Almost one quarter of the respondents attended at least one technical workshop or field day that included wildlife management, indicating a high level of interes t. One way of increasing attendance at workshops is to insure the topics are interesting to the target audience. These results suggest that although most ranchers indicated they were either unlikely or neither unlikely or likely to participate in mos t workshop topics, the

PAGE 42

42 most popular ones were those that integrated wildlife management with grazing or agricultural systems planning. Although agency personnel overestimated rancher participation in workshops for most topics, their comments in the opene nded questions section regarding technical workshops elucidate some good ideas on topics and formats that would interest ranchers, including integrating wildlife management into routine ranch activities and preferred species management. Other answers about actual workshop field days or logistics should also be considered in workshop or field day planning including minimizing classroom and digital slide show presentations, using rancher leaders to conduct presentations, holding workshops and field days on r anches with successful wildlife programs, providing food, and timing programs when there is less ranch work. Ranchers generally agreed with statements that wildlife workshops and field days were a good way for ranchers to learn how to conduct wildlife man agement on their ranch and that ranchesr are generally knowledgeable about wildlife. As suggested by agency personnel, they should utilize respected rancher leaders in workshop and field day design and implementation as ranchers generally think their fell ow ranchers are conversant on wildlife management topics. It is encouraging that agency personnel were able to accurately assess many rancher perceptions of wildlife management and that ranchers scored the intrinsic benefits of wildlife highly. By underst anding ranchers, agency personnel should be well suited to tailor wildlife management programs to their needs be able to start meaningful discussions and connect with them on contemporary wildlife management issues. It is therefore vital that agency leaders continue soliciting the ideas from agency personnel and cattle ranchers when developing, modifying, and implementing wildlife technical

PAGE 43

43 and financial assistance programs. Although ranchers were not overly concerned with wildlife damage, it is still pru dent for agencies to address issues with problem wildlife, but it is important that they focus on preferred species and positive wildlife benefits to demonstrate how wildlife management can be integrated on working ranches without negatively affecting oper ations and profit. Currently, most ranchers do not realize economic benefits from wildlife enterprises, yet they and agency personnel see the potential for them. If ranchers call upon agency personnel to assist them with wildlife enterprises, this should be encouraged, yet it might be premature to develop statewide or national programs to actively promote them. Further investigation is needed in the southeastern US including detailed economic studies that show supply and demand, willingness to pay, profi tability, and broader conservation value of wildlife enterprises before agencies design programs promoting them as viable alternatives or addi tions to a working cattle ranch.

PAGE 44

44 Table 2 1. Perceptions of ranchers and agency personnel of w ildlife benefits o n ranches in the southeastern U nited States 2009. Wildlife benefit a Ranchers ( n = 1,093) Agencies ( n = 52) Factor 1b i ntrinsic benefits Factor 2b e conomic benefits a SE a SE Maintain healthy land and environment 3.4 0.0 3.5 0.1 0.884 0.033 Conservation 3.3 0.0 3.1 0.1 0.874 0.057 Enjoyment for future generations 3.4 0.0 3.5 0.1 0.845 0.069 Family wildlife watching 3.1 0.0 2.8 0.1 0.766 0.009 Family hunting opportunities 2.6 0.0 3.4 0.1 0.590 0.251 Crop and pasture polli nation 2.4 0.0 2.2 0.1 0.558 0.128 Family fishing opportunities 2.3 0.0 2.7 0.1 0.536 0.249 Conduct guided hunts 1.1 0.0 1.7 0.1 0.079 0.781 Ecotourism enterprises 1.1 0.0 1.5 0.1 0.114 0.759 Lease land to others for hunting 1.3 0.0 2.9 0.2 0.044 0.725 Income from mitigation banking 1.2 0.0 2.0 0.1 0.145 0.724 Enjoyment by the general public c 1.6 0.0 1.5 0.1 0.354 0.318 Eigenvalues b 3.962 2.487 Variance explained (%) b 33 21 Chronbachs d 0.859 0.753 a Ordered response categories: 1 = not a benefit, 2 = slight benefit, 3 = moderate benefit, 4 = serious benefit, and 5 = extreme benefit. b Factor analysis with a Varimax rotation ( n = 1,145). c Variable excluded from analyses due to low factor loadings. d Data with factor loadi ngs

PAGE 45

45 Table 2 2. Rancher and agency personnel perceptions of w ildlife problems on ranches in the southeastern U nited States 2009. Wildlife problem a Ranchers ( n = 1 093) Age ncies ( n = 52) Factorb w ildlife problems a SE a SE Feed damage 1.7 0.0 2.5 0.1 0.730 Equipment damage 1.3 0.0 1.8 0.1 0.701 Crop damage 2.0 0.0 2.8 0.2 0.687 Spread disease 1.4 0.0 1.8 0.1 0.660 Tree damage 1.5 0.0 1.9 0.1 0.650 Building damage 1.3 0.0 1.6 0.1 0.648 Const raints from endangered species regulations 1.3 0.0 2.7 0.2 0.627 Burrows or tunnels in fields 2.1 0.0 2.3 0.1 0.615 Fence damage 1.8 0.0 2.0 0.1 0.609 Water control system damage 1.4 0.0 2.1 0.1 0.604 Spread unwanted plants 1.8 0.0 2.3 0.1 0.603 Lives tock predation 1.9 0.0 2.9 0.2 0.601 Eigenvalue b 5.007 V ariance explained (%) b 42 Chronbachs 0.871 a Ordered response categories: 1 = not a problem, 2 = slight problem, 3 = moderate problem, 4 = serious problem, and 5 = extreme problem. b Factor analysis with a Varimax rotation ( n = 1,145).

PAGE 46

46 Table 2 3. Rancher and agency personnel perceptions of wildlife likability on ranches in the southeastern U nited States 2009. Wildlife type a Ranchers Agencies ( n = 52) Fact or 1b enjoyed Factor 2b damaging Factor 3b mixed a SE n a SE Turkey 4.4 0.0 1001 4.5 0.1 0.849 0.085 0.109 Quail 4.5 0.0 1015 4.5 0.1 0.817 0.006 0.110 Deer 4.1 0.0 1024 4.3 0.1 0.790 0.196 0.074 Waterfowl c 4.0 0.0 922 3 .9 0.1 0.727 0.092 0.274 Fish 4.3 0.0 896 4.0 0.1 0.723 0.012 0.166 Songbirds 4.4 0.0 1055 3.6 0.1 0.591 0.110 0.407 Rats and other rodents 1.6 0.0 1057 2.0 0.1 0.086 0.767 0.049 Alligators or crocodiles 2.1 0.0 610 2.8 0.1 0.208 0.742 0.029 Predat ory mammals d 1.9 0.0 1030 1.7 0.1 0.058 0.679 0.284 Snakes and lizards 2.5 0.0 1070 2.5 0.1 0.028 0.632 0.454 Predatory birds e 3.4 0.0 1070 2.7 0.1 0.056 0.095 0.801 Endangered species 3.3 0.0 788 2.6 0.2 0.256 0.120 0.646 Turtles and tortoises 3.2 0.0 1034 3.0 0.1 0.170 0.301 0.638 Small to medium sized mammals f 3.7 0.0 1085 3.1 0.1 0.395 0.014 0.570 Eigenvalues b 3.721 2.179 2.379 V ariance explained (%) b 27 16 17 Chronbachs g 0.854 0.698 0.714 a Ordered response categ ories: 1 = extremely dislike, 2 = dislike, 3 = neither dislike or like, 4 = like, 5 = extremely like, and N/A not applicable. N/A responses excluded from analyses. b Factor analysis with a Varimax rotation ( n = 542). c e.g. ducks and geese. d e.g. coyotes panthers, mountain lions, bears, bobcats, etc. e e.g. hawks, eagles, owls, etc. f e.g. rabbits, raccoons, etc. g Data with factor loadings

PAGE 47

47 Table 24 Rancher and agency personnel perceptions of how likely ranchers were to consider wildlife management when ranching in the southeastern U nited States 2009. Likeliness to consider wildli fe management when a Ranchers Agencies ( n = 51 ) Factor b wildlife management a SE n a SE Rotating cattle between pastures or paddocks 2.7 0.0 1039 2.5 0.1 0.886 Managing hayfields 2.9 0.0 981 2.5 0.1 0.871 S electing grasses, t rees, or shrubs to plant 3.0 0.0 1053 2.5 0.1 0.868 Building fences 3.0 0.0 1053 2.3 0.1 0.840 Setting stocking rates 2.4 0.0 1002 2.4 0.1 0.792 Eigenvalue c 3.630 V ariance explained (%) b 73 Chronbachs 0.905 a O rdered res ponse categories: 1 = never, 2 = seldom, 3 = some of the time, 4 = most of the time, 5 = all of the time, and N/A not applica ble. N/A responses excluded from analyses. b F actor analysis with a Varimax rotation ( n = 967).

PAGE 48

48 Table 2 5. Ran cher and agency personnel perceptions of how interested or likely cattle ranchers are to participate in financial assistance programs and selected workshop or field day topics in the southeastern U nited States 2009. Program or workshop topic Ranchers ( n = 1093) Agencies ( n = 51) Factorc workshop a SE b SE Cost share financial assistance programs a 3.1 0.0 3.7 0.2 General wildlife habitat workshops or field days b 2.7 0.0 3.3 0.1 0.904 Game species management workshops or fiel d days b 2.7 0.0 3.7 0.1 0.902 Agricultural systems planning workshops or field days that included wildlife management b 2.9 0.0 3.9 0.1 0.899 Grazing systems management workshops or field days for both wildlife and cattle b 3.2 0.0 4.1 0.1 0.817 Endangere d species management workshops or field days b 2.6 0.0 2.3 0.1 0.803 Wildlife enterprise (e.g. hunting leases, ecotourism, husbandry, etc.) workshops or field days b 2.4 0.0 3.6 0.1 0.712 Watchable wildlife (e.g. songbirds) workshops or field days b 2.6 0.0 2.2 0.1 0.704 Eigenvalue c 4.755 V ariance explained (%) c 68 Chronbachs 0.919 a O rdered response categories: 1 = very uninterested, 2 = uninterested, 3 = neither uninterested or interested, 4 = interested, and 5 = very interes ted. b O rdered response categories: 1 = very unlikely, 2 = unlikely, 3 = neither unlikely or likely, 4 = likely, and 5 = very likely. c F actor analysis with a Varimax rotation ( n = 1144).

PAGE 49

49 Table 2 6. C attle rancher agreement with statements about cost share programs, wildlife workshops and field days, and wildlife on their ranch in the southeastern U nited States 2009. Statement a Ranchers ( n = 1093) a SE Cost share financial assistance programs benefit cattle ranchers. 3.7 0.0 Co st share financial assistance programs are good for land and wildlife conservation. 3.8 0.0 Cost share programs are a worthwhile use of government funds. 3.7 0.0 Generally, ranchers like having government staff of their property. 2.5 0.0 Ranchers think the cost share application process is complicated. 3.4 0.0 Ranchers think cost share contracts are restrictive. 3.4 0.0 Workshops and field days are a good way for ranchers to learn how to conduct wildlife management on their ranch. 3.7 0.0 If ranchers have wildlife enterprises like hunting leases, guided hunts, and ecotourism, it can contribute to their ranch. 3. 6 0.0 Establishing wildlife enterprises is important to my ranch. 2.9 0.0 Seeing wildlife in my fields, pastures, and property is important t o me. 4.0 0.0 Wildlife activities like bird watching, hunting, and fishing on my ranch are important to me. 4.1 0.0 Having both wildlife and cattle on my ranch is important to me. 4.0 0.0 a O rdered response categories: 1 = strongly disagree, 2 = disagre e, 3 = neither disagree or agree, 4 = agree, and 5 = strongly agree.

PAGE 50

50 CHAPTER 3 EXPLAINING THE INTEGRATION OF WILDLIFE MANAGEMENT AND CATTLE OPERATIONS IN THE SOUTHEASTERN UNITED STATES Introduction Livestock management is one of the most widespread land uses in the US with 326,000,000 ha or 35% of the land in the US devoted to the practice (Lubowski et al. 2006). The amount of grazing land has slowly but steadily declined since 1945, with substantial conversion to environmental, wildlife, and recreatio nal uses. Grassland habitats, including pastures, hay fields, and grazable woodlands, can support diverse communities of common and listed wildlife species due to the proliferation of graminoids and forbs (Willcox 2010) These types of vegetation are readily consumed by both livestock and wildlife, attract diverse assemblages of insects and insectivores, provide thermal and escape cover, and supply nesting material. Grassland systems can be suitably managed for both wildlife and cattle though rotational grazing, fencing, stocking rate manipulation, and other forms of vegetation management (West 1993, Payne and Bryant 1994, Morrison and Humphrey 2001). Contrary to national trends, grazing lands have increased by 1,200,000 ha in the Southeast mostly at th e expense of croplands ( Lubowski et al. 2006). With this land use becoming more prominent, it may be prudent for wildlife managers and conservationists in this region to increase outreach efforts with ranchers about integrating wildlife management and cat tle operations in this expanding land use. An important step in designing programs to promote integrating wildlife management with cattle operations is to determine underlying factors that explain why ranchers do or do not consider wildlife management in t heir ranching activities. Human dimensions of wildlife researchers are increasingly employing explanatory behavioral

PAGE 51

51 models grounded in social psychology to predict intention and participation in a variety of behaviors (Manfredo et al. 1995, Rossi and Arm strong 1999, Hrubes et al. 2001). To date, there have been no behavioral studies that explain cattle rancher behavior with respect to wildlife management on their ranches. This study addresses a gap in the human dimensions of wildlife knowledge by modeli ng cattle rancher wildlife behavior using a prominent social psychology theory. Theoretical Framework The Theory of Planned Behavior is a social psychology model that explains individuals intentions to enact behaviors (Ajzen 1988). It has been successful ly applied to human dimensions of wildlife research by explaining behaviors related to wildlife recreation (Rossi and Armstrong 1999, Hrubes et al. 2001, Martin and McCurdy 2009) and ranchers by predicting participation in conservation easements (Brain 2008). According to the model, behavioral intent is predicted by three explanatory variables: 1) attitude toward the behavior, 2) subjective norms, and 3) perceived behavioral control. These constructs are commonly estimated by calculating a general belief variable for each component multiplied by the perceived strength or degree of influence the belief has on the individual (Ajzen 1988). The first variable, attitude toward the behavior, is elicited by asking several questions about the perceived outcome of engaging in a behavior and then corresponding attitudinal strength questions that evaluate the importance of the outcome to the respondent. The final attitude toward the behavior is then calculated using the following equation (Fishbein and Ajzen 1975, A jzen 1988):

