<%BANNER%>

Local Communities and Protected Areas

Permanent Link: http://ufdc.ufl.edu/UFE0041986/00001

Material Information

Title: Local Communities and Protected Areas Social Dimensions of Pro-Environmental Engagement in Retezat National Park, Romania
Physical Description: 1 online resource (207 p.)
Language: english
Creator: Buta, Natalia
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: civic, community, environmental, protected, rural
Tourism, Recreation, and Sport Management -- Dissertations, Academic -- UF
Genre: Health and Human Performance thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Local communities and protected areas: Social dimensions of pro-environmental engagement in Retezat National Park, Romania A lack of community interest and participation in biodiversity conservation has been discussed as a major constraint for natural resources management in Romania and other areas of the world. Given weak understandings of Romanian rural residents relationships with neighboring environments, this study examined various social dimensions of community and their potential to facilitate or hinder local pro-environmental behaviors. Three research questions were addressed in this study. First, how levels and types of community attachments, conservation attitudes, connections to nature and perceived collective environmental responsibility facilitate or hinder attitudes towards pro-environmental civic engagement and pro-environmental civic behavioral intentions. Secondly, whether different facets of community attachment are distinctively predicted by length of residence, social interaction, and socio-demographic characteristics. Lastly, differences in affective and attitudinal environmental responses were assessed among young adults, middle age adults, and older adults residing in rural Romania. Quantitative and qualitative data were collected in nine rural communities in Romania, communities neighboring Retezat National Park (RNP). Cross-sectional data was collected from 260 residents during June October 2009 using face to face interviews and mail surveys. Also, in depth-interviews were conducted with 24 community members representing the nine communities selected for study. Structural Equation Model (SEM) was used to assess the hypothesized relationships between constructs. Multivariate analysis of variance (MANOVA) was employed to assess differences in affective and attitudinal environmental responses among age groups. The qualitative data was analyzed using a grounded theory approach to inquiry focusing on an in depth theoretical conceptualization of the community attachment construct. The results revealed the hierarchical structure of the relationships between social and park attachments, connections to nature and conservation attitudes and local environmental identity accounting for the relationships between these constructs. Local environmental identity and perceived collective environmental responsibility were found to have a significant direct impact on attitudes towards pro-environmental civic engagement, which ultimately were found to influence pro-environmental civic behavioral intentions. Furthermore, social interaction was found to have a stronger effect on park attachment, while length of residence had a stronger association with attachment to the social environment. Four distinct dimensions of attachment emerged from the textual analysis: attachments to the natural, social, institutional and cultural environments. In addition, younger residents ( < 31) were found to have weaker affective bonds with nature and the national park neighboring their communities. This research extends current theoretical understandings of social predictors and inter-relationships between attachments, connections to nature, conservation attitudes, and perceived environmental responsibilities in areas rich in natural resources and their attitudinal and behavioral implications. Furthermore, this study complements contemporary understandings of the community attachment construct and affective and attitudinal environmental responses in rural contexts. Practical implications are discussed.
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 Natalia Buta.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Holland, Stephen.
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: UFE0041986:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041986/00001

Material Information

Title: Local Communities and Protected Areas Social Dimensions of Pro-Environmental Engagement in Retezat National Park, Romania
Physical Description: 1 online resource (207 p.)
Language: english
Creator: Buta, Natalia
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: civic, community, environmental, protected, rural
Tourism, Recreation, and Sport Management -- Dissertations, Academic -- UF
Genre: Health and Human Performance thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Local communities and protected areas: Social dimensions of pro-environmental engagement in Retezat National Park, Romania A lack of community interest and participation in biodiversity conservation has been discussed as a major constraint for natural resources management in Romania and other areas of the world. Given weak understandings of Romanian rural residents relationships with neighboring environments, this study examined various social dimensions of community and their potential to facilitate or hinder local pro-environmental behaviors. Three research questions were addressed in this study. First, how levels and types of community attachments, conservation attitudes, connections to nature and perceived collective environmental responsibility facilitate or hinder attitudes towards pro-environmental civic engagement and pro-environmental civic behavioral intentions. Secondly, whether different facets of community attachment are distinctively predicted by length of residence, social interaction, and socio-demographic characteristics. Lastly, differences in affective and attitudinal environmental responses were assessed among young adults, middle age adults, and older adults residing in rural Romania. Quantitative and qualitative data were collected in nine rural communities in Romania, communities neighboring Retezat National Park (RNP). Cross-sectional data was collected from 260 residents during June October 2009 using face to face interviews and mail surveys. Also, in depth-interviews were conducted with 24 community members representing the nine communities selected for study. Structural Equation Model (SEM) was used to assess the hypothesized relationships between constructs. Multivariate analysis of variance (MANOVA) was employed to assess differences in affective and attitudinal environmental responses among age groups. The qualitative data was analyzed using a grounded theory approach to inquiry focusing on an in depth theoretical conceptualization of the community attachment construct. The results revealed the hierarchical structure of the relationships between social and park attachments, connections to nature and conservation attitudes and local environmental identity accounting for the relationships between these constructs. Local environmental identity and perceived collective environmental responsibility were found to have a significant direct impact on attitudes towards pro-environmental civic engagement, which ultimately were found to influence pro-environmental civic behavioral intentions. Furthermore, social interaction was found to have a stronger effect on park attachment, while length of residence had a stronger association with attachment to the social environment. Four distinct dimensions of attachment emerged from the textual analysis: attachments to the natural, social, institutional and cultural environments. In addition, younger residents ( < 31) were found to have weaker affective bonds with nature and the national park neighboring their communities. This research extends current theoretical understandings of social predictors and inter-relationships between attachments, connections to nature, conservation attitudes, and perceived environmental responsibilities in areas rich in natural resources and their attitudinal and behavioral implications. Furthermore, this study complements contemporary understandings of the community attachment construct and affective and attitudinal environmental responses in rural contexts. Practical implications are discussed.
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 Natalia Buta.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Holland, Stephen.
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: UFE0041986:00001


This item has the following downloads:


Full Text

PAGE 1

1 LOCAL COMMUNITIES AND PROTECT ED AREAS: SOCIAL DIMENSIONS OF PRO ENVIRONMENTAL ENGAGEMENT IN RETEZAT NATIONAL PARK, ROMANIA By NATALIA BUTA A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN P ARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 Natalia Buta

PAGE 3

3 To my parents

PAGE 4

4 ACKNOW LED GMENTS This dissertation was made possible with the support and encouragement of many ind i viduals I will always cherish their support. First and foremost, I am truly grateful for the unfailing trust and guidance given by my advisor, Dr. Stephen Holland. I am extremely indebted for his tremendous help. I would a lso like to acknowledge my disse rtation committee members, Dr. Kyriaki Kaplanidou, Dr. Mark A. Brennan, and Dr. Alan D. Cook, for their advic e, encouragement, and continuous support Furthermore, m y special thanks go to Dr. Algina who helped me very much with the stru ctural equation mode l analysis. I am also very grateful to the people in the communities where I conducted my study. I truly enjoyed every minute I spent in their communities. Their kindness, openness and willingness to participate in the study have been invaluable and truly inspiring. I am also very thankful for the support received from the Center for European Studies at the University of Florida who awarded me a travel grant to conduct my dissertation work. I am truly grat eful for their financial support It is with great a ppreciation that I acknowledge the faculty of the RPL Department at Central Michigan University. They introduced me to t his profession and continuously encouraged me through my doctoral studies. I would like to give special thanks to Dr. Roger L. Coles who has been very caring, supportive and helpful in numerous ways. Finally, I would like to a cknowledge my family. I thank my parents for their constant care and encouragement I thank my brothers and sisters who were always there whenever I needed advice an d always ins pired me through their ambition s I feel truly fortunate to have such great individuals around me.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 8 LIST OF FIGURES ................................ ................................ ................................ ....................... 10 ABSTRACT ................................ ................................ ................................ ................................ ... 11 CHAPTER 1 INTRODUCTI ON ................................ ................................ ................................ .................. 13 Background ................................ ................................ ................................ ............................. 13 Problem Statement ................................ ................................ ................................ .................. 17 Purpose of the Study ................................ ................................ ................................ ............... 20 Dissertation Format ................................ ................................ ................................ ................ 22 2 METHODS ................................ ................................ ................................ ............................. 25 Research Design ................................ ................................ ................................ ..................... 25 Data Collection Procedures ................................ ................................ ................................ .... 25 Quantitative Data Collection ................................ ................................ ........................... 25 Instrument ation ................................ ................................ ................................ ......... 28 Reliability and validity ................................ ................................ ............................. 29 Qualitative Data Collection ................................ ................................ ............................. 29 Data Analysis ................................ ................................ ................................ .......................... 31 Quantitative Data Analysis ................................ ................................ .............................. 31 Descriptive statistics ................................ ................................ ................................ 34 Confirmatory factor analysis ................................ ................................ .................... 39 Qualitative Data Analysis ................................ ................................ ................................ 47 3 PRO ENVIRONMENTAL CIVIC ENGAGEMENT IN RETEZAT NATION AL PARK, ROMANIA ................................ ................................ ................................ ................. 65 Introduction ................................ ................................ ................................ ............................. 65 Conceptual Framework ................................ ................................ ................................ ........... 67 Meth odology ................................ ................................ ................................ ........................... 70 Description of the Study Area ................................ ................................ ......................... 70 Measures of Model Components ................................ ................................ ..................... 72 Pro environmental civic behavioral intentions ................................ ........................ 72 Community attachment ................................ ................................ ............................ 74 Connection to nature ................................ ................................ ................................ 77 Conservation attitudes ................................ ................................ .............................. 78 Perceived collective environmental responsibility ................................ ................... 80

PAGE 6

6 Data Collection ................................ ................................ ................................ ................ 81 Data Analysis ................................ ................................ ................................ ................... 83 Results ................................ ................................ ................................ ................................ ..... 84 Discussion ................................ ................................ ................................ ............................... 89 Conclusions ................................ ................................ ................................ ............................. 94 4 A MIXED METHOD INVESTIGATION OF COMMUNITY ATTACHMENT ............... 102 Introduction ................................ ................................ ................................ ........................... 102 Conceptual Framework ................................ ................................ ................................ ......... 105 Methodology ................................ ................................ ................................ ......................... 111 Stu dy Area ................................ ................................ ................................ ..................... 111 Survey Data Collection and Analysis ................................ ................................ ............ 112 Qualitative Data Collection and Analysis ................................ ................................ ..... 115 Survey Data Results ................................ ................................ ................................ .............. 116 Qualitative Data Results ................................ ................................ ................................ ....... 119 Discussion and Conclusions ................................ ................................ ................................ 127 5 THE IMPLICATIONS OF AGE ON AFFECTIVE AND ATTITUDINAL ENVIRONMENTAL RESPONSE ................................ ................................ ....................... 136 Introduction ................................ ................................ ................................ ........................... 136 Literature Review ................................ ................................ ................................ ................. 138 Connection to Nature ................................ ................................ ................................ ..... 139 Attachment to the Park ................................ ................................ ................................ .. 140 Conservation Attitudes ................................ ................................ ................................ .. 142 Attitudes towards Pro environmental Civic Engagement ................................ ............. 143 Perceived Environmental Responsibility ................................ ................................ ...... 144 Methods ................................ ................................ ................................ ................................ 145 Study Area ................................ ................................ ................................ ..................... 145 Meas urements ................................ ................................ ................................ ................ 146 Data Collection ................................ ................................ ................................ .............. 147 Data Analysis ................................ ................................ ................................ ................. 1 48 Results ................................ ................................ ................................ ................................ ... 150 Discussion ................................ ................................ ................................ ............................. 154 Conclusions ................................ ................................ ................................ ........................... 160 6 CONCLUSIONS ................................ ................................ ................................ .................. 166 Findings ................................ ................................ ................................ ................................ 166 Implications and Future Research ................................ ................................ ........................ 169 Limitations ................................ ................................ ................................ ............................ 173 Conclusions ................................ ................................ ................................ ........................... 174 APPENDIX A STUDY QUESTIONNAIRE ................................ ................................ ................................ 176

PAGE 7

7 B STUDY QUESTIONNAIRE (IN ROMANIAN) ................................ ................................ 184 C SEMI STRUCTURED INTERVIEW GUIDE ................................ ................................ ..... 193 D SEMI STRUCTURED INTERVIEW GUIDE (IN ROMANIAN) ................................ ...... 194 LIST OF REFERENCES ................................ ................................ ................................ ............. 195 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ....... 207

PAGE 8

8 LIST OF TABLES Table page 1 1 Communes and villages adjacent to Retezat National Park ................................ ............... 24 2 1 Sampling frame for local community residents based on households ............................... 48 2 2 Demographic characteristics of the respondents ................................ ................................ 49 2 3 Descriptive statistics social attachment ................................ ................................ ............. 52 2 4 Descriptive statistics social interaction ................................ ................................ .............. 53 2 5 Descriptive statistics connection to nature ................................ ................................ ......... 54 2 6 Major purpose for visiting the park ................................ ................................ ................... 55 2 7 Descriptive statistics park attachment ................................ ................................ ................ 55 2 8 Descriptive statistics conservation attitudes ................................ ................................ ...... 56 2 9 Descriptive statistics attitudes towards pro environmental civic engagement .................. 58 2 10 Descriptive statistics pro environmental civ ic behavioral intentions ................................ 59 2 11 Descriptive statistics perceived environmental responsibility ................................ ........... 59 2 12 Goodness of fit indices f or each construct ................................ ................................ ......... 60 2 13 Reliability and validity of the community attachment CFA model ................................ ... 61 2 14 Reliability and validity of the social interaction CFA model ................................ ............ 61 2 15 Reliability and validity of the connection to nature CFA model ................................ ....... 62 2 16 Reliability and val idity of the conservation attitudes CFA model ................................ ..... 63 2 17 Reliability and validity of the attitudes towards pro environmental civic engagement CFA model ................................ ................................ ................................ ......................... 64 2 18 Reliability and validity of the pro environmental civic behavioral intentions CFA model ................................ ................................ ................................ ................................ .. 64 3 1 Fit indices for measurement model and SEM model ................................ ......................... 97 3 2 Summary results for measurement model ................................ ................................ .......... 97 3 3 Correlations among factors (based on the measurement model) ................................ ..... 100

PAGE 9

9 4 1 Fit indices for measurement model and SEM model ................................ ....................... 133 4 2 Summary results for measurement model ................................ ................................ ........ 133 4 3 Correlations among factors (based on the measurement model) ................................ ..... 134 5 1 Goodness of fit indices for each construct ................................ ................................ ....... 162 5 2 Reliability and validity of the measurements (CFA models) ................................ ........... 163 5 3 MANOVA of age ................................ ................................ ................................ ............. 165

PAGE 10

10 LIST OF FIGURES Figure page 1 1 Map of Retezat National Park, Romania ................................ ................................ ........... 23 3 1 Proposed research model ................................ ................................ ................................ ... 96 3 2 A structural equati on model test ................................ ................................ ...................... 101 4 1 A structural equation model test ................................ ................................ ...................... 135

PAGE 11

11 Abstract of Dissertation Presented to the Graduate School of the University of in P artial Fulfillment of the Requirements for the Degree of Doctor of Philosophy LOCAL COMMUNITIES AND PROTECT ED AREAS: SOCIAL DIMENSIONS OF PRO ENVIRONMENTAL ENGAGEMENT IN RETEZAT NATIONAL PARK, ROMANIA By Natalia Buta August 2010 Chair: Stephen Holland M ajor: Health and Human Performance A lack of community interest and participation in biodiversity conservation has been discussed as a major constraint for natural resources management in Romania and other areas of the world Given weak understandings of neighboring environments t his study examined various social dimensions of community and their potential to facilitate or hinder local pro environmental behaviors. Three research questions were addressed in thi s study. First, how levels and types of community attachment s conservation attitudes, connections to nature and perceived collective environmental responsibility facilitate or hinder attitudes towards pr o environmental civic engagement and pro environment al civic behavioral intentions. Secondly, whether different facets of community attachment are distinctively predicted by length of residence, social interaction, and socio demographic characteristics. Lastly, differences in affective and attitudinal envir onmental responses were assessed among young adults middle age adults, and older adult s residing in rural Romania Quantitative and qualitative data were collected in nine rural communities in Romania, communities neighboring Retezat National Park (RNP) Cross sectional data was collected from 260 residents during June October 2009 using face to face interviews an d mail survey s Also,

PAGE 12

12 i n depth interviews were conducted with 24 community members representing the nine communities selected for study. S truc tural Equation M odel (SEM) was use d to assess the hypothesized relationships between constructs Multivariate analysis of variance (MANOVA) was employed to assess differences in affective and attitudinal environmental responses among age groups. The qualit ative data was analyzed using a grounded theory approach to inquiry focusing on an in depth theoretical conceptualization of the community attachment construct. The results revealed the hierarchical structure of the relationship s between social and park at tachments, connection s to nature and conservation attitudes and l o cal environmental identity account ing for the relationship s between these constructs. Local environmental identity and perceived collective environmental responsibility were found to have a significan t dir ect impact on attitudes towards pro environmental civic engagement which ultimately were found to influence pro environmental civic behavioral intentions Furthermore, s ocial interaction was found to have a stronger effect on park attachmen t, while length of residence had a stronger association with attachment to the social environment. Four distinct dimensions of attachment emerged from the textual analysis : attachment s to the natural, social, institutional and cultural environment s In add ition, y ounger residents (<31) were found to have wea ker affective bonds with nature and the national park neighboring their communities. T his research extend s current theoretical understandings of social predictors and inter relationships between attachme nts, connection s to nature, conservation attitudes and perceived environmental responsibilities in areas rich in natural resources and their attitudinal and behavioral implications Furthermore, this study comp lements contemporary understandings of the co mmunity attachment construct and affective and attitudinal environmental response s in rural contexts Practical implications are discussed.

PAGE 13

13 CHAPTER 1 INTRODUCTION Background Romania is situated in the southeastern part of Central Europe, sharing borders with the Republic of Moldova, Ukraine, Hungary, Serbia, and Bulgaria. The population of Romania is approximately 21.5 million people (2006 estimate), of which approximately 44.8% live in rural areas (Romanian National Institute of Statistics, 2007). In siz e, Romania is slightly larger than the US state of Oregon. In terms of physical features, with 91,700 square miles (237,500 sq km), the Romanian landscape includes a variety of geographical areas. Approximately 31% of Romanian territory is covered by mount ains (the Carpathians), 36% by hills and orchards, and 33% by plains and meadows (Romanian National Institute of Statistics, 2007). The three main tourist attractions in Romania are the Danube River, the Black Sea, and the Carpathian Mountains. Romania, du e to legislative measures during the communist era (when the majority of land was owned by the biodiversity and natural ecosystems (Ioras, 2003). The conservation focus in terms of natural resources management is indicated by the presence of a large variety of forest fauna species, including 60% of all European brown bears and 40% of the European wolves. After 1989, when major changes in land ownership started to b e implemented, the threats to biodiversity sharply approximately 6.6% of the country's land mass had protected status, including three biosphere reserves, thirteen nat ional parks and seven natural parks (Kuijs & Bergh, 2006). The first national park in Romania was established in 1935 in the Retezat Mountains. In 1979, Retezat National Park (RNP) was declared an International Biosphere Reserve under the

PAGE 14

14 UNESCO Man and Bi osphere program (Ioras, 2003). The Romanian Government has established protected areas over the years, but they persistently lacked management guidelines and enforcement for resource protection (Cutumisu, 2003). In 1999, the first national project for esta blishing a national system of protected areas was initiated, with a focus on incorporating biodiversity concerns into the planning and management of Romanian protected areas. Retezat National Park was the first park in Romania with a management system in p lace (van Hal, 2006). Retezat National Park received its PAN (Protected Area Network) Certification in 2004. The PAN Parks mission is to establish a network of effectively managed and independently verified protected areas in Europe, that successfully inte grate nature conservation and tourism objectives (PAN Parks, 2009). Retezat National Park is located in the southwestern Carpathians, Hunedoara County, and the total surface of the park is 38,138 ha (RNP Management Plan, 2008). Within the park, t here are m ore than twenty mountain peaks of 2,000 meters or higher (the highest p eak is Peleaga at 2,509 meters, ~8,232 ft.) in addition to eighty lakes of glacial origin. There are more than 1,100 species of plants, over 50 species of mammals including roe deer, c hamois, lynx, bear, and otter and 168 recorded bird species including the golden eagle. Due to changes in elevation, there a variety of forest types, but the predominant ones are the beech forest located between 800 and 1200 m, mixed forest between 1200 an d 1400 m, and the spruce fir forest between 1400 and 1800 m (RNP 2008). Two major hydrological basins bisect the park's land scape, Mures and Jiul de Vest. In terms of management, Retezat National Park encompasses three different areas with different poli cies and regulations assigned (RNP Management Plan, 2008). The Gemenele Scientific Reserve (1,600 ha; 4%) represents the central zone or the scientific zone, an area

PAGE 15

15 managed by the Romanian Academy where only scientific research activities are allowed. The Integral Protection Zone (18,000 ha; 47%) is the second largest area in the park and the major activities allowed in this area are passive recreation related, as well as some other activities such as traditional grazing (occurring under the establishing r egulations of the park, as stated in the park management strategy). The remaining park area (49%) is represented by a buffer zone where sustainable forest harvesting and tourism development are permitted. Operationally, the park administration is responsib le for overseeing a series of hiking trails, designated camping sites, and two visitor information centers. A total area of 6.5 ha (<1%) of the park has tourist infrastructure. The park area is primarily wilderness, visitors driving to the fringe of the pa rk and hiking to the major attractions in the park. There are no paved roads within the park. Figure 1 1 depicts a map of the park and its current touristic infrastructure. The park co management framework includes, in addition to the park full time mana gers a Scientific Cou ncil and a Consultative Council The Scientific Council consists of scientists that represent the Romanian Academy, while the Consultative Council is represented by key stakeholders of the Retezat National Park area (local communities tourism operators, local businesses, etc.) The Consultative Council provides recommendations regarding park management activities but does not have decision making authority (van Hal, 2006). The management framework initiated by Retezat National Park is perceived as being a model for other protected areas in Romania. A large portion of the park area (17,564 ha, 46%) is owned by the state and is administered by the National Forestry Administration, through its local ROMSILVA agency (Hunedoara Forest Direct orate). Local associations own the r eminder of the land (20,574 ha). R ural communities adjacent to Retezat National Park belong to five communes encompassing 43

PAGE 16

16 villages with a total population estimated at 14,009 adult residents. A commune is an administr ative division in Romania encompassing one or more villages that share similar economic, socio cultural, geographic and demographic conditions. Table 1 1 depicts the villages adjacent to RNP and the communes they belong to Twenty six villages have grazing rights to alpine meadows, and their rights are administered through the local councils of the five communes to which the villages belong or by local associations (Kuijs & Bergh, 2006). Of these, three communes are primarily important from a management per spective, due to their close proximity to the park and their land ownership and use of resources in the park (use of buffer zone resources): Rau de Mori, Salasu de Sus, and Campu lui Neag (van Hal, 2006). Similarly, the influence of the park on these villa ges is not negligible; the park policies and regulations have a major impact on the livelihoods of the people living in these communities. The daily livelihoods within these communities are sustained through agricultural practices, with only a quarter of t he population being involved in industrial fields (Cutumisu, 2003). Communities rely on park resources primarily for grazing and the use of other natural resources such as wood, forest fruits, mushrooms, and medicinal plants. The major management concerns as it relates to conservation, are related to overgrazing of the pasture areas and i llegal wood harvesting (RNP Management Plan, 2008). Tourism impacts are also viewed as a threat to the conservation goals of the park, as over the last ten years, an incr ease in the number of ported (van Hal, 2006). Previous research conducted in the area discussed a variety of challenges for park management and surrounding communities. van Hal (2006) emphasized the growing concerns of local people due to the increased restrictions imposed on grazing by the park administration for conservation purposes. These restrictions were strongly viewed as having great impact on the

PAGE 17

17 landowners control over their private lands. van Hal ( 2006), based on an investigation conducted to better understand co management in the area, observed a lack of a common interest in conservation in the area. The author described this situation as being directly linked to a lack of effective cooperation and management. Similarly, Kuijs & Bergh (2006) emphasized the lack of a conservation attitude in the communities surrounding Retezat National Part and a lack of care for the environment. These two studies used a qualitative approach in understanding park man agement efforts, depicting primarily the voices of the park administrative staff and park stakeholders (e.g. lodge owners, travel operators, mayors of local communities) but not directly the views of local residents. The Retezat National Park employed a co management framework for the management and planning of the park's natural resources. The effectiveness of the management practice is still questioned, as the local communities' interest and participation in conservation is perceive d as being limited. Con sequently it is essential to better understand the major factors that shape community residents' conservation attitudes and behaviors, and what community residents' see as their role in protecting the natural resources of the area. Problem Statement The l iterature on parks and protected areas management underscores that successful management endeavors and sustainability depend on the cooperation and support of local communities (Augustyn, 1998; Brandon & Wells, 1992; Cottrell & Cutumisu, 2006 ; de Beer & Ma rais, 2005; Hall, 2004; Ioras, 2003). Historically, parks and protected areas have been viewed as islands of biodiversity conservation with little or no networks or connections to the human dimensions (Cutumisu, 2003). Currently, local communities adjacent to national parks and protected areas are perceived as having a major role in achieving conservation and sustainability ideals, due to their strong connections to the surroundi ng environments and resources. Manfredo

PAGE 18

18 et al. (2004) emphasized that people li ving in such communities make use of, develop meanings and attachments to, and are affected by the conditions of neighboring natural environments. Furthermore, Agrawal (2001) underscored that g iving consideration to the social and economic environment in the management process is essential to assuring equity as well as effe ctive management Actively involving local communities in the management of protected areas has been associated with an increased awareness in terms of the benefits of biodiversity cons ervation, a more responsible use of resources, and ultimately enhanced livelihoods and welfare of local people (Ioras, 200 3; Pagdee et al 2006). Heltberg (2001) argued that formal and informal community based resource management institutions play a posit ive role in redu cing resource use and dependence However, efforts to involve local communities are often challenged either by the urgency of implementing biodiversity conservation projects or the lack of community motivation to be involved or even the lac k of knowledge on how and to what e xtent to involve local people. Often in the literature, local communities adjacent to parks and protected areas are perceived as being passive, lacking initiative and care for the environment, and having expectations for established authorities to take the active role in terms of resource management (d e Beer & Marais, 2005). This assertion has been constantly made as it regards local communities adjacent to national parks and protected area in Romania, but little is empir ically known about the Romanian context, and especially about the institutional and social factors that shape the local communities attitudes and behaviors towards protected areas. The institutional capacity for conservation and protected area management in Romania is still at a nascent stage, although efforts are being directed towards defining the country's nature

PAGE 19

19 conservation agenda. In 1999, the first models for protected area management were established in Romania, models primarily framed using a part representation in the management process. Stanciu (2002) underscored that the legal framework for protected area management in Romania encourages the development of co management systems that sustain conservation objectives and bring benefits to the park stakeholders. Co management and participatory decision making initiatives have been encouraged over the years but modest success has been achieved in terms of management effectiveness. Excessive exploitation of nat ural resources is still a real threat to biodiversity conservation in Romania. The lack of community interest and participation in biodiversity conservation has been discussed as being a major constraint for natural resources management in Romania. This si tuation has been primarily attributed to a low sense of community and collective responsibility that characteri zes Romanian rural communities (PJB Associates, 2006 ) and also to the legacy of the communist system, primarily based on a centralized political structure, which la sted until 1989 (Oostenbrink & Kosterink, 2005). However, there is scarce empirical evidence that supports these assertions. Based on a study conducted in the Retezat National Park (the first park in Romania with a co management system in place) to better understand the co management framework employed by the park, van Hal (2006) emphasized a lack of common interest in conservation in the area. The author described this situation as being directly linked to a lack of effective cooperatio n and management. Similarly, Kuijs & Bergh (2006) emphasized a lack of conservation attitude in communities surrounding Retezat National Part as well as a lack of care for the environment. These two studies used a qualitative approach in understanding the park administrative efforts, depicting primarily the voices of those directly involved in management and to a lesser extent the views of citizens from local communities.