PAGE 52

52 in which is the calculated attitude toward the behavior. is the sum of all pairs of the strength or degree of importance of the outcome, and a corresponding the outcome evaluat ion of engaging in the behavior The second explanatory variable, subjective norms, is a measure of how important people, groups, and organizations that interact with respondents would influence respondent decisions to engage in the behavior. As with attitudes a bove, subjective norms are elicited by measuring normative beliefs of important social groups and motivation of the respondent to comply with those groups. Subjective norms are calculated similarly to attitudes by substituting paired measures of normative beliefs and motivation to comply in the equation. The final component, perceived behavioral control, is the component that makes the Theory of Planned Behavior unique to its predecessor, the Theory of Reasoned action, which contained only attitudes and su bjective norms (Fishbein and Ajzen 1975, Ajzen and Fishbein 1980). Perceived behavioral control is a measure of the simplicity or difficulty of conducting the behavior. It measures the perception of respondents who do not have total volitional control ov er the behavior and may lack skills, opportunities, or resources to conduct it (Ajzen 2002). As with both other components, perceived behavioral control is measured in paired questions; the first evaluates the belief that the control can positively or neg atively affect the behavior, and the second, the strength of this belief in relation to respondents ability to enact the behavior. The equation for this component is identical to those above, except paired control beliefs and belief strength are used ins tead of attitudinal and subjective norm measures. After calculating the

PAGE 53

53 three explanatory variables, they are regressed with behavioral intent as the dependent variable to predict the behavior. Methods Rancher Survey Beef cattle producers and University of Florida livestock extension faculty were interviewed in spring 2008 to develop a self administered mail survey. T he survey was designed from interview results and data obtained from focus groups in a concurrent study of private landowners in FL (Willcox et al. 2010b). Ten human dimensions of agriculture, wildlife, and forestry experts, 5 wildlife agency personnel, and 5 ranchers in FL and GA reviewed the draft survey. I improved drafts and resent them to reviewers several times for testing before final izing the survey for content validity The final survey contained 35 numbered questions over 9 pages (Appendix A) A n initial screening question asked if the respondent was currently ranching. Ranchers were also asked questions about behaviors integrati ng wildlife management into future cattle management activities, and attitudes, subjective norms, and perceived behavior controls related to those behaviors with 5point scales using ordered responses (i.e., 1 = not a benefit to 5 = extreme benefit; 1 = not a problem to 5 = extreme problem; 1 = extremely dislike to 5 = extremely like; and 1 = strongly disagree to 5 = strongly agree; Dillman et al. 2009). Other questions asked ranchers about land use, ranch characteristics, and demographics. In addition to using th e se data for ranching behavior analyses, I compared demographic and ranch size data with the 2007 Census of Agriculture to assess nonresponse error (USDA 2009). A list of 24,049 nondairy cattle producers was obtained from the Farm Index of CAS In c. in September 2008 (www.cas online.com). L ists from UFs Cooperative

PAGE 54

54 Extension Service augmented the Florida list because the purchased frame was low (227 ranchers) when evaluated with the 15,717 farms with beef cattle reported in the USDAs National Ag ricultural Statistics Service 2002 Census of Agriculture (USDA 2004). The Mississippi sampling frame was also small, but additional lists proved unattainable to increase it. I next condensed the list by zip code to my study area in the Southern Coastal P lain and Southeastern Plains of Omernicks Level III Ecoregions, resulting in 15,023 addresses (Commission for Environmental Cooperation 1997). Florida was stratified by north and south at approximately 29N because the agricultural systems and environment differ, with the south containing more planted and native range and north having more pine timber and native forest (Clouser et al. 200 7, Willcox et al. 2010b). North and south Florida were later combined as rancher results did not differ between the north and south, unlike the previous studies involving landowners and farmers. I drew random samples of 500 ranchers from each state and two Florida strata, except for Mississippi as the sampling frame contained only 271 ranchers, and in that case, all Miss issippi ranchers were surveyed. A n additional 500 ranchers were drawn from Georgia to conduct a financial incentive effect methodological experiment (Willcox et al. 2010 a ). My final sample contained 2,771 addresses. I conducted a five wave mailing follo wing the Tailored Design Method (Dillman et al. 2009), sending preletters, initial surveys, and reminder postcards in November 2008, replacement surveys in January 2009, and final replacement surveys in February 2009. Data Analyses Missing data were imputed using the expectationmaximization function with 20 iterations and 0.001 convergence to reduce uncertainty from missing values and use completedata tec hniques (Shafer and Graham 2002, SYSTAT 2007). I used likelihood

PAGE 55

55 ratio ( LR ) analyses to assess nonres ponse error by comparing demographic and ranch size data with the 2007 Census of Agriculture (USDA 2009). Principal components analysis with a Varimax rotation and Chronbachs were used to assess reliability of the dependent behavioral intent variable and independent variables measuring attitudes, subjective norms, and perceived behavioral control. The dependent variable was calculated by taking the mean scores of five questions that elicited future intention to consider wildlife management when conducting five cattle management activities (setting stocking rates, choosing when to rotate cattle among pastures or paddocks, selecting vegetation to plant, managing hayfields, and building fences; answer choices included 0 = never, 1 = seldom, 2 = some of the time, 3 = most of the time, and 4 = all of the time). As answer choices included N/A not applicable because not all ranchers have some of the land uses present on their property (e.g., hayfields or multiple pastures or paddocks), a dummy variable was created for each question assigning 0 to not applicable choices and 1 to all other choices. Dummy variables were regressed with the calculated behavioral intent variable to determine N/A effects, and if found significant, were controlled in later regression analyses. Independent variables for attitudes, subjective norms, and perceived behavioral control were calculated using Theory of Planned Behavior protocols (Ajzen 1988). Ranch diversity, ranch size, and frequency of collaborating with governmental and nonprofit conservation groups were also regressed with the dependent variable as farms with a variety of crops and farmers who were frequently consulted technical experts have been found to more frequently adopt complex conservation prac tices (Lambert et al. 2006). To measure ranch diversity, I constructed a ranch diversity index by coding

PAGE 56

56 dummy variables for ranchers indicating if they had property in each of 11 different uses (planted timber, grazed native forest, ungrazed native fores t, improved pasture, hayfields, silvopastures, sod, annual row crops, perennial groves or orchards, and wetlands). The mean of these 11 dummy variables was then calculated, resulting in a ranch diversity index potentially ranging from 01. Likeliness to collaborate with six government and nonprofit groups were analyzed using factor analysis and saved factor scores were then used in the regression model. I used forced and stepwise ( FTE = 0.05, FTR = 0.10) threestage hierarchical regression to determine relationships between behavioral intent (dependent variable) and the independent variables. I fit the model by entering all three theory components first, the likeliness to work with agricultural and wildlife agencies and ranch characteristics second, and a global model that also contained demographics third. Significant theory components were rechecked using correlation matrixes and general linear models with interaction terms to investigate potential moderator effects (Ajzen and Fishbein 2005) and multi collinearity (Appendix B). SYSTAT 12 was used for all analyses and statistical significance concluded at P homogeneity of variance, and those violating test assumptions were rank transformed pri or to analysis (Conover 1980). Results Response Demographics, and Ranch Characteristics Of the 2,771 surveys mailed, 1,634 ranchers responded, resulting in adjusted response rates of: overall = 59%, AL = 57%, FL = 72%, GA = 49% and MS = 56% (discounting undeliverable surveys; Chapter 2). After accounting for those returned blank or respondents answering no to the screening question indicating they were no

PAGE 57

57 longer active ranchers, I had 1,093 usable surveys. I then compared my results with the 2007 Census of Agriculture (USDA 2009) on equivalent demographic (i.e., gender, age, and ethnic group) and ranch size variables, and found nonresponse error in all my samples. Across all states, there were fewer ranchers younger than 35 years old ( LR = 80.73, df = 8, n = 69,008, P < 0.001) and ranches smaller than 57 ha ( LR = 1443.69, df = 11, n = 68,566, P < 0.001) than the 2007 Census. Additionally, Florida had more women ( LR = 23.67, df = 1, n = 17,281, P < 0.001) and Mississippi had fewer other (than white) ethnic group respondents ( LR = 21.24, df = 5, n = 15,139, P < 0.001) than the 2007 Census. I found no differences in any other comparisons. Therefore, care should be taken when generalizing results to the entire population as the data best reflects the perceptions of producers 35 yrs old and those that have ranches 57 ha. Respondents were 87% male, 96% white, and averaged 61.0 0. 4 ( SE) yrs old. They owned and leased 597.7 43.5 ha, owned properties for 52.5 1.2 yrs, managed 110.2 12.3 cattle, had raised cattle on their current property for 43.5 1.0 yrs, and their mean ranch diversity index was 0.4 0.0. Likeliness to work with different types of agencies factored into two groupings labeled agricultural organizations (NRCS, Farm Service Agency, and Cooperative Extension Service) and wildlife organizations (USFWS, state wildlife agency, and various nongovernmental wildlife groups) explaining 40% ( Chronbachs = 0.85) and 40% ( Chronbachs = 0.88) of the total variance, respecti vely (Table 3 1). Forty percent graduated from college with at least an associates degree, 23% had some college courses 32% graduated from high school, and 6% had less than a high school diploma. Five percent of ranchers reported

PAGE 58

58 household incomes >$200,000; 3% at $150,000199,000; 15% at $100,000149,000; 41% at $50,00099,000; 25% at $25,00049,000; and 11% at $0 24,999. Theory of Planned Behavior Components Analyses Behavioral intent The individual wildlife management behaviors of considering wildlif e when setting stocking rates; choosing when to rotate cattle between pastures or paddocks; selecting grasses, trees, shrubs or other vegetation to plant; managing hayfields; and building fences; ( Table 2 3 and 24 ) were analyzed as a group for reliability with one factor extracted that explained 63% of the total variance and had a Chronbachs = 0.85, indicating high reliability among the behav ioral intent responses. Twenty two respondents indicated N/A for all activities and were excluded from further analyses The remaining ranchers averaged 2.8 0. 0 on the dependent variable indicating they will consider wildlife management when conducting future ranching activities some of the time. Attitudes The three calculated attitudes were descr ibed in one factor explaining 69% of the total variance with a Chronbachs = 0.85, demonstrating sufficient reliability. Ranchers scored high on both attitudinal strength and outcome evaluations, indicating general agreement that they considered wildlife important and thought wildlife could be beneficial on ranches (Table 3 2). Overall attitude toward the behavior was then calculated as suggested by Ajzen (1988) resulting in a mean attitude score of 60.1 0.6 on a scale of 4 to 100, with hi gher numbers indicating more positive attitudes.

PAGE 59

59 Subjective norms The calculated subjective norms also factored together, explaining 70% of the total variance with a Chronbachs = 0.86, indicating adequate reliability. All normative belief a nd motivation to comply mean scores were encouraged by various social groups to incorporate wildlife management into ranching activities and were positively influenced by these groups (Table 3 3). Overall subjective norms towar d the behavior was then calculated as suggested by Ajzen (1988) resulting in a mean subjective norm score of 56.9 0.5 on a scale of 4 to 100, with higher numbers indicating more positive subjective norms. Perceived behavioral control Calculated perceived behavioral control variables factored together explaining 57% of the variance with a Chronbachs = 0.74, indicating construct reliability. Perceived behavioral control means were generally lower than attitudes and subjective norms, ranging from 2.7 3.2 (Table 3 4). Overall perceived behavioral control toward the behavior was then calculated as suggested by Ajzen (1988) resulting in a mean score of 35.5 0.3 on a scale of 4 to 100, with higher numbers indicating more positive subjective nor ms. Main Analys es The forced model was significant at each hierarchical step (Table 3 5). Once stepwise elimination procedures were included, attitude toward the behavior, subjective norms, likeliness to work with a wildlife agency in the future, and age were positively and income negatively associated with behavioral intent, explaining 19% of the variation (Table 3 6). One theory component, perceived behavioral control, was dropped from the final stepwise model along with most ranch and rancher character istics. Attitudes,

PAGE 60

60 subjective norms, and likeliness to work with a wildlife agency had the most effect on the dependent variable ( = 0.247, 0.177, and 0.130, respectively). Discussion Similar to other published human dimensions of wildlife studies using the Theory of Planned Behavior with hunter (Hrubes et al. 2001) and backpacker (Martin and McCurdy 2009) behavior and cattle rancher participation in conservation easements (Brain 2008), my results show only attitude and subjective norm components of the theory proved important. It is probable perceived behavioral control was not important to ranchers decisions to consider wildlife management in cattle operations as these decisions are l argely under volitional control, meaning outside factors do not limit participation in the behavior. As volitional control increases, the importance of perceived behavioral control to predict the behavior decreases (Madden et al. 1992, Hrubes et al. 2001). From a theoretical standpoint, my results suggest that the Theory of Reasoned Action (Fishbein and Ajzen 1975; Ajzen and Fishbein 1980), which is the precursor to the Theory of Planned Behavior and does not include perceived behavioral control, was more parsimonious than the Theory of Planned Behavior in explaining cattle rancher wildlife behavior. In practical terms, this means that although perceived behavioral controls described in the variable (i.e., land, cost, technical knowledge, and time) might be important to conducting wildlife management on ranches, ranchers do not perceive that these factors limit them from integrating wildlife management in their operations. Aside from small negative ef fects of income and positive effects of age, demographics and ranch characteristics were not important to ranchers behavioral intentions to integrate wildlife management into cattle operations. Brain (2008) had

PAGE 61

61 similar findings in her research explaining rancher participation in conservation easement programs in Florida using the Theory of Planned Behavior, finding that demographics and ranch size did not predict easement participation. These results differ from comparable demographic and land use measur es in studies involving private landowners and farmers generally, with larger farm sizes positively affecting the number of conservation behaviors adopted (Lambert et al. 2006); age negatively, and income, farm size, and education positively contributing t o conservation easement enrollment (Mashour 2004); and parcel size positively influencing enrollment in farmland preservation programs (Nickerson and Lynch 2001). My results indicate a universal framework could be designed for programs that seek to increase wildlife management on ranches, but it is always important to adapt these programs locally so they appropriately serve the type of operations and demographics in any given area. Examining the three most important model components more closely gives insi ght as to how wildlife management agencies and nongovernmental organizations could influence ranchers to consider wildlife management integration in their operations. Ranchers agreed with a general belief statement that wildlife and cattle can cohabitate a working ranch. They also thought that having wildlife on ranches could result in positive outcomes such as establishing wildlife enterprises, knowing the land is healthy and productive, and participating in wildlife recreational activities. Most corresponding attitudinal strength statements also scored highly (i.e., about wildlife enterprises being important to respondents. This statement scored 2.9, indicating that ranchers neither disagreed nor agreed. This phenomeno n mirrors other results from this survey that examined rancher perceptions of wildlife benefits, where