PAGE 20

20 Taking into consideration that Romania is still at an incipient stage in the process of shaping its nature conservation approach, there is an emerging need to better understand the major factors that shape community conservation attitudes and behaviors. Thus, in view of the increasingly recognized importance of local engagement in protecte d area management and the realities of the Romanian context, this study examined the extent to which community social dimensions could explain conservation attitudes and pro environmental civic attitudes and behavioral intentions. Taking into account that the linkage between the institutional and social dimensions of community in Romania is still viewed as fairly weak (Cottrell & Cutumisu, 2006), this study primarily fo cuses on the social dimensions of pro environmental engagement. As previously emphasized, in the Romanian context, assertions have been made that the lack of environmental engagement is a consequence of a low sense of community and collective environmental responsibility. Community attachment has been previously identified in the literature as a predictor of environmental engagement ( Brehm, Eisenhauer, & Krannich, 2004) and environmental responsibility has been linked to pro environmental engagement ( Garling et al. 2003 ). Furthermore, concerns have been raised over the attitudes people residin g adjacent to Retezat National Park have towards conservation and their care for the environment. Therefore, this study examined these constructs, focusing on investigating the major determinants of community attachment and the nature of the relationship b etween community attachment, conservation attitudes, connections to nature, perceived environmental responsibility, attitudes towards p ro environmental civic engagement and pro environmental civic behavioral intentions. Purpose of the Study The purpose of this study is to examine the social dime nsions of community and their potential to facilitate or hinder local engagement i n pro environmental behaviors. Three research questions were addressed in this study. First, how levels and types of community attach ment,

PAGE 21

21 connections to nature, conservation attitudes, and perceived collective environmental responsibility facilitate or hinder attitudes towards p ro environmental civic engagement and pro environmental civic behavioral intentions. Secondly, this study que stioned if different facets of community attachment are distinctively predicted by length of residence, social interaction, and socio demographic characteristics. Lastly, differences in affective and attitudinal environmental responses were assessed among young adults middle age adults, and older adult s residing in rural Romania The study questions were examined in the context of nine rural communities adjacent to Retezat National Park, Romania. Retezat National Park was the first area designated (1935) as a national park in Romania and also the first park in Romania with a co management system in place. This national park is viewed as being a management model for other parks and protected areas in Romania (van Hal, 2006). The rural communities adjacent t o Retezat National Park were well suited for this study because they are representative of other rural communities in Romania that are making use of and are affected by the conditions of the neighboring protected areas. The research findings have implicati ons and relevance for other protected areas in the country, as well as abroad. This study seeks to i mprove our understanding of community social dimensions that facilitate or inhibit pro environmental behavioral intentions. By increasing our understanding of the theoretical bases for public engagement and our ability to account for them, greater knowledge to assist community program development emerges To accomplish the study purpose, qualitative and quantitative data collection techniques were employed. D ata were collected through in depth interviews and surveys with community members.

PAGE 22

22 Dissertation Format This dissertation is presented in six chapters. Following the introduction and a more detailed description of the study methods, are chapters devoted to answering the three major research questions of the study. A discussion of the content contained within each chapter follows. Chapter two depicts the methods employed in this study. The data collection and analysis process are described in greater detail. Descriptive statistics for all the constructs included in the study are provided. Furthermore, the results of the measurement model for constructs included in the study are reported. Chapter three examines the structural relationships between community at tachment, connection s to nature, conservation attitudes, perceived collective environmental responsibility, attitudes towards p ro environmental civic engagement and pro environmental civic behavioral intentions. A structural equation model was proposed and the hypothesized relationships were tested using MPLUS version 5.21. Chapter four explores the intricacies of community attachment using a multidimensional conceptualization of the construct, attachment to the social and natural environment. Th is chapter focuses on assessing if different facets of community attachment are distinctively predicted by length of residence, social interaction, and socio demographic characteristics. Furthermore, it elaborates on current understanding of community attachment by q ualitatively exploring the extent to which we could speak of other facets of the community attachment construct. Chapter five investigates whether affective and attitudinal environmental related constructs differ among young adults middle age adults, and older adult s residing in rural Romania Citizens of differing ages were raised under different political and park status stages in Romania. Affective response captured connections with nature and attachments to the park, while

PAGE 23

23 attitudinal response included measures of attitudes towards conservation and attitudes towards pro environmental civic engagement behavior s Chapter six conceptually identifies further research streams based on the study findings and discussions In particular, the importance of testi ng the proposed structural relationships under different contexts is emphasized as well as practical applications Figure 1 1. Map of Retezat National Park Romania

PAGE 24

24 Table 1 1. Communes and villages adjacent to Retezat National Park Commune Villages Adu lt Population 1. Salasu de Sus Salasu de Sus* Coroiesti Malaiesti Nucsoara Ohaba de sub Piatra Paros Pestera Rau Mic Rau Alb Salasu de Jos Zavoi 2,490 2. Rau de Mori Rau de Mori* Brazi Clopotiva Ohaba Sibistel Ostrov Ostrovel Ostrovu Mic Sibistel Sus eni Unciuc Valea Biljii 3,314 3. Santamaria Orlea Santamaria Orlea* Balomir Barastii Hategului Bucium Orlea Ciopeia Sacel Sanpetru Subcetate Vadu 3,198 4. Pui Pui* Baiesti Federi Galati Hobita Ohaba Ponor Rau Barbat Rusor Serel Fizesti Uric Po nor 4,527 5. Uricani Town Campu lui Neag 480 TOTAL 14,009

PAGE 25

25 CHAPTER 2 METHODS Research Design This study examin ed community social dimensions and their ability to facilitate or inhibit pro environmental behavioral intentions. To accomplish the study go als, this research was conducted in nine communities adjacent to Retezat National Park, Romania using a mixed methods approach, quanti tative and qualitative methods for data collection and data analysis being employed A series of steps were taken to gath er information relevant to better un derstand ing the social dimensions of pro environmental engagement in Retezat National Park. Primarily, two major data collection phases were carried out simultaneously to answer the research questions of this study. A qu alitative approach to data collection, entailing in depth/ key informant interviews with co mmunity members was conducted. Concurrent to the in depth/ key informant interviews, a household survey of a sample of local communities adjacent to Retezat National Park was conduc ted. In the following sections the data collection and analysis procedures for these two phases are described in more detail. Data Collection Procedures Quantitative Data Collection The theoretical population of this phase included all ad ults living in rural communities adjacent to protected areas in Romania. The accessible population for the study was adult community members of 43 rural communities adjacent to Retezat National Park, Romania, villages that belong to five communes. A commun e is an administrative division in Romania encompassing one or more villages that share similar economic, socio cultural, geographic and

PAGE 26

26 demographic conditions. For administrative purposes, one village in each commune is designated as the major village, wh ere all the local administrative authorities are located. The total adult population in the villages adjacent to Retezat National Park was estimated at 14,009 adult residents. In order to more thoroughly examine the determinants of pro environmental civic behavioral intentions in the communities neighboring Retezat National Park, and due to budget/ resource limitations, nine communities were selected using multistage sampling to be included in the final sampling frame. Considering that the rural communitie s adjacent to Retezat National Park belong to five communes, a decision was made to select two villages dissimilar in terms of size from each commune, to assure representation of each commune in the final sample. Furthermore, taking into account that each commune has a village designated as the major village (the largest village in the commune), where all the public offices are located, a decision was made to include these villages in the final sample frame (one commune had only one village) Thus, at first the following five major villages were included in the sampling frame and excluded from further sampling: Salasu de Sus, Rau de Mori, Santamaria Orlea, Pui, and Campu lui Neag. Secondly, from four commune s one more village was randomly selected Using a table listing all the villages adjacent to the park, each village was numbered and using a table of random numbers a village was selected from each commune. Accordingly, the following four villages were selected to be included in the final sample: Nucsoa ra, Unciuc, Sacel Sanpetru, Hobita. The sampling frame included the total number of households in the nine communities adjacent to RNP, but the final unit of analysis was the individuals residing in the households. The nine villages selected have a popula tion of 4,232 persons residing in 1,159 private households. A systematic sampling method with a random start was used to select participants for the face to

PAGE 27

27 face interviews, the sampling interval being established based on the number of households in the s ampling frame for each community divided by sample size needed in each community. The person in the household of age of 18 or older was asked to participate in the study, and the questions were asked directly to the respondents and recorded by the intervie wer. Bernard (2000) suggested that a sample size depends on several aspects as follows: the heterogeneity of the population from which the elements are chosen; how many population subgroups (independent variables) the researcher wants to deal with simulta neously in the analysis; the size of the phenomenon that the researcher is trying to detect and how precise you want your sample statistics (or parameter estimator) to be. In terms of the heterogeneity of the population from which the elements are chosen, a fairly heterogeneous sample was anticipated in terms of the variables of interest for the study, due to the sampling approach (systematic random sampling). De Vaus (2001) emphasized the need for sufficient variation in the sample on the key variables of the study. Considering that this study focuses on collecting data from different communities, with d ifferent population sizes, it was expected to have an acceptable le vel of variation in the sample. Based on the accessible population size, estimated at ap proximately 14,009 adult residents, the minimum number of completed questionnaires needed was determined to be 576 The accuracy of sample statistics are expressed in terms of the level of confidence that the statistics fall within a specified interval fro m the parameter. This sample size was found to be sufficient to limit sampling error and be statistically representative of the population at a confidence level of .95 and confidence interval of 4, wh ich means within plus or minus 4 percentage points of th e population parameter.

PAGE 28

28 Taking into account the possibility of a low response rate, in addition to the sample determined of 576, an approximately 50% sample was calculated to be included in the study (Steele et al., 2001). Thus, the total number of people that was planned to be asked to participate in the study was established at 850. The number of questionnaires to be conducted in each community was calculated based on the number of households in each community. The total number of households in the nine communities was estimated at 1 159. For example, given a population of 205 households in Pui, approximately 153 residents were attempted to be asked to participate in the study Similarly, at a total of 100 households in a community (7 % of the estimated to tal number of households in the nine villages), approximately 60 households were planned to be contacted to participate in the study. The sampling frame is depicted in Table 2 1 Instrumentation Participants in this study were asked to express their opinio ns on a series of questions about their community as well as Retezat National Park. The instrument consisted of fixed choice, partially closed ended and open ended questions ( Appendix A). The final questionnaire included four broad sections: opinions about the community and local life conditions; feelings about Retezat National Park and nature in general; engagement in environmental protection in Retezat National Park; and demographics. More specifically, the community and local life conditions section incl uded six questions (e.g. social attachment scale, social interaction scale, q uality of life perceptions, level of involvement in community affairs). The section capturing feelings about Retezat National P ark and nature in general contained ten questions ( e.g. connection to nature scale, park attachment scale, conservation attitudes, park visitation patterns). The section on engagement in environmental protection in Retezat National Park included four questions (e.g. attitudes towards pr o environmental ci vic engagement scale, pro environmental civic b ehavioral intentions scale, perceived en vironmental responsibility items ).

PAGE 29

29 The demographics sect ion consisted of ten questions measuring length of residence, household size gender, age, family status, educati onal level, employment, and household income. Community attachment was measured using a scale (multiple items) ; the same approach was used to measure conservation attitudes, connection s to nature, attitudes towards pr o environmental civic engagement perc eived environmental responsibility and pro environmental civic behavioral intentions. The instruments employed are described in greater detail in the subsequent chapters. Reliability and validity Predominantly scales were employed to measure the study cons tructs, in order to reduce random error. In cases were scales were used, reli ability measures as well as confirmatory factor analyses were conducted to determine their accuracy. To assess face and content validity, multiple strategies were followed. At fi rst, three university professors reviewed the survey instrument and also provided feedback in terms of the extent the empirical measures adequately reflect ed the real meanings of the concepts under consideration. Secondly, the principal investigator (a nat ive of Romania) translated the English language version of the questionnaire in Romanian. Thirdly, a Romanian student translated the Romanian language version of the questionnaire back to English, and after revisions were made, further verified the accurac y of the translation. Lastly, at the study site, park management staff was asked to complete the survey and comment on the questionnaire content, design, clarity, wording and format. Based on the feedback, minor adjustments were made to the questionnaire. Appendix B depicts the Romanian language version of the study questionnaire. Qualitative Data Collection In depth interviews were conducted with community members in each of the nine communities selected for study. A grounded theory approach to inquiry was followed to allow

PAGE 30

30 for an in depth theoretical conceptualization of the community attachment construct. Furthermore, an examination of the feelings, attitudes, and behaviors that people living adjacent to a national park have towards their communities and the protected area was undertaken. A non probability sampling approach was employed, this phase of the study being focused on understanding the experiences of the respondents and not generalizing to the larger population. A referral sampling technique was employed to identify participants, with a focus on including individuals highly involved in their community and also individuals with low levels of local engagement. Park management employees and public officials were asked to provide names of community me mbers from the nine selected communities that are interested and highly engaged in community affairs. Based on a first interview in each of the nine communities, the principal investigator asked the respondents to provide names of other community members t hat might participate in the study and have dissimilar levels of interest and engagement in their community. Data was collected from 24 community members representing the nine communities adjacent to Retezat National Park. Creswell (2007) recommends that r esearchers interview 20 to 30 individuals when a grounded theory approach to qua litative inquiry is employed. Cons idering that this study involved multiple communities, approximately three interviews were conducted in each community. A semi structured int erview with an interview guide was employed to organize the discussion. The discussion primarily evolved around four major questions, respondents being asked to: (1) describe their community to someone who has never been there; (2) describe what is the mos t important thing to them about their community; (3) describe their feelings and attitudes towards the park; and (4) describe the things they do or they are willing to do for their commun ity and the park ( Appendix C ). The Romanian version of the interview guide is

PAGE 31

31 presented in Appendix D. The interviews were conducted by the principal investigator in the native language and were tape recorded, when permission to record the interview was obtained from the interviewee. All the interviews were translated (into English), transcribed and analyzed by the principal investigator. Data Analysis Quantitative Data Analysis Data analys e s w ere performed to address the three research questions of the study. The data for two of the research questions was analyzed using str uctural equation model (SEM) analysis, while for the third research question multivariate analysis of variance ( MANOVA ) was employed. The structural equation model analysis encompassed several stages. First, descriptive statistics were computed for the var iables used in the study using the Statistical Package for the Social Sciences (SPSS) version 18.0. Second, the data collected was screened and the critical assumptions underlying the statistical techniques employed by the study were assessed. Third, a two step data analysis was employed to assess the hypothesized relationships among the research construct s (Anderson & Gerbing, 1988). A s part of this process, individual items were examined using Confirmatory Factor Analysis (CFA) and the measurement model f or constructs included in the study was estimated using MPLUS version 5.21 to determine how well the indicators captured their specific constructs and the ability of the respondents to differentiate between constructs (Hair et al., 2006). This was followed by an assessment of the S tructural Equation M odel (SEM) assessing the hypothesized relationships between constructs. SEM was assessed using MPLUS version 5.21 using the WLSMV (weighted least squares mean and variance adjusted) method of estimation, method recommended for categorical and ordinal data (Muthen et al., 1997).

PAGE 32

32 The fit of the measurement model and the structural equation model were assessed using multiple criteria. T he chi square test of model fit divided by the degrees of freedom was used as a reference criteria ( <3.0; Kline, 2005 ) supplemented by the Root Mean Square Error of Approximation (RMSEA), Wei ghted Root Mean Square Residual (WRMR) Comparative Fit Index (CFI) and Tucker Lewis Index ( TLI ) RMSEA values equal to or less than .06 is indic ative of a good model fit values between .08 and .10 indicate acceptable mod el fit, and values higher than .1 0 are considered to be indicative of po or fit (Browne & Cudeck 199 2; MacCallum, Browne & Sugawara 1996). CFI and TLI values equal to or greater than .95 also indicate good model fit (Hu & Bentler, 1999) The criterion for WRMR is a value less than 1.00 (Yu, 2002 ). Cronbac estimate the reliability of confirmatory factors. Evidence of internal consistency is provided by 1994). Fornell & Larker (1981) suggested that a composite reliability (CR) greater than .7 0 as adequate. For convergent validity, the average var iance extracted (AVE) coefficient has been widely used. The AVE value that is greater than .5 0 is deemed accep table (Bagozzi, 1994; Fornell & Larker, 1981). Another criterion for convergent validity is factor loadings. T statistics for factor loadings ) greater than 1.96 at a significance level of p < 0.05 are regarded as significant. Hatcher (1994) indicated that attributes with factor loadings lower than .40 should be excluded from further analysis Furthermore, Kline (2005 ) emphasiz ed the importance of using good psychometric characteristics that will each also have relatively s tandardized factor loadings (i e. > .60). Thus, in this study the criteria for factor loadings were set at .60. For obtaining discriminant validity, the corr elations between variables should be less than .85 (Kline, 2005).

PAGE 33

33 Marsh, Craven, & Debus (1991) underscored that when a model has been misspecified (poor model fit), the researcher has to respecify the model. One way to respecify the model is to delete ind icators and the other option is to allow errors to correlate, and decisions should be supported by theory or rationale (Joreskog 1993). Residuals and modification indices are two main sources to identify misfits in the model. Standardized residuals exceed ing 2.58 are a considered to be a baseline for respecification. Another type of information relevant for respecification is values of modification indices for parameter estimates constrained to equal zero. Large modification indices (> 3.85) and statistic ally significant at the .05 level may often be associated with the largest expected change value (Saris, Satorra, & Sorbom, 1987). On the baseline of large residuals and large expected chan ge values, the researcher can determine how the model can proceed t o be respecified. The literature suggests that l arge modification indices should be considered for elimination, and if they are so critical that it is hard for them to be eliminated, the next largest modificatio n index should be evaluated and considered fo r respecification ( Saris, Satorra, & Sorbom, 1987 ). To assess group differences in terms of affective and attitudinal environmental response, three multivariate analysis of variance (MANOVA) models were performed. Data analysis encompassed several stages. First, descriptive statistics were computed for the variables used in the study using the Statistical Package for the Social Sciences (SPSS) version 18.0. Second, the data collected was screened and the critical assumptions underlying the statistical techn iques employed were assessed. Third, individual items were examined using Confirmatory Factor Analysis (CFA) and the measurement model for constructs included in the study was estimated using MPLUS version 5.21 Lastly three multivariate analysis of varia nce (MANOVA) models

PAGE 34

34 were assessed. hoc tests with Tukey statistics were evaluated in this stage. The following sections depict the descriptive statistics for the variables used in the study, followed by an a ssessment of the study measurements using Confirmatory Factor Analysis (CFA) with a focus on determining how well the indicators captured their specific constructs Descriptive statistics Participants. Cross sectional data was collected from 260 residents during June October 2009 using face to face interviews (68% response rate) and mail survey (9% response rate). The average age was 45.0 years, with almost one quarter of the respondents (23.7%, n = 59) being between 18 and 30 years and slightly more than one third of the respondents (36.1%, n = 90) being between 31 and 50 years. Respondents who were above 51 years were the most represented group (40.2%, n = 100). Of the respondents, 54.2% ( n = 135) were males and 46.8% ( n = 114) were females. The average length of residency was 37.36 years ( SD = 18.269), with 40.1% ( n = 101) residing in their community between 20 to 40 years, 27.4% ( n = 69) between 41 to 60 years and 20.6% ( n = 52) between 1 to 20 years. The majority of the respondents (65.1% n = 162 ) wer e mar ried or partnered, 48.6% ( n = 121) had one to five children under 18 years, and on average, the number of adults per household was three. Almost one third of the respondents (31.6% n = 79 ) indicated high school as the highest level of education atta ined, 14% ( n = 35) of the respondents had some college or a college degree and 14 % ( n = 35) had an elementary school education or less. About one third of respondents (33.8% n = 82 ) reported a monthly househol d income between 1,000 and 2,000 RON (about U S $330 $660) and 30.6% ( n = 74) indicated a monthly household income of more than 2,000 RON (about US $661). When their current employment status was asked, slightly less than one third of

PAGE 35

35 the respondents (29.5%, n = 74) reported they were retired, 10.8% ( n = 27) were students, while a small percentage of the respondents identified themselves as being unemployed (2.4%, n = 6). The majority of respondents (70.6% n = 173 ) indicated that they do not have any property rights (ownership or land use rights) in R etezat National Park. Almost all respondents (96.0%, n = 241 ) reported they do not personally receive any income from the park or its visitors. Similarly, th e majority of the respondents (97.2%, n = 244) reported their immediate family does not receive any income from the park and its visitors. The demographic characteristics are depicted in Table 2 2 Local life. The majority of the respondents evaluated the quality of life in their community as average (70.5%, n = 182), with an overall mean ( M ) of 3.07 an d standard deviation ( SD ) equal to .61, where 1 = Poor; 2 = Above poor; 3 = Average; 4 = Above average; 5 = Excellent. Almost half of the respondents (49.6%, n = 127) indicated they are somewhat active in community activities or events, while only 2.7% ( n = 7) reported as being extremely active in community activities and events. The level of involvement in community activities and events had a mean of 2.77 and standard deviation of .92, where 1 = Not active at all; 2 = Not very active; 3 = Somewhat active; 4 = Very active; 5 = Extremely active. Social attachment. The social dimension of community a ttachment was measured using 12 items on a five point Likert scale ranging from 1 to 5, where 1 = Strongly disagree; 2 = Disagree; 3 = Neutral; 4 = Agree; 5 = Str ongly agree. he associations that I have with other people in this community mean a lot to me M = 4.52, SD = .76 ) I feel like I belong in this community M = 4.50, SD = I agree with most people in this

PAGE 36

36 community about what is important in life M = 3.61, SD = 1.30). These re sults are presented in Table 2 3 Social interaction. Social interaction was measured using the f requency of interaction do you see or meet with at least one of the following types of people? Family, Close Friends, the respondents were given response options of: (1) never, (2) a few times a year, (3) once a month, (4) a few times a month, (5) once a week, (6) ( M = 6.43, SD = 1.01) was the most highly rated M = 5.55, SD M = 1.47, SD M = 2.71 SD = 1.82). Table 2 4 depicts the results for the social interaction items. Connection to nature. Connection s to nature were measured using 18 items on a five point Likert scale ranging from 1 to 5, where 1 = Strongly disagree; 2 = Disagree; 3 = Neutral; 4 = Agree; 5 = St rongly agree. The items captured five dimensions of connection to nature: admiration (3 items), spirituality (3 items), identity (4 items), sorrow (2 items), restoration (3 items), and fear (3 items). hen surrou nded by M = 4.91, SD = .33 ) I feel sorrow because w e're destroying M = 4.75, SD = .76 ). The lowest rated item M = 1.60, SD = 1.18), followed b I have too much fear of nature to hike in remote natural M = 2.10, SD = 1.51). These re sults are presented in Table 2 5 Park visitation. The majority of the respondents (91.5%, n = 227) reported they have been at least once inside Retezat Natio nal Park. Slightly more than half of the respondents (51.5%, n =