PAGE 62

62 economic benefits from wildlife were not valued by ranchers but intrinsic benefits such as family hunting opportunities and the enjoyment of wildlife by future generations were rated as having moderate to serious benefits (Chapter 2). Although economic analyses indicate ranchers can substantially benefit from wildlife enterprises (e.g., leasing hunting rights; Livengood 1983) and they are promoted by wildlife professionals (Benson et al. 1999; Brown and Messmer 2009), it remains unclear why ranchers believe that enterprises can positively contribute to ranches, yet do not engage in the behavior. Research regarding hunting access to private lands indicates landowners may restrict access to their property to friends and fami ly for reasons of liability, privacy, and safety (Brown and Messmer 2009). As intrinsic benefits were important to ranchers it is possible that due to the personal wildlife enjoyment, r anchers would rather restrict access to their property for intrinsic benefits over financial gain. Future research on wildlife enterprises should include questions to investigate if intrinsic wildlife benefits supersede potential wildlife recreational ent erprises. Regardless, as a whole, ranchers reliably indicated that having wildlife on the ranch would result in positive intrinsic outcomes and the affirmative attitudes related to these outcomes were also strong. Wildlife agencies and nongovernmental organizations should consider promoting programs that foster these positive attitudes over the economic benefits which should in turn increase the integration of wildlife management on working cattle ranches. Subjective norms also predict why ranchers integrate wildlife management into ranch activities. Therefore, programs that increase positive wildlife attitudes on working ranches should use important rancher social networks to increase their effect. The normative belief statements indicated that ranchers thought their families, friends and

PAGE 63

63 neighbors, fellow ranchers, and agency staff generally encouraged and approved of ranchers managing for wildlife. Ranchers were also strongly motivated to comply with these beliefs as they agreed with statements indi cating that the opinions and approval of family, friends and neighbors, fellow ranchers, and agency staff mattered to them. Therefore, wildlife programs that influence key members of respected social groups and utilize them in conduct ing outreach and extension activities should maximize program effect. P ractices that integrated wildlife management on ranches should spread through the ranching community if influential members of rancher social networks adopted them and they influence their peers (Rogers 20 03) Likeliness to work with a wildlife organization in the future also affected rancher wildlife behavioral intent. This has interesting programming implications as prior to regression, the agricultural and wildlife organizations factored separately with ranchers generally more likely to work with agricultural than wildlife organizations. These results might validate recent trends in Florida and Georgia where organizations involved in wildlife management and conservation are increasingly allying with agr icultural agencies. For example, in Florida, the NRCS has recently funded private lands wildlife biologist positions through the Florida Fish and Wildlife Conservation Commission to assist them in private landowner assistance programs and they are frequently conducting joint projects (J.S. Sanders, Florida Fish and Wildlife Conservation Commission, personal communication). By combining efforts and collaborating on programs, wildlife and agricultural organizations should mutually benefit as wildlife organi zations might get more exposure to ranchers through agricultural organization

PAGE 64

64 contacts and agricultural organizations can tap wildlife expertise to realize their wildlife program goals. As attitudes and subjective norms explained rancher intentions, it is possible a social marketing approach would be appropriate to positively influence wildlife management on cattle ranches (McKenzieMohr and Smith 1999, Jacobson et al., 2006). Also, because likeliness to work with wildlife organizations increased behavioral intent, wildlife organizations are well situated to explore this approach. S ocial marketing practices can build upon positive benefits and attitudes associated with a behavior and channel these messages through important community leaders to create soci al norms that reinforce the behavior. The target behaviors and means to achieve them must be effectively communicated and other tools, including the use of prompts, written or verbal commitments, reducing barriers, providing feedback, and carefully applied incentives, can help motivate people to act Wildlife and conservation agencies and nongovernmental organizations are already using many of these tools including commitments to voluntary conservation programs (e.g., the USDA Forest Stewardship Program), financially incentivizing wildlife friendly practices (e.g., the USDA Wildlife Habitat Incentives Program), and removing barriers to wildlife management by conducting technical assis tance workshops and fields days. They are also fostering social norms by recognizing prominent landowners who have adopted wildlifefriendly practices, and providing visual prompts to conduct wildlifefriendly practices using calendars, bumper stickers, and signs ( McKenzie Mohr and Smith 1999). Human dimensions of wildlife researchers should continue recent trends that use predictive social science theories like the Theory of Planned Behavior to investigate

PAGE 65

65 behaviors important to wildlife management. By applying results of these studies, agencies can design and modify program s with objectives to change wildlife management behaviors in stakeholders. The results of this study should enable a gencies and nongovernmental organizations working to increase wildlife management in tegration in cattle operations to design programs that foster sustainable attitudes towards beneficial wildlife practices by tapping existing important social networks to make wildlife friendly ranching practices a social norm.

PAGE 66

66 Table 3 1. Cattle rancher ( n = 1,093) likeliness to work with organizations over the next fiv e years in the southeastern United States 2009. Organization a SE b Factor 1 agricultural organization c Factor 2 wildlife organization c Farm Service Agency 3.7 0.0 0.913 0.184 NRCS 3.4 0.0 0.801 0.349 Cooperative Extension Ser vice 3.8 0.0 0.799 0.262 Non profit wildlife management and conservation groups d 2.8 0.0 0.147 0.868 State fish and wildlife agency 3.2 0.0 0.306 0.865 USFWS 3.0 0.0 0.411 0.816 a Ordered responses: 1 = very unlikely, 2 = unlikely, 3 = neither unlikely or likely, 4 = likely, 5 = very likely. b SE = 0.0 were <0.1 c Factor analysis with a Varimax rotation. d Examples included the National Wild Turkey Federation, Ducks Unlimited, Quail Unlimited, The Nature Conservancy, The Audubon Society, and the World W ildlife Fund.

PAGE 67

67 Table 3 2. Cattle rancher ( n = 1,093) agreement with outcome evaluation and attitudinal strength statements about combining wildlife management with beef cattle operations in the southeastern United States 2009. Statement a SE b c SE b If ranchers have wildlife enterprises like hunting leases, guided hunts, and ecotourism, it can contribute to their ranch. d 3.6 0.0 Establishing wildlife enterprises is important to my ranch. e 2.9 0.0 10.9 0.2 By seeing wildlife in their fields, pastures, and property, ranchers can know the land is healthy and productive. d 3.9 0.0 Seeing wildlife in my fields, pastures, and property is important to me. e 4.0 0.0 15.9 0.2 If ranchers have wildlife on their proper ty, they can participate in activities like bird watching, hunting, and fishing. d 4.1 0.0 Wildlife activities like bird watching, hunting, and fishing on my ranch are important to me. e 3.8 0.0 16.1 0.2 Wildlife and cattle can live together on a working ranch. d 4.1 0.0 Having both wildlife and cattle on my ranch is important to me. e 4.0 0.0 17.1 0.2 a Ordered responses: 1 = strongly disagree, 2 = disagree, 3 = neither disagree or agree, 4 = agree, 5 = strongly agree. b SE = 0.0 were <0.1 c Calculated attitude scores (outcome evaluationsd x attitudinal strengthe) potentially ranged from 1 to 25 with higher scores indicating a more positive attitude. d Outcome evaluation. e Attitudinal strength.

PAGE 68

68 Table 3 3. Cattle rancher ( n = 1,093) agreement with normative belief and motivation to comply statements about combining wildlife management with beef cattle operations in the southeastern United States 2009. Statement a SE b c SE My family would approve of people like me managing for wildlife. d 3.8 0.0 The opinions of my family matter to me. e 4.2 0.0 16.3 0.2 My friends and neighbors encourage people like me to manage for wildlife. d 3.5 0.0 My friends and neighbors views matter to me. e 3.7 0.0 13.1 0.2 Rancher co lleagues would approve of people like me managing for wildlife. d 3.7 0.0 The approval of fellow ranchers matters to me. e 3.7 0.0 14.0 0.1 State and federal agency staff (e.g. extension agents, USDA, FSA, NRCS, wildlife and natural resource departments, state foresters, etc.) encourage ranchers to manage for wildlife. d 3.7 0.0 Agency staff opinions matter to me. e 3.6 0.0 13.6 0.1 a Ordered responses: 1 = strongly disagree, 2 = disagree, 3 = neither disagree or agree, 4 = agree, 5 = strongly agree. b SE = 0.0 were <0.1 c Calculated subjective norm scores (normative beliefd x motivation to complye) potentially ranged from 1 to 25 with higher scores indicating a stronger normative influence. d Normative belief. e Motivation to comply.

PAGE 69

69 Table 3 4. Cattle rancher ( n = 1,093) agreement with control belief and control power statements about combining wildlife management with beef cattle operations in the southeastern United States 2009. Statement a SE b c SE Managing for wil dlife requires a lot of land. d 3.1 0.0 Because I have enough land, I am able to manage for wildlife. e 3.2 0.0 10.3 0.1 Managing for wildlife is expensive. d 2.9 0.0 Because I have enough funds, I am able to manage for wildlife. e 2.8 0.0 8.0 0.1 Mana ging for wildlife requires a lot of technical knowledge. d 2.7 0.0 Because I have knowledge of wildlife management, I am able to do it. e 3.2 0.0 8.7 0.1 Managing for wildlife requires a lot of time. d 2.8 0.0 Because I have enough time, I am able to m anage for wildlife. e 3.0 0.0 8.4 0.1 a Ordered responses: 1 = strongly disagree, 2 = disagree, 3 = neither disagree or agree, 4 = agree, 5 = strongly agree. b SE = 0.0 were <0.1 c Calculated perceived behavioral control scores (control beliefd x control powere) potentially ranged from 1 to 25 with higher scores indicating a stronger influence of the perceived behavioral control. d Control belief (reverse coded). e Control power.

PAGE 70

70 Table 3 5. Forced hierarchical regression model explaining how likely cattle ranchers ( n = 1,071) are to consider wildlife management in beef cattle operations in the southeastern United States 2009. Independent variables B t P R 2 Step 1: theory components <0.001 0.164 Attitude 0.014 0.229 6.340 <0.001 S ubjective norm 0.013 0.196 5.599 <0.001 Perceived behavioral control 0.002 0.021 0.689 0.491 Step 2: theory components and ranch characteristics <0.001 0.185 Attitude 0.012 0.208 5.735 <0.001 Subjective norm 0.012 0.177 4.973 <0.00 1 Perceived behavioral control 0.002 0.018 0.619 0.536 Total land <0.001 0.064 1.552 0.121 Number of cattle <0.001 0.107 2.566 0.010 Years raising cattle 0.001 0.022 0.539 0.590 Years owned property <0.001 0.011 0.275 0 .783 Work with agricultural agency 0.035 0.032 1.081 0.280 Work with wildlife agency 0.120 0.110 3.836 <0.001 Ranch diversity 0.183 0.039 1.329 0.184 Step 3: theory components, ranch characteristics, and demographics <0.001 0.202 Attitude 0.014 0.234 6.372 <0.001 Subjective norm 0.011 0.163 4.614 <0.001 Perceived behavioral control 0.002 0.025 0.829 0.407 Total land <0.001 0.060 1.470 0.142 Number of cattle <0.001 0.083 2.003 0.045 Years r aising cattle 0.001 0.020 0.486 0.627 Years owned property <0.001 0.001 0.027 0.979 Work with agricultural agency 0.058 0.053 1.765 0.078 Work with wildlife agency 0.142 0.130 4.510 <0.001 Ranch diversity 0.216 0.046 1.577 0.1 15 Age 0.007 0.076 2.575 0.010 Income 0.079 0.086 2.751 0.006 Education 0.022 0.032 1.050 0.294

PAGE 71

7 1 Table 3 6. Final stepwise regression model explaining how likely cattle ranchers ( n = 1,071) are to consider wildlife management in beef cattle operations in the southeastern United States 2009a. Independent variables B Attitude 0.015 0.247 Subjective norm 0.012 0.177 Likeliness to work with a wildlife agency or non profit group 0.142 0.130 Income 0.082 0.089 Age 0.007 0.0 82 a F = 19.078, P < 0.001, R 2 = 0.192

PAGE 72

72 CHAPTER 4 EXPLAINING CATTLE RANCHER PARTICIPATION IN COST SHARE FINANCIAL AND TECHNICAL ASSISTANCE PROGRAMS IN THE SOUTHEASTERN UNITED STATES Introduction Since the mid1980s, federal and state legislation has ena bled wildlife conservation and natural resource agencies to fund private lands conservation programs. The pioneer projects were described in the conservation provisions of the 1985 Food Securities Act, more commonly referred to as the Farm Bill. The Farm Bill conservation provisions are implemented by the United States Department of Agricultures (USDA) Natural Resources Conservation Service (NRCS) and Farm Service Agency. This legislation expanded conservation ideas of conserving soil and erodable lands from purely a food production standpoint to also include wildlife, water, and air quality. These programs fall loosely into three categories: financial assistance, technical assistance, and land retirement (Lambert et al. 2006). Most state wildlife and natural resource conservation agencies have variants of each type of program and funding is usually dependent on federal and state legislation (J. S. Sanders, Florida Fish and Wildlife Conservation Commission, personal communication). Land retirement prog rams involve long term or permanent land sequestration from many farming activities and are usually tied directly to the land deed or title, thus permanently limiting future land developments. These programs, although extremely beneficial for conservation and wildlife, appeal to a different type of agricultural producer usually those nearing retirement or low production, than financial and technical assistance programs (Lambert et al. 2006).