PAGE 37

37 123 ) mentioned they have been inside the park a few times during the past 12 months. Nearly third of the respondents (29.3%, n = 70 ) did not visit the park during the past 12 months. When ask ed was the major purpose for visiting the park, the respondents mentioned hiking (79.5% n = 190 ), camping (32.6% n = 78) or collecting non timber forest products (30.5% n = 73 ) as the major purpose for their visit Table 2 6 depicts in more detail the m ajor reasons for visiting the park. When asked to identify for what reasons the park is important to them, the respondents mentioned a great place to visit with family and friends (87.0%, n = 215), s cenery, its unique landscapes, plants, and animals (8 5.8%, n = 212), outdoor recreation (hiking, climbing, etc.) (84.2%, n = 208), b enefits it brings to our community (48.5%, n = 120), u se of natural resources (timber, mushrooms, medicinal plants, fishing, etc.) (42.5%, n = 105) and l ivestock grazing ( 34.8%, n = 86). Park attachment. using nine items on a five point Likert scale ranging from 1 to 5, where 1 = Strongly disagree; 2 = Disagree; 3 = Neutral; 4 = Agree; 5 = Strongly agree. T he items included captured two dimensions of park attachment, i dentity (5 items) and dependence (4 items). Of the nine items, Retezat National Park means a lot to me M = 4.54, SD I enjoy living near Retezat National Park M = 4.54, SD = .75 Retezat National Park is very important to me M = 4.33, SD I get many personal benefits out of living near Retezat National Park M = 3.72, SD I identify strongly with the Retezat National Park M = 3.78, SD = 1.18). These results are presented in Table 2 7 Conservation attitudes. based on their reactions to 25 items on a five point Likert sca le ranging from 1 to 5, where 1 =

PAGE 38

38 Strongly disagree; 2 = Disagree; 3 = Neutral; 4 = Agree; 5 = Strongly agree. These items captured four dimensions of conservation attitudes: conservation awareness (6 items), conservation benefits (7 items), land use persp ectives (7 items), and management considerations (5 items). t is important to have the Retezat National Park for the survival of various plants and animal species M = 4.91, SD = .42) was the most highly rated, followed t i s necessary to set aside some land for the protection of plants and animals M = 4.89, SD Retezat National Park is a waste of land M = 1.36, SD eople who own land / have property rights in t he park should be allowed to use resources as they wish M = 1.92, SD = 1.33). Table 2 8 depicts these results. When asked who should establish the rules and regulations for land management in rk Administration ( 89.6%, n = 215), local councils (56.3%, n = 135), and landowners (40.0%, n = 96). Only seven respondents (2.9%) mentioned there is no need to establish strict regulations for administering the land within the park. Attitudes towards pro environmental civic engagement Attitudes towards pro environmental civic engagements were measured using nine items on a five point Likert scale ranging from 1 to 5, where 1 = Not at all effective; 2 = Never effective; 3 = Sometimes effective; 4 = Often e ffective ; 5 = Always effective Of the nine earning about the environment ( M = 4.59, SD nvesting time to learn about the park and environmental protection M = 4.36, SD articipating in educational programs about the environment M = 4.36, SD articipating in a worksho p on how to reduce my dependence on park resources M = 3.87, SD = 1.15). Overall, t ringing tourists to the p ark M = 4.41, SD = .89) and

PAGE 39

39 oting for public officials that show interest in environmental issues as being often effective in protecting the environment in Retezat National Park Table 2 9 depicts these results. Pro environmental civic behavioral intentions. Pro environmental civic behavioral intentions were measured using eight items on a five point Likert scale ranging from 1 to 5, where 1 = Very unlikely; 2 = Somewhat unlikely; 3 = Neither likely nor unlikely; 4 = Somewhat like ly; 5 = Very likely. ttend a public presentation about Retezat National Park M = 4.02, SD ive my input into park management decisions M = 3.94, SD = 1.16). The lowest rated item was articipate in a worksho p on how to reduce my dependence on park resources M = 3.23, SD = 1.39). These results are presente d in Table 2 10 Perceived environmental responsibility. Perceived environmental responsibility was measured using four items on a five point Likert scale ranging from 1 to 5, where 1 = Strongly disagree; 2 = Disagree; 3 = Neutral; 4 = Agree; 5 = Strongly agree. The items captured personal environmental responsibility, collective environmental responsibility, and governmental environ a uthorities, together with the citizens, are responsible for protecting the environment in Retezat National Park M = 4.67, SD = .81) was the most highly rated uthorities, rather than the citizens, are responsible for protecting the environment in Retezat National Park ( M = 4.30, SD = 1.07). Table 2 11 depicts these results. Confirmatory factor a nalysis A first order confirmatory factor analysis (CFA) of each construct measu red in the study was used to assess the reliability and validity of the constructs and which variables should be included in the models based on good fits. Thus, the CFA of the following constructs was

PAGE 40

40 assessed: community attachment, social interaction, connection t o nature, conservation attitudes, attitudes towards pro environmental civic engagement and pro environmental civic behavioral intentions. Confirmatory factor analysis (CFA) of community attachment. Community attachment was conceptualized as encompassing attachment t o the social environment (social attachment ) as well as the natural environment (attachment to the park). In thi s study, the social dimension of community attachment was measured using a multiple item scale including 12 items. The attachment to the natural environment (the park in this study), was measured using a multi item scale encompassing two distinct dimensions of attachment, identity (5 items) and dependence (4 items) For the purposes of this study first and foremost one measurement mode l with thr ee factors was tested using confirmatory factor analysis The multifaceted three factor model (social attachment park identity, park dependence ) assumed that the underlying factors of community attachment are three distinct but related factors, which are cause of their observed measures Two items which were negative items were reserve coded before running the confirmatory factor analysis. The results of the CFA with three factors and 21 items revealed poor fit, the chi square or 2 / df ratio (3. 68: 2 = 198.57 df = 54 p < .00 1) was higher than the suggested criteria (i.e., <3.0; Kline 2005). CFI (.94 ), RMSEA ( .102 ) and WRMR ( 1.162) yielded a poor model fit with the exception of TLI ( .97) which suggested an adequate fit. An investigation of the f actor loadings revealed that factor loading for one item here are many things I would like to change about this community was statistically insignificant and four items f the people in this community were planning something, I'd think of it as so mething WE were doing rather than THEY were doing f I needed advice about something, I could go to someone in this community here are things going on in this

PAGE 41

41 community that I am not proud of I feel no commitment to Retezat National Park had fa ctor loadings lower than .60. Thus, a decision was made to eliminate these five items from further analysis due to their low factor loadings. Two more items verall, I am very attached to this community I agree with most people in this community about what is important in life were eliminated from further analysis based on a n investigation of the modification indices that showed the model could be improved if we allow these items to correlate to other two items very similar in working O ne item ( I w ouldn't substitute other places for living near Retezat National Park was eliminated because it was found as a wea k measure of the park dependence factor, due to high correlation with the other two factors in the model. Furthermore, the correlation betw een p ark identity and park dependence ( r = .90 ) was high er than the recommended criteria of .85 (Kline, 2005), suggesting a weaker differentiation of these constructs by this population. Consequently, scale items were collapsed and one measure of park atta chment was retained including seven items. The CFA model with six items measuring social attachment and seven items measuring park attachment, revealed good model square/ df ratio (2.87: 2 = 77.402 df = 27 p < .00 1) was lower than the suggest ed criteria (i.e., <3.0; Kline 2005). CFI (.98 ), TLI (.99), RMSEA ( .085 ) and WRMR ( .842) yielded a good model fit and all the item (indicator) loadings were significant ( p < .001) and ranged from .65 to .95 that provides strong evidence of convergent valid ity In terms of discriminant validity the correlation between social attachment and park attachment was lower than .85, being equal to .56. The fit indices for the community attachment measurement mo del with three factors are depicted in Table 2 12 The item (indicator) loadings, reliability coefficients (Cronbach alphas), composite reliability coefficients (CR), and convergent validity results (AVE) were above the recommended criteria and are portrayed in Table 2 13

PAGE 42

42 Confirmatory factor analysis (CFA) o f s ocial interaction. Social interaction was measured using 10 items encompassing three underlying factors: interaction with strong ties (3 items), interaction with weak ties (3 items ), and interaction with authorities (4 items). The initial CFA model reve square/ df ratio (2.16: 2 = 47.604 df = 22 p < .00 1) was lower than the suggested criteria (i.e., <3.0; Kline 2005). CFI (.96 ), TLI (.96), RMSEA ( .068 ) and WRMR ( .680) yielded a good model fit However, the factor loadings fo r two items ommunity groups (e.g. church) were low (<.60) and the reliability coefficient for the interaction with strong ties factor was lower than the recommended criteria (.70), Cr onbach alpha being equal to .58 Consequently, the two items and the factor capturing interaction with strong ties were eliminated from further analysis. After eliminating these items, the RMSEA was high (.102). Thus, based on an evaluation of the modification indices a new model was proposed for social intera ction encompassing interaction with friends (2 items), inte raction with public officials (1 item), and park interactions (3 item s ). In this model the factor loading for the one item capturing interaction with public officials was set at 1.00. The results o f embers of non governmental organizations loading and thus considered for elimination. The fit indices for the social interaction measu rement model with three factors (a total of 5 items) revealed good fit. square/ df ratio (1.96: 2 = 5.890 df = 3 p > .05) was lower than the suggested criteria (i.e., <3.0; Kline 2005). CFI (.99 ), TLI (.99), RMSEA ( .062 ) and WRMR ( .242) yielded a good model fit These re sults are depicted in table 2 12 In terms of discriminant validity, the correlations between interactions with friends and interactions with the park was .40, interactions with friends and interactions with public officials was .43, interactions with the park and interactions with public officials was .52 all being lower than .85. The item (indicator)

PAGE 43

43 loadings, reliability coefficients (Cronbach alphas), composite reliability coefficients (CR), and convergent validity results (AVE) were above the recommended criteria and are portrayed in Table 2 1 4 Confirmatory factor analysis (CFA) of connection to nature. The initial CFA model of connection s to nature included 18 items encompassing five factors: admiration (3 items), spirituality (3 items), identity (4 items), sorrow (2 items), restoration (3 it ems), and fear (3 items). The multifaceted six factor model assumed that the underlying factors of nature connection are six distinct but related factors, which are c ause of their observed measures. The results of the initial CFA model revealed adequate mo square/ df ratio (2.69: 2 = 96.73 df = 36 p < .00 1) was lower than the suggested criteria (i.e., <3.0; Kline 2005). CFI (.98 ), TLI (.98), RMSEA ( .083 ) and WRMR ( .883) yielded a good model fit However, the reliability coefficients for thr ee fact ), sorrow (. 42 ), and fear ( .67 ) were low W hen analyzing the factor loadings one item nature fear had a low factor loading. A fter elimina ting this item the reliabili ty for the items left measuring nature fear was low .69 ) Thus, the three factors (awe, sorrow, and fear) were eliminated from further analysis. The results of the new model with three factors (identity, spirituality, and restorati on) revealed a high RMSEA ( .111 ). An investigation of the modification indices showed that one ature provides dimension, the results showing the model could be improved if this item would be allo wed to load on the identity and restoration dimensions. Thus, this item was eliminated from further analysis. Furthermore, inter factor correlations were high, the correlation between nature restoration and nature identity (.87), and between nature restora tion and nature spirituality (.93)

PAGE 44

44 were above the suggested criteria of .85 (Kline, 2005) and suggesting inability of the sample to differentiate between factors Therefore, scale i tems were collapsed and tested as a one dimensional measure of connection t o nature After collapsing the items, RMSEA (.122) showed poor model fit. Further investigation of the modification indices showed two items that strongly eeling part of n ature is a spir m y love for natur further analysis. The fit indices for the connection to nature measurement model with one factor revealed acceptable model fit, the square/ df ratio (3.01: 2 = 30.094 df = 10 p < .0 01 ) being close to the suggested criteria (i.e., <3.0; Kline 2005). CFI (.99 ), TLI (.99), RMSEA ( .091 ) and WRMR ( .600) yielded an acceptable model fit These re sults are depicted in table 2 12 The item (indicator) loadings, reliabi lity coefficients (Cronbach alphas), composite reliability coefficients (CR), and convergent validity results (AVE) were above the recommended criteria and are portrayed in Table 2 15 Confirmatory factor analysis (CFA) of conservation attitudes. For the p resent study one a priori four factor model was pro posed as a measure of conservation attitudes. The multifaceted four factor model assumed that the underly ing factors of conservation attitudes are fou r distinct but related factors: conservation awareness (6 items) benefits (7 items) land use perspectives (7 items) and m anagement considerations (5 items) which are cause of their observed measures The initial CFA model with four factors encompassing a total of 25 items, had a poor model fit the square/ df ratio (4.22: 2 = 257.385 df = 61 p < .001) being higher than the suggested criteria (i.e., <3.0; Kline 2005). CFI (.79 ), TLI (.90), RMSEA ( .115 ) and WRMR ( 1.355) yielded a poor model fit. An investigation of the factor loadings revealed fou r items ( f overgrazing continues in the park, all the animals will soon disappear he benefits

PAGE 45

45 from the park usually outweigh negative consequences t is good if some land within the park is allocated to local people Retezat National Park is for visitors, those who enjoy hiking and wildlife viewing ) with low factors loadings that were dropped. Furthermore, four more items Retezat National Park is a waste of land ersonally, I support the rules and regulations established by the park adminis tration t he quality of the air is higher because of living near the Retezat National Park area he park resources help local waters stay pure for our community were considered for elimination based on modification indices that showed these items be ing a weak measure of their constructs, the results showing that the model could be improvement if we allow these items to load on multiple latent factors Furthermore, the land perspective factor was eliminated from further analysis due to its weak associ ation with the other factors The land use perspectives factor did not load on the higher order model capturing conservation attitudes and was eliminated from further analysis. The fit indi ces for the conservation attitudes measurement model with three fa ctor s revealed good model fit, the square/ df ratio (2.23: 2 = 60.155 df = 27 p < .001) being lower than the suggested criteria (i.e., <3.0; Kline 2005). CFI (.96 ), TLI (.98), RMSEA ( .071 ) and WRMR ( .780) yielded a good model fit These re sults are depicted in table 2 12 In terms of d iscriminant validity, the correlations between conservation awareness and conservation management was .63 conservation awareness an d conservation benefits was .52 conservation benefits and conservation management was .58, all being lower than .85. The it em (indicator) loadings, reliability coefficients (Cronbach alphas), composite reliability coefficients (CR), and convergent validity results (AVE) were above the recommended criteria and are portrayed in Table 2 1 6

PAGE 46

46 Confirmatory factor analysis (CFA) of attitudes towards p ro environmental civic engagement For the present study one a priori model was proposed as a measure of attitudes towa rds p ro environmental civic engagements A one factor model was proposed that ass umed that the underlying attributes ( 12 items) of attitudes tow ard p ro environmental civic engagement load on one factor The initial CFA model with 12 items, had a poor model fit the square/ df ratio (6.87: 2 = 171.714 df = 25 p < .001) being higher than the suggested criteria (i.e., <3.0; Kline 2005). CFI (.91 ), TLI (.97 ), RMSEA ( .196 ) and WRMR ( 1.219 ) yielded a poor model fit. educing use of park resources rin ging tourists to the park v oting for public officials that show interest in environmental issues were eliminated from further analysis due to low factor loadings. Furthermore, four items ttending public presentations about the Retezat National Park earning about the environmen articipating in educational programs about the environment articipation in a local organization that is involved in park protection were strongly associated with other four items and the modification indices revealed that the model could be improved if we allow the errors of these items to correlate. Considering the items were very similar in working, a decision was made to eliminate four of the items from further analysis. The fit indices for the attitudes towards p ro environmental civic engagement measurement model with one factor (5 items) square/ df ratio (1.42: 2 = 7.105 df = 5 p > .001) being lower than the suggested criteria (i.e., <3.0; Kline 2005). CFI (.997 ), TLI (.998), RMSEA ( .042 ) and WRMR ( .289) yielded a good model fit These re sults are depicte d in table 2 12 The item (indicator) loadings, reliability coefficients (Cronbach alphas), composite reliability coefficients (CR), and convergent validity results (AVE) were above the recommended criteria and are portrayed in Table 2 1 7

PAGE 47

47 Confirmatory fa ctor analysis (CFA) of pro environmental civic behavioral intentions. A one factor model was proposed that ass umed that the underlying attributes of pro environmental civic behavioral intentions load on one factor (8 items). The initial CFA model with eigh square/ df ratio (10.99 : 2 = 131.931 df = 12 p < .001) being higher than the suggested criteria (i.e., <3.0; Kline 2005). CFI (. 93 ), TLI (.97), RMSEA ( .203 ) and WRMR ( 1. 224 ) yielded a poor model fit. Based on an inves tigation of the modification indices three items were found to be strongly associated with other items that were articipate in a public meeting related to Retezat National Park e actively involved in an orga nization that supports park management efforts xpress my concerns about park management to elected officials further analysis. The fit indices for the pro environmental civic behavioral intentions measurement model with one factor (5 items) square/ df ratio (1.21: 2 = 6.045 df = 5 p > .05) being lower than the suggested criteria (i.e., <3.0; Kline 2005). CFI (.999 ), TLI (.999), RMSEA ( .029 ) and WRMR ( .279) yielded a good model fit These re su lts are depicted in table 2 12 The item (indicator) loadings, reliability coefficients (Cronbach alphas), composite reliability coefficients (CR), and convergent validity results (AVE) were above the recommended criteria and are portrayed in Table 2 18 Q ualitative Data Analysis In depth with 24 community members representing the nine communities adjacent to Retezat National Park were conducted. Interviews were analyzed using a grounded theory ap proach (Strauss & Corbin, 1990, 1998) allowing for unique the oretical categ ories to emerge from the data. A series of steps specific to grounded theory research were followed in analyzing the data collected through the in depth interviews. At first, the principal investigator highlighted

PAGE 48

48 significant statements, sent ences as well as quotes, depicted from the respondent narratives. This stage was be followed by a clustering of meanings, the major themes in the narratives being identified in this stage. T he principal investigator coded the data for its major categories of information. This stage was followed by a process of linking categories across subjects and cases focusing on a deeper understanding of the interrelationships between categories and the context or setting that influences the revealed experiences. The f inal findings provide a theoretical conceptualization of the community attachment construct and depict how is to live as part of these communities, the shared feelings, attitudes, and be haviors people have towards their community and the neighboring protec ted area. Table 2 1. Sampling frame for local community residents based on households Communities Households (N) Sampling Frame Asked to Participate in the Study (On site/Mail) Completed Surveys (On site/ Mail) Santamaria Orlea 162 119 105 (40/65) 36 (32 /4) Sacel Sanpetru 196 128 87 (53/34) 23 (20/3) Rau de Mori 101 77 75 (75/0) 61 (61/0) Unciuc 55 43 30 (30/0) 25 (25/0) Salasu de Sus 150 111 54 (54/0) 33 (33/0) Nucsoara 70 51 32 (32/0) 28 (28/0) Pui 205 153 134 (26/108) 28 (15/13) Hobita 50 34 20 (20/0) 14 (14/0) Campu lui Neag 170 128 43 (20/23) 12 (11/1) Total: 1159 844 580 (350/230) 260 (239/21)

PAGE 49

49 Table 2 2 Demographic characteristics of the respondents Frequency Percent (%) Age 18 to 30 years 59 23.7 31 to 50 years 90 36.1 51 and ab ove 100 40.2 Total 249 100.0 Mean (SD) 45.02 16.405 Gender Male 135 54.2 Female 114 45.8 Total 249 100.0 Length of residence 1 to 20 years 52 20.6 21 to 40 years 101 40.1 41 to 60 years 69 27.4 61 to 80 years 28 11.1 Over 80 years 2 0.8 To tal 252 100.0 Mean (SD) 37.36 18.267 Family Status Single 64 25.7 Married/ Partnered 162 65.1 Divorced/ Separated 9 3.6 Widowed 14 5.6 Total 249 100.0 Household Adults 1 18 7.2 2 67 26.9 3 56 22.5 4 58 23.3 5 33 13.3 6 15 6.0 7 1 0.4 8 1 0.4 Total 249 100.0 Mean (SD) 3.30 1.395

PAGE 50

50 Table 2 2. Continued Frequency Percent (%) Minors (Under 18 years old) 0 128 51.4 1 61 24.5 2 52 20.9 3 6 2.4 4 1 0.4 5 1 0.4 Total 249 100.0 Mean (SD) .77 .933 Education None 2 0.8 Primar y school (1 4) 6 2.4 Elementary school (5 8) 27 10.8 Professional/ vocational school 37 14.8 Some high school (9 10) 14 5.6 High school graduate (9 12) 79 31.6 Post high school 34 13.6 Some college 7 2.8 College degree 28 11.2 Advanced degree 16 6 .4 Total 250 100.0 Household Income Almost no income 15 6.2 Less than 250 RON 10 4.1 Between 250 and 499 RON 14 5.8 Between 500 and 999 RON 47 19.4 Between 1000 and 1499 RON 48 19.8 Between 1500 and 2000 RON 34 14.0 More than 2000 RON 74 30.6 Total 242 100.0 Employment Retired 74 29.5 Not working outside of home 15 6.0 Unemployed 6 2.4 Student 27 10.8 Working in agriculture/ owning farming land 13 5.2 Working in industry 35 13.9 Working in commerce, tourism, and other services 7 2.8 Technician, supervisor 27 10.8 Personnel with higher qualifications 21 8.4 Business owner, entrepreneur 8 3.2 Other (please specify) 18 7.2 Total 251 100.0

PAGE 51

51 Table 2 2. Continued Frequency Percent (%) Property Ownership/Rights in RNP No 173 70.6 Yes 72 29.4 Total 245 100.0 Personal Income from RNP No 241 96.0 Yes 10 4.0 Total 251 100.0 Immediate Family Income from RNP No 244 97.2 Yes 7 2.8 Total 251 100.0

PAGE 52

52 Table 2 3 Descriptive statistics social attachment Social attachment Percent* Mean SD 1 2 3 4 5 % % % % % Overall, I am very attached to this community 0.8 3.5 8.6 25.8 61.3 4.43 .851 I feel like I belong in this community 0.8 3.5 7.5 21.7 66.5 4.50 .842 The associations that I have with other people in this community me an a lot to me 0.4 2.0 8.2 24.2 65.2 4.52 .762 If the people in this community were planning something, I'd think of it as something WE were doing rather than THEY were doing 0.8 6.4 14.3 30.7 47.8 4.18 .958 If I needed advice about something, I could go to someone in this community 5.9 5.9 8.2 27.1 52.9 4.15 1.166 I agree with most people in this community about what is important in life 8.8 14.0 16.0 29.6 31.6 3.61 1.298 Given the opportunity, I would move out of this community 56.0 8.1 10.9 11.7 13.3 2.18 1.520 I feel loyal to the people in this community 3.1 2.8 18.9 24.0 51.2 4.17 1.034 There are things going on in this community that I am not proud of 6.5 9.0 12.1 30.2 42.2 3.92 1.222 I plan to remain a resident of this community for a number of years 2.4 5.1 8.6 16.5 67.5 4.42 1.004 There are many things I would like to change about this community 1.6 5.5 5.9 29.9 57.1 4.35 .933 I like to think of myself as similar to the people who live in this community 6.3 10.7 9.1 27.3 46.6 3.97 1.249 *No te: 1 = Strongly disagree, 2 = Disagree, 3 = Neutral, 4 = Agree, 5 = Strongly agree.

PAGE 53

53 Table 2 4 Descriptive statistics social interaction Social i nteraction Percent* Mean SD 1 2 3 4 5 6 7 % % % % % % % Immediate family (parents, siblings) 1.2 11. 3 5.7 9.7 8.1 13.4 50.6 5.55 1.847 Extended family (cousins, uncles) 3.2 26.6 8.9 16.9 18.1 20.2 6.0 4.05 1.728 Acquaintances 0.4 12.6 5.7 21.9 20.2 20.6 18.6 4.85 1.606 Close friends 1.2 6.5 3.6 16.5 17.7 21.4 33.1 5.40 1.573 Neighbors 0.4 0.0 2.5 3.7 6.2 21.0 66.3 6.43 1.007 Community groups (e.g. church) 6.1 12.2 13.0 15.4 33.7 14.2 5.3 4.22 1.584 Public officials 6.5 33.5 17.7 11.7 13.3 7.7 9.7 3.54 1.790 Retezat National Park Staff 29.0 35.5 7.7 9.3 6.9 5.2 6.5 2.71 1.821 Tourists 16.1 38.3 8.9 15.3 8.1 4.8 8.5 3.09 1.816 Members of non governmental organizations 64.3 28.9 3.4 2.6 0.9 0.0 0.0 1.47 .758 Others 26.3 15.8 21.1 10.5 10.5 0.0 15.8 3.26 2.104 *Note: 1 = never, 2 = a few times a year, 3 = once a month, 4 = a few times a month, 5 = o nce a week, 6 = more than once a week, and 7 = everyday.

PAGE 54

54 Table 2 5 Descriptive statistics connection to nature Connection to n ature Percent* Mean SD 1 2 3 4 5 % % % % % Admiration I have seen things in nature that were so amazing; they j ust filled me with wonder 0.8 0.4 2.9 22.9 73.1 4.67 .634 The pow er of nature is just incredible 1.6 0.8 4.9 16.5 76.1 4.65 .759 The ma gnitude of nature is impressive 0.4 1.7 2.5 15.8 79.6 4.73 .633 Spirituality Nature provides me with a spiritu al connection 3.3 2.9 9.9 25.9 58.0 4.33 .994 My feelings for nature have influenced my spiritual beli efs 5.4 4.1 12.9 28.2 49.4 4.12 1.125 Feeling part of n ature is a spiritual experience 4.2 2.9 11.3 23.9 57.6 4.28 1.055 Identity My love for n atur e is a big influence in my life 2.0 2.4 6.9 26.5 62.0 4.44 .883 I am connected to nature much like I'm connected to my family 2.9 3.7 7.0 23.8 62.7 4.40 .974 Na ture is a huge part of who I am 2.1 4.1 6.6 26.7 60.5 4.40 .932 I often feel a sense of o neness w ith the natural world around me 2.9 5.4 9.1 26.0 56.6 4.28 1.028 Sorrow I feel sorrow because w e're destroying too much nature 2.5 1.2 1.2 9.0 86.1 4.75 .764 Seeing how much nature is being d estroyed affects me emotionally 2.1 1.7 5.0 17. 0 74.3 4.60 .832 Restoration When surrou nded by nature, I feel at peace 0.0 0.0 1.2 6.5 92.2 4.91 .327 Listening to the wind go through the trees calms my mind 4.6 2.5 8.7 21.6 62.7 4.35 1.051 When I'm alone in a natural area, I have this feel in g of complete calm 2.1 1.7 5.4 24.8 66.1 4.51 .841 Fear A lot of nature just scares me 75.6 6.6 4.5 8.7 4.5 1.60 1.184 Remot e natural areas make me nervous 43.1 6.7 10.0 20.1 20.1 2.67 1.643 I have too much fear of nature to hike in remote natur al areas 58.3 12.0 4.1 13.2 12.4 2.10 1.506 *Note: 1 = Strongly disagree, 2 = Disagree, 3 = Neutral, 4 = Agree, 5 = Strongly agree.

PAGE 55

55 Table 2 6 Major purpose for visiting the park T able 2 7 Descriptive statistics park attachment Park a ttachment Percent* Mean SD 1 2 3 4 5 % % % % % Identity Retezat National Park means a lot to me 0.8 1.2 9.7 19.8 68.4 4.54 .784 I feel no commitment to Retezat National Park 49.6 14.3 9.8 13.1 13.1 2.26 1.497 I am very attached to Retezat National Park 2.0 4.5 15.9 28.6 49.0 4.18 .992 Retezat National Park is very important to me 2.0 2.0 11.8 29.0 55.1 4.33 .910 I identify strongly with the Retezat National Park 6.6 6.6 23.5 29.2 34.2 3.78 1.178 Dependence I get many personal benefits out of living near Retezat National Park 8.9 9.3 17.5 28.9 35.4 3.72 1.280 I wouldn't substitute other places for living near Retezat National Park 3.7 5.3 11.5 19.3 60.2 4.27 1.093 I enjoy living near Retezat National Park 1.2 0.4 7.0 25.5 65.8 4.54 .750 I get lots of satisfaction out of living near Retezat National Park 3.3 2.4 15.5 31.8 46.9 4.17 .996 *Note: 1 = Stron gly disagree, 2 = Disagree, 3 = Neutral, 4 = Agree, 5 = Strongly agree. Frequency Percent Collecting wild resources 73 30.5 Hiking 190 79.5 Cam ping 78 32.6 Work 36 15.1 I never go there 6 2.5 Other 33 13.8

PAGE 56

56 Table 2 8 Descriptive statistics conservation attitudes Conservation a ttitudes Percent* Mean SD 1 2 3 4 5 % % % % % Awareness It is important to have the Retezat Na tional Park for the survival of various plants and animal species 0.4 0.4 0.8 4.5 93.9 4.91 .415 It is necessary to set aside some land for the protection of plants and animals 0.0 0.4 1.2 7.0 91.4 4.89 .381 Retezat National Park is our country's pride 0 .0 0.8 0.8 8.6 89.8 4.87 .481 Retezat National Park being protected is important for the benefit of our future generations 0.4 0.0 0.4 10.8 88.3 4.87 .418 The illegal cutting of trees in t he park should be s trictly regulated 2.1 2.1 3.7 8.7 83.5 4.69 .81 3 If overgrazing continues in the park, all the animals will soon disappear 11.6 9.4 13.7 26.6 38.6 3.71 1.367 Land use perspectives What people and their livestock need is more important than protecting wild animals and plants 37.6 8.7 28.5 16.1 9.1 2.50 1.370 It is good if some land within the park is allocated to local people 20.6 10.5 14.3 26.1 28.6 3.32 1.497 Retezat National Park is a waste of land 84.5 5.4 3.3 2.9 3.8 1.36 .964 People who own land / have property rights in t he park should be allowed to use resources as they wish 58.1 17.4 8.9 6.4 9.3 1.92 1.328 Retezat National Park is for visitors, those who enjoy hiking and wildlife viewing 1.7 2.5 5.4 14.6 75.8 4.60 .837 The economic stability of communities is more important than pro tecting park resources 36.9 14.5 20.3 14.5 13.7 2.54 1.452

PAGE 57

57 Table 2 8. Continued Conservation attitudes Percent* Mean SD 1 2 3 4 5 % % % % % Management Retezat National Park is managed success fully for the benefit of future generations 4. 3 6.8 10.2 26.4 52.3 4.16 1.123 Personally, I support the rules and regulations established by the park administration 2.6 3.4 6.8 17.9 69.4 4.48 .949 Retezat National Park is managed successfully for a wide range of uses and values, not just tourism 7.4 5.2 16.1 26.5 44.8 3.96 1.220 Retezat National Park management does a good job at protecting the natural resources in the park 3.4 6.0 10.3 28.8 51.5 4.19 1.066 The citizens from the communities around the park have enough say in how the park is managed 14.7 16.5 24.2 27.3 17.3 3.16 1.304 Benefits The benefits from the park usually outweigh negative consequences 6.1 5.7 21.8 32.3 34.1 3.83 1.145 My community benefits from being near the Retezat National Park 8.4 4.6 13.0 31.8 42.3 3.95 1.222 H aving the Retezat National Park near my home benefits me and my family 7.6 2.9 13.4 37.4 38.7 3.97 1.151 My community is a more beautiful place to live because we are living near Retezat National Park 1.3 0.9 8.1 28.6 61.1 4.47 .787 The quality of the ai r is higher because of living near the Retezat National Park area 0.4 0.4 1.7 16.7 80.8 4.77 .536 The park resources help local waters stay pure for our community 2.9 2.5 7.5 22.5 64.6 4.43 .948 The tourists who come to the area are useful to we who live in adjacent communities Retezat National Park Administration 5.8 11.2 17.0 27.0 39.0 3.82 1.227 *Note: 1 = Strongly disagree, 2 = Disagree, 3 = Neutral, 4 = Agree, 5 = Strongly agree.

PAGE 58

58 Table 2 9 Descriptive statistics attitudes towards pro environmental civic engagement Attitudes pro environmental civic engagement Percent* Mean SD 1 2 3 4 5 % % % % % Attending public presentations about the Retezat National Park 2.5 1.2 24.0 29.8 42.6 4.09 .967 Participating in public meetings related to Retez at National Park 2.1 4.1 23.2 32.8 37.8 4.00 .983 Participating in a community project addressing environmental concerns in the area 0.8 1.7 15.4 32.9 49.2 4.28 .844 Learning about the environment 0.4 1.3 6.7 22.5 69.2 4.59 .709 Investing time to learn about the park and environmental protection 0.8 3.3 11.6 27.8 56.4 4.36 .874 Participating in educational programs about the environment 1.3 3.3 10.4 28.3 56.7 4.36 .890 Participating in a worksho p on how to reduce my dependence on park resources 4.2 9.3 19.5 28.8 38.1 3.87 1.149 Participation in a local organization that is involved in park protection 2.5 4.6 17.1 27.1 48.8 4.15 1.024 Investing personal time to get involved with the park 2.5 5.8 24.8 30.2 36.8 3.93 1.034 Reducing use of park resources 6.4 6.8 17.9 26.1 42.7 3.92 1.207 Bringing tourists to the park 1.3 2.5 12.6 21.8 61.9 4.41 .893 Voting for public officials that show interest in environmental issues 3.3 5.8 12.9 22.1 55.8 4.21 1.087 Note: 1 = Not at all effective, 2 = Never effecti ve, 3 = Sometimes effective, 4 = Often effective, 5 = Always effective.

PAGE 59

59 Table 2 10 Descriptive statistics pro environmental civic behavioral intentions Behavioral i ntentions Percent* Mean SD 1 2 3 4 5 % % % % % Participate in a public meeting re lated to Retezat National Park 7.4 7.4 10.7 39.1 35.4 3.88 1.189 Attend a public presentation about Retezat National Park 5.8 5.3 9.9 39.5 39.5 4.02 1.110 Participate in a community project addressing environmental concerns 6.6 6.6 16.9 30.6 39.3 3.89 1. 190 Invest time to learn more about the park and environmental protection 7.0 9.1 12.8 33.1 38.0 3.86 1.221 Give my input into park management decisions 5.0 9.6 11.7 33.9 39.7 3.94 1.163 Be actively involved in an organization that supports park managem ent efforts 9.9 12.4 17.8 31.8 28.1 3.56 1.288 Participate in a worksho p on how to reduce my dependence on park resources 17.0 14.5 18.3 28.6 21.6 3.23 1.389 Express my concerns about park management to elected officials 9.2 4.2 14.2 36.3 36.3 3.86 1.218 Visit the park at least twice 5.8 4.1 9.5 24.9 55.6 4.20 1.142 Visit the park at least six times 26.7 8.5 11.9 18.6 34.3 3.25 1.630 Note: 1 = Very unlikely, 2 = Somewhat unlikely, 3 = Neither li kely nor unlikely, 4 = Somewhat likely, 5 = Very likely. Table 2 11 Descriptive statistics perceived environmental responsibility Environmental r esponsibility Percent* Mean SD 1 2 3 4 5 % % % % % Personally, I have no responsibility for protecting the environment in Retezat National Park 37.3 18.4 11 .0 8.1 25.0 2.65 1.627 Every citizen in my community must take responsibility for protecting the environment in Retezat National Park 3.0 1.3 6.4 19.9 69.5 4.52 .901 Authorities, rather than the citizens, are responsible for protecting the environment in Retezat National Park 3.4 7.1 5.0 24.8 59.7 4.30 1.072 Authorities, together with the citizens, are responsible for protecting the environment in Retezat National Park 2.1 2.1 2.5 13.0 80.3 4.67 .807 *Note: 1 = Strongly disagree, 2 = Disagree, 3 = Neutr al, 4 = Agree, 5 = Strongly agree.

PAGE 60

60 Table 2 12 Goodness of fit indices for each construct CFA models 2 / df RMSEA WRMR CFI TLI Community attachment Original 198.57/54 0.102 1.162 0.944 0.972 Respecified 77.40/27 0.085 0.842 0.988 0.992 Social inte raction Original 47.604/22 0.068 0.680 0.962 0.962 Respecified 5.890/3 0.062 0.244 0.994 0.987 Connection to nature Original 96.73/36 0.083 0.883 0.983 0.988 Respecified 30.09/10 0.091 0.600 0.991 0.994 Conservation attitudes Original 257.39/61 0.115 1.355 0.786 0.902 Respecified 60.16/27 0.071 0.780 0.964 0.982 Attitudes towards pro environmental engagement Original 171.71/25 0.156 1.219 0.911 0.965 Respecified 7.105/5 0.042 0.289 0.997 0.998 Pro environmental behavioral in tentions Original 131.93/12 0.203 1.224 0.933 0.966 Respecified 6.045/5 0.029 0.279 0.999 0.999 Note. 2 = chi square test statistic; df = degree of freedom; RMSEA = root mean square error of approximation; WRMR = weighted mean score residual; CFI = comparative fit index; TLI = Tucker Lewis Index.

PAGE 61

61 Table 2 13 Reliability and validity of the community attachment CFA model Factors and items Mean SD CR AVE Social attachment .773 .876 .544 I feel like I belong in this community 4.50 .842 .74 The associations that I have with other people in this community mean a lot to me 4.52 .762 .68 Given the opportunity, I would move out of this community 3.82 1.520 .67 I feel loyal to the people in this community 4.17 1.034 .78 I plan to remain a resident of this community for a number of years 4.42 1.004 .87 I like to think of myself as similar to the people who live in this community 3.97 1.249 .66 Park attachment .903 .949 .729 Retezat National Park means a lot to me 4 .54 .784 .91 I am very attached to Retezat National Park 4.18 .992 .93 Retezat National Park is very important to me 4.33 .910 .95 I identify strongly with the Retezat National Park 3.78 1.178 .84 I get many personal benefits out of liv ing near Retezat National Park 3.72 1.280 .65 I enjoy living near Retezat National Park 4.54 .750 .88 I get lots of satisfaction out of living near Retezat National Park 4.17 .996 .80 Note. = t statistic (> 1.96) at a significance level of p Table 2 14 Reliability and validity of the social interaction CFA model Factors and items Mean SD CR AVE Interactions wi th friends .750 .786 0.647 Acquaintances 4.85 1.606 0.772* Close Friends 5.40 1.573 0.836* Interactions with public officials 3.54 1.790 set@1.00 N/A N/A N/A Interactions with park .682 .773 .602 Retezat National Park Staff 2.71 1.821 0.8 06* Tourists 3.09 1.816 0.745* Note. = t statistic (> 1.96) at a significance level of p

PAGE 62

62 Table 2 15 Reliability and validity of the connection to nature CFA model Factors and items Mean SD CR AVE Connection to na ture .871 .932 .665 I am connected to nature much like I'm connected to my family 4.40 .974 .88 Nature is a huge part of who I am. 4.40 .932 .90 I often feel a sense of oneness with the natural world around me. 4.28 1.028 .93 My feelings for nature have influenced my spiritual beliefs 4.12 1.125 .90 When surrounded by nature, I feel at peace 4.91 .327 .68 Listening to the wind go through the trees calms my mind 4.35 1.051 .74 When I'm alone in a natural area, I have this fe eling of complete calm 4.51 .841 .64 Note. = t statistic (> 1.96) at a significance level of p

PAGE 63

63 Table 2 1 6 Reliability and v alidity of the conservation attitudes CFA model Factors and items Mean SD CR AVE Conservation awareness .741 0.921 0.704 It is important to have the Retezat National Park for the survival of various plants and animal species 4.91 .415 0.87 I t is necessary to set aside some land for the protection of plants and animals 4.89 .381 0.89 Retezat National Park is our country's pride 4.87 .481 0.84 Retezat National Park being protected is important for the benefit of our future generations 4.87 .418 0.93 The illegal cutting of trees in the park should be discouraged/ strictly regulated 4.69 .813 0.63 Conservation management .834 0.881 0.651 Retezat National Park is managed successfully for the benefit/enjoyment of future gener ations 4.16 1.123 0.84 Retezat National Park is managed successfully for a wide range of uses and values, not just tourism 3.96 1.220 0.79 Retezat National Park management does a good job at protecting the natural resources in the park 4.19 1.066 0.90 The citizens from the communities around the park have enough say in how the park is managed 3.16 1.304 0.68 Conservation benefits .767 0.859 0.606 My community benefits from being near the Retezat National Park 3.95 1.222 0.85 Hav ing the Retezat National Park near my home benefits me and my family 3.97 1.151 0.84 My community is a more beautiful place to live because we are living near Retezat National Park 4.47 .787 0.76 The tourists who come to the area are useful to we who live in adjacent communitie s 3.82 1.227 0.66 Note. = t statistic (> 1.96) at a significance level of p <0.05;

PAGE 64

64 Table 2 17 Reliability and validity of the attitudes towards p ro environmental civic engagement CFA model Factors and items Mean SD CR AVE Attitudes pro environmental civic engagement .823 0.880 0.596 Participating in public meetings related to Retezat National Park 4.00 .983 0.82 Participating in a community project addressing environmental concerns in the area 4.28 .844 0. 73 Investing time to learn about the park and environmental protection 4.36 .874 0.81 Participating in a worksho p on how to reduce my dependence on park resources 3.87 1.149 0.71 Investing personal time to get involved with the park 3.93 1.0 34 0.78 Note. = t statistic (> 1.96) at a significance level of p Table 2 18 Reliability and validity of the pro environme ntal civic behavioral intentions CFA model Factors and items Mean SD CR AVE Pro environmental behavioral intentions .844 0.893 0.626 Attend a public presentation about Retezat National Park 4.02 1.110 0.85 Participate in a community project addressing environmental concerns 3.89 1.190 0.84 Invest time to learn more about the park and environmental protection 3.86 1.221 0.86 Give my input into park management decisions 3.94 1.163 0.71 Participate in a workshop on how to reduce m y dependence on park resources 3.23 1.389 0.68 Note. = t statistic (> 1.96) at a significance level of p

PAGE 65

65 CHAPTER 3 PRO ENVIRONMENTAL CIVIC ENGAGEMENT IN RETEZA T NATIONAL PARK, ROMANIA Introduction A synthesis of the wider literature o n parks and protected areas management reveals a persistent argument that successful management endeavors and sustainability depend on the cooperation and support o f local communities ( Augustyn 1998; Brandon & Wells 1992; de Beer & Marais 2005 ; Hall 20 04 ). Historically, parks and protected areas have been viewed as islands of biodiversity conservation with little or no connections to the human dimensions (Cutumisu 2003). Currently, local communities adjacent to national parks and protected areas are pe rceived as having a major role in achieving conservation and sustainability ideals, due to their permanent interactions and attachments to surrounding environments (Manfredo et al. 2004). The community development literature further underscores the intert wined relationship between community well being and ecological well being (Wilkinson 1991). Linkages between the two dimensions have been hypothesized focusing on explaining ways in which natural resource conditions may contribute or detract from individu al and collective well being (Manfredo et al. 2004) or how community development reduces the probability of occurrence of actions that degrade ecological well being (Wilkinson 1991). On a similar note, Agrawal (2001) emphasized the importance of giving c onsideration to social and economic environments in the management process, to assure equity, as well as effective management. Actively involving local communities in the management of protected areas has been repeatedly associated with increased awareness in terms of the benefits of biodiversity conservation, a more responsible use of resources, and ultimately enhanced livelihoods and welfare of local peoples (Pagdee et al. 2006). However, efforts to involve local communities are often challenged either b y the urgency

PAGE 66

66 of implementing biodiversity conservation projects or the lack of community motivation to be involved or even a lack of knowledge on how and to what extent to involve local people. Often communities adjacent to protected areas are perceived a s being passive, lacking initiative and environmental sensitivity, and have beliefs that established authorities will take responsib ility for resource management (de Beer & Marais 2005). This assertion is regularly made for local communities adjacent to R omani an protected areas (Kuijs & Bergh 2006; van Hal 2006), but little is empirically known about the Romanian context, and especially about the institutional and socio p rotected areas. Co management of protected areas and participatory decision making initiatives has been encouraged over the years in Romania but only modest success has been achieved in terms of management effectiveness. Excessive exploitation of natural resources is still a real threat to biodiversity conservation in Romania. The lack of community interest and participation in biodiversity conservation has been discussed as being a major constraint for natural resources management in Romania. This situati on has been primarily attributed to a low sense of community and collective responsibility that characterize Romanian rural communities (PJB Associates, 2006 ) and also to the legacy of the communist system, primarily based on a centralized political struct ure, which las ted until 1989 (Oostenbrink & Kosterink 2005). However, there is only scarce empirical evidence, much of it anecdotal that supports these assertions. Taking into consideration that Romania is still at an incipient stage in the process of sh aping its nature conservation approaches, there is an emerging need to better understand the major factors that shape community conservation attitudes and pro environmental civic

PAGE 67

67 engagements. Thus, in view of the increasingly recognized importance of local engagement in protected area management and the realities of the Romanian context, this study examined the extent to which community social dimensions could explain attitudes towards pr o environmental civic engagement and pro environmental civic behaviora l intentions. Taking into account that the linkage between the instructional and social dimensions of community in Romania is still view ed as fairly weak (Cottrell & Cutumisu 2006), this study primarily f ocuses on the social dimensions of pro environmenta l civic engagement. The study addressed the following question: How levels and types of community attachment, conservation attitudes, connections to nature and perceived collective environmental responsibility facilitate or hinder attitudes towards pr o env i ronmental civic engagement and pro environmental civic behavioral intentions? Conceptual Framework Different theoretical socio psychological models were proposed to account for relevant behavioral patterns. Among others, the theory of reasoned action (TR A), the theory of planned behavior (TPB) and the theory of interpersonal behavior (TIB) are all well established social psychological theories belonging to the school of cognition ( Barr & Gilg 2007) Many similarities exist between these models, the main differences pertaining to the predictors used in explaining social behaviors. Fishbein and result of intention to engage in a specific behavior. Additionally, they argued that be havioral intentions are a result of the attitude a person has towards the behavior (behavior belief), perceived social pressure to engage in a particular behavior (normative belief) and strength of the desire to comply with the norm. The TPB adds one more variable to the model, perceived perform the behavior) (Ajzen 1991).

PAGE 68

68 determ inants similar to those found in the TPB. Gagnon et al. (2003) emphasized the broader purpose of TIB as compared to TPB by considering that this model includes determinants that are not accounted for in the other behavior models. The factors that are uniqu e to TIB are role beliefs, habit and affect (Bamberg & Schmidt 2002). Based on the TIB model, behavior is determined by three dimensions: intention, facilitating conditions and habit. In TIB, intention is determined by five constructs: affect, perceived c onsequences, perceived social norms, perceived normative belief and self identity. These concepts have been defined by Triandis (1980) and used in conducting research (Gagnon et al. 2003; Zhang, In bakaran, & Jackson 2006) as follows: affect was described as an emotional factor associated with the behavior; perceived consequences refer to the evaluation of the possible consequences of the behavior. The perceived social norms were defined as normative and role beliefs. Normative beliefs include the opinions individual. Role beliefs reflect the extent to which an individual thinks someone of a similar social position should or should not behave. The other normative compon ent of the TIB is the personal normative belief that represents the feeling of personal responsibility regarding the performance or not of a given behavior. Finally, self identity refers to the degree of congruence mself or herself and the characteristics he or she associates with the realization of the behavior. Comparative studies between the three socio psychological models showed that the TIB had increased predictive powers, and ultimately explained more intentio nal variance due to the unique elements included in the model, habit, affect and role beliefs (Bamberg & Schmidt, 2002). In the case of this study, the actual behavior was not measured, the focus being on

PAGE 69

69 assessing various factors and their influence on be havioral intentions. Thus, measures of habit as well as facilitating conditions were not included. As previously mentioned, the TIB also includes an emotional component of attitude that influences behavioral intentions, a component that has been less studi ed in the context of environmental behaviors. Based on the tenets of the theory of interpersonal behavior and sociological and ecological literature, a research model examining determinants of attitudes towards pro environmental civic engagement and pro e nvironmental civic behavioral intentions was proposed and tested (Figure 3 1) The factors included in the model were chosen based on the literature as well as their relevance for the study site. This study hypothesized based on theory that at different le vels of attitude (behavior belief), affect, and role beliefs, dissimilar levels of behavior intentions will be observed. Thus, the constructs explored were attitudes towards behavior and its predicting factors, conservation attitudes, role beliefs, affect and ultimately their relationships with behavioral intentions. Consequently, research hypotheses were developed as follows: H1: Attitudes towards pro environmental civic engagement positively impacts pro environmental civic behavioral intentions. H2: Perce ived collective environmental responsibility positively impacts pro environmental civic behavioral intentions. H3: Community attachment positively impacts pro environmental civic behavioral intentions. H4: Community attachment positively impacts conservati on attitudes. H5: Conservation attitudes positively impacts attitudes towards pro environmental civic engagement. H6: Connection to nature positively impacts conservation attitudes.