PAGE 73

73 Financial and technical assistance programs in the Farm Bill aim to improve conservation on working ranchlands and farmlands. The main f inancial and technical assistance programs that involve wildlife conservation are the Environmental Quality Incentives Program (EQIP) and Wildlife Habitat Incentives Program (WHIP). In these programs, the federal government allocates funds to the NRCS. The allocations are then used for technical assistance programs that inc lude conducting technical trainings and workshops, onfarm conservation planning initiatives, and educational materials. They are also used to financially incentivize onfarm projects by sharing the cost of approved conservation activities. Environmental Quality Incentives Program allocations have more than tripled from an initial $200 million/year in the 1996 Far m Bill to $ 627 million /year in 2003 (www.nrcs.usda.gov/programs/eqip) This trend has continued in recent years with WHIP more than doubling allocations from $21 million in 2003 to $57 million in 2008 (www.nrcs.usda.gov/Programs/whip) and EQIP nearly doubling again since 2003 with $1.2 billion allocated in 2008 ( www.nrcs.usda.gov/programs/eqip). Of all funds disbursed to agriculturalists, livestock producers have received 60% of EQIP funds (Lambert et al. 2006). As livestock producers, specifically cattle ranchers, are one of the key recipients of assistance, ranchland wildlife conservation can be improved as agencies technically support and fund many wildlifebeneficial ranch activities including prescribed grazing, prescribed burning, planting wildlifefriendly vegetation, and habitat restoration (Haufler and Ganguli 2007). Assuming financial and technical assistance programs are generally beneficial to wildlife conservation, it is important that agency field staff are able to enroll ranchers in them to maximize wildlife benefits, especially as program

PAGE 74

74 funding continues to increase. It is therefore imperative that the humandimensions of conservation assistance programs are investigated to better understand participation in these programs To date, the studies addressing participation in conservation programs have focused on the financial incentive aspects, specifically, socioeconomic descriptions of the participants (Lambert et al. 2006), economic valuations related to EQIP contract withdrawals (Cattaneo 2003), and program cost effectiveness (Claassen et al. 2008). This study addresses a gap in behavioral research relating to participation in financial and technical assistance program s by describing rancher participation in these programs using a promi nent social psychology model the Theory of Planned Behavior (see Chapter 3 for detailed framework description). Methods Rancher Survey I interviewed beef cattle producers and University of Florida livestock extension faculty in spring 2008 to design a sel f administered mail survey. Data from other interview s and focus groups in a concurrent study of private landowners in FL were also used to construct the survey (Willcox et al. 2010b). I sent survey drafts to 10 human dimensions of agriculture, wildlife, and forestry experts, 5 wildlife agency personnel, and 5 ranchers in FL and GA for review Drafts were improved and resent to reviewers several times for testing before finalization for content validity The resultant instrument contained 35 numbered questions over 9 pages (Appendix A) I asked an initial screening question to confirm the respondent was currently ranching. Ranchers were also asked questions about behavioral intentions related to future participation in cost share financial assistance programs and technical workshops and field days on various wildlife topics and attitudes, subjective norms, and perceived behavior controls related

PAGE 75

75 to those behavioral intentions with 5 point scales using ordered responses (i.e., 1 = very unlikely to 5 = v ery likely, and 1 = strongly disagree to 5 = strongly agree; Dillman et al. 2009). I also asked ranchers questions about past participation in programs, likeliness to work with agencies, land use, ranch characteristics, and demographics. Preceding the section related to cost share financial assistance programs, a brief paragraph was included explaining cost share programs to respondents unfamiliar with the term. In addition to using those data for ranching behavior analyses, demographic and ranch size data were contrasted with the 2007 Census of Agriculture to assess nonresponse error (USDA 2009). In September 2008, I purchased a sampling frame of 24,049 nondairy cattle producers f rom the Farm Index of CAS Inc. (www.cas online.com). T he Florida sampling frame was augmented with lists from UFs Cooperative Extension Service, as the purchased frame was low (227 ranchers) when compared with the 15,717 farms with beef cattle reported in the USDAs National Agricultural Statistics Service 2002 Census of Agric ulture (USDA 2004). The Mississippi list was also small, but supplemental lists could not be procured. The list was reduced by zip code to the study area in the Southern Coastal Plain and Southeastern Plains of Omernicks Level III Ecoregions, resulting in 15,023 addresses (Commission for Environmental Cooperation 1997). I stratified Florida by north and south at approximately 29N because the agricultural systems and the environment differ, with the south containing more planted and native range and nor th having more pine timber and native forest (Clouser et al. 200 7, Willcox et al. 2010b). N orth and south Florida were later combined as differences did not exist unlike the previous studies involving landowners and farmers. R andom

PAGE 76

76 samples of 500 ranche rs were drawn from each state and the two Florida strata, except for Mississippi as the sampling frame contained only 271 ranchers In that case, all Mississippi ranchers received surveys A n additional 500 ranchers from Georgia were drawn to conduct a f inancial incentive effect methodological experiment (Willcox et al. 2010a). My final sample contained 2,771 addresses. I conducted a five wave mailing following the Tailor ed Design Method (Dillman 2009), sending preletters, initial surveys, and reminder postcards in November 2008, replacement surveys in January 2009, and final replacement surveys in February 2009. Data Analyses I imputed m issing data using the expectationmaximization function with 20 iterations and 0.001 convergence to reduce uncertainty from missing values and use completedata tec hniques (Shafer and Graham 2002, SYSTAT 2007). L ikelihood ratio ( LR ) analyses were used to assess nonresponse error by comparing demographic and ranch size data with the 2007 Census of Agriculture (USDA 2009). Principal components analysis with a Varimax rotation and Chronbachs were used to assess reliability of the workshop behavioral intent variable and independent variables measuring attitudes, subjective norms, and perceived behavioral contr ol. Factor scores from the workshop behavioral intent variable were used in regression models. I calculated i ndependent variables for attitudes, subjective norms, and perceived behavioral control were using Theory of Planned Behavior protocols (Ajzen 1988) except for attitudes toward financial assistance programs as they were directly measured (Francis et al. 2004, Ajzen 2006). In that case, attitudinal factor scores were employed in the regression models Ranch diversity, ranch size, and frequency of collaborating with governmental and nonprofit conservation groups were also

PAGE 77

77 regressed with the dependent variable, as farms with a variety of crops and farmers who were frequently consulted technical experts have been found to more frequently adopt complex conservation practices (Lambert et al. 2006). A ranch diversity index was constructed by coding dummy variables for ranchers indicating if they had property in each of 11 different uses (i.e. planted timber, grazed native forest, ungrazed native forest, improved pasture, hayfields, silvopastures, sod, annual row crops, perennial groves or orchards, and wetlands). The mean of these 11 dummy variables was then calculated, resulting in a ranch diversity index potentially ranging from 01. Likeliness to co llaborate with six government and nonprofit groups were analyzed using factor analysis and factor scores were then used in the regression model s. Also, past participation in workshops and cost share financial assistance program variables were used in the regression model s as past participation can effect behavior ( Ouellette and Wood 1998 ). I used forced and stepwise ( FTE = 0.05, FTR = 0.10) threestage hierarchical regression to determine relationships between behavioral intent (dependent variables) and the independent v ariables. I fit the model with all three theory components first, adding ranch characteristics, likeliness to work with wildlife and agricultural agencies, and past program participation second, and a final global model that also included demographics third. Significant theory components were rechecked using correlation matrixes and general linear models with interaction terms to investigate potential moderator effects (Ajzen and Fishbein 2005) and multicollinearity (Appendices C and D). SYSTAT 12 was used for all analyses and statistical significance concluded at P

PAGE 78

78 homogeneity of variance, and those violating test assumptions were rank transformed pri or to analysis (Conover 1980). Results Response Demographics, and Ranch Characteristics Of the 2,771 surveys mailed, 1,634 ranchers responded, resulting in adjusted response rates of: overall = 59%, AL = 57%, FL = 72%, GA = 49% and MS = 56% (discounting undeliverable surveys; Chapter 2). After accounting for those returned blank or respondents answering no to the screening question indicating they were no longer active ranchers, I had 1,093 usable surveys. I then compared my results with the 2007 Census of Agriculture (USDA 2009) on equivalent demographic (i.e., gender, age, and ethnic group) and ranch size variables, and found nonresponse error in all my samples. Across all states, there were fewer young ranchers less than 35 years old ( LR = 80.73, df = 8, n = 69,008, P < 0.001) and ranches smaller than 57 ha ( LR = 1443.69, df = 11, n = 68,566, P < 0.001) t han the 2007 Census. Additionally, Florida had more women ( LR = 23.67, df = 1, n = 17,281, P < 0.001) and Mississippi had fewer other (than white) ethnic group respondents ( LR = 21.24, df = 5, n = 15,139, P < 0.001) than the 2007 Census. I found no differences in any other comparisons. Therefore, care should be taken when generalizing results to the entire population as the data best reflects the perceptions of producers 35 yrs old and those that have ranches 57 ha. Respondents were 87% male, 96% white, and averaged 61.0 0. 4 ( SE) yrs old. They owned and leased 597.7 43.5 ha, owned properties for 52.5 1.2 yrs, managed 110.2 12.3 cattle, had raised cattle on their current property for 43.5 1.0 yrs, and their mean ranch diversity index was 0.4 0.0. Likeliness to work with different types of agencies factored into two groupings we labeled agricultural organizations

PAGE 79

79 (NRCS, Farm Service Agency, and Cooperative Extension Service) and wildlife organizations (USFWS, state wildlife agency, and various nongovernmental wildlife groups) explaining 40% ( Chronbachs = 0.85) and 40% ( Chronbachs = 0.88) of the total variance, respectively (Table 4 1). Forty percent graduated from college with at least an associat es degree, 23% had some college courses 32% graduated from high school, and 6% had less than a high school diploma. Five percent of ranchers reported household incomes >$200,000; 3% at $150,000199,000; 15% at $100,000149,000; 41% at $50,00099,000; 25 % at $25,00049,000; and 11% at $024,999. In the past, 40% of ranchers had participated in cost share programs in the past and 23% in technical workshops or field days that included wildlife management. Theory of Planned Behavior Components Analyses Beha vioral intent Intention to participate in a cost share financial assistance pr ogram averaged 3.1 0.0 on a 5point scale. Likeliness to participate in a variety of potential workshop or field day topics averaged 2.7 0.0 (general wildlife habitat manage ment, game species management, agricultural systems planning that included wildlife management, grazing systems management for livestock and wildlife, endangered or listed species management, watchable wildlife such as songbirds, and wildlife enterprise es tablishment) and factored into one variable explaining 68% of the variance with a Chronbachs = 0.919 ( Table 25 ). Attitudes The three questions directly measuring attitudes regarding cost share financial assistance programs were grouped into one factor explaining 82% of the total variance with a Chronbachs = 0.89, demonstrating sufficient reliability (Table 4 1). The

PAGE 80

80 attitude scores were also all beneficial to ranchers, are good for the land and wildlife, and a worthwhile use of government funds. Regarding technical workshops and field days, ranchers scored high (3.7) on the outcome evaluation question that workshops and field days are a good way for ranchers to learn about wildlife management, but the attitudinal strength of the outcome evaluation was lower (3.2) (Table 4 1). Overall attitudes toward the behavior was then calculated as suggested by Ajzen (1988) resulting in a mean score of 12.3 0. 1 on a scale of 1 to 25, with higher numbers indicating more positive attitudes. Subjective norms Cost share financial assistance program subjective norms factored together, explaining 73% of the total variance with a Chronbachs = 0.87, indicating adequate reli ability. Cost share program normative belief sta tements were generally lower ( 3.3) than motivation to comply mean scores ( encouraged by various social groups to incorporate wildlife management into ranching activitie s and were positively influenced by these groups (Table 4 2). Overall subjective norms toward the behavior was then calculated as suggested by Ajzen (1988) resulting in a mean subjective norm score of 46.7 0.4 on a scale of 4 to 100, with higher numbers indicating more positive subjective norms. Workshop and field day technical assistance program subjective norms also factored together, explaining 75% of the total variance with a Chronbachs = 0.88, indicating adequate reliability. Again, ranch ers were somewhat encouraged ( 3.2) by various groups to participate in technical assistance programs, and were highly influenced by these groups (Table 4 2). Overall subjective norms toward the behavior was then calculated as suggested by Ajzen

PAGE 81

81 (1988) resulting in a mean subjective norm score of 45.2 0.4 on a scale of 4 to 100, with higher numbers indicating more positive subjective norms. Perceived behavioral control Cost share financial assistance program perceived behavioral control measures factored together explaining 50% of the total variance ( Chronbachs = 0.6 5 ; Table 4 3). Overall perceived behavioral control toward the behavior was then calculated as suggested by Ajzen (1988) resulting in a mean score of 25.4 0.3 on a scale of 3 to 75, with higher numbers indicating more positive subjective norms. Workshop and field day perceived behavioral control means also factored together accounting for 58% of the variance ( Chronbachs = 0.63; Table 4 3). Overall perceived behavioral control toward the behavior was then calculated as suggested by Ajzen (1988) resulting in a mean score of 33.1 0.3 on a scale of 3 to 75, with higher numbers indicating more positive subjective norms. Main Analyses The forced model was significa nt at each hierarchical step for both financial and technical program behavioral intentions (Tables 4 4 and 4 5 ). Once stepwise elimination procedures were included, the predictor variables present in the final stepwise cost share model differed from the workshop and field day model ( Tables 4 6 and 4 7 ). In the final stepwise cost share program model, one theory component, perceived behavioral control, was dropped, but several other predictor variables, including past participation in cost share programs, likeliness to work with an agricultural agency, and farm diversity, were all positively associated with behavioral intent. The only demographic variable remaining, age, was negatively associated with cost share program behavioral intent. All three theor y components positively

PAGE 82

82 contributed to the stepwise workshop and field day model, as well as likeliness to work with a wildlife agency or nonprofit group, likeliness to work with an agricultural agency, past participation in a wildlife workshop or field day, and education. Number of years raising cattle and age were negatively associated with workshop and field day behavioral intent. Discussion The Theory of Planned Behavior afforded good explanations of behavioral intent in both cost share financial assi stance programs and technical workshops and field days with one major difference, perceived behavioral control was important to technical but not cost share financial assistance programs. Other research involving southeastern ranchers integrating wildlife management with cattle activities (Chapter 3) or participating in conservation easements (Brain 2008) also found perceived behavioral control unimportant when predicting rancher behavioral intent. It has been theorized that perceived behavioral control b ecomes less important when the behavior is largely under volitional control, meaning that the decision or ability to conduct the behavior rests fully with the individual (Madden et al. 1992, Hrubes et al. 2001). Aside from ranchers not receiving announcem ents or literature from agencies regarding technical workshops and field days, financial assistance programs, or conservation easements; participating in cost share programs, attending workshops and field days, integrating wildlife management into cattle o perations, and enrolling in a conservation easement appear to be under volitional control. That is, it is ultimately an individual ranchers decision to engage in these behaviors, largely unmitigated by outside controls. A striking difference between att ending a workshop or field day and the other behaviors is that the act of participating in a cost share program, placing land in a conservation

PAGE 83

83 easement, or integrating wildlife management into cattle operations involves intense, long term commitments, whereas attending a workshop or field day is a short term behavior that does not necessarily have long lasting implications to ranchers or ranch operations. This difference has not yet been described in the literature, but warrants further investigation by s ocial psychologists comparing low vs. high commitment and long vs. short term behaviors through direct experimental or indirect metanalytical research using the Theory of Planned Behavior. Examining the important theory components more closely, subjective norms were most influential, followed by perceived behavioral control and attitudes, respectively, in explaining behavioral intent to participate in workshops and field days. For agencies or nonprofit groups wanting to increase participation in technical workshops or field days, it is crucial they widely promote and announce the events using existing social networks, as ranchers would be more inclined to participate when encouraged by important groups like their families, friends and neighbors, fellow ranchers, and agency personnel. By doing so, it would concurrently address aspects of perceived behavioral control and subjective norms. With respect to attitudes, the other major theory component explaining participation in technical workshops and field days, ranchers generally had positive attitudes about the activity, indicating workshops and field days are a good way for ranchers to learn about wildlife management. These positive attitudes should be fostered by agencies as continuing technical workshop and field day events with relevant and engaging topics should further increase positive attitudes and future participation.