PAGE 70

70 Methodology Description of the Study Area Retezat National Park (RNP) was designated in 1935, being the first national park established in Romania. In 1979, RNP was declared an International Biosphere Reserve under the UNESCO Ma n and Biosphere program a nd in 2004 RNP received its Protected Area Network (PAN) Certification Retez at National Park is located in the southwestern Carpathians, Hunedoara County, and the total surface of the park is 38,138 ha (RNP Management Plan 2008). Within the park, there are more than twenty mountain peaks 2,000 meters or hig her in addition to eig hty lakes of glacial origin. There are more than 1,100 species of plants, over 50 species of mammals including roe deer, chamois, lynx, bear, and otter and 168 recorded bird species including the golden eagle In 1999, the first models for protected area management were established in Romania, the management process (Stanciu 2002). RNP was the first park in Romania with a management system in place (van Hal 2006) and the management framework initiated by RNP is perceived as being a model for other protected areas in Romania. The RNP co management framework includes, in addition to the park administration, a Scientific Council and a Consultative Council. The S cientific Council consists of scientists that represent the Romanian Academy, while the Consultative Council is represented by key stakeholders of the RNP area (local communities, tourism operators, local businesses, etc.) The Consultative Council provides recommendations regarding park management activities but does not have decision making ability (van Hal 2006). A large portion of the park area (17,564 ha, 46%) is owned by the state, while local associations own the remainder of the land (20,574 ha). Tw enty six villages have grazing rights to alpine meadows, and their rights are administered either through community based

PAGE 71

71 associations or the local councils of the five communes to which the vil lages belong (Kuijs & Bergh 2006). A commune is an administra tive division in Romania encompassing one or more villages that share similar economic, socio cultural, geographic and demographic conditions. Of the five, three communes are primarily important from a management perspective, due to their close proximity t o the park and their land ownership and use of resources in the park (use of buffer zone resources): Rau de Mori, Salasu de Sus, and Campu lui Neag (van Hal 2006). The total population of these three communes was estimated at 6,837 inhabitants (Kuijs & Be rgh, 2006). Communities rely on park resources primarily for grazing and the use of other natural resources such as wood, non timber forest products, mushrooms, and medicinal plants. The major management concerns, as it relates to conservation, are relate d to overgrazing of the pasture areas and illegal wood h arvesting (RNP Management Plan 2008). Previous research conducted in the area discussed a variety of challenges for park management and surrounding communities. van Hal (2006) emphasized the growing concerns of local people due to the increased restrictions imposed on grazing by the park administration for conservation purposes. These restrictions were strongly viewed as having great impact on the landowners control over their private lands. van Hal ( 2006) observed a lack of a common interest in conservation in the area, as a result of rudimentary local cooperation in ma nagement. Furthermore, Kuijs & Bergh (2006) using a qualitative approach in understanding the park management efforts, emphasized the lack of a conservation attitude in the communities surrounding Retezat National Part and a lack of care for the environment. These two studies depicted primarily the voices of the park administrative staff and park stakeholders (e.g. lodge owners, travel o perators, mayors of local communities) but not directly the views of local residents.

PAGE 72

72 The Retezat National Park employs a co management framework for the management and planning of the park's natural resources. The effectiveness of the management practice is still questioned, as the local communities' interest and participation in conservation is perceived as being limited. Therefore, it is essential to better understand the major factors that shape community residents' conservation attitudes and behavior i ntentions. Thus, this study examined the attitudes that people living adjacent to Retezat National Park have towards conservation, and how these attitudes are affected by community attachment and nature connections and further shape attitudes towards pr o e nvironmental civic engagement and pro environmental civic behavioral intentions. Measures of Model Components Pro environmental civic behavioral intentions Environmental behavior has been studied by social scientists from different disciplines, including p sychology, education, communication, and env ironmental studies. Kollmuss & Agyeman (2002) defined pro environmental behavior as behavior deliberately aimed at Gagnon et al. (2003) desc ribed behavioral intention regarding the performance of a given behavior. The theoretical socio psychological models that come to account for behavior, postulate that our behaviors and intentions to behave in a certain way ar e strongly correlated. Furthermore, attitudes towards the behavior are perceived to have a strong influence on behavioral intentions. Attitudes were primarily defined as an overall positive or negative evaluation of performing the behavior (Fielding et al. 2008). In this study, the actual performance of the behavior was not measured, the focus was on understanding the predictors of behavioral intentions.

PAGE 73

73 Pro environmental behaviors can take different forms, and accordingly, different factors can different ly shape or influence each specific type of behavior. The literature makes a distinction between political behaviors, as compared with behaviors that relate with consumption (e.g. energy saving, green consumer behavior) (Aoyagi Usui, Vinken & Kuribayashi 2003). Stern (2000) provides a more detailed classification of pro environmental behaviors, identifying several different types of pro environmental behaviors such as: environmental activism, environmental citizenship behavior (active involvement in envir onmental issues, public participation that with an influence on policy making, decision making) and private sphere environmentalism (e.g. consumer behavior, automobile, energy use, green consumerism etc.). B ental behaviors, the environmental behaviors of interest in this study, relate to what the author describes as environmental citizenship behaviors. Primarily, the focus is on behaviors that relates to public participation and active involvement in local en vironmental concerns. Chavis & Wandersman (1990) argued that three important components influence an individual's civic participation, specifically, their perception of the environment, one's social relations, and one's perceived control and empowerment wi thin the community. The author viewed sense of place as a factor that mobilizes these factors. According to the social participation literature, the most important factors that shape involvement have been social ties and networks (Granoveter 1973; Luloff 1990), and social interaction with community members ( Luloff & Swanson 1995 ; Wilkinson 1991 ), which were found to ultimately define co mmunity attachment (Kasarda & Ja nowitz, 1974). In the context of this study, two variables were tested for their direc t influence on pro environmental civic behavioral intentions: community attachment, as an emotional investment in place and its people,

PAGE 74

74 and perceived collective environmental responsibility. Furthermore, a measure of attitudes towards pro environmental civ ic engagement was suggested to mediate the relationship between conservation attitudes and pro environmental civic behavioral intentions. Attitudes towards pro environmental civic engagements were measured using twelve items adapted from previous work don e by Halpenny (2006) and Garling et al. (2003). The items were measured on a five point Likert scale ranging from 1 to 5, where 1 = Not at all effective; 3 = Sometimes effective; 5 = Always effective. Pro environmental behavioral civic intentions were meas ured using eight items adapted from previous studies conducted by Gar ling et al. (2003); Theodori & Luloff (2002); and Halpenny (2006). The items were measured on a five point Likert scale ranging from 1 to 5, where 1 = Very unlikely; 3 = Neither likely no r unlikely; 5 = Very likely. Community attachment Community attachment has been examined in a variety of disciplines, primarily focusing on understanding the major determinants of community attachment as well as the implications of community attachment to social and ecological well being. Hummon (1990) defined community attachment as an emotional connection to a place that emerges in the context of residence and belonging. Some of the major determinants of community attachment discussed in the literature ar e length of residence, participation in community activities and groups, and local ties and networks ( Beggs, Hurlbert & Haines 1996; Brehm 2007 ; Brehm, Eisenhauer & Krannich 2004; Kasarda & Janowitz 1974; Theodori & Luloff 2002 ). T he collective act ion and engagement literature views attachment to a community as an important predictor of action and engagement that ultimately translates into well being at the individual and community levels ( Brennan 2007 ; Theodori 2000; Wilkinson 1991). The relatio nship between community attachment and pro environmental civic behavior has received

PAGE 75

75 reduced empirical testing. Theodori & Luloff (2002) emphasized that the focus of earlier studies was primarily on understanding behavior as a consequence of differences in socio demographic characteristics and attitudes, l ess being known in terms of the sentimental and emotional attachments (based on social ties, as well as ties to the natural environment). Community attachment has been commonly attributed to social relati ons and their contribution to the emotional bond people have with their communities. Thus, it has been primarily operationalized as a unidimensional construct, capturing emotional responses concerning th e social environment (Gursoy & Rutherford 2004). Hum mon (1990) argued that community attachment appears to be most strongly rooted in involvement in local social relations, but he also acknowledged that the built and natural environment might also contribute to such emotional ties if perceived in favorable terms. Cross (2003) also emphasized the need to operationalize community attachment as a multidimensional construct, the unidimentionality approach limiting the depth of information captured and our ability to distinguish how different dimensions of attach ment might shape community behavior and action differently. Academic investigation has been conducted looking at the relationship between social attachment and attachment to the natural environment and collective action, open communication, attitudes towa rd relevant environmental concerns and local conflicts over land use ma nagement (Brehm, Eisenhauer, & Krannich 2006; Brehm, Eisenhauer & Krannich 2004). Social attachment was found to have different implications on the independent variables as compared w ith the natural environmental attachment, providing evidence of the relevance of understanding community attachment as a reflection of attachment to the social and natural environment. Brehm's (2007) argued that in communities where natural amenities are a bundant,

PAGE 76

76 attachments to their community, regardless of length of residence. This assertion was supported by a study conducted by McCool & Martin (1994). Thus, a need for studies focused on better understanding specific place attributes (natural environment versus social) to which people are attached and the mechanisms through which such attachments are formed emerged. Extensive literature exists that focuses on place based attachments, sense of place and place attachment (Altman & Low 1992; Kyle et al. 2004; Williams et al. 1992; Williams & Vaske 2003), concepts that can be directly linked to the idea of attachment to the natural environment in the context of commu nity attachment. Not surprisingly, Mesch & Manor (1998) views place attachment as a subjective evaluation of the features of the physical, as well as the social environment that has behavioral implications. Positive relationships have been found between pl ace attachment and specific en vironmental behaviors ( Bott, Cantrill & Myers 2003 ; Vaske & Kobrin 2001 ). These studies highlight the importance of better understanding the relationships that exist between place based attachments and community attachment and ultimately their role in shaping attitudes and behavioral intentions. Thus, this study employed an assessment of community attachment based on attachment to the social environment and attachment to the natural environment (the protected area neighborin g the local communities participating in the study). The social dimension was measured using ten items on a five point Likert scale ranging from 1 to 5, where 1 = Strongly disagree; 3 = Neutral; 5 = Strongly agree. The item s were adapted from Theodori & Ma yfield (2008). Attachment to the park was assessed using the conceptualization employed in the place attachment literature, more specifically the two dimensions of attachment: place dependence and place identity ( Kyle et al. 2004 ; Williams & Roggenbuck 1 989; Williams et al. 1992 ). Park was measured using nine items on a five point Likert scale ranging from 1 to 5, where 1 =

PAGE 77

77 Strongly disagree; 3 = Neutral; 5 = Strongly agree. Consequently, it is expected tha t community attachment has a hierarchical structure, encompassing attachment to the social environment and attachment to the park. Connection to nature Connection to nature is often portrayed as a sense of oneness with nature. The connection people have wi th nature has been discussed in the literature as an important factor in explaining environmentally responsible behavior (Schultz et al. 2004). The feelings people have for nature shape personal identity, values and attitudes, and ultimatel y influence beh avior (Driver & Ajzen 1996; Mannell 1996; Roggenbuck & Driver 2000). Connection to Nature (CTN) primarily encompasses biospheric values, which have been found as building a strong foundation for developing value based concerns and increased motivation for environmentally responsible behavior (Schultz et al. 2004). The main assumption connected to nature, they intrinsically care for it and act to protect and preserve it (Pennisi 2007). The complexity of human connection to nature derives from its multiple underpinnings, which relate to self concept and identity, as well as cognitive, affective and behavioral components. Pennisi (2007) depicted values and ide ntity as the two core aspects of connection to nature. The interconnectivity between personal identity, values, attitudes, and ultimately behaviors has been previously accounted for in the literature (Clayton 2003; Clayton & Opotow 2003; Hitlin 2003; Va ske & Donnelly 1999). Connection to nature was found as a strong predictor or attitudes, which ultimately shape behaviors. Therefore, connection to nature as a reflection of personal identity and values represents an important dimension for explaining pro environmental attitudes and behaviors. To better understand the relationships people living

PAGE 78

78 nature was assessed in this study. Furthermore, the impact of conn ection to nature on conservation attitudes was examined. Several scales measuring connection to nature have been developed. These scales include the Connectedness to Nature Sca le (CNS) (Mayer & Frantz, 2004 ), single item Venn diagram measure of Inclusiven ess of Self (Schultz 2002), a measure of environmental identity (Clayton 2003), and a computer dependent Implicit Association test (Schultz et al. 2004). The first three measures assume the construct is unidimensional, a belief that has been contradicte d by recent work done by Pennisi (2007). Thus, the scale developed by Pennisi (2007) was employed. The items included captured five dimensions of connection to nature: admiration (3 items), spirituality (3 items), identity (4 items), sorrow (2 items), rest oration (3 items), and fear (3 items). The eighteen items were measured on a five point Likert scale ranging from 1 to 5, where 1 = Strongly disagree; 3 = Neutral; 5 = Strongly agree. Therefore, it is expected that connection to nature has a hierarchical s tructure capturing the five dimensions of connection to nature previously proposed in the literature. Conservation attitudes Jennings & Nickerson (2006) defined attitudes as an enduring predisposition toward edisposition is translated in the way p eople think, feel, and behave. The importance of understanding attitudes evolves from the r elationship between attitudes and behavior al intentions (Ajzen 1991). Researchers frequently examined the attitudes towards c literature on attitudes towards conservation, and protected area s is fairly notable. The focus of inquiry was primarily on understanding the main predictors that shape attitudes and to a lesser

PAGE 79

79 extent investigations were conducted on the consequences of having positive or negative attitudes towards either the protected area, or conservation. The perceived benefits achieved from conservation are often acknowledged as strong predi 2005). However, study finding s not always found t he relationship between benefits and support for conservation noteworthy, other fact ors being suggested as having higher predictive power in t erms of con servation attitudes (Walpole & Goodwin, 2001) interactions with the authorities of a protected area could play an important role in shaping co nservation attitudes (Kappelle 2001; Robertson & Lawes 2005). As previously mentioned predominantly directed towards understanding antecedents of conservation attitudes rather than the consequences of having positive or negative attitudes. Dolisca et al. (2009) found forest conservation behaviors to be directly influenced by conservation attitudes. However, Infield & Namara ( 2001) conducted an assessment of a community conservation program in Uganda, in an effort to evaluate the impacts of the program on community attitudes and observed an effect of the program on conservation attitudes but behaviors were not greatly changed based on program participation. In the case of this study, attitudes seemed to be vulnerable to poor behavior s on the part of the park staff and law enforcement actions. Thus, the relationship between conservation attitudes and pro environmental behaviors requires further investigation considering that positive conservation attitudes might not always lead to anticipated behaviors. T o better account for changes in pro environmental civic behavioral intentions, this study employ ed a measure of conservation attitudes, but also a measure of attitudes towards pro environmental civic

PAGE 80

80 engagement s. T h o se attitudes that strictly relate to the expected behavior as suggested by theory, have a higher predictive power in terms of behavioral intentions. Different measures have been employed in the literature to assess community attitudes towards conservation, primarily m ultidimensional scales bein g used Based on previous studies, r espondents' attitudes towards co nservation were measured on the bases o f their reactions to 25 items, capturing four dimensions of conservation attitudes: conservation awareness (6 items), conservation benefits (7 items) land use perspectives (7 items) and man agement considerations (5 items) ( Infield & Namara, 2001; McFarlane & Boxall 2003 ; Nguyen, 2007 ). These items were measured on a five point Likert scale ranging from 1 to 5, where 1 = Strongly disagree; 3 = Neutra l; 5 = Strongly agree. Therefore, it is hypothesized that conservation attitudes have a hierarchical structure capturing conservation awareness, conservation benefits, land use perspectives and conservation management. Perceived collective environmental re sponsibility Perceived environmental responsibility has been identified in several behavioral models as a significant determinant of behavioral intentions ( Ajzen 1991; Hines et al. 1987; Stern & Oskamp 1987; Triandis 1980). Fransson & Garling (1999) de fined environmental responsibility general, or specific environmental concerns. A strong relationship has constantly been found between environmental respon sibility and behaviors that have implications for the well being o f the environment ( Garling et al. 2003 ; Hines et al. 1987; Van Liere & Dunlap 1978; Vining & Ebreo 1992 ). The social context has been viewed as shaping ascribed collective responsibility for pro environmental engagement. Garling et al. (2003) identified pro environmental behavioral intentions as being a reflection of personal norms, ascribed responsibility, and awareness of consequences for oneself, others, and

PAGE 81

81 for the biosphere. Therefor e, perceived environmental responsibility has been primarily suggested as being a strong predictor for behaviors that have collective implications (e.g. participation in public meetings), as compared to behaviors that relate to self interest (e.g. reducing w ater use). Furthermore, Barr & Gilg (2007) found the social context and alternative perceptions of trust and responsibilities within localities as having a mediating role in shaping public understandings of sustainability and environmental issues. In thi s study, perceived environmental responsibility is viewed as an important predictor of pro environmental behavioral intentions, considering the Romanian social context that has been shaped by historical events that primarily emphasized public/governmental responsibility as compared to collective responsibility. In this study, perceived collective environmental responsibility was examined on the basis responsibility. The questions were previously employed by Garling et al. (2003) and were assessed using a five point Likert scale ranging from 1 to 5, where 1 = Totally disagree; 3 = Neither agree nor disagree; 5 = Totally agree. Data Collection R ural communities adjacent t o Retezat National Park belong to five communes encompassing 43 villages with a total population estimated at 14,009 adult residents. Two villages dissimilar in size from each commune, to assure representation of each commune in the final sample were selec ted. Nine communities (one commune had only one village) were selected for this study using multistage random sampling. The sampling frame included the total number of households in the nine communities adjacent to RNP, but the final unit of analysis was t he individuals residing in the households. The nine villages selected have a population of

PAGE 82

82 4,232 persons residing in 1,159 private households. A sampling frame based on households as opposed to individuals was found to be more suitable due to the complexit y of the survey. To assess face and content validity, multiple strategies were followed. At first, three university professors reviewed the survey instrument and also provided feedback in terms of the extent the empirical measures adequately reflect the r eal meanings of the concepts under consideration. Secondly, a Romanian student translated the Romanian language version of the questionnaire back to English, and after revisions were made, further verified the accuracy of the translation. Lastly, at the st udy site, park management staff was asked to complete the survey and comment on the questionnaire content, design, clarity, wording and format. Based on the feedback, adjustments were made to the questionnaire. Cross sectional data was collected from 260 residents during June 2009 October 2009 using face to face interviews (68% response rate) and mail survey (9% response rate). A systematic sampling method with a random start was used to select participants for the face to face interviews, the sampling i nterval being established based on the number of households in the sampling frame for each community divided by sample size needed in each community. Considering the small size of these communities and the challenge of finding people at home, in general ev ery other household was selected. The person in the household age 18 or older was asked to participate in the study, and the questions were asked directly to the respondents and recorded by the interviewer. In addition, a mail survey was sent to households where the residents were not at home, even after multiple visits at different times during the day. A total of 230 surveys were sent by mail to the population of four communities using home address information from the phone book. The low response to the mail method revealed the lack of feasibility of this method for the residents living in these communities.

PAGE 83

83 Participants in this study were asked to express their opinions on a series of questions about their community as well as Retezat National Park. The questionnaire measured the following constructs: community attachment, connection to nature, conservation attitudes, perceived collective environmental responsibility, attitudes towards pro environmental civic engagement, and pro environmental civic behavi oral intentions. In addition to these constructs, several socio demographic characteristic variables (age, gender, education, occupation, residency, length or residency, income) were included. Data Analysis Data analys e s w ere performed following three stag es. First, descriptive statistics were computed for the variables used in the study using the Statistical Package for the Social Sciences (SPSS) version 18.0. Second, the data collected was screened and the critical assumptions underlying the statistical t echniques employed by the study were assessed. Third, a two step data analysis was employed to assess the hypothesized relationships among the research construct s (Anderson & Gerbing 1988). A s part of this process, individual items were examined using Con firmatory Factor Analysis (CFA) and the measurement model for constructs included in the study was estimated using MPLUS version 5.21 to determine how well the indicators captured their specific constructs and the ability of the respondents to differentiat e between constructs (Hair et al. 2006). This was followed by an assessment of the S tructural Equation M odel (SEM) assessing the hypothesized relationships between constructs. SEM was assessed using MPLUS version 5.21 using the WLSMV (weighted least squar es mean and variance adjusted) method of estimation, method recommended for categorical ordinal data (Muthen et al. 1997). Finally, the direct and indirect effects as well as the moderator effects be tween constructs were assessed. The fit of the measurem ent model and the structural equation model were assessed using multiple criteria. T he chi square test of model fit divided by the degrees of freedom was used as

PAGE 84

84 a reference criteria supplemented by the Root Mean Square Error of Approximation (RMSEA), Wei g hted Root Mean Square Residual (WRMR) Comparative Fit Index (CFI) and Tucker Lewis Index ( TLI ) RMSEA values equal to or less than .06 is indicative of a good model fit values between 0.08 and 0.10 indicate acceptable model fit, and values higher than 0. 1 0 are considered to be in dicative of poor fit (Browne & Cudeck 199 2; MacCallum, Browne & Sugawara 1996). CFI and TLI values equal to or greater than 0.95 also indicate good model fit (Hu & Bentler 1999) The criterion for WRMR is a value less than 1.0 0 (Yu, 2002 ). After adjustments (specified below), the models utilized in reporting the findings met the minimum standards listed here. Results The residents sampled resided in nine communities adjacent to Retezat National Park. The average age was 45.0 ye ars, with almost one quarter of the respondents (23.7%) being between 18 and 30 years and slightly more than one third of the respondents (36.1%) being between 31 and 50 years. Respondents who were above 51 years were the most represented group (40.2%). Fi fty four percent were males and 46% percent were females. The average length of residency was 37 years. The majority of the respondents (65.1%) were married or partnered, and on average, the number of adults per household was three. Almost one third of the respondents (31.6%) indicated high school as the highest level of education attained, 14% of the responden ts had some college or a college degree and 14 % had an elementary school education or less. About one third of respondents (33.8%) reported a monthly household income between 1,000 and 1,999 RON (about US $330 $660) and 30.6% indicated a monthly household income of more than 2,000 RON (about US $661). The majority of respondents (70.6%) indicated that they do not have any property rights (ownership or land use rights) in Retezat National Park.

PAGE 85

85 Individual constructs and their measurements were examined using c onfirmatory f actor a nalysis (CFA). Marsh, Craven & Debus (1991) underscored that w hen a model has been misspecified (poor model fit) the researc her has to respecify the model. One way to respecify the model is to delete indicators and the other option is to all ow errors to correlate, and decisions should be supported by theory or rationale (Joreskog 1993). In this study, the measurement model for each construct was examined and modifications were made based on residuals assessment and modification indices. As a result of initial CFA tests, several items in various factors were dropped due to their low factor loadings. Ultimately, for each construc t those items where retained that were substantive in size and had significant factor loadings. In this stage, the proposition that connection to nature has a hierarchical structure capturing five dimensions was not supported, inter factor correlations be ing high and suggesting inability of the sample to differentiate between factors. Therefore, scale items were collapsed and ultimately seven items were retained as a one dimensional measure of connection to nature. This finding questions the multidimension ality of the connection to nature construct. Similarly, the correlation between p ark identity and park dependence was high suggesting a weaker differentiation of these constructs by this population. Consequently, scale items were collapsed and one measure of park attachment was retained including seven items. Furthermore, for the conservation attitude scale, one of the four factors (land use perspectives) did not load on the higher order model and was eliminated from further analysis. After assessing the me asurement model for each construct and a good fit to the data was observed, t he fit indices for the total measurement model were examined. The measurement model including nine factors (social atta chment, park attachment, connection to nature conservation awareness, conservation benefits, conservation management, attitudes pro

PAGE 86

86 environmental civic engagement, pro environmental civic behavioral intentions, and perceived collective environmental resp onsibility) was tested. The fit indices for a total measureme nt model with nine factors revealed good fit ( T able 3 1 ) 2 = 167.159, df = 92, p < .001) was lower than the suggested criteria (i.e., <3.0; Kline 2005). CFI (.97), TLI (.98), RMSEA (.056) and WRMR (.898) yielded a good model fit and all the item (indicator) loadings were significant (p < .001) and ranged from .61 to .94 that provides strong evidence of convergent validity ( T able 3 2 ) recomm ended lev el of .70 (Nunnally & Bernstein 1994), ranging from .71 (perceived collective environmental responsibility) to .90 (park attachment) and composite reliability (CR) above the recom mended level of .70 (Fornell & Larker 1981) ranging from .84 (perceived co llective environmental responsibility) to .95 (park attachment). Also included in T able 3 2 are the average variance extracted (AVE) estimates with recommended levels of .50 or higher indicating (Bagozzi 1994; Fornell & Larker 1981 ). All values exceeded the recommended level ranging from .54 (social attachment) to .73 (park attachment). All the inter correlations among latent factors were less than the suggested threshold of .85 (Kline 2005), ranging from .24 to .71 and being a strong evidence of discriminant validity ( T able 3 3 ). These findings reveal that the proposed measurement model satisfied all the psychometric requirements, thus the measures were adequate for further analysis. A hierarchical model wa s tested with social attachment and park att achment set to load on a second order factor, community attachment, and conservation awareness, conservation benefits and conservation management were also set to load on a conservation attitudes factor. A specif ica tion problem was encountered and t he results revealed high inter correla tions between

PAGE 87

87 the latent constructs, conservation attitudes and community attachment (.91 ). Also each second order factor had high correlations with some of the first order factors that d id not define the second order factor. In fact, community attachment was more highly correlat ed with nature connection ( .82), conser vation awareness ( .79 ), and conservation ben efits ( .73 ) than with social attachment ( .65 ). Also conservation attitudes were more highly correlated with park attachment ( .78 ) than with cons ervation management attitudes ( .63 ). This situation indicated that these factors collectively were a measure of a latent factor, which based on the literature most likely implies a measu re of local environmental identity. Environmental identity has been discussed as reflecting a sense of connection to the social and natural environment, based on history and emotional attachment that ultimately affects the ways in which we perceive and a ct toward the world (Clayton & Opotow 2003). Th us, six factors were set to load collectively on a second ord er factor measuring local environmental identity. The confirmatory factor analysis revealed the hierarchical structure of the independent variable s, a single factor local environmental identity explaining the relationships between the first order factors. Results for the measurement model with local env ironmental identity as a second order factor revealed good fits. The square/ df ratio (2.18: 2 = 189.316, df = 87 p < .001) was lower than the suggested criteria (i.e., <3.0; Kline 2005). CFI (.96), TLI (.98), RMSEA ( .067 ) and WRMR ( 1.054 ) yielded a good model fit and all factor loadings were significant and substantia l in size. The f actor loadings on local environmental identity ranged from .5 8 (conservation management) to .86 (park attachment). The results of final CFA tests yield that all item (indicator) loadings for each factor were significant ( p < .0 0 1) and range d from .61 to .94 providi ng strong evidence of convergent validity.

PAGE 88

88 Based on tenets of the theory of interpersonal behavior that incorporates self identity as a predictor of attitudes, four hypotheses were reformulated (H3; H4; H5; H6) and a structural e quation model was tested with local environmental identity influencing attitudes towards pro environmental civic engagement, and pro environmental civic behavioral intentions being influenced by perceived collective environmental responsibility and attitud es towards pro environmental civic engagement. SEM analysis was performed to examine the overall model as well as individual tests of the hypothesized relationships among the latent constructs. The hypothesized model to the data resulted in adequate fit wi square / df ratio (2.24: 2 = 194.749, df = 87, p < .00 1) being lower than the suggested criteria (i.e., <3.0; Kline 2005). The goodness of fit indices CFI (.96), TLI (0.98), RMSEA ( .069 ) and WRMR ( 1.075 ) revealed acceptable fits All factor loadings were significant and subst antial in size. Support for the hypotheses was examined via the significance of the individual path coefficients ( F igure 3 2) Significant path coefficients were found between local environmental identity and attitudes towards p ro environmental civic engagement (standardized coefficient of .657; p < 0.001 ), and between attitudes towards pro environmental civic engagement and pro environmental civic behavioral intentions (standardized coefficient of .645; p < 0.001). P erceived col lective environmental responsibility did not have a significant impact on pro environmental civic behavioral int entions (standardized coefficient of .112, p > 0.05 ). A decision was made to respecify the model and include a path from perceived collective en vironmental responsibility to attitudes towards pro environmental civic engagement. The results s howed a reasonable improvement in square 2 = 187.532, df = 87, p < .00 1) lower than the suggested criteria. CFI (.96), TLI (.98), RMSEA ( .067 ) and WRMR ( 1.055 ) revealed good model fits Significant path coefficients were found between local environmental identity

PAGE 89

89 and attitudes pro environmental civic engagement (standardized coefficient of .450; p < 0.00 1 ); between perceived collective environmental responsibility and attitudes pro environmental civic engagement (standardized coefficient of .349; p < 0.00 1 ); and between attitudes towards pro environmental civic engagement and pro environmental behav ioral intentions (standardized coefficient of .722; p < 0.00 1 ). Furthermore, local e nvironmental identity was found to indirectly influence pro environmental ci vic behavioral intentions through attitudes towards pro environmental civic engagement (standardized coefficient of 0.325 ; p < 0.001). The significant indirect effect suggests that attitudes towards pro environmental civic engagement mediate the relationsh ip between local environmental identity and pro environmental civic behavioral intentions. When pro environmental civic engagement and local environmental identity were set to impact on pro environmental civic behavioral intentions, the path coefficient fr om engagement to behavioral intentions was significant (standardized coefficient of .660; p < 0.001), while the path coefficient from local environmental identity to behavioral intentions was insignificant (standardized coefficient of .061; p > 0.05). Dis cussion The primary findings of this study are twofold. First, the results underscored the hierarchical structure of the relationship between social and park attachments, connection to nature and conservation attitudes. Local environmental identity was fou nd to account for the relationship between these constructs. Consequently, four hypotheses were reformulated (H3; H4; H5; H6) and a direct impact of local environmental identity on attitudes towards pro environmental civic engagement was proposed. Second, this study highlights the interplay of variables that affect pro environmental civic behavioral intentions. Local environmental identity and perceived collective environmental responsibility were found to have a significant direct

PAGE 90

90 impact on attitudes towar ds pro environmental civic engagement that ultimately shape pro environmental civic behavioral intentions. Clayton & Opotow (2003) argued that strong attachments and connections with nature can contribute to the formation of identities in environmental co ntexts This sentiment supports the paradigm that environmental identities emerge from interactions with the natural world and from socially constructed understandings of oneself and others (Chawla 1999). Identity is a notion directly linked to self conce pt and involves beliefs about who we are and the way we organize information about ourselves. It has been emphasized that we do have multiple ways of organizing information about ourselves, resulting in multiple identities that ultimately vary in their sal ience and importance according to the immediate context and expected behaviors (Clayton & Opotow 2003). Consequently Kihlstrom et al. ( 1988 ) talked about natural environments and environmental issues and how they directly tie to our core identity due to their relevance to our sense of self, emotional arousal, and their connections to other aspects of life that have personal significance. This does not mean a negation of the social aspects of identity, identity emerging in a social context and taking into account social interdependence and cultural aspects (Snyder & Canto 1998). It has been emphasized that for many people, their identity lies in the natural and the social world they are part of; that is, in the relationships they have to the natural and s ocial world. In this study, it was found that attachment to the social environment and the park, connection to nature and attitudes towards conservation are strongly embedded and they define a local environmental identity. This identification of the local population with the surrounding environment might be due to the enduring interaction of the local people with the environment, their way of life being strongly dependent on the natural and social environment. In addition, the

PAGE 91

91 park adjacent to these communi ties has a long history of conservation and protection and its magnificence and relevance for the community seems to be embedded in the cognitive representation of self. The park could be viewed as a local stimuli in the definition of the self in this cont ext. Identity is viewed more salient when individuals of groups undergo geographical, social, and psychological shifts. In the case of this study, the local population is very stable, generation after generation living in these communities and thus an argu ment can be made that the local environmental identity is vivid. This study underscores the importance of attachment and nature connections and conservation attitudes in shaping local environmental identities, and the need to sustain such connections thro ugh social interactions as well as interaction with natural environments. Social interactions play an important role in defining attachment and thus any initiatives that bring people together in order to act together, can greatly contribute to local enviro nmental identity. A weaker contribution to this latent construct, local environmental identity seems to be made by social attachments and attitudes people have about the conservation management approach in the area. From a social perspective, it could be a rgued that given the changing social environment in Romania characterized by socio economic instabilities, local social identity in these communities has been weakened over time. Furthermore, conservation management attitudes do not strongly define identit y. This might suggest a weaker association with the management and a development of identity on the grounds of nature connections, park attachment and conservation attitudes which might not necessarily be in agreement with the current management philosophy The notion of other has been discussed in relation to environmental identity, the argument being that identity can elicit strong connections that can take an intensified meaning and create a (community members)

PAGE 92

92 (park administration) possibly emerge (Clayton & Opotow 2003). Thus, the weak identification of the local population with the management approach could be an indication that local institutions responsible for conservation ma nagement do not truly represent what the locals believe and their connections and understanding of the environment. The wider literature on e nvironmental identity underscores that this construct should be understood as a product as well as a force, being a representation of beliefs about self in an environmental context and a behavior motivator. The interconnectivity between personal identity, values, attitudes, and ultimately behaviors has been previously accounted for in the literature (Clayton 2003; Cla yton & Opotow 2003; Hitlin 2003; Vaske & Donnelly 1999). The argument being that local environmental identity provides people with a sense of connection and a sense of being part of a larger whole, and the extent to which we see ourselves as part of the group, ultimately influences our intentions to act. The theory of interpersonal behavior incorporates self identity erception of himself or herself, as one of the major predictors of behavioral intentions (Gagnon et al. 2003; Zhang, In bakaran & Jackson 2006). This relationship is based on the argument that our definition of ourselves ultimately defines our actions. T his study found a direct impact of local environmental identity (as a reflection of attachment, connection to nature, a nd conservation attitudes) on attitudes towards pro environmental civic engagement. Furthermore, the relationship between environmental identity and pro environmental behavioral civic intentions was found to be mediated by attitudes towards civic engagemen t. environment the more positive attitudes they have towards engagement that ultimately shapes behavioral intentions. For the most part, residents were born and raised in t heir respective

PAGE 93

93 communities and from a socio psychological perspective the local environment is an integral part of their identity that ultimately structures their attitudes towards pro environmental civic engagement. This finding substantiates current un derstanding of what has been previously described as community identity and its implications for environmental values and attitudes (Pol 2002; Van Vugt, 2001). Perceived collective responsible behaviors to guard natural resources were also found to have a direct impact on attitudes towards pro environmental civic engagement. This finding is when people feel a strong collective responsibility for the well being of the environment and have cooperative relationships with one another (Dolisca et al. 2009; Kaplan, 2000; White & Runge 1994 ). A direct relationship between collective responsibility and behavioral intentions wa s not supported in this study. Attitudes t owards pro environmental civic engagement were found to mediate the relationship between local environmental identity, perceived collective environmental responsibility and pro environmental civic behavioral intentions. This underscores the importance of a ttitudes in the context of pro environmental civic engagement, identity and perceived responsibility influencing attitudes but not behavioral intentions directly. Similarly, Dolisca et al. (2009) found conservation behavior to be directly influenced by res involved (Lepp 2006). This study did not test such a concept, but the strength of environm ental identity seems to negate feelings of lack of ability or knowledge to engage. The strength of local environmental identity facilitates positive attitudes towards civic engagement and ultimately willingness to get involved in local pro environmental in itiatives. This further indicates the need

PAGE 94

94 change in their communities and the environments they are attached and connected to. The Retezat National Park admi nistration should take into account the strong connection people in these communities have to their environment and create opportunities for them to express their opinions for sustainable management of the protected area. As emphasized previously, identity the ability to reach effective communication and collaborations weakens. The residents of these communities developed strong attachments and meanings to the su rrounding environment and their existence is embedded in a permanent interaction with their environment, and should not be overlooked by management. The conceptualization of local environmental identity could be argued might be a characteristic of the popu lation in this study and the local environment which is rich in natural resources amenities. Thus, this study should be repeated under different conditions, culturally, socially and environmentally to test the environmental identity conceptualization. This study provides evidence that with stronger local environmental identity and collective environmental environmental engagement will likely increase, ultimately shaping behavioral intentions. Furthermore, this study substantiates the knowledge on rural communities and their interactions with neighboring natural environments, highlighting the major predictors of pro environmental civic engagement intentions. Conclusions Findings from this study show residen with the theory of interpersonal behavior. The model supported the hypothesis that a more positive attitude toward the environment increases the probability of civic engagement. Factors such as local enviro nmental identity, attitudes, and collective environmental responsibility play a

PAGE 95

95 mediator role in stimulating behavioral intentions. This study contradicts beliefs that people living adjacent to protected areas do not care about the environment and do not have an interest in having or bringing their contribution to protecting the neighboring environment. Such communities strongly identify with their environment and d evelop meanings, attitudes and attachments that are hard to disentangle and ultimately even harder to change. Results show that pro environmental civic behavioral intentions can be improved by providing villagers information about the importance of civic engagement, and also by strengthening local environmental identities which could be done by encouraging social interaction as well as interaction with the environment (the park, in this case). Furthermore, providing increasing understanding of the benefits of the park, the benefits of conservation, and higher management transparency and cooperation can ultimately generate greater identification of the local population with the local environment. Thus, this study emphasizes the need to create opportunities f or local people to utilize and benefit from the park so that ultimately they become an integrated part of the management and conservation stewards. Projects that require involvement, interaction, and sharing of knowledge and information (about benefits an d management approach) should be supported and constantly encouraged and implemented throughout the area. Furthermore, collective responsibility was found to impact on attitudes towards pro environmen tal civic engagement. Pretty & Smith (2004) suggest that trust and connectedness in community are necessary for determining individual actions to achieve engagement and ultimately positive biodiversity outcomes. Recommendations include using meetings and informative messages to strengthen intentions for local e ngagement and increase knowledge and awareness of conservation benefits as well the responsibilities derived from

PAGE 96

96 living adjacent to the park. Efforts to better inform local residents how current park rules and management strategies sustain the natural env ironment to better achieve the environmental goals that the residents expressed, and how they can be involved in making specific decisions facilitating those sho uld further enhance synergistic partnerships. Figure 3 1. Proposed research model

PAGE 97

97 Table 3 1. Fit indice s for measurement model and SEM m odel Construct 2 df 2 / df RMSEA CFI Measurement model 167.16 92 1.817 0.056 0.970 Measurement Model with Second Order Factor 189.32 87 2.177 0.067 0.959 SEM A 194.75 87 2.239 0.069 0.957 SEM B 187.53 87 2. 156 0.067 0.960 Table 3 2. Summary results for measurement model Factors and items Mean SD CR AVE Social attachment 1 .773 .872 .537 I feel like I belong in this community 4.50 .842 .72 The associations that I have with other people in this community mean a lot to me 4.52 .762 .63 Given the opportunity, I would move out of this community a 3.82 1.520 .61 I feel loyal to the people in this community 4.17 1.034 .84 I plan to remain a resident of this community for a number of yea rs 4.42 1.004 .88 I like to think of myself as similar to the people who live in this community 3.97 1.249 .68 Park attachment 1 .903 .950 .733 Retezat National Park means a lot to me 4.54 .784 .90 I am very attached to Retezat National P ark 4.18 .992 .92 Retezat National Park is very important to me 4.33 .910 .94 I identify strongly with the Retezat National Park 3.78 1.178 .87 I get many personal benefits out of living near Retezat National Park 3.72 1.280 .65 I enjoy living near Retezat National Park 4.54 .750 .88 I get lots of satisfaction out of living near Retezat National Park 4.17 .996 .80

PAGE 98

98 Table 3 2. Continued Factors and items Mean SD CR AVE Connection to nature 1 .871 .930 .661 I am connected to nature much like I'm connected to my family 4.40 .974 .90 Na ture is a huge part of who I am 4.40 .932 .89 I often feel a sense of oneness w ith the natural world around me 4.28 1.028 .93 My feelings for nature have influenced my spiritual beliefs 4.12 1.125 .90 When surrounded by nature, I feel at peace 4.91 .327 .67 Listening to the wind go through the trees calms my mind 4.35 1.051 .70 When I'm alone in a n atural area, I have this feeling of complete calm 4.51 .841 .65 Conservation awareness 1 .727 0.896 0.687 It is important to have the Retezat National Park for the survival of various plants and animal species 4.91 .415 0.86 It is necessary to set aside some land for the protection of plants and animals 4.89 .381 0.87 Retezat National Park is our country's pride 4.87 .481 0.94 The illegal cutting of trees in the park should be strictly regulated 4.69 .813 0.62 Conservation manage ment 1 .834 0.877 0.645 Retezat National Park is managed successfully for the benefit/enjoyment of future generations 4.16 1.123 0.91 Retezat National Park is managed successfully for a wide range of uses and values, not just tourism 3.96 1.220 0.7 4 Retezat National Park management does a good job at protecting the natural resources in the park 4.19 1.066 0.90 The citizens from the communities around the park have enough say in how the park is managed 3.16 1.304 0.63

PAGE 99

99 Table 3 2. Conti nued Factors and items Mean SD CR AVE Conservation benefits 1 .767 0.862 0.613 My community benefits from being near the Retezat National Park 3.95 1.222 0.82 Having the Retezat National Park near my home benefits me and my family 3.97 1.151 0.83 My community is a more beautiful place to live because we are living near Retezat National Park 4.47 .787 0.84 The tourists who come to the area are useful to we who live in adjacent communitie s 3.82 1.227 0.62 Attitudes pro environmental civic engagement 2 .823 0.8 80 0.596 Participating in public meetings related to Retezat National Park 4.00 .983 0.81 Participating in a community project addressing environmental concerns in the area 4.28 .844 0.75 Investing time to learn about the park and environmental protection 4.36 .874 0.82 Participating in a worksho p on how to reduce my dependence on park resources 3.87 1.149 0.68 Investing personal time to get involved with the park 3.93 1.034 0.79 Pro environmental civic behavioral intentions 3 .8 44 0.893 0.628 Attend a public presentation about Retezat National Park 4.02 1.110 0.89 Participate in a community project addressing environmental concerns 3.89 1.190 0.80 Invest time to learn more about the park and environmental protection 3. 86 1.221 0.84 Give my input into park management decisions 3.94 1.163 0.75 Participate in a worksho p on how to reduce my dependence on park resources 3.23 1.389 0.67

PAGE 100

100 Table 3 2. Continued Factors and items Mean SD CR AVE Perceived coll ective environmental responsibility 4 .707 0.843 0.729 Every citizen in my community must take responsibility for protecting the environment in Retezat National Park 4.52 .901 .90 Authorities, together with the citizens, are responsible for protect ing the environment in Retezat National Park 4.67 .807 .80 Note. = t statistic (> 1.96) at a significance level of p a Variable reverse coded for consistency directionality of items; 1 Measured on a 5 point scale where 1 = Strongly disagree; 2 = Disa gree; 3 = Neutral; 4 = Agree; 5 = Strongly agree; 2 Measured on a 5 point scale where 1 = Not at all effective; 2 = Never effective; 3 = Sometimes effective; 4 = Often effective; 5 = Always effective; 3 Measured on a 5 point scale where 1 = Very unlikely; 2 = Somewhat unlikely; 3 = Neither likely nor unlikely; 4 = Somewhat likely; 5 = Very likely. 4 Measured on a 5 point scale where 1 = Totally disagree; 2 = Somewhat disagree; 3 = Neither agree not disagree; 4 = Somewhat agree; 5 = Totally agree. Table 3 3. Correlations among factors (based on the measurement model) 1 2 3 4 5 6 7 8 9 1. Social attachment 1.00 2. Park attachment 0.559* 1.00 3. Connection to nature 0.605* 0.676* 1.00 4. Conservation awareness 0.491* 0.713* 0.524* 1.00 5. Conser vation management 0.380* 0.420* 0.434* 0.626* 1.00 6. Conservation benefits 0.409* 0.672* 0.388* 0.486* 0.583* 1.00 7. Attitudes pro environmental civic engagement 0.341* 0.510* 0.563* 0.492* 0.357* 0.555* 1.00 8. Pro environmental civic behavioral inte ntions 0.242 0.460 0.407 0.358 0.154** 0.382 0.704 1.00 9. Perceived collective environmental responsibility 0.281** 0.433 0.468 0.672 0.398 0.482 0.632 0.413 1.0 0 *Correlation significant p < .001; **Correlation significant p < .05.

PAGE 101

101 Figur e 3 2. A structural equation model test

PAGE 102

102 CHAPTER 4 A MIXED METHOD INVES TIGATION OF COMMUNIT Y ATTACHMENT Introduction Rural communities constantly gathered academic attention due to their continuous exposure to external pressures determining local economic and social instabilities. Flora and Flora (1990) emphasized that in order to achieve sustainability in such communities it is imperative to build the social structure, maintain the population base and engage residents for community action. Community attac hment is one of the key concepts discussed as being at the roots of community action that ultimately fosters community development (Wilkinson, 1986). Consequently, community attachment has been examined in a variety of disciplines, focusing on operationali zing the construct, understanding its major determinants as well as implications for community well being. Hummon (1990) defines community attachment as an emotional connection to a place that emerges in the context of residence and belonging. Predominant ly, the community attachment literature views the social relations within a community at the foundation of the emotional bond people have with their communities. Consequently, community attachment was operationalized as a unidimensional construct, general measures of feeling at home in the community, knowing y) being widely used (Gursoy & Rutherford, 2004). Over the years, one more dimension has been acknowledged to frame the comm unity attachment concept, attachment to the natural environment. Cross (2001), Stedman (2002, 2003), Beckley (2003), and Brehm (2007) emphasized in their work the importance of the natural environment in shaping community attachment. Hummon (1990) argued t hat community attachment is most strongly rooted in involvement in local social relations, but he also

PAGE 103

103 acknowledged that the built and natural environment might also contribute to such emotional ties. Furthermore, Cross (2003) underscored that a unidimensi onal approach to community attachment limits the depth of information captured and our ability to distinguish how different dimensions of attachment might shape community behavior and action differently. The multidimensionality of the construct has been s carcely assessed, only a few studies proposing measures of attachment to the social and natural environment and further investigating their predictors. Generally, the major determinants of community attachment discussed in the literature are length of resi dence, participation in community activities and groups, local ties and networks, and various socio demographic characteristics, such as age, family status, income, education ( Beggs, Hurlbert & Haines, 1996; Brehm, Eisenhauer, & Krannich, 2004; Brehm, 2007 ; Kasarda & Janowitz, 1974; Theodori & Luloff, 2002; ). The influence of length of residence on community attachment has been primarily attributed to social interaction and social integration components that define residency, but this relationship was not always found to be significant. McCool & Martin (1994) identified that newcomers were more highly attached to their community than long term residents. An argument was made that this situation might be indicative of a tendency for newcomers to be attached to biophysical or landscape features of place, as opposed to social networks and local relationships. Consequently, it can be argued that attachment to the natural environment can be equally strong in forming an emotional investment in community. Brehm et al. (2006) examined the relationship between length of residence and attachment to the social and natural environment. The study findings indicate that the strength of social attachments to the community were significantly different between residents who had lived in the area more than ten years and those who had not. In contrast, there was not a significant

PAGE 104

104 and supportive of prior prepositions underscoring the importance of assessing the multiple facets of community attachment and their predictors. In terms of socio demographic characteristics and their impact on community attachment, Kasarda & Janowitz (1974 family status, and number of children, indicative of emotional investment in community. Others emphasized the importance of education and income in e xplaining attachment ( Brehm et al., 2006 ; Gursoy & Rutherford, 2004 ). To date, the strength of such relationships is still not very clear and little is known how different socio demographic characteristics relate to attachment to the social and natural environment. Based on a qualitative st udy, Brehm's (2007) speaks to the complexity of developing a discrete separation of the natural environment dimension from the social dimension, yet clearly demonstrates that the natural environment does play an important role in the overall formation of c ommunity attachment. The author argues that in communities where natural amenities are abundant and often on a grand scale, they can play an important role in the development of a of length of residence. Brehm (2007) advocates for studies focused on better understanding specific place attributes (natural environment versus social) to which people are attached and the mechanisms through which such attachments are formed. To date, v arious research studies discussed the complexity of the community attachment concept and the contribution of the social and natural environment in shaping emotional connections to community. It is relevant to notice the majority of these studies were condu cted in

PAGE 105

105 North America, multiple assertions being proposed in terms of community attachment and its dimensions but only a few being actually empirically tested under different conditions and settings. In the mist of irresolute findings and emerging need to empirically test the community attachment dimensionality under different conditions and settings, this study examined the social and natural dimensions of community attachment in the context of nine rural communities in Romania, communities adjacent to a n ational park. Two distinct facets of community attachment were assessed in order to examine if they are distinctively predicted by length of residence, social interaction and socio demographic characteristics. Furthermore, this study qualitatively explored the extent to which we could speak of other dimensions of community attachment in these communities, questioning the existence of various pillars that support the emotional bond people have with their place of residence. Conceptual Framework Community att achment is a complex and integrating sociological construct that depicts the emotional connections among people and their communities. Feelings of belonging and identity with a place of residence are viewed as the core elements that permeate personal emoti onal responses. Community attachment is perceived as providing social and psychological attachment for residents that ultimately encourage their participation in community based activities ( Brennan, 2007 ; Theodori, 2000 ). Consequently, t he c ollective actio n and engagement literature constantly identified attachment to a community as an important pr edictor of action and active participation that ultimately translates into well being at the individual and community levels (Wilkinson, 1991). The sociological literature on community attachment primarily views social relations within a community at the foundation of the emotional bond people have with their communities. The importance of interpersonal connections (strong ties to friends and kin), involvement in

PAGE 106

106 formal community organizations, and affective and emotional feelings towards the local community has been constantly emphasized as pillars of community attachment ( Beggs, Hurlbert, & Haines, 1996; Kasarda & Janowitz 1974; Luloff and Swanson, 1995;). Howev er, more recently other ways of defining the attachment construct have been proposed. Cross (2001), Stedman (2002, 2003), Beckley (2003), and Brehm (2007) emphasized in their work the importance of the natural environment in shaping community attachment. S uch assertions are grounded in studies that examined attachment in terms of affinity to a physical place rather than the web of interpersonal relationships ro oted in a community (Stokols & Shumaker, 1981 ; Williams et al. 1992 ). Extensive literature exists that focuses on place based attachments, sense of place and place attachment (Altman & Low, 1992; Kyle et al., 2004; Williams et al. 1992; Williams & Vaske, 2003 ). Place attachment has also been discussed as being a subjective evaluation not only of the physical features of ones environment, but also tangentially integrating a personal assessment of t he social environment (Mesch & Manor, 1998). Consequently, it could be argued that community attachment integrates an emotional response to the natural envir onment and should be viewed as a variation of sense of place and place attachment, constructs primarily r ooted in the connections people have with some physical locations in the natural world. The previous remarks come to highlight the importance of bette r understandi ng the interplay between place based attachments and community attachment, and the overall role played by the natural environment in the overall assessment of the community attachment. The sociological analyses have been deficient at capturing the influence of the natural environment on community attachment, even though it did constantly highlight the interlinked relationship between community and ecological well being and the ability of the natural environment to