PAGE 84

84 The important theory components to cost share program behavioral intent, subjective norms and attitudes, should also be used by ag encies to promote participation in these programs. Ranchers generally indicated that they thought cost share programs were beneficial to ranchers, wildlife, and the land while being a worthwhile use of government funds. Therefore, campaigns that continue to support positive attitudes toward cost share programs should also increase participation in them. As with participation in workshops and field days described above, subjective norms can also be utilized when promoting cost share programs. As well as direct ing promotion of cost share programs through established social networks, cost share program adoption should flow though social networks so long as they are perceived as beneficial by the ranching community (Rogers 2003). In addition to theory components past participation in programs were important in both cost share assistance and technical workshop and field day participation. Past behavior has been shown to be a strong predictor of future acts if the behavior becomes habitual or when people cons ciously deliberate before enacting a future behavior (Ouellette and Wood 1998). Due to the relative infrequency of enrolling in cost share programs or attending workshops or field days, it is doubtful these activities could become habitual. In cases wher e the past behavior is sporadic or intermittent, past behavior works in combination with other variables such as attitudes, subjective norms, and perceived behavioral control, to explain behavioral intent (Ouellette and Wood 1998). The addition of past be havior to the Theory of Planned Behavior model has been debated because to argue that we behave the way we do now because we preformed the behavior in the past begs the question as to why we previously behaved

PAGE 85

85 that way (Ajzen and Fishbein 2005). In both points of view, the main measured variable was frequency of the past behavior that focuses on habits, but in the case of my research, past behavior was calculated as a binary (yes or no) variable due to the infrequency of repeated past participation and t he doubt cost share program or workshop participation could ever become habitual. My research indicates that infrequent past behavior explains infrequent future behavioral intent, but more research is needed to determine if this trend exists in other situations. Regardless, from a conservation programming perspective, past participation in cost share and technical programs was positively correlated with behavioral intent, meaning that many who participated in the past will continue to do so in the future. Therefore, care should be taken to keep past participants engaged so they will continue to participate. Also, when promoting programs, special care should be taken to target nonparticipants, as once they participate, they are more likely to do so again. Likeliness to work with agricultural agencies was important to both cost share participation and likeliness to participate in wildlife workshops and field days. Additionally, likeliness to work with wildlife agencies explained participation in workshops and field days. These results were unsurprising as substantial interaction among ranchers, agencies, and nonprofit groups is required to participate in technical and financial assistance programs, as agencies sponsor them. With respect to cost share pr ograms, as likeliness to participate in them was not related to wildlife agency interactions, wildlife agencies with financial assistance programs should look to their colleagues in agricultural agencies for advice and consider collaborative efforts to pro mote them. In Florida and other southeastern states, joint efforts have been initiated

PAGE 86

86 among agricultural and wildlife agencies and nonprofit groups to collaboratively streamline financial and technical assistance programs to maximize effects by using the expertise and networks available to both agricultural and wildlife organizations ( J. S. Sanders, Florida Fish and Wildlife Conservation Commission, personal communication). Negative associations among age and behavioral intent were found for both cost sh are programs and technical workshops and field days. Similar trends have been shown nationally in agriculturalists, where younger farmers participated more often than older farmers in working lands programs like EQIP ( Lambert et al. 2006). Although it is important that all members of s ociety are included in taxpayer funded assistance programs, it is encouraging that younger ranchers are more inclined to participate in programs as the future of ranchland wildlife conservation lies with them. Another possi bility is that older ranchers are already knowledgeable and experienced, and technical workshops and field days would have greater appeal to younger ranchers. In a similar vein, number of years raising cattle was negatively associated with workshop and fi eld day participation. T herefore, the less experienced ranchers are more likely to seek knowledge by participating in technical assistance activities. Ranch diversity was positively associated with participation in cost share financial assistance program s. This could be explained as more diverse operations could potentially enact a variety of approved conservation practices just due to the diversity of the operation, whereas a more monotypic operation might not have as many opportunities. Finally, education was positively associated with participation in technical workshops and field days. Although unsurprising, care should be taken by wildlife and agricultural agencies

PAGE 87

87 attempting to reach parity participation, as less educated ranchers might be overlooked in these programs. In conclusion, as these types of technical and financial assistance programs have become the mainstay of state and federal natural resource agencies, it is important to investigate participation in them with robust social science theories to improve upon their success. Studies such as this can be used to improve program content and delivery to increase participation in them. Additionally, continued longitudinal repetition of studies similar to this should enable agencies to evaluate program effectiveness over time.

PAGE 88

88 Table 4 1. Cattle rancher ( n = 1,093) agreement statements directly measuring attitudes towards cost share financial assistance programs and indirectly measuring attitudes towards technical workshops and field days thro ugh outcome evaluation and attitudinal strength st atements in the southeastern United States 2009. Statement a SE b c SE Cost share financial assistance programs Cost share programs benefit cattle ranches 3.7 0.0 Cost sh are programs are good for land and wildlife conservation 3.8 0.0 Cost share programs are a worthwhile use of government funds 3.7 0.0 Technical assistance workshops and field days c Workshops are a good way for ranchers to learn how to c onduc t wildlife their ranch. d 3.7 0.0 If I attend wildlife workshops, I will be able to conduct wildlife management on my ranch. e 3.2 0.0 12.3 0.1 a Ordered responses: 1 = strongly disagree, 2 = disagree, 3 = neither disagree or agree, 4 = agree, 5 = strong ly agree. b SE = 0.0 were <0.1 c Calculated attitude scores (outcome evaluationd x attitudinal strengthe) potentially ranged from 1 to 25 with higher scores indicating a stronger influence of the attitude. d Outcome evaluation. e Attitudinal strength.

PAGE 89

89 Table 4 2. Cattle rancher ( n = 1,093) agreement statements indirectly measuring subjective norms towards cost share financial assistance programs and workshops and field days through normative belief and motivation to comply sta tements in the southeastern United States 2009. Statement a SE b c SE Cost share financial assistance programs My family encourages people like me to apply for cost share programs 3.0 0.0 The opinions of my family matter to me. e 4.2 0.0 12.6 0.1 My friends and neighbors encourage people like me to enter into cost share programs. 2.9 0.0 My friends and neighbors views matter to me. e 3.7 0.0 10.8 0.1 Other ranchers encourage p eople like me to apply for cost share programs. 3.0 0.0 The app roval of fellow ranchers matters to me. e 3.7 0.0 11.2 0.1 State and federal agency staff (e.g. extension agents, USD A, FSA, NRCS, wildlife and natural resource departments, state foresters, etc.) encourage ranchers to apply for cost share programs. d 3.3 0.0 Agency staff opinions matter to me. e 3.6 0.0 12.0 1.1 Technical assistance workshops and field days c My family encourages people like me to attend technical field days and workshops. 2.7 0.0 The opinions of my family matter to me. e 4.2 0.0 11.6 0.1 My friends and neighbors encourage people like me to attend technical field days and workshops. 2.6 0.0 My friends and neighbors views matter to me. e 3.7 0.0 9.9 0.1 Other ranchers encourage people like me to attend technical field days and workshops. 2.7 0.0 The approval of fellow ranchers matters to me. e 3.7 0.0 10.2 0.1 State and federal agency staff (e.g. extension agents, USD A, FSA, NRCS, wildlife and natural resource departments, state foresters, etc.) encourage ranchers to attend technical field days and workshops. d 3.2 0.0 13.5 0.1

PAGE 90

90 Table 42. Continued. Statement a SE b c SE Agency staff opinions matter to me. e 3.6 0.0 a Ordered responses: 1 = strongly disagree, 2 = disagree, 3 = neither disagree or agree, 4 = agree, 5 = strongly agree. b SE = 0.0 were <0.1 c Calculated attitude scores (outcome evaluationd x attitudinal strengthe) potentially ranged from 1 to 25 with higher scores indicating a stronger influence of the attitude. d Outcome evaluation. e Attitudinal strength.

PAGE 91

91 Table 4 3. Cattle rancher ( n = 1,093) agreement statements indirectly measuring perceived behavioral control towards cost share financial assistance programs and workshops and field days through agreement with control belief and control power statements in the southeastern United States 2009. Statement a SE b c SE Cost share financial assistance programs Generally, ranchers like having government staff on their property. d 2.5 0.0 Beca use I do not like having government agency staff on my property, I do not apply for cost share programs. ef 3.3 0.0 8.6 0.1 Ranchers think that the cost share application process is complicated. df 2.6 0.0 Because the cost share application process is co mplicated, I do not consider applying for them. df 3.0 0.0 8.4 0.1 Ranchers have heard of cost share programs. d 3.5 0.0 Because I have not heard of cost share programs, I do not consider applying for them. df 3.3 0.0 11.9 0.2 Ranchers think cost share p rograms are restrictive. df 2.6 0.0 Because cost share programs are restrictive, I do not consider applying for them. df 3.1 0.0 8.3 0.1 Technical assistance workshops and field days State and federal agencies make announcements to ranchers about wi ldlife workshops. d 3.2 0.0 Because I receive announcements about wildlife workshops, I am able to attend them. e 2.8 0.0 9.3 0.1 Generally, ranches are suitable for wildlife management. d 3.6 0.0 Because my ranch is suitable for wildlife management, I attend workshops. e 3.0 0.0 11.3 0.1 Generally, ranchers are knowledgeable about wildlife management. d 3.4 0.0 Because I already know enough about wildlife management, I do not attend workshops. ef 3.7 0.0 12.5 0.1 a Ordered responses: 1 = strongly dis agree, 2 = disagree, 3 = neither disagree or agree, 4 = agree, 5 = strongly agree. b SE = 0.0 were <0.1

PAGE 92

92 c Calculated perceived behavioral control scores (control beliefd x control powere) potentially ranged from 1 to 25 with higher scores indicating a stro nger influence of the perceived behavioral control. d Control belief. e Control power. f Reverse coded.

PAGE 93

93 Table 4 4. Forced three step hierarchical regression model explaining how likely cattle ranchers ( n = 1,093) are to participate in cost sh are financial assistance pr ograms in the southeastern United States 2009. Independent variables B F P R 2 Step 1: theory components <0.001 0.149 Attitude 94.917 0.309 9.320 <0.001 Subjective norm 2.748 0.133 4.018 <0.001 Perceived behavioral control 0.628 0.021 0.672 0.502 Step 2: theory components and ranch characteristics <0.001 0.1 80 Attitude 79.476 0.259 7.589 <0.001 Subjective norm 1.662 0.080 2.343 0.019 Perceived behavioral control 1.131 0.037 1.217 0.224 Past participation in cost share programs 78.003 0.125 4.070 <0.001 Work with agricultural agency 24.693 0.080 2. 516 0.012 Work with wildlife agency 13.627 0.044 1.570 0.117 Ranch diversity 62.359 0.048 1.637 0.102 Total land 0.002 0.019 0.460 0.646 Number of cattle 0.023 0.030 0.727 0.468 Years raising cattle 0.137 0.014 0.347 0.729 Years owned property 0.212 0.028 0.705 0.481 Step 3: theory components, ranch characteristics, and demographics <0.001 0.192 Attitude 75.945 0.247 7.252 <0.001 Subjective norm 1.766 0.086 2.503 0.012 Perceived behavioral control 0.324 0.011 0.342 0.732 Past p articipation in cost share programs 76.633 0.123 4.021 <0.001 Work with agricultural agency 18.597 0.061 1.874 0.061 Work with wildlife agency 7.848 0.026 0.894 0.372 Ranch diversity 68.444 0.052 1.798 0.073 Total land 0.002 0.027 0.661 0.509 Num ber of cattle 0.016 0.021 0.506 0.613 Years raising cattle 0.216 0.022 0.548 0.584 Years owned property 0.127 0.017 0.424 0.672

PAGE 94

94 Table 44. Continued. Independent variables B F P R 2 Age 2.920 0.117 3.968 <0.001 Income 8.801 0.034 1.100 0.272 Education 6.118 0.031 1.042 0.298

PAGE 95

95 Table 45. Forced three step hierarchical explaining how likely cattle ranchers ( n = 1,093) are to participate in wildlife management workshops and field days in the southeastern United States 20 09. Independent variables B t P R 2 Step 1: theory components <0.001 0.383 Attitude 0.031 0.138 4.784 <0.001 Subjective norm 0.022 0.322 10.738 <0.001 Perceived behavioral control 0.028 0.277 9.093 <0.001 Step 2: theory components and ranch characteristics <0.001 0.455 Attitude 0.023 0.105 3.814 <0.001 Subjective norm 0.018 0.257 8.828 <0.001 Perceived behavioral control 0.020 0.195 6.490 <0.001 Past participation in workshops and field days 0.347 0.147 5.956 <0.001 Work with agricultural agency 0.160 0.160 6.4 58 <0.001 Work with wildlife agency 0.168 0.168 7.127 <0.001 Ranch diversity 0.104 0.024 1.032 0.302 Total land <0.001 0.003 0.083 0.934 Number of cattle <0.001 0.009 0.280 0.780 Years raising cattle 0.002 0.068 2.056 0.040 Years owned p roperty <0.001 <0.001 0.015 0.988 Step 3: theory components, ranch characteristics, and demographics <0.001 0.466 Attitude 0.021 0.096 3.498 <0.001 Subjective norm 0.018 0.266 9.209 <0.001 Perceived behavioral control 0.019 0.190 6.400 <0.001 P ast participation in workshops and field days 0.317 0.134 5.446 <0.001 Work with agricultural agency 0.143 0.143 5.757 <0.001 Work with wildlife agency 0.151 0.151 6.374 <0.001 Ranch diversity 0.107 0.025 1.068 0.286 Total land <0.001 0.004 0.125 0 .900 Number of cattle <0.001 0.025 0.728 0.467 Years raising cattle 0.002 0.063 1.926 0.054 Years owned property <0.001 0.011 0.351 0.726