PAGE 107

107 support social integration (Wi lkinson, 1991). Brehm (2007) argued for a dual appraisal of community attachment, attachment to the social and natural environment and in the same time questioned the extent to which the specific natural attributes people are attached to can be distinguish ed from the social environment To date, studies did show that separati ng these two dimensions is feasible and suitable as they seem to have di stinct predictors and attitudinal behavioral implications. Kasarda & Janowitz (1974) were among the first sociol ogists to explore the major factors that contribute to community attachment. The authors attributed community attachment primarily to the prevalence of a complex system of ties and networks within community that supports social interaction. Large area and density was associated with limited social interaction and decreased community particip ation and attachment. Luloff & Swanson (1995) and Beggs, Hurlbert, & Haines (1996) also speak of the importance of social interaction, interaction with friends, relative s and others in creating the emotional bond at the foundation of community attachment. Furthermore, length of residence in a community was persistently viewed as a benchmark of community attachment ( Beggs, Hurlbert, & Haines, 1996; Goudy 1990; Payne & Sc haumleffel; 2008; Theodori, 2004 ). The influence of length of residence on community attachment has primarily been attributed to social interaction and social integration components that define residency. However, the correla tion between length of residenc e and attachment, while statistically significant, was not always found to be strong. McCool & Martin (1994) identified that newcomers were more highly attached to their community than long term residents. Such findings are indicative that one can decide t o live in a community an d become

PAGE 108

108 attached to it rapidly, thus questioning assertions that local ties and networks, and social relations in general are the primary stimuli of community attachment. The community attachment literature also highlights various socio demographic characteristics relevant for understanding emotional investment in community. Kasard a & Janowitz (1974) discussed the importance of achieved social position (income and education) and stage in the life cycle (age, family status and numbe r of children) in explaining community attachment. Previous studies assessed the role of achieved social position in shaping community atta chment ( Brehm et al. 2006 ; Gursoy & Rutherford, 2004; Goudy, 1990; Kasarda & Janowitz, 1974; Stinner et al. 1990 ). Beggs, Hurlbert, & Haines (1996) found that persons with higher levels of education had weaker local ties than persons with lower levels of education do and income had a negative effect on local sentiments. Furthermore, Goudy (1990) emphasized that age pla ys a role in affecting local bonds and sentiments but usually of less importance than length of residence. The inability of age measures to capture parenting and marital status was highlighted by Beggs, Hurlbert, & Haines (1996) and further argued for incl uding measures of family status when assessing attachment. While socio demographic measures and their role in shaping community attachment have been previously studied, findings have not always been conclusive. To date, several studies employed a multidime nsional assessment of community attachment and tested how core predictors of attachment correlate with distinct dimensions of community attachment. Using a quantitative assessment, Brehm et al. (2004) found social attachment and attachment to the natural e nvironment as two distinct dimensions of the broader concept of community attachment. The results showed religious affiliation, length of residence and social involvement strongly associated with social attachment, while not significantly

PAGE 109

109 associated with l evels of attachment to the natural environment. Furthermore, a significant connection between collective action, perceptions of open communication and social attachment was found However, the relationship between natural environment attachment and collect ive action and open communication was not significant. Brehm et al. (2006) identified household income as a better predictor of attachment to the natural environment, an argument being made that those who have money tend to have more interest in natural am enities. In addition, Brehm et al ( 2006 ) and Cross (2003) identified sentimental and emotional attachments to local communities, based on social ties as well as ties to the physical environment as having a direct influence on levels of environmental conce rn and local conflicts over land use management The ability to differentiate between the two dimensions of community attachment was questioned by Brehm (2007) in a study using a mixed method approach. While the quantitative analyses presented a more disc rete picture of social and natural environment dimensions, the qualitative results of the analyses revealed attachment to the natural environment as a distinct dimension of community attachment, as well as a facet of community attachment that is more embed ded in the social dimension of attachment. The ability of the natural environment to support certain life styles and social connections was underscored when trying to explain the embedded nature of the social and natural dimensions. In addition, it is ques tionable to what extent attachment to the natural environment could be a direct consequence of a social instability within the community. Thus, Brehm (2007) emphasized the need to further investigate the complex dimensionality of the community attachment c oncept using multiple methods. C urrent understanding s of the community at tachment dimensionality demonstrate the need for further research in various community contexts to clarify the role of the social and natural environment in the assessment of communit y attachment. Understanding the underpinnings of

PAGE 110

110 various dimensions of attachment could further provide deeper knowledge of the mechanisms through which different facets of community attachment contribute to the formation of social groups and ultimately in fluence community action. Based on current theoretical conceptualization of community attachment, this study assessed the structural relationships between social attachment and natural attachment (the national park neighboring the local communities partici pating in the study) and social interaction, length of residence and socio demographic characteristics (age, family status, number of children under 18 years, education, and income ) The factors included in the model were chosen based on the literature as well as their relevance for the study site. Consequently, following findings from previous studies research hypotheses were developed as follows: H1: Social interaction positively impacts attachment to the social and natural environment. H2: Length of resi dence positively impacts attachment to the social and natural environment. H3: Age positively impacts attachment to the social and natural environment. H4: Family status positively impacts attachment to the social and natural environment. H5: Number of chi ldren under 18 years old negatively impacts attachment to the social and natural environment. H6: Education negatively impacts attachment to the social and natural environment. H7: Income negatively impacts attachment to the social and natural environment. Brehm et al. ( 2006 ) underscored that community attachment might have other facets not examined, such as cultural traditions and beliefs, economic linkages and activities, and political engagement. Thus, a qualitative approach was employed to further asses s the dimensionality of the community attachment construct.

PAGE 111

111 Methodology Study Area Retezat National Park (RNP) was designated in 1935, being the first national park established in Romania. In 1979, RNP was declared an International Biosphere Reserve under the UNESCO Ma n and Biosphere program a nd in 2004 RNP received its Protected Area Network (PAN) Certification Retezat National Park is located in the southwestern Carpathians, Hunedoara County, and the total surface of the park is 38,138 ha (RNP Management Plan 2008). Within the park, there are more than twenty mountain peaks of 2,000 meters or hig her in addition to eighty lakes of glacial origin. There are more than 1,100 species of plants, over 50 species of mammals including roe deer, chamois, lynx, be ar, and otter and 168 recorded bird species including the golden eagle A large portion of the park area (17,564 ha, 46%) is owned by the state, while local associations own the reminder of the land (20,574 ha). Of the 43 villages adjacent to Retezat Nati onal Park, 26 villages have grazing rights to alpine meadows, and their rights are administered either through community based associations or the local councils of the five communes to whic h the villages belong (Kuijs & Bergh 2006). The total population of these five communes was estimated at 14,006 inhabitants. Communities rely on park resources primarily for grazing and the use of other natural resources such as wood, non timber forest products, mushrooms, and medicinal plants. Due to reduced level of i nterest and engagement in community affairs, a low sense of community and collective responsibility has been constantly emphasized as characterizing Romanian rural communities ( Oostenbrink & Kosterink, 2005; PJB Associates, 2006). However, there is scarce empirical evidence that supports these assertions and little is known about the determinants and level of attachment in these rural communities in Romania. Therefore, this study will investigate the major determinants of

PAGE 112

112 community attachment and the multid imensionality of this construct in the context of nine rural communities adjacent to Retezat National Park, using both a quantitative and qualitative approach. Survey Data Collection and Analysis R ural communities adjacent to Retezat National Park belong t o five communes encompassing 43 villages with a total population estimated at 14,009 adult residents. Two villages dissimilar in size from each commune, to assure representation of each commune in the final sample were selected. Nine communities (one commu ne had only one village) were selected for this study using a multistage random sampling. The sampling frame included the total number of households in the nine communities adjacent to RNP, but the final unit of analysis was the individuals residing in the households. The nine villages selected have a population of 4,232 persons residing in 1,159 private households. Cross sectional data was collected from 260 residents during June October 2009 using face to face interviews (68% response rate) and mail su rvey (9% response rate). A systematic sampling method with a random start was used to select participants for the face to face interviews, the sampling interval being established based on the number of households in the sampling frame for each community di vided by sample size needed in each community. Considering the small size of these communities and the challenge of finding people at home, in general every other household was selected. The person in the household of age of 18 or older was asked to partic ipate in the study, and the questions were asked directly to the respondents and recorded by the interviewer. In addition, a mail survey was sent to those households where the residents were not found at home, even after multiple visits at different times during the day. A total of 230 surveys were sent by mail to the population of four communities using home

PAGE 113

113 address information from the phone book. The low response ( n = 21) to the mail method revealed the lack of feasibility of this method for the resident s living in these communities. Participants in this study were asked to express their opinions on a series of questions about their attachment to the community, social interaction, length of residence and several socio demographic characteristics This stu dy employed an assessment of community attachment based on attachment to the social environment and attachment to the natural environment. The social dimension was measured using ten items on a five point Likert scale ranging from 1 to 5, where 1 = Strongl y disagree; 3 = Neutral; 5 = Strongly agree. The item s were adapted from Theodori & Mayfield (2008). Cross (2003) emphasized that attachment to the natural environment can also be viewed as an attachment to the bioregion and its attributes, the park in the case of this study Attachment to the park was assessed using the conceptualization employed in the place attachment literature, more specifically the two dimensions of attachment: place dependence and place identity ( Kyle et al. 2004 ; Williams & Roggenb uck 1989; Williams et al. 1992 ). Place dependence centers on a functional attachment to place, while place identity reflects attachment to Retezat National Park was measured using nine items on a five point Likert scale ranging from 1 to 5, where 1 = Strongly disagree; 3 = Neutral; 5 = Strongly agree. Social interaction was measured using the frequency of interaction with various types of people in the community with at least one of the following types of people? Family, Close Friends, Acquaintances, few times a year, (3) once a month, (4) a few times a month, (5) once a week, (6) more than once a week, and (7) everyday. The respondents were asked to report for how long they have lived in

PAGE 114

114 their community. Five categories of length of residence were create d (1=One to 20 years; 2=21 to 40 years; 3=41 to 60 years; 4=61 to 80 years; 5=Over 80 years). Individual level socio demographics included age (1=18 to 30 years; 2=31 to 50 years; 3=51 years and above), family status (1=Single; 2=Partnered); number of chil dren under 18 years (frequency from 1 to 5 children), educational attainment (1=Elementary school or less; 2=High school graduate; 3=Post high school education), and monthly household income ( 1=Almost no income to 7=More than 2,000 RON; ~US $661). Data an alyses were performed following three stages. First, descriptive statistics were computed for the variables used in the study using the Statistical Package for the Social Sciences (SPSS) version 18.0. Second, the data collected was screened and the critica l assumptions underlying the statistical techniques employed by the study were assessed. Third, a two step data analysis was employed to assess the hypothesized relationships among the research construct s (Anderson & Gerbing 1988). A s part of this process individual items were examined using Confirmatory Factor Analysis (CFA) and the measurement model for constructs included in the study was estimated using MPLUS version 5.21 to determine how well the indicators captured their specific constructs and the ability of the respondents to differentiate between constructs (Hair et al. 2006). This was followed by an assessment of the S tructural Equation M odel (SEM) assessing the hypothesized relationships between constructs. SEM was assessed using MPLUS version 5.21 using the WLSMV (weighted least squares mean and variance adjusted) method of estimation, method recommended for categorical ordinal data (Muthen et al. 1997). After adjustments (specified in the results section), the models utilized in reporting the findings exceeded minimum standards of acceptability for model fits.

PAGE 115

115 Qualitative Data Collection and Analysis In depth interviews were conducted with community members in each of the nine communities selected for study. A grounded theory approach to inqui ry was followed to allow for an in depth theoretical conceptualization of the community attachment construct. A referral sampling technique was employed to identify participants, with a focus on including individuals highly involved in their community and also individuals with low levels of local engagement. Park management employees and public officials were asked to provide names of community members from the nine selected communities that are interested and highly engaged in community affairs. Based on a first interview in each of the nine communities, the principal investigator asked the respondents to provide names of other community members that might participate in the study and have dissimilar levels of interest and engagement in their community. Dat a was collected from 24 community members representing the nine communities adjacent to Retezat National Park. Creswell (2007) recommends that researchers interview 20 to 30 individuals when a grounded theory approach to qualitative inquiry is employed. A semi structured interview with an interview guide was employed to organize the discussion. Following Brehm (2007) approach to investigate the multiple facets of community attachment, the discussion begun with a very general discussion of their community. T he respondents were asked to describe their community to someone who has never been there, followed by a question that asked them to more specifically talk about things they are attached to or care about the most in their community. The interviews were co nducted by the principal investigator in the native language and were tape recorded, with the exception of one interview where the respondent declined permission to record the interview. All the interviews were transcribed and analyzed by the principal inv estigator. Interviews were analyzed using a groun ded theory approach (Strauss &

PAGE 116

116 Corbin, 1990, 1998) allowing for unique theoretical categories to emerge from the data. A series of steps specific to grounded theory research were followed in analyzing the d ata collected through the in depth interviews. At first, the principal investigator coded the data for its major categories of information. This stage was followed by a process of linking categories, focusing on a deeper understanding of the interrelations hips between categories. Survey Data Results The residents sampled resided in nine communities adjacent to Retezat National Park. The average age was 45.0 years, with almost one quarter of the respondents (23.7%) being between 18 and 30 years and slightly more than one third of the respondents (36.1%) being between 31 and 50 years. Respondents who were above 51 years were the most represented group (40.2%). Fifty four percent were males and 46% percent were females. The average length of residency was 37 ye ars. The majority of the respondents (65.1%) were married or partnered, 48.6 percen t had one to five children less than 18 years old and on average, the number of adults per household was three. Almost one third of the respondents (31.6%) indicated high s chool as the highest level of education attained, 14% of the responden ts had some college or a college degree and 14 % had an elementary school education or less. About one third of respondents (33.8%) reported a monthly household income between 1,000 and 1 ,999 RON (about US $330 $660) and 30.6% indicated a monthly household income of more than 2,000 RON (about US $661). The majority of respondents (70.6%) indicated that they do not have any property rights (ownership or land use rights) in Retezat National Park. Individual constructs and their measurements were examined using confirmatory factor analysis (CFA). As a result of initial CFA tests, several items in various factors were dropped due to their low factor loadings. Ultimately, for each construct, th ose items where retained that were substantive in size and had significant loadings on the factor. In this stage, the correlation

PAGE 117

117 between p ark identity and park dependence was found to be high suggesting a weaker differentiation of these constructs by this population. Consequently, scale items were collapsed and one measure of park attachment was retained including seven items. Furthermore, due to a high correlation between age and length of residence and multicollinearity concerns, age as an observed varia ble was eliminated from further analysis (H 3 was not assessed) Education and income were also found to be highly correlated, thus education was assessed as a predictor of attachment to the social environment, while income was examined as a predictor of at tachment to the natural environment. After assessing the measurement model for each construct and a good fit to the data was observed, the fit indices for the total measurement model were examined. The measurement model including five latent factors (socia l attachment, park attachment, interaction with friends, interactions with public officials and park interactions) and five observed variables (length of residence, family status, number of children under 18 years, level of education, and income) was teste d. The fit indices for a t otal measurement model with five latent factors revealed g ood fit (Table 4 1 square/ df ratio (2.08 2 = 133.12, df = 64 p < .001) was lower than the suggested criteria (i.e., <3.0; Kline 2005). CFI (.98), TLI (.98), RM SEA (.064 ) and WRMR (. 764 ) yielded a good model fit and all the item (indicator) loadings were significant ( p < .001) and ranged from .67 to .95 providing strong evide nce of convergent validity (T able 4 2). Evidence of internal consistency is provided by recomm ended level of .70 (Nunnally & Bernstein 1994), ranging from .68 (interaction with park) to .90 (park attachment) and composite reliability (CR) above the recom mended level of .70 (Fornell & Larker 1981), ranging from 79 (interactions with friends) to .96 (park attachment). Also included in T able 4 2 are the average variance extracted (AVE) estimates with

PAGE 118

118 recommended level s (Bagozzi 1994; Fornell & Larker 1981). All values exceeded the re commended level ranging from .55 (social attachment) to .73 (park attachment). All the inter correlations among latent factors were less than the suggested threshold of .85 (Kline 2005), ranging from .23 to .57 and being a strong evidenc e of discriminant validity (T able 4 3). These findings reveal that the proposed measurement model satisfied all the psychometric requirements, thus the measures were adequate for further analysis. A hierarchical model was tested with interaction with friends, public officials and park set to load on a second order factor, social interaction. The fit indices for the model revealed good fit, social interaction being founds to have a hierarchical structure and this model was used fo r further analysis. The square/ df ratio (2.09 2 = 139.99, df = 6 7 p < .001) was lower than the suggested criteria (i.e., <3.0; Kline 2005). CFI (.98 ), TLI (.98 ), RMSEA ( .065 ) and WRMR ( .826 ) yielded a good model fit and all the item (indicator) load ings were significant ( p < .001), the factor loadings for the sec ond order model ranging from .67 (interactions with park) to .80 (interactions with public officials). SEM analysis was performed to examine the overall model as well as individual tests of the hypothesized relationships among the latent constructs. The hypothesized model to the data resulted in adequate fit wi square / df ratio (2.05 2 = 139. 0 9 df = 6 8, p < .00 1) being lower than the suggested criteria (i.e., <3.0; Kline 2005). The goodness of fit indices CFI (.98), TLI (0.98), RMSEA (.063 ) and WRMR ( .827 ) revealed acceptable fit All factor loadings wer e significant and subst antial in size. Support for the hypotheses was examined via the significance of the individual path coefficients (Figure 4 1) Statistically s ignificant path coefficients were found between : social

PAGE 119

119 interaction and attachment to the s ocial environment (standardized coefficient of .402; p < 0.001 ) and attachment to the park ( standardized coefficient of .482 ; p < 0.001) (H1 supported); length of residence and attachment to the social environment ( standardized coefficient of .318 ; p < 0.0 01 ) and attachment to the park ( standardized coefficient of .195 ; p < 0.05) (H2 supported). Family status did not have a statistically significant impact on attachment to the social environment ( standardized coefficient of .131 ; p > 0.05) and attachment to the park ( standardized coefficient of .110 ; p > 0.05) (H4 not supported). A significant negative path coefficient was found between the number of children under 18 and attachment to the social environment (s tandardized coefficient of .154 ; p < 0.05 ), whi le this variable did not have a statistically significant impact on attachment to the park ( standardized coefficient of .098 ; p > 0.05) (H5 partially supported). Furthermore, a significant negative path coefficient was found between level of education and attachment to the social environment (st andardized coefficient of 0.224 ; p < 0.05 ) (H6 partially support ed), and between income and attachmen t to the p a rk (s tandardized coefficient of .283 ; p < 0.001 ) (H7 partially supported). The model explained 38% of the variance in attachment to the social environment, and 37% of attachment to the park. Qualitative Data Results construct, several facets of attachme nt emerging from the textual data. These themes were found to best represent the meanings and communalities of the shared narratives. Four thematic categories emerged from the narratives that depict distinct facets of community attachment and in the same t ime speak to the complexity of the construct. For a better depiction of the thematic with the quotes are fictive.

PAGE 120

120 Attachment to the social e nvironment One t narratives captured the social environment within the community as a discrete dimension of attachment. Respondents mentioned the importance of the social connections within the community in shaping feelings of emotio nal attachment. The narratives underscored that strength of the community as well as personal attachment to the community emerges from the care and respect community members have for each other. they are, meaning there is this connection between people based on respect, of love in the end, behaved even though others say that people from the mountain areas are uncultivated, and I to work to survive and in exchange they had that kindness because ones without the ev en now people greet each other. (Ioan) Respondents through their work responsibilities show their c are and attachment to fellow community members. Feelings of attachment to the social environment translate to behavior directed at protecting the social structure of the community. I care about the people, their well being, because of this we even allow t hem to pick up non have penalt ies, fines, given to the local population, we are in an open partnership with them, we helped them with timber for house consumption, and with th is we answer to their requests. (Ioana) unity, and the following quote further reveals that certain dimensions of attachment might have higher personal relevance than other. Furthermore, attachment and care for one aspect of the community determines positive connections and relevance for other d imensions of community that strongly relate to the social life.

PAGE 121

121 I also care very much about the cultural and historical values we have in the commune, becaus e thanks Goodn (Horea) Attachment to the social environment also relates to the family background, respondents attaching strong meaning to the place due to genealogical roots in the area. Notions such as ly emphasized in the narratives. I have strong roots in the area, being a local, my whole family is from the area, the properties I inherited from my parents, grandp arents, and so for (Florenta) I have lots of relatives here, my grandparents where from here, my parents, my have relatives in this village. (Adrian) Attachment to the natur al e nvironment The natural environment emerged as a distinct dimension of attachment, the respondents talked about the natural environment in very distinct and i ts beauty, the whole local landscape contributing to the connections people have to their natural environment. Respondents noted how the natural environment supports their livelihood and the livelihood of others in the community. The relationships develope d between the residents and nature seems to emerge from the ability of the natural environment to provide opportunities for relaxation and self actualization. on the hill, (Alexandru) Respondents reflected on different aspects of the natural environment that permeates their emotional investment in the community. Th different features in the natural environment, including the Retezat National Park adjacent to

PAGE 122

122 their community. The natural beauty of the park was emphasized by the respondents in their narratives. The church something that cannot be described in words, so if you go there it cannot be that there are insect s, birds, that cannot be found anywhere else and there is forest there is vegetation and there is so much silence and if you go there you recharge your batteries like nowhere else, like (Miruna) Of course, Retezat National Park, yes, so Retezat and consequently the National m (George) The respondents in thei r description of the natural environment constantly depicted the importance of nature through their life and primarily how much it meant for their childhood. Furthermore, the ability of nature to provide and support their existence was also highlighted, be ing viewed at the foundation of the community well being. The area and the location of the community provided opportunities for constant interaction with the natural environment, being an object of appreciation derived from constant activities strongly con nected with the natural environment. I love the nature I spent my childhood with and being with the cattle while grazing since I was 8, 9 years old, our parents sent us with the cattle have beautiful hills here, near the village, the pa sture is near the village, the plains are close to the village, we were going there to help our parents with the land, on the hills we spent most of the time with the cattle, that was when all of us were orest, the water bodies, the wildlife in the forest, from school since I was a kid I loved these things which I still admire, and it happened for me, I was lucky to (Verde) Remarks capturing attachment to the natural environme nt depicted the personal significance of the current place of residence as a place of birth. Experiences accumulated over the years in the context of residence and interaction with the natural environment encroach feelings of attachment and care.

PAGE 123

123 I care a bout it as a place of my origin (a birthplace) and because of its beauty being adjacent to Retezat, being adjacent to some very beautiful water bodies (rivers), Streiu, can I say, as any birthplace to which you have been attached ones, it a ttracts you as it attracted me when I came back here after my retirement and I started to build a house, but being adjacent to Retezat, not with the park, not only with the park, with Reteza t in his whole integrity is passion and a great enjoym ent considering that is place of origin) that brought me here, I still live some disappoin tments from this (Veronica) Further more, respondents talked about their affinity for the land and continuous investment in continuing activities that support their livelihood. Primarily, they emphasized the importance they attach to their property, the land and the agricultural practices th ey engage in. In addition, taking care of the livestock has been and continues to be an essential aspect of the local life. Much of the emotional connections with the land seem to emerge more from an identity dimension, and less a dependence response, an a spect reflected by comments that underscore personal efforts to cultivate the land and taking care of the livestock even though these activities are not always efficient or beneficial. d after 40 years of work at Paroseni thermal power station and still I love to work, having (Minodora) the (Alun) It has been transmitted from generation to generation this love for taking care of the animals and even now there is this p reoccupation (Ion) Attachment to the institutional e nvironment The institutional environment was depicted as playing a role in shaping attachment to the community. Respondents emphasized their care and concern for the local institutions, such as scho ol, church, clinic any the business environment that provides opportunities for jobs. The importance of the local church was emphasized in the narratives, as an institutional place that brings people together, people being

PAGE 124

124 emotionally involved with the ch urch, volunteering for it and putting their efforts to be a better place for the community to meet. use to sing as well, this is what I am attached to because during the reli gious (Mihai) looking; we have done some renovations at school and the local cultural center, working on this astructure that upsets us the most, we (Paula) Furthermore, institutional attachment evolves from personal occupation. Respondents mentioning the school and the local clinic, as well as the business environment as having spe cial meaning for them and the benefits if brings to the well being of the community. school to be bet ter than at my house, unfortunately I could not accomplish the school, and other local (Bogdan) I care about everything; I care about everything t hat moves around here starting from the smallest thing, from a small bar, a job for somebody, to some of the most primarily in school, church, local council, the city hall local forest center, all of them affect us in one way or another and motivate us to have an attitude or (Florenta) The respondents emphasized the importance of the institutional environment in bringing people together. Aspects of community (e.g. elements of infrastructure) that gather common interest were emphasized. working in the natural areas, for example foresters, hunters, they still care about the good and beautif fixing the road, water, the pasture, and everything we us e in common, because this was we share in commo (Verde)

PAGE 125

125 Att achment to the cultural e nvironment The cultural environment within the community was depicted as contributing to the emotional connections with the place. An emphasis was attributed to the local traditions and customs that historically defined and streng thened the social environment. The local residents do identify themselves through their past and history, and value its contribution to the local identity and see the importance of keeping them vivid and transmitting them from generation to generation. I think is important to protect/ guard our archaic way of life, our old, our old traditions which are very beautiful and I care a lot about them, these old undertakings I think it will be a shame to be lost because they are not that valuable financially, the y do not bring money and it takes a lot of time, and people are more future will look like for (Eugen) and I love who is considered to be the strongest (the most vivid) express ion of the Romanian history a lot, I like it very much considering that we are in a place with a lot of history, her e in our garden we have the ruins of a Romanian aristocrat, we found coins, we found objects who are currently at the museum in Deva, from the museum they came to us, you can see these ones here, these are fo ssils of dinosaurs well, because while I was with the animals in the mountains I found these fossils and things, and I still admire and love them, I end up remain ing the custodian of the (Verde) Attachment to the cultural environment has the ability to determine personal interest in engaging in activities that will ultimately protect the local traditions and further instill community pride. We try to maintain our traditions, folklore, or to bring them back into actuality, because there was a period of a few years now when these act ivities have been totally ignored, not taken care of, this is what I care about, to put into value everything th at is beautiful in our commune. (Horea) Interlinked nature of the community attachment d imensions It could be noticed from the respondents narr atives that the dimensionality of the community attachment construct

PAGE 126

126 captures in addition to distinct dimensions, the interplay of the different facets of attachment. This suggests that such dimensions are distinct but in the same time they do share common connotations, links and interrelationships existing between the different facets of the construct. The natural and social dimensions are strongly linked, primarily considering the importance of the natural resources for sustaining livelihoods. The people are very good, people with fear of God, hard working people, assiduous, they like to take care of the livestock they care about the forest, because this was there occupation here, the parents raised us this way and they grew up like this as well, in the future we want to protect as much as we can, to protect the nature the most because it gives us everything (Stefan) The area in its entity encompasses not only the natural environment, but also the social structure which is an integral part of the environ mental elements that sustain the welfare of these communities. Much of the attachment evolves from love for nature and love of the socio cultural life, elements rooted in years of persistence in the area that give distinctiveness and enforce connections to the community in its entity. Primarily, the beauty are people from the spread, who persisted village was totally destroyed. (Stefan) For a short period of time, 20 years, I have been gone from the area in a different attached to the land so to speak, to the area, of wha t the whole mountain represents (Alexandru) The multiple facets of community attachment even if distinct they do come together and define an overall personal attachment to the community. The respondents narratives emphasize that commun ity attachment is embedded in the multiple dimensions of community that ultimately create the emotional bond at the foundation of attachment. Distinct objects of commun ity, even if sometimes such objects are not easily distinguished.