PAGE 96

96 Table 45. Continued. Independent variables B t P R 2 Age 0.007 0.087 3.663 <0.001 Income 0.004 0.005 0.205 0.838 Education 0.036 0.057 2.313 0.021

PAGE 97

97 Table 4 6. Final stepwise model explaining how likely cattle ranchers ( n = 1,093) are to participate in cost share financial assistance programs in the southeastern United States 2009a. Independen t variables B Attitude 75.280 0.245 Subjective norm 1.877 0.091 Past participation in cost share programs 74.755 0.120 Likeliness to work with an agricultural agency 18.309 0.060 Ranch diversity 74.886 0.057 Age 3.013 0.121 a F = 41.936, P <0.001, R 2 = 0.188 Table 4 7. Final stepwise model explaining how likely cattle ranchers ( n = 1,093) are to participate in wildlife management workshops and field days in the southeastern United States 2009a. Independent variables B Attitude 0.022 0.099 Subjective nor m 0.018 0.264 Perceived behavioral control 0.019 0.191 Likeliness to work with a wildlife agency or non profit group 0.153 0.153 Likeliness to work with an agricultural agency 0.147 0.147 Past participation in a workshop or field day 0.318 0.135 Numbe r of years raising cattle 0.002 0.054 Age 0.007 0.085 Education 0.036 0.056 a F = 104.471, P < 0.001, R 2 = 0.465

PAGE 98

98 CHAPTER 5 EFFECTS OF TOKEN FINANCIAL INCENT IVES ON RESPONSE RATES AND ITEM NONRESPONSE FOR MAIL SURVEYS Introduction Incentives have a well established track record for improving response rates of mail surveys, second only to using a series of contacts tailored to the recipients ( Dillman et al. 2009). Previous studies have shown that the likelihood of the survey being returned can be i ncreased by including a few dollars with the survey request ( Lesser et al. 2001, Jobber et al. 2004, Groves et al. 2006). Over the last few years, the amount of the typical incentive has increased, in part to compensate for the decline in the real value o f smaller amounts For example, recent studies by Messer and Dillman (2009) and Smyth et al. ( in press ) tested the effect of a $5 incentive. What is not known is whether a smaller amount retains the necessary symbolism to invoke a more generalized social exchange, as hypothesized by Dillman et al. (2009), to improve response rates I f a small financial incentive is still effective, it is important to know whether the higher response rate improves data quality (i.e., reduces nonresponse error or item nonr espo nse). Item nonresponse can be exacerbated for questions that are perceived to be sensitive, and frequently occurs when respondents are asked about income, politics, or religion (Dillman et al. 2009). In human dimensions of wildlife studies, item nonr esponse can be acute when asking private landowners questions about threatened or endangered species (Brook et al. 2003). I test the effectiveness of a small cash incentive using a $1 Jefferson coin as an incentive in a wildlife survey administered to a s ample of cattle producers

PAGE 99

99 Methods I purchased a sampling frame of 9,193 non dairy cattle producers in Georgia found on the Farm Index of CAS Inc. during September 2008 (www.cas online.com) I then reduced the list by zip code to my study area in the Sout heastern Plains and Southern Coastal Plain of Omernicks Level III Ecoregions, resulting in 3,789 addresses (Commissio n for Environmental Cooperation 1997) I drew two 500 address random sa mples from this list: one experimental list that received coins and one control list without coins. The only difference between the two groups was that the experimental group had a $1 Jefferson coin glued to the top right corner of the cover letter and a paragraph in the cover letter stating: We have included a token o f our appreciation for your thoughtful participation. Attached to this letter is the Thomas Jefferson $1 coin issued as part of the new presidential dollar coin series We selected President Jefferson because, in addition to his accomplishments as one of our founding fathers, he was also a champion of agriculture and a pioneer in wildlife ecology. I did not include additional incentives or text referring to incentives in subsequent mailings. I employed a fivewave mailing that included a preletter, surv ey with cover letter and business reply envelope, reminder postcard, second survey with a different cover letter and business reply envelope for nonrespondents, and final survey with a different cover letter and business reply envelope for nonrespondents ( Dillman et al. 2009 ) I strayed from Dillman et al.s (2009) Tailored Design Method due to budget constraints, choosing to use a financial incentive rather than first class postage, stamped return envelopes, and certified mail to increase response. I mai led preletters, initial surveys, and reminder postcards in November 2008, replacement questionnaires in January 2009, and final replacement questionnaires in February 2009. Although Dillman et al.

PAGE 100

100 (2009) suggest not mailing surveys over holiday periods, r anchers and livestock extension agents suggested this time frame as cattle production work is not as intense during these months when compared with the rest of the year. The questionnaire consisted of 142 items over nine pages (Appendix A) printed in a booklet format. The survey began by asking an initial screening question Do you own beef cattle? Subsequent questions explored rancher attitudes toward wildlife, wildlife problems, wildlife conservation behaviors, participation in financial and technical a ssistance programs, ranch characteristics and demographics To assess nonresponse error, I compared demographic and ranch characteristic variables among experimental and control groups and the actual beef cattle rancher population in Georgia according to the United States Department of Agriculture National Agricultural Statistics Service 2007 Census of Agriculture using tables related to demographics of the North American Industry Classification S ystem for Beef Cattle Ranching and Farming 112111 ( USDA 200 9 ) I also compared item nonresponse between experimental and control groups for all demographic variables, two potentially controversial questions regarding threatened or endangered species, and wildlife questions regarding attitudes toward songbirds, wi ld turkey ( Meleagris gallopavo) white tailed deer ( Odocoileus virginianus ) and wildlife crop damage. I used likelihood ratio for all comparisons and considered differences significant at P Results Of the 1,000 questionnaires mailed, four were returned undeliverable, two from the experimental and two from the control group (Table 5 1). The experimental group returned 267 questionnaires for a 54% response rate, while the control group returned 224 surveys for a 45% response rate. The experimental group had a higher response

PAGE 101

101 rate than the control group, with a nine percentage point increase response when a coin incentive was received, LR (1, n = 996) = 7.44, p = .006 = .09 The effe ct was most prominent during the first mailing when the coin was included, with the experimental group response rate nine percentage points higher than the control This difference reduced to two and one percentage points on the second and third replaceme nt questionnaire mailings respectively. After accounting for multiple mailings and associated costs, the experimental group cost $0.19 more per response than the control group. I found no differences between experimental (70%) and control (66%) groups ans wering Yes to the screening question, indicating they currently owned beef cattle, LR (1, n = 433) = 1.03, p = .340 = .01 With respect to demographic questions, I collapsed ethnic group bi ns into White and Other as the number of responses (other than White) was not enough for practical comparisons. I found no differences between experimental and control groups for any of the demographic and ranch size variables measured (Table 5 2). The only directly comparable data among my samples and the 20 07 Census of Agriculture were for age, sex, ethnicity, and farm size (Table 5 3). For both experimental and control groups, sex did not differ among my samples and the census, but farm size and age did, with my samples generally having fewer farms with < 70 acres in size and a smaller number of respondents < 45 years than the census. Additionally, ethnic group differed between control and census. Item nonresponse ranged from 210% for demographic variables and 610% for questions related to threatened or endangered species, deer, wild turkey, songbirds, and wildlife crop damage, with no differences detected between experimental and control groups

PAGE 102

102 (Table 5 4). Item nonresponse for a sensitive question about income (910%) was higher than other demographic questions about age (23%), sex (4 5%), ethnicity (24%), and education (34%). Similarly, item nonresponse rates for sensitive questions about endangered species (610%) were higher than questions about crop damage (67%), songbirds (7%), deer (6%), and wild turkey (6%). Discussion Although some researchers have increased token financial incentives amounts to $5 recent ly (Messer and Dillman 2009, Smyth et al. in press), it appears that a smaller amount still serves to increase the response rate. The inc lusion of a $1 coin resulted in a nine percentage point increase in response, but this also is less than the average increase of 19 percentage points for a $2 incentive (Lesser et al. 2001) and the 17 percentage point increase for a $5 incentive (Messer and Dillman 2009). While the incentive increased the response rate, it did so without creating differences between the experimental and control groups for any demographic or farm size variables. Of data that were comparable with the 2007 Census of Agriculture, demographics were not different for sex, but the control group differed from the census on ethnic group. Both the incentive and the control groups did not record as many smaller ranches or younger ranchers. In practical terms, it appears the main benefit of using the coin incentive was that it generated more data for analysis and reduced ethnic group nonresponse error Item nonresponse was essentially unaffected by the $1 coin. This is similar to other studies that conducted item nonresponse experime nts using prepaid monetary incentives (Davern et al. 2003, Messer 2009, Shettle and Mooney 1999). For demographics, the sensitive question regarding income had higher item nonresponse

PAGE 103

103 than questions regarding sex, age, ethnicity, and education. This was also true for wildlife issues regarding threatened or endangered species, as item nonresponse was higher for those questions than questions about general wildlife crop damage and attitudes about songbirds, deer, and wild turkey. In this case, token financ ial incentives did not decrease item nonresponse for sensitive issues, and were not effective in enticing ranchers to express their opinions about threatened or endangered species. Ultimately, the decision to use financial incentives is often based upon the additional cost. In my case, the cost of responses receiving a coin was only $0.19 per response higher than those not receiving the coin. Although a subjective conclusion, the additional $0.19 appears to be a reasonable amount to pay for the nine percentage point increase in response and a reduction in ethnicity nonresponse error. Although not the case in my research, some human subjects research review panels require additional justification before project approval when using financial incentives in m ail surveys, therefore increasing labor costs and project rejection risks. As the incentive effect is most prominent during the initial questionnaire mailing, financial incentives could boost response rates for short duration research that will not include second or third replacement questionnaire mailings The initial investment to include financial incentives with a survey may seem high, but the long term benefits can outweigh the cost.

PAGE 104

104 Table 51. Response rates and costs for wildlife questionnaires mailed to cattle ranchers with and without a one dollar coin incentive. Experimental Control Mailed Responded Response rate Mailed Responded Response rate Initial survey a 498 161 32% 498 115 23% Replacement survey b 337 66 20% 383 66 17% Second r eplacement survey b 271 40 15% 317 43 14% Total 1106 267 54% 1198 224 45% Unit cost c $1.39 $1.39 Printing and mailing cost $1537.34 $1665.22 Incentives $498 $0 Total cost $2035.34 $1665.22 Cost/response $7.62 $7.43 a Four addresses, two experimental and two control, were undeliverable and not counted as mailed b Number of replacement surveys mailed is a close approximation using the previously mailed number previously responded because occasionally mailings and responses overlapped in transit, making exact replacement mailed and received calculations difficult c Unit cost for printing and mailing one 8.5 x 11 survey with cover letter and business reply envelope by nonprofit mail excluding preletters, postcards, incentives, and return business reply postage

PAGE 105

105 Table 5 2 Demographic and ranch size variables for wildlife questionnaires mailed to cattl e ranchers with and without a one dollar coin incentive Coin % No Coin% LR P Age Group (years) (7, n = 300) = 3.77 0 .806 0 .11 Under 35 0 0 35 44 6 5 45 49 8 8 50 54 10 13 55 59 17 14 60 64 14 14 65 69 19 15 70 and older 26 32 Sex (1, n = 294) = 0 .79 0 .373 0 .05 Male 90 87 Fe male 10 14 Ethnic Group (1, n = 299) = 2.34 0 .126 0 .09 White 95 98 Other 5 2 Ranch Size (10, n = 296) = 5.58 0 .849 0 .14 Under 50 acres 11 9 50 69 acres 5 5 70 99 acres 14 19 100 139 acres 10 13 140 179 acres 10 8 180 2 19 acres 10 6 220 259 acres 5 7 260 499 acres 20 21 500 999 acres 10 6 1000 1999 acres 6 6 >2000 1 2 Income (5, n = 278) = 2.38 0 .794 0 .09 $0 24999 15 13 $25000 49999 28 31 $50000 99999 35 34 $100000 149999 16 19 $150000 199999 3 1 >$200000 5 3

PAGE 106

106 Table 52. Continued. Coin % No Coin% LR P Education (6, n = 296) = 1.30 0 .972 0 .07 < high school 7 6 High school 34 37 Some college 22 25 Associates degree 4 4 Bachelors deg ree 21 18 Masters degree 9 6 Doctorate 4 5

PAGE 107

107 Table 5 3 Demographic and ranch size variables for the 2007 Census of Agriculture compared with wildlife questionnaires mailed to cattl e ranchers with and without a one dollar coin incentive. Censu s % Census Vs Coin LR P Cramers V Census Vs No Coin LR P Cramers V Age Group (years) (8, n = 15968) = 403.22 <0 .001 0 .46 (8, n = 15924) = 409.90 < 0 .001 .54 Under 35 5 35 44 10 45 49 9 50 54 11 55 59 13 60 64 15 65 69 14 70 and older 24 Sex (1, n = 15964) =. 0 14 0 .705 0 .00 (1, n = 15922) = .73 0 .394 .01 Male 89 Female 11 Ethnic Group (1, n = 15967) = 0 .08 0 .780 0 .00 (1, n = 15922) = 4.42 0 .04 White 95 Other 5 Ranch Size (10, n = 15965) = 121.50 <0 .001 0 .09 (10, n = 16096) = 707.47 < 0 .001 .08 Under 50 acres 40 50 69 acres 12 70 99 acres 11 100 139 acres 11 140 179 acres 6 180 219 acres 5 220 259 acres 3 260 499 acres 8 500 9 99 acres 3

PAGE 108

108 Table 53. Continued. Census % Census Vs Coin LR P Cramers V Census Vs No Coin LR P Cramers V 1000 1999 acres 1 >2000 1 Table 5 4 Demographic and threatened or endangered species question item nonresponse percentage comparisons for wildlife questionnaires mailed to cattl e ranchers with and without a one dollar coin incentive. Experimental % Control % LR (1, n = 293) P Age 2 3 0 .61 0 .436 0 .05 Sex 4 5 0 .07 0 .815 0 .02 Ethnicity 2 4 1.31 0 .253 0 .07 Inco me 9 10 0 .18 0 .673 0 .03 Education 3 4 0 .23 0 .635 0 .03 Problems caused by constraints from endangered species regulations 9 10 0 .04 0 .844 0 .01 Attitudes about threatened or endangered species on your property 8 6 0 .39 0 .532 0 .04 Problems caused by wildl ife damaging crops 7 6 0 .29 0 .593 0 .03 Attitudes about songbirds on your property 7 7 0 .00 0 .985 0 .00 Attitudes about deer on your property 6 6 0 .02 0 .898 0 .01 Attitudes about wild turkey on your property 6 6 0 .02 0 .898 0 .01

PAGE 109

109 CHAPTER 6 CONCLUSIONS Thi s study presented a breadth of knowledge regarding rancher and agency perspectives on wildlife management and participation in cost share and financial assistance programs. With respect to the rancher and agency personnel comparisons of perceptions they corresponded on issues that included intrinsic wildlife benefits and liked wildlife types. Agencies should build upon these similarities and develop programs that promote rancher preferred wildlife and intrinsic wildlife benefits. Although it may be prem ature to develop widespread programs that focus on wildlife enterprises as ranchers currently dont perceive they benefit from them, the topic warrants further investigation as both agency staff and ranchers see the potential for them. F uture research cou ld include cost benefit analyses, willingness to pay, and rancher wildlife enterprise needs assessment research before designing and promoting wildlife enterprise programs on a broad scale. While agency personnel overestimated how ranchers perceive wildlife damage, the majority of ranchers already control unwanted species. This overestimation might be caused by agencies receiving complaints without a counterbalance of positive wildlife management reports. Agencies should continue to address wildlife problems with ranchers to improve their relationship with ranchers and increase positive attitudes about agencies, which should improve or increase wildlife management on ranches as the level of trust increases. The social psychology model, the Theory of Planned Behavior, proved a useful tool to predict rancher participation in wildlife technical field days and workshops as all three components: attitudes, subjective norms, and perceived behavioral control, contributed to the model. The model was less predicti ve of ranchers intentions to

PAGE 110

110 integrate wildlife management into their c attle operations and participate in cost share financial incentives programs as one theory component, perceived behavioral control did not add to the model. It is possible that short t erm or lower commitment behaviors like participating in workshops work better with the model than long term higher commitments like changing ranching practices or enrolling in cost share programs, and this aspect of the theory should be further explored by social psychologists. From a practical standpoint, agencies should consider approaches like social marketing or diffusion of innovations to promote programs, because increasing attitudes and using existing social networks should increase participation in them. Past experience in a program was also a strong predictor of participation in cost share programs and workshops and field days. O nce ranchers partook of an agency sponsored assistance program, they were more likely to do so again in the future, thus indicating a high level of satisfaction with cu rrent programs. Therefore agencies should maintain current programs with relevant topics to continue engaging past participants for their future fidelity while actively promoting programs to nonparticipants to reach a larger audience. The experiment testing if a small financial incentive increases response rates and decreases nonresponse error in human dimensions of wildlife surveys was successful. Especially if researchers have limited time frames and can only conduct one or two mailings, small financial incentives should increase response rates and decrease non response error. It is unlikely the financial incentive will decrease item nonresponse for sensitive questions about endangered species or househ old income, so for such topics, other strategies should be explored to reduce item nonresponse.

PAGE 111

111 In conclusion, the research presented in this document is largely exploratory, but can be used as a baseline by which to monitor future agency program impacts. These results should be useful for agencies to improve wildlife management assistance programs for cattle ranchers and monitor future program success .

PAGE 112

112 APPENDIX A SOUTHEAST CATTLE RANCHER WILDLIFE MANAGEMENT SURVEY

PAGE 113

113

PAGE 114

114

PAGE 115

115

PAGE 116

116

PAGE 117

117

PAGE 118

118

PAGE 119

119

PAGE 120

120

PAGE 121

121

PAGE 122

122

PAGE 123

123

PAGE 124

124 APPENDIX B BIVARIATE CORRELATIONS OF ALL VARIABL ES INCLUDED IN SOUTHEASTERN UNITED STATES RANCHERS LIKELINESS TO CONSIDER WILDLIFE MANAGEM ENT WHEN CONDUCTING RANCH ACTIVITIES MODEL, 2009. Variables a 1 b 2 b 3 b 4 b 5 6 7 8 9 b 10 b 11 b 12 13 b Behavi oral intent (dependent) b r P Attitude b r P 0.377 <0.001 Subjective norm b r P 0.322 <0.001 0.605 <0.001 Perceived behavioral control b r P 0.129 <0.001 0.342 <0.001 0.251 <0.001 Total land r P 0.010 0.756 0.037 0.224 0.030 0.315 0.053 0.082 Number of Cattle r P 0.036 0.241 0.042 0.168 0.031 0.308 0.050 0.097 0.741 <0.001 Years raising cattle r P 0.053 0.081 0.030 0.328 0.095 0.02 0.026 0.389 0.104 <0.001 0.150 <0.001 Years owned pro perty r P 0.042 0.173 0.016 0.603 0.045 0.139 0.114 <0.001 0.053 0.080 0.066 0.029 0.720 <0.001 Work with agricultural agency b r P 0.109 <0.001 0.214 <0.001 0.257 <0.001 0.095 0.002 0.027 0.367 0.064 0.35 0.130 <0.001 0.124 <0.001 Work with wi ldlife agency b r P 0.186 <0.001 0.215 <0.001 0.198 <0.001 0.049 0.103 0.040 0.185 0.027 0.381 0.015 0.614 0.007 0.810 <0.001 1.000 Ranch diversity b r P 0.063 0.040 0.095 0.002 0.058 0.055 0.059 0.051 0.080 0.008 0.109 <0.001 0.182 <0.001 0.153 <0.00 1 0.218 <0.001 0.091 0.003 Age r P 0.027 0.374 0.224 <0.001 0.056 0.063 0.072 0.017 0.012 0.699 0.053 0.077 0.117 <0.001 0.127 <0.001 0.117 <0.001 0.165 <0.001 0.021 0.498 Income b r P 0.060 0.049 0.136 <0.001 0.087 0.004 0.148 <0.001 0.150 <0.001 0.190 <0.001 0.044 0.150 0.007 0.819 0.129 <0.001 0.116 <0.001 0.154 <0.001 0.223 <0.001 Education r P 0.021 0.486 0.112 <0.001 0.075 0.013 0.074 0.015 0.063 0.038 0.096 0.001 0.029 0.331 0.015 0.623 0.154 <0.001 0.072 0.018 0.055 0.069 0.138 <0.001 0.381 <0.001 a All variables checked for multicollinearity. As no values for r exceeded 0.741 (R 2 = 0.549; number of cattle vs. total land), no R 2 values were close to 1.0, indicating an absence of multicollinearity (Agresti and Finlay 1997). b Al l theory components significant with the dependent variable ( P (Ajzen and Fishbein 2005). The strongest potential relationship (attitudes vs. work with wildlife agency, r = 0.215) was modeled as an interaction term and did not contribute to the model ( P = 0.093).

PAGE 125

125 APPENDIX C BIVARIATE CORRELATIONS OF ALL VARIABL ES INCLUDED IN SOUTHEASTERN UNITED STATES RANCHERS LIKELINESS TO PARTICIPATE IN COST SHARE PROGRAMS, 2009. Variables a 1 b 2 b 3 b 4 b 5 6 7 8 9 b 10 b 11 b 12 13 b 14 b Behavioral intent (dependent) b r P Attitude b r P 0.369 <0.001 Subjective norm b r P 0.282 <0.001 0.504 <0.001 Perceived behavioral control b r P 0.135 <0.001 0.354 <0.001 0. 348 <0.001 Past cost share participation r P 0.259 <0.001 0.323 <0.001 0.318 <0.001 0.256 <0.001 Total land r P 0.058 0.056 0.020 0.505 0.042 0.165 0.003 0.909 <0.001 0.995 Number of cattle r P 0.072 0.018 0.026 0.386 0.026 0.388 0.005 0.874 0.052 0.087 0.741 <0.001 Years managing cattle r P 0.054 0.077 0.049 0.107 0.131 <0.001 0.026 0.391 0.090 0.003 0.104 0.01 0.150 <0.001 Years working property r P 0.030 0.318 0.042 0.167 0.082 0.007 0.020 0.520 0.093 0 .002 0.053 0.080 0.066 0.029 0.720 <0.001 Work with agricultural agency r P 0.253 <0.001 0.397 <0.001 0.385 <0.001 0.222 <0.001 0.284 <0.001 0.027 0.367 0.064 0.035 0.130 <0.001 0.124 <0.001 Work with wildlife agency r P 0.101 0.001 0.129 <0.0 01 0.173 <0.001 0.049 0.106 0.041 0.170 0.040 0.185 0.027 0.381 0.015 0.614 0.007 0.810 <0.001 1.000 Ranch diversity r P 0.125 <0.001 0.084 0.006 0.090 0.003 0.149 <0.001 0.198 <0.001 0.080 0.008 0.109 <0.001 0.182 <0.001 0.153 <0.001 0.218 <0.001 0. 091 0.003 Age r P 0.155 <0.001 0.090 0.003 0.010 0.748 0.149 <0.001 0.024 0.432 0.012 0.699 0.053 0.077 0.117 <0.001 0.127 <0.001 0.117 <0.001 0.165 <0.001 0.021 0.498 Income r P 0.075 0.014 0.111 <0.001 0.051 0.094 0.029 0.342 0.087 0.004 0.150 <0.001 0.190 <0.001 0.044 0.150 0.007 0.819 0.129 <0.001 0.116 <0.001 0.154 <0.001 0.223 <0.001 Education r P 0.086 0.004 0.087 0.004 0.040 0.189 0.073 0.016 0.061 0.045 0.063 0.038 0.096 0.001 0.029 0.331 0.015 0.623 0.154 <0.001 0.072 0.018 0 .055 0.069 0.138 <0.001 0.381 <0.001 a All variables checked for multicollinearity. As no values for r exceeded 0.741 (R2 = 0.549; number of cattle vs. total land), no R2 values were close to 1.0, indicating an absence of multicollinearity (Agresti and Finlay 1997). b All theory components significant with the dependent variable ( P (Ajzen and Fishbein 2005). The strongest potential relationship (attitude vs. past cost share participation, r = 0.323) was modeled as an interaction term and did not contribute t o the model ( P = 0.066. Age was also checked as an interaction with attitude, as logically, it is possible younger ranchers might be more inclined to have positive attitudes about new programs,, but the interaction was also not sign ificant ( P = 0.414).

PAGE 126

126 APPENDIX D BIVARIATE CORRELATIONS OF ALL VARIABL ES INCLUDED IN SOUTHEASTERN UNITED STATES RANCHERS LIKELINESS TO PARTICIPATE IN A VARIETY OF WOR KSHOP TOPICS, 2009. Variables a 1 b 2 b 3 b 4 b 5 6 7 8 9 b 10 b 11 b 12 13 b 14 b Behavioral intent (dependent) b r P Attitude b r P 0.437 <0.001 Subjective norm b r P 0.545 <0.001 0.487 <0.001 Perceived behavioral control b r P 0.529 <0.001 0.512 <0.001 0.562 <0.001 Past workshop participation r P 0.380 <0.001 0.233 <0.001 0 .318 <0.01 0.370 <0.001 Work with farming agency r P 0.346 <0.001 0.277 <0.001 0.273 <0.001 0.320. <0.001 0.196 <0.001 Work with wildlife agency r P 0.311 <0.001 0.156 <0.001 0.251 <0.001 0.204 <0.001 0.153 <0.001 <0.001 1.000 Ranch diversity r P 0.145 <0.001 0.126 <0.001 0.088 0.004 0.146 <0.001 0.133 <0.001 0.218 <0.001 0.091 0.003 Total land r P 0.034 0.262 0.017 0.578 0.044 0.145 0.065 0.032 0.082 0.007 0.027 0.367 0.040 0.185 0.080 0.008 Number of cattle r P 0.032 0.290 0.002 0.937 0.040 0.187 0.068 0.025 0.087 0.004 0.064 0.035 0.027 0.381 0.109 <0.001 0.741 <0.001 Years owned cattle r P 0.010 0.750 0.006 0.846 0.091 0.003 0.090 0.003 0.067 0.027 0.130 <0.001 0.015 0.614 0.182 <0.001 0.104 0.001 0.150 <0.001 Years owned land r P 0.002 0.938 0.011 0.722 0.044 0.144 0.059 0.050 0.004 0.894 0.124 <0.001 0.007 0.810 0.153 <0.001 0.053 0.080 0.066 0.029 0.720 <0.001 Age r P 0.207 <0.001 0.097 0.001 0.063 0.037 0.119 <0.001 0.120 <0.001 0.117 <0.001 0.165 <0.001 0.021 0.498 0.021 0.699 0.053 0.077 0.117 <0.001 0.127 <0.001 Income r P 0.155 <0.001 0.101 0.001 0.093 0.002 0.125 <0.001 0.137 <0.001 0.129 <0.001 0.116 <0.001 0.154 <0.001 0.150 <0.001 0.190 <0.001 0.044 0.150 0.007 0.819 0. 223 <0.001 Education r P 0.172 <0.001 0.138 <0.001 0.067 0.026 0.101 0.001 0.153 <0.001 0.154 <0.001 0.072 0.018 0.055 0.0069 0.063 0.038 0.096 0.001 0.029 0.331 0.015 0.623 0.138 <0.001 0.381 <0.001 a All variables checked for multicollinearity. As no values for r exceeded 0.741 (R2 = 0.549; number of cattle vs. total land), no R2 values were close to 1.0, indicating an absence of multicollinearity (Agresti and Finlay 1997). b All theory components significant with the dependent variable ( P (Ajzen and Fishbein 2005). The strongest potential relationship (attitude vs. past workshop participation, r = 0.380) was modeled as an interaction term and did not contribute to the model ( P = 0.212). Education was also checked as an interaction with attitude, as logically, those with higher education might have positive attitudes about attending technical workshops, but the interaction was also not significant ( P = 0.206).

PAGE 127

127 LIST OF REFERENCES Agresti A., and B. Finlay. 1997. Statistical methods for the social sciences. Third edition. Prentice Hall, Upper Saddle River, New Jersey, USA. Aiken, R. 2005. Private and public land use by hunters: Addendum to the 2001 national sur vey of fishing, hunting, and wildlifeassociated recreation. United States Fish and Wildlife Service Report 2001 8, Washington, D.C., USA. Ajzen, I. 1988. Attitudes, personality, and behavior. The Dorsey Press, Chicago, Illinois, USA. Ajzen, I. 2002. Perceived behavioral control, self efficacy, locus of control, and the theory of planned behavior. Journal of Applied Social Psychology 32:665683. Ajzen, I. 2006. Constructing a TpB questionnaire: conceptual and methodological considerations. Univ ersity of Massachusetts ( http://www.people.umass.edu/aizen/pdf/tpb.measurement.pdf ) Ajzen, I., and M. Fishbein. 1980. Understanding attitudes and predicting social behavior. Prenti ce Hall, Edgewood Cliffs, New Jersey, USA. Ajzen I., and M. Fishbein. 2005. The influence of attitudes on behavior. Pages 173221 in D. Albarracin, B. T. Johnson, and M. P. Zanna, editors. The handbook of attitudes. Lawrence Erlbaum Associates Inc., Mahwah, Jew Jersey, USA. Bearss, C. E. 1969. Redwood National Park History Basic Data. United States Department of the Interior. Washington, D.C. Benson D. 1998. Enfranchise landowners for land and wildlife stewardship: examples from the western United States. Human Dimensions of Wildlife 3:5968. Benson, D. E., R. Shelton, and D. W. Steinbach. 1999. Wildlife stewardship and recreation on private lands. 2008, Reprint. Texas A&M University Press. College Station, Texas, USA. Bernard, H. R. 2000. Social research methods: qualitative and quantitative approaches. Sage Publications Ltd. Thousand Oaks, California, USA. Brain, R. G. H. 2008. Predicting engagement in a conservation easement agreement. Dissertation, University of Florida, Gainesville, USA. Brook, A., M. Zint, and R. DeYoung. 2003. Landowners responses to an Endangered Species Act listing and implications for encouraging conservation. Conservation Biology 17: 16381649.

PAGE 128

128 Brown, T. L., and T. A. Messmer. 2009. Trends in acc ess and wildlife privatization. Pages 275288 in M. J. Manfredo, J. J. Vaske, P. J. Brown, D. J. Decker, and E. A. Duke, editors. Wildlife and society: the science of human dimension. Island Press, Washington, D.C., USA. Burke, W. W. 2008. Organizati on change theory and practice. Second Edition. Sage Publications Ltd. Thousand Oaks, California, USA. Cattaneo, A. 2003. The pursuit of efficiency and its unintended consequences: contract withdrawals in the Environmental Quality Incentives Program. Review of Agriucltural Economics 25, 449469. Claassen, R., A. Cattaneo, and R. Johansson. 2008. Cost effective design of agri envi ronmental payment programs: United States experience in theory and practice. Ecological Economics 65, 737752. Clouser R. L., R. Muraro, and L. Racevskis. 2007. 2006 Florida land value survey. Electronic Data Information Source, University of Florida FE687, Gainesville, Florida, USA. Commission for Environmental Cooperation. 1997. Ecological regions of North Americ a: t oward a common perspective. CEC Secretariat, Montreal, Canada. Conley, J. L, M. E. Fernandez Gimenez, G. B. Ruyle, and M. Brunson. 2007. Forest Service grazing permittee perceptions of the Endangered Species Act in southeastern Arizona. Rangeland E cology and Management 60:136145. Conover, M. R. 1994. Perceptions of grass roots leaders of the agricultural community about wildlife damage on their farms and ranches. Wildlife Society Bulletin 22:94100. Conover, M. R. 1998. Perceptions of Americ an agricultural producers about wildlife on their farms and ranches. Wildlife Society Bulletin 26:597604. Conover, W. J. 1980. Practical nonparametric statistics. John Wiley and Sons, Inc. New York, New York, USA. Davern, M., T. H. Rockwood, R. She rrod, and S. Campbell, S. 2003. Prepaid monetary incentives and data quality in faceto face interviews: Data from the 1996 survey of income and program participation incentive experiment. Public Opinion Quarterly 67 139147. Dillman, D. A., J. D. Smyth, and L. M. Christian 2009. Internet, mail and mixedmode surveys: the tailored design method. John Wiley and Sons Inc, New York, New York, USA.

PAGE 129

129 Enck, J. W., and D. J. Decker. 1997. Examining assumptions in wildlife manageme nt: a contribution of human dimensions inquiry. Human Dimensions of Wildlife 2:5672. Fernandez Gimenez, M. E., S. J. McClaran, and G. Ruyle. 2005. Arizona permittee land management agency employee attitudes toward rangeland monitoring by permittees. Rangeland Ecology and Management 58:344351. Fishbein, M., and I. Ajzen. 1975. Belief, attitude, intention, and behavior: and introduction to theory and research. AddisonWesly Publishing, Reading, Massachusetts, USA. Florida Fish and Wildlife Conserv ation Commission. 2006. 2006/2007 Programs of the Florida Fish and Wildlife Conservation Commission. http://myfwc.com/about/FWC2006 2007Programs.pdf (accessed 9/2007). Francis, J. J., M. P. Eccles, M. Johnston, A. Walker, J. Grimshaw, R. For, E. F. S. Kaner, L. Smith, and D. Bonetti. 2004. Constructing questionnaires based on the Theory of Planned Behavior: a manual for health services researchers. Center for Health Services Research, Univ ersity of Newcastle, Newcastle upon Tyne, United Kingdom. Groves, R. M., M. P. Couper, S. Presser, E. Singer, R. Tourangeau, G.P. Acosta, and L. Nelson. 2006. Experiments in producing nonresponse bias. Public Opinion Quarterly 70:720736. Haufler, J. B., and A. C. Ganguli. 2007. Benefits of Farm Bill grassland conservation practices to wildlife. Pages 5769 in J. B. Haufler, editor. Fish and wildlife response to Farm Bill conservation practices. The Wildlife Society, Bethesda, Maryland, USA. Hrub es, D., I. Ajzen, and J. Daigle. 2001. Predicting hunting intentions and behavior: An application of the theory of planned behavior. Leisure Sciences, 23, 165 178. Irby, L. R., J. Saltiel, W. E. Zidack, and J. B. Johnson. 1997. Wild ungulate damage: p erceptions of farmers and ranchers in Montana. Wildlife Society Bulletin 25:320329. Jacobson, S. K., M. D. McDuff, and M. C. Monroe. 2006. Conservation education and outreach techniques. Oxford University Press, Oxford, United Kingdom. Jobber, D., J Saunders, and V. Mitchell. 2004. Prepaid monetary incentive effects on mail survey response. Journal of Business Research 57:2125. Kautz, R. S. and J. A. Cox 2001. Strategic habitats for biodiversity conservation in Florida. Conservation Biology 15:5577.

PAGE 130

130 Kellert, S.R. 1979. Public attitudes toward critical wildlife and natural habitat issues. United States Government Printing Service, Washington, D.C., USA Kellert, S. R. 1980. Activities of the American public relating to animals. United States Government Printing Service, Washington, D.C., USA Lambert, D., 1979. Shenandoah National Park Administrative History. http://www.nps.gov/history/history/online_books/s hen/admin.pdf (accessed 9/2007). Lambert, D., P. Sullivan, R. Claassen, and L. Foreman. 2006. Conservation compatible practices and programs: who participates? USDA Economic Research Report 14, Washington, D.C., USA. Leopold, A. 1959. A sand coun ty almanac: with essays on conversation from Round River. Ballantine Books, New York, New York, USA. Lesser, V.M., D. A. Dillman, J. Carlson, F. Lorenz, R. Mason, and F. Willits. 2001. Quantifying the influences of incentives on mail survey response rates and their effect on nonresponse error. Proceedings of the Annual Meeting of the American Statistical Association August 59, 2001. Likert, R. 1967. The human organization. McGraw Hill, New York, New York, USA. Livengood, K. R. 1983. Value of big game for hunting leases: the hedonic approach. Land Economics 59:287 291. Lubowski, R.N., M. Vesterby, S. Bucholtz, A. Baez, and M. J. Roberts 2002. Major Uses of Land in the United States, 2002. United States Department of Agriculture, Washington, D.C., USA. Lubowski, R. N., M. Vesterby, S. Bucholtz, A. Baez, and M. J. Roberts. 2006. Major Uses of Land in the United States, 2002. USDA, Washington, D.C., USA. Madden, T. J., P. S. Ellen, and I. Ajzen. 1992. A comparison of the theory of plann ed behavior and the theory of reasoned action. Personality and Social Psychology Bulletin 18:39. Manfredo, M.J., J.J. Vaske, and D.J. Decker. 1995. Human dimensions of wildlife management: basic concepts. Pages 275288 in R. L. Knight and K. J. Gutzw iller, editors. Wildlife and recreationalists. Island Press, Covelo, California, USA. Martin, S. R., and K. McCurdy. 2009. Wilderness food storage in Yosemite: using the theory of planned behavior to understand backpacker canister use. Human Dimensio ns of Wildlife 14:206218.

PAGE 131

131 Mashour, T. M. 2004. Assessing landowner perceptions and prices of conservation easements in Florida. Thesis, University of Florida, Gainesville, USA. McKenzieMohr, D., and W. Smith. 1999. Fostering sustainable behavior: an introduction to community based social marketing. New Society Publishers, Gabriola Island, British Columbia, Canada. Messer, D. L. 2009. Improving Survey Response in Mail and Internet General Public Surveys Using Address Based Sampling and Mail Cont act Procedures Thes is, Washington State University, Pullman, USA. Messer, D. L, and D. A. Dillman. 2009. Improving the effectiveness of mail contact procedures to obtain survey response over the Internet for general public household surveys. Paper pres ented at the annual meeting of the American Association for Public Opinion Research, Hollywood, CA, May 15, 2009. Miller, C. A., L. K. Campbell, J. A. Yeagle, and R. J. Williams. 2002. Results of studies of Illinois hunters, landowners, and participants in Access Illinois. Human Dimensions Program Report SR 02 01. Illinois Natural History Survey, Champaign, Illinois, USA. Morrison, J.L., and S. R. Humphrey. 2001. Conservation value of private lands for crested caracaras in Florida. Conservation Biol ogy 15:675684. National Parks Service. 2006. History of the National Parks Service. http://www.nps.gov/history/history/hisnps/NPSHistory/timeline.htm (accessed 9/2007). N ickerson, C. J., and L. Lynch. 2001. The effect of farmland preservation programs on farmland prices. American Journal of Agricultural Economics 83:341 351. Ouellette, J. A., and W. Wood. 1998. Habit and intention in everyday life: the multiple processes by which past behavior predicts future behavior. Psychological Bulletin 124: 5474. Payne, N. F., and F. C. Bryant. 1994. Techniques for wildlife habitat management of uplands. McGraw Hill Inc., New York, New York, USA. Rice, M. B, W. B. Ballard E. B. Fish, D. B. Wester, and D. Holderman. 2007. Predicting private landowner support toward recolonizing black bears in the Trans Pecos region of Texas. Human Dimensions of Wildlife 12:405415. Riley, S. J., and D. J. Decker. 2000. Wildlife stake holder acceptance capacity for cougars in Montana. Wildlife Society Bulletin 28:931939.

PAGE 132

132 Rogers, E. M. 2003. Diffusion of innovations. Fifth edition. The Free Press, New York, New York, USA. Saltiel, J., and L. R. Irby. 1998. Perceptions of game dam age in Montana by resource agency personnel and agricultural producers. Wildlife Society Bulletin 26:8491. Rossi, A. N., and J. B. Armstrong. 1999. Theory of reasoned action vs. theory of planned behavior: testing the suitability and sufficiency of a popular behavioral model using hunting intentions. Human Dimensions of Wildlife 4:4056. Saltiel, J., and L. R. Irby. 1998. Perceptions of game damage in Montana by resource agency personnel and agricultural producers. Wildlife Society Bulletin 26: 84 91. Schafer, J. L., and J. W. Graham. 2002. Missing data: our view of the state of the art. Psychological Methods 7: 147177. Scott, J. M., F. W. Davis, R. G. McGhie, R. G. Wright, C. Groves, and J. Estes. 2001. Nature reserves: do they capture the full range of Americas biological diversity? Ecological Applications 11:9991007. Seimer, W. F., and T. L. Brown. 1993. Public access to private land for hunting in New York: A study of landowners. Human Dimensions Research Unit Report 934. Depart ment of Natural Resources, Cornell University, Ithaca, New York, USA. Shands, W. E., and R. G. Healy 1977. The lands nobody wanted. The Conservation Foundation, Washington, D.C., USA. Shettle, C., and G. Mooney. 19 99. Monetary incentives in United States Government surveys. Journal of Official Statistics 15:231250. Smyth, J.D., D. A. Dillman, L. M. Christian, and A. C. ONeil ( in press). Using the Internet to survey small towns and communities: Limitations and possibilities in the early 21st cent ury. American Behavioral Scientist. USDA. 2004. 2002 Census of Agriculture. Washington, D.C., USA. USDA. 2009. 2007 Census of Agriculture: Washington, D.C., USA. USDI 2007. National Atlas Map Maker. < http://www.nationalatlas.gov/ >. A ccessed Sep tember 2007. Vance, D.R. 1976. Changes in land use and wildlife populations in southeastern Illinois. Wildlife Society Bulletin 4:11 15. West, N. E. 1993. Biodiversity in rangelands. Journal of Range Management 46:213.

PAGE 133

133 Willcox, A. S., W. M. Giuliano, and G. D. Israel. 2010a Effects of token financial incentives on response rates and item nonresponse for mail surveys. Human Dimensions of Wildlife 15:in press. Willcox, A. S., W. M. Giuliano, C. Wynn, and J. S. Sanders, 2010b Wildlife management on private lands in Florida. Proc eedings of the Annu al Conference of Southeast ern Association of Fish and Wildlife Agencies 63:in press. Willcox, E. V. 2010. Wildlife and habitat responses to prescribed burning, roller chopping, and grazing of Flori da rangelands and pasture. Dissertation, University of Florida, Gainesville, USA. Williams, C. K., G. Ericsson, and T. Heberlein. 2002. A quantitative summary of attitudes towards wolves and their reintroduction (19722000). Wildlife Society Bulletin 30:575584.

PAGE 134

134 BIOGRAPHICAL SKETCH Adam Sage Willcox was raised in Scottsville, V irginia. He graduated with a Bachelor of Arts from the Univ ersity of Virginia majoring in environmental sciences and minoring in English language and l iterature. He then serv ed as an Agroforestry E xtension Agent with the United States Peace Corps in Southwest Cameroon. Adam remained in Cameroon conducting participatory rural appraisals and hunting research in communities around the Banyang Mbo Wildlife Sanctuary for the Wildl ife Conservation Society. Adam received his Master of Science with distinction in Conservation Biology from the Durrell Institute of Conservation and Ecology at the University of Kent, U.K. He then worked as the Community Conservation Advisor for the Fra nkfurt Zoological Society in Tanzania during the initial two years of the European Commissionfunded Mahale Ecosystem Management Project. Adam is an applied social scientist with keen interests in integrating wildlife management into rural private land us es, public lands outreach, and wildlife extension.