PAGE 127

127 not live anywhere els (Mihnea) Supported by the previous remarks, it ca n be argued that the community attachment construct has a hierarchical structure, different environmental dimensions being at the foundation of defining emotional investment in the community. Discussion and Conclusions The multidimensionality of the commun ity attachment construct was assessed in this study. More specifically, the focus was on identifying if different facets of community attachment (attachment to the social and natural environment) are distinctively predicted by length of residence, social i nteraction, and socio demographic characteristics. This analysis was complemented by a qualitative exploration of the community attachment construct to investigate the extent to which we could speak of other dimensions of community attachment. The quantit ative results indicate that attachment to the social and natural environment are distinctively predicted by variables commonly associated with community attachment. Social interaction was found to have a stronger effect on park attachment, while length of residence had a stronger association with attachment to the social environment. The reduce effect of length of residence on park attachment further supports previous assertions that one can decide to live in a community an d rapidly become attached to the n atural environment (McCool & Martin, 1994 ; Brehm et al., 2006). Furthermore, the stronger association of social interaction with park attachment could be the result of social groups and shared experiences that relate to the park and the natural environment in general. The park surrounding these rural communities and the natural resources in general has strong implications for the residents due to their relevance for their livelihoods. Activities, such as agriculture, livestock raising, and even hiking have the ability to bring people together and interact on topics of common interest. This assertion emerged for the

PAGE 128

128 an integral part of their existence. Family st atus was not found to hav e a significant effect on attachment to the social environment and park attachment This fi nding suggests that being in a relationship does not make any difference for the attachment people held toward their social and natural envi ronment. Ultimately, other factors having greater ability in predicting personal feelings of attachment and belonging. In addition, a significant negative effect was found between the number of children under 18 years in the household and the strength of a ttachment to the social environment. Drawing on the social network literature and community interaction, which suggest that stronger ties (family ties) have weaker influence on community attachment ( Beggs, Hurlbert, & Haines, 1996), it could be argued that larger families with children will tend to devote more of their time to their family and have weaker level of interaction in the community. Previous literature underscored that the stronger the family ties are the weaker the level of interaction in the co mmunity is and thus ultimately a reduced level of attachment to the community ( Brennan, 2006; Bridger & Alter, 2008; Summers, 1986; Wilkinson, 1991). The number of children in the household was found not to have an effect on attachment to the park, further reinforcing that attachment to community lay in the multiple facets of the community and various personal conditions distinctly facilitating the development of attachment to different dimensions of the community. Education and income, as a reflection of s ocial position in the community were two other measures assessed in this study. Education was found to have a significant negative effect on attachment to the social environ ment, while income had a significant negative influence on attachment to the park. Previous literature underscored that local concentration of network ties

PAGE 129

129 should be higher among persons of lower social position (e.g., persons with lower level of education and income) ( Beggs, Hurlbert, & Haines, 1996 ; Campbell et al., 1986; Goudy 1990). Thus, the implication of education and income on attac hment has consistent effects, education being inversely related to attachment to the social environment and income having a negative influence on attachment to the park Beggs, Hurlbert, & Haines (1996) found income as having a negative net effect on local sentiments, suggesting that individuals with higher incomes might hold higher expectations for their community and its leaders, and therefore, they may tend to be more critical of the community than le ss affluent persons. In the cas e of this study, income was found to significantly affe ct park attachment. Ultimately, the results suggest that social position does play a role in shaping attachment to the social and natural environment s with education hav ing greater impact on social attachment and income influencing park attachment. The results of the quantitative part of the study revealed the distinct nature of the two dimensions of community attachment, different predictors being more powerful than othe rs in predicting attachment to the social and natural environment. Consequently, the importance of incorporating measures of attachment to the na tural environment in the broader assessment of community attachment is emphasized. In various contexts and sett ings, the natural environment might be a stronger dimension of attachment having the ability to generate powerful emotional responses. Simply acknowledging that the natural dimension does exist and as is most often embedded in the social environment, as su ggested by Brehm (2007) is not sufficient, there is much more information that can be captured by integrating assessments of various dimension of community attachment. This study reveals that park attachment, as a reflection of attachment to the natural en vironment, is strongly correlated with attachment to the social environment emphasizing its role in tangentially supporting the social dynamics that are at the foundation of

PAGE 130

130 the community. The correlation between the two dimensions further supported the sh ared meanings that the two dimensions capture, finding strongly supported by the qualitative part of the study. T he multidimensionality of the community attachment construct and the intertwined nature of the relationship between the emerging dimensions we narratives Four distinct dimensions of attachment emerged from the textual analysis, attachment to the natural, social, institutional and cultural environment. While distinct, these dimensions of attachment do have an inte rtwined nature that makes them worthy of attention and further assessment. Brehm (2007) talked about th e complexity of the construct, and primarily discussed two dimensions of attachment, the natural and social environment and the tangled relationship of t he two dimensions. This study brings into attention two more dimensions of attachment, the institutional and cultural environment. Respondents emphasized their attachment to the social and natural environment, their connections being shaped by various expe riences some that go back to their childhood when family tasks required greater interaction with the social and the natural environment. Various activities (e.g. agriculture, livestock) were depicting as being at the roots of attachment, their ability to b ring people together and closer to the social and natural these activities have been lost over the years in these communities, thus a weakened interaction being the c urrent reality with direct implications for the strength of the connections with the social and natural environment. Thus, such interactions should be encouraged in the future, primarily for children and their connection and interactions with the surroundi ng environment. Attachment and care was also found to emerge from job responsibilities that are locally based, suggesting that people that have local responsibilities do develop strong connections to

PAGE 131

131 the medium that relates to their job environment. Such feelings of attachment emerging from direct engagement in activities that relate to job responsibilities might ultimately have implication for empowerment, local engagement and intentions to act on local concerns. Thus, it is important to provide employmen the community (e.g. local school, hospital, national park). Furthermore, t he residents of the communities adjacent to the park have strong roots in the area, generation after generation liv ing in the same community fact that shows their strong connection with the land and their commitment to the area. On the basis of this study, a refinement of the community attachment measurement is needed that better captures the multidimensionality of th e construct. This study emphasizes the importance of understanding how different dimensions of attachment are shaped and distinctively contribute to community attachment. The importance of depicting the connections people h ave to the multiple dimensions of their environment derives from current understanding that attachment is relevant for development, planning purposes, and conflict resolution. People are attached to various community attributes that they care about and ultimately want to protect. Thus, as sessments focusing on understanding the multiple facets of community attachment can provide deeper understandings of community aspects and their contribution to local attachment. Informed programs could be developed aimed at strengthening various dimension s of the community that lack attachment and public care through the establishment of channels of interaction and communication. This further suggests a need to understand how various dimensions of community attachment are formed, how are they f luctuating a nd how separately predict various attitudinal and behavioral patterns.

PAGE 132

132 L istening to the local residents and understanding their attachments within the community context so that informed decision could be made and in the long run avoid conflicts is paramou nt Warren (1987) emphasized the importance of understanding the particular local geographical attributes, due to their implication for community togetherness that ultimately is a crucial element in the formation and continuation of co mmunity. Thus, considering the multiple facets of the community attachment construct is questionable how people become attached on dimension over the other in the context of the community, and what happens when people are not socially attached to their env ironment, what forms of attachment do they develop. belonging and connection with the surrounding environment. Residence and rootedness were constantly viewed as key elemen ts that define community attachment, and personal interest in the local. This study tested a series of variables in their ability to predict two distinct dimensions of community attachment, and the results showed that different personal and social characte ristics have distinct influence on community attachment dimensions. Thus, much more information was captured in relation to community attachment. The correlation between the two dimensions (the natural and social environment) further supports the intertwin ed relationship between the two dimensions. This was supported by the qualitative analysis, which revealed four dimensions of attachment that share common meanings and interconnections. Based on these results, the importance of developing a measurement tha t focuses on the hierarchical nature of the construct is emphasized. The multiple dimensions of attachment building on each other to ultimately define an overall attachment to the community, with its foundations being rooted in multiple local aspects.

PAGE 133

133 Tab le 4 1. Fit indices for measurement model and SEM model Construct 2 df 2 / df RMSEA CFI Measurement M odel 133 12 64 2.0 8 0.06 4 0.9 76 Measurement Model with Second Order Factor 139 99 6 7 2. 09 0.0 65 0.9 75 SEM 139 09 6 8 2. 05 0.0 63 0.9 75 Table 4 2. Sum mary results for measurement m odel Factors and items Mean SD CR AVE Social attachment .773 891 546 I feel like I belong in this community 4.50 .842 .76 The associations that I have with other people in this community mean a lot to me 4.52 .762 .69 Given the opportunity, I would move out of this community 3.82 1.520 .67 I feel loyal to the people in this community 4.17 1.034 .77 I plan to remain a resident of this community for a number of years 4.42 1.004 .87 I like to t hink of myself as similar to the people who live in this community 3.97 1.249 .65 Park attachment .903 .967 .808 Retezat National Park means a lot to me 4.54 .784 .95 I am very attached to Retezat National Park 4.18 .992 .94 Retezat Nati onal Park is very important to me 4.33 .910 .95 I identify strongly with the Retezat National Park 3.78 1.178 .66 I get many personal benefits out of living near Retezat National Park 3.72 1.280 .8 4 I enjoy living near Retezat National Park 4.54 .750 .87 I get lots of satisfaction out of living near Retezat National Park 4.17 .996 .80 Interaction s with friends .750 786 .6 50 Acquaintances 4.85 1.606 .85 Close Friends 5.40 1.573 .76 Interactions with p ublic officials 3 .54 1.790 .83 N/A N/A N/A Interactions with park .682 .7 79 .6 41 Retezat National Park Staff 2.71 1.821 .93 Tourists 3.09 1.816 .65 Note. = t statistic (> 1.96) at a significance level of p coefficients; CR = composite reliability; AVE = average variance extracted.

PAGE 134

134 Table 4 3 Correlations among factors (based on the measurement model) 1 2 3 4 5 6 7 8 9 10 11 1. Social attachment 1.00 2. Park attachment .561 1.00 3. Le ngth r esidenc e .416 .230 1.00 4. Family status .186 .149 .247 1.00 5. Kids under 18 .126** .081 155* .104 1.00 6. Level of e ducation .198 ** .078 .093 .020 .143 1.00 7. Income .054 .233* .007 .060 130 .302 1.00 8. Friends interaction .274 .340 .036 .004 .056 .054 .042 1.00 9. P ublic officials interaction .311 .387 .041 .004 .064 .062 .047 .565 1.00 10. Park interaction .261* .325 .034 .003 .054 .052 .040 .474 .540 1.00 11. Social interact ion .388 .483 .051 .005 .080 .077 .059 .705* .802* .673* 1.0 0 *Correlation significant p < .001; **Correlation significant p < .05.

PAGE 135

135 Figure 4 1. A structural equation model t est

PAGE 136

136 CHAPTER 5 THE IMPLICATIONS OF AGE ON AFFECTIVE AND ATTITU DINAL ENVIRO NMENTAL RESPONSE Introduction Romania is an Eastern European country going through a transition from a communis tic to a democratic society with a market based economy. Institutional renewal, massive restructuring and privatization are characterizing this period. In this process demographic and social economic changes are taking place with direct impact s (Roman & Roman, 2003). Romani a is a country rich in natural and cultural resources. Due to a historical focus on prot ecting natural habitats, Romania still has a large variety of forest fauna species, including 60% of all European brown bears and 40% of the European wolves (Ioras, 2003) After 1989, when major changes in land ownership started to be implemented, threats to biodiversity sharply increased. Excessive exploitation of natural resources became a real threat to biodiversity in Romania. Furthermore, Law 1 of 2000 and 247 of 2005 provided legal rights for use and administration of forests and alpine pastures by pr ivate owners and local associations. This situation generated increased concerns over habitat fragmentation and preservation, especially on forest lands because of land ownership ch anges (UNDP 2001). The literature on parks and protected areas management underscores that successful management endeavors and environmental sustainability depend on the cooperation and support of local communities (Brandon & Wells, 1992; de Beer & Marais, 2005; Hall, 2004 ). Actively involving local communities in the management of protected areas has been associated with an increased awareness in terms of the benefits of biodiversity conservation, a more responsible use of resources, and ultimately enhanced livelihoods and welfar e of local people ( Pagdee et al 2006). The lack of community interest and participation in biodiversity conservation has been discussed as a major constraint for natural resources management in Romania. This situation has

PAGE 137

137 been primarily attributed to a low sense of community and collective responsibilit y that characterizes Romanian rural com munities (PJB Associates, 2006). Furthermore, r esearchers participation in local decision making in Romania to a history of top down, centralized system p olitical system ( Oostenbrink & Kosterink, 2005 ; Popescu, 1993 ). Based on a study conducted in Retezat National Park to better understand the co management framework employed by the park, van Hal (2006) emphasized a lack of common interest in conse rvation i n the area. Similarly, van den Kuijs & Bergh (2006) emphasized a lack of conservation attitude in communities surrounding Retezat National Park as well as a lack of care for the environment. These two studies used a qualitative approach in understanding th e park administrative efforts, depicting primarily the voices of the park administrative staff and park stakeholders (e.g. lodge owners, travel operators, mayors of local communities) but not directly the views of citizens from local communities. To date, little is known about the they might differ across age groups. neighboring environ ments, this study seeks to explore the affective and attitudinal responses of people living adjacent to a national park towards their neighboring natural environments. Considering lack of care for the environment has been previously questioned, the connect ions people living adjacent to the park have with nature and their attachments to the park were assessed. Furthermore, a better understanding of attitudes towards conservation and p ro environmental civic engagement is warrant. Thus, measures of attitudes t owards conservation,

PAGE 138

138 p ro environmental civic engagement and perceived environmental responsibility were assessed to capture attitudinal response. Roberts and Simpson (1993) viewed the communism period as having a strong influence relations with social and natural environments. Thus, to better depict the current situation, age group differences were assessed in terms of connection to nature, park attachment, conservation attitudes, and attitudes towards pro environm ental civic engag ement and perceived environmental responsibility. This study questions the extent to whic h young adults (18 to 30 years) which h ave lived primarily in a post communist period show diffe rent response patterns towards nature as compared with middle age adult s ( 31 to 50 years) and older adults (51 to 8 0 years) Three multivariate analysis of variance ( MANOVA) models were assessed. The first model captured an affective response to nature and the park, the second model assessed attitudes towards conservation, an d the third model captured behavioral attitudes related to p ro environmental civic engagement and perceived environmental responsibility. No hypotheses are made regarding group differences due to the unique context of the study and lack of empirical preced ents on which to base them. A qualitative exploration and introspection depicting the current situation as well as historical insights complement the quantitative findings. Literature Review tions are a result of the attitudes towards the behavior (behavior belief), perceived social pressure to engage in a particular behavior (normative belief) and strength of the desire to comply with the norm. of affect in shaping behavioral intentions. The environmental behavior literature has employed several constructs to assess pro environmental behavior intentions. Affective as well as attitudinal related constructs were employed. Connections to nature and place attachments have been constantly linked to pro environmental

PAGE 139

139 behaviors, due to an affective component stimulating motivation and commitment for engagement. To better account for changes in pro environmental behavioral intentions, general measures of attitudes have been employed, as well as measures of attitudes towards specific behaviors. Those attitudes that strictly relate to the expected behavior have been found as having a higher predictive power in terms of behavioral intentions. Consequently, t his study assessed general attitudinal measure, and attitudes strictly related to the expected behavior ( pro environmental civic engagement ). The following para graphs discuss the five constructs employed in this study. Connection to Nature Connection to nature is often portrayed as a sense of oneness with nature. The connection people have with nature has been discussed as being an important factor in explaining environmentally responsible behavior (Schultz et al., 2004). The feelings people have for nature shape personal identity, values and attitudes, and ultimately influence behavior (Driver & Ajzen, 1996; Mannell, 1996; Roggenbuck & Driver, 2000). The main ass umption made in regard to nature, they intrinsically care for it and act to protect and preserve it (Pennisi, 2007). Pennisi (2007) depicted values and identity as the two core aspects of connection to nature. The interconnectivity between personal identity, values, attitudes, and ultimately behaviors has been previously accounted for (Clayton, 2003; Clayton & Opotow, 2003; Hitlin, 2003; Vaske & Donnelly, 1999). In t he values attitudes hierarchy, values are considered to be more stable and 1999). Therefore, connection to nature as a reflection of personal identity and values rep resents an important dimension for explaining pro environmental behavior.

PAGE 140

140 Several scales measuring connection to nature have been developed. These scales include the Connectedness to Nature Scale (CNS) (Mayer & Frantz, 2004), single item Venn diagram meas ure of Inclusiveness of Self (Schultz, 2002), and a measure of environmental identity (Clayton, 2003). These measures capture the construct as unidimensional, fact that has been contradicted by recent work done by Pennisi (2007). Furthermore, Perrin & Bena ssi (2009) measure of emotional connection. Thus, the scale developed by Pennisi (2007) was employed. Louv (2006) underscored a weakened connection of the younge r generation with the natural environment, due to technological advances as well as lack of interactions with nature. Furthermore, several authors argued that, as societies developed, humans got more distant, both psychologically and physically, from the n atural world (Sheldrake, 1999; Vining, 2003). Attachment to the Park Extensive literature exists that focuses on place based attachments, sense of place and place attachment (Altman & Low, 1992; Kyle et al., 2004; Williams et al., 1992; Williams & Vaske, 2 003). Place attachment is defined as an emotional bond between people and places (Altman & Low, 1992). The personal bond between people and settings has constantly gathered academic attention due to the meanings, values, beliefs, and symbols that people as sociate with places and their relevance for identity and other personal benefits ( Brown & Perkins, 1992; Stedman, 2003 ; Tuan, 1974 ). The relationship between people and places has been discussed as relying on subjective evaluations of social features (Wol doff, 2002), physical features (Stokols & Shumaker, 1981), or both (Brown & Perkins, 1992; Mesch & Manor, 1998 ; Riger & Lavrakas, 1981). Lawton (1990) emphasized that highly attached residents are often older and have stronger relations and interactions wi th community members Brown & Perkins (1992) viewed place attachments as

PAGE 141

141 being nourished by interactions with the social and natural environment. Rollero & De Piccoli (2010) underscored the role of age for place attachment, but at the same time argued tha t such relationships might be mediated by length of residence. However, other studies show that feelings of attachment to place can develop for physical spaces with which individuals have had recent contact (Bonaiuto et al., 1999; Harris, Brown, & Werner, 1996: Mesch & Manor, 1998 ). Positive relationships have been found between place attachment and attitudes toward relevant environmental concerns or local conflicts over land use management (Cross, 2003; Vorkinn & Riese, 2001; Williams et al., 1992; Brehm e t al., 2006 ) and specific environmental behaviors (Bott, Cantrill, & Myers, 2003 ; Vaske and Kobrin, 2001 ). The social action literature provides consistent evidence that strong attachment correlates to increased motivation for action, local participation a nd civic behaviors with direct implications for the well being of the social and natural environment (Lewicka, 2005; Van Vugt, 2001). P lace attachment has been assessed at different level s of scale. Vorkinn & Riese (2001) measured place attachment as a ge neral measure of attachment to municipality, as well as a measure of attachment to five areas neighboring these local municipalities. Cross (2003) talked about attachment to the bioregion and its attributes, for example a local park. Manzo (2005) argued th at the types of places individuals find important represent a broad range from physical settings, to built environments such as houses, streets, certain buildings, and non residential indoor settings, to natural environments such as lakes, parks, trails, f orests, and mountains. Considering that fewer studies have been conducting looking at the attachment people have towards a park neighboring their communities, this study employed a measure of attachment to a park and not an overall attachment to a place of residence.

PAGE 142

142 Notions of dependence and identity have been associated with attachment to a place. Thus, resource based sociologist s primarily employed a measure of place attachment that captured two dimensions of attachment: place dependenc e and place ident ity ( Kyle et al. 2004 ; Williams et al., 1992 ). Place dependence has been discussed as centering on a functional attachment to place, while place identity reflecting an emotional attachment that relates to the symbolic meaning of the place. Conservation Att itudes Jennings & Nickerson (2006) defined attitudes as an enduring predisposition toward feel, and behave. The importance of understanding attitudes evolve s from the relationship between attitudes and behavior (Ajzen, 1991). Researchers have adjacent to protected areas. The perceived benefits achieved from conservation are often acknowl (2005) confirmed that perceived benefits from conservation determined strong positive attitudes. However, Walpole & Goodwin (2001) identified that even though some of the respondents were highly benefiting from conservation through tourism, a positive relationship between the attained benefits from tourism and support for conservation was not observed as expected. Thus, the literature on people park relationships suggests t he importance of giving consideration to the type of experiences residents have with the park and its authorities. protected area could play an important role in shap ing conservation attitudes (Kappelle, 2001; Robertson & Lawes, 2005). Ormsby & Kaplin (2005) identified interaction and benefits as two

PAGE 143

143 conservation in general ha ve been viewed as being quite heterogeneous in nature ; being representative of group affiliation as well as socio demographic characteristics (Infield & Namara, 2001; Weladji, Moe, & Vedeld, 2003). Different measures have been employed in the literature to assess community attitudes towards conservation. Multidimensional scales have primarily been used, incorporat ing conservation awareness dimensions, perceptions on conservation benefits, and management considerations ( Infield & Namara, 2001; McFarlane & Boxall, 2003 ; Nguyen, 2007 ). Attitudes towards Pro environmental Civic Engagement Kollmuss & Agyeman (2002) defined pro environmental behavior as behavior consciously s Pro environm ental behaviors can take different forms, and accordingly, different factors can differently shape or influence each specific type of behavior. The literature makes a distinction between political behaviors, as compared with behaviors that relate with cons umption (e.g. energy saving, green consumer behavior s ) (Aoyagi Usui, Vinken, & Kuribayashi, 2003). Stern (2000) provides a more detailed classification of pro environmental behaviors, identifying several different types of pro environmental behaviors such as: environmental activism, environmental citizenship behavior (active involvement in environmental issues, public participation that with an influence on policy making, decision making) and private sphere environmentalism (e.g. consumer behavior, automobi le, energy use, green consumerism etc.). Stern (2000) underscored that environmental citizenship behavior is primarily influenced by general environmental attitudes, beliefs, and perceptions of costs and benefits. S ocio demographic characteristics were gen erally found to have a weak direct influence on behavior, but the literature shows that personal characteristics do have a significant influence on values and attitudes ( McFarlane & Boxall, 2003) Theodori & Luloff (2002) found that highly educated,

PAGE 144

144 young, with higher income respondents, and those with liberal political ideologies, were more likely to have proactive positions on environmental issues. Furthermore, those with proactive positions on environmental issues were reported as being more pro environm entally engaged. behaviors of interest in this study, relates to what the author describes as environmental citizenship behaviors. Primarily, the focus is on behaviors tha t relates to public participation and active involvement in local environmental concerns. Perceived Environmental Responsibility The literature suggests that people with a greater sense of personal responsibility are more likely to engage in environmental ly responsible behaviors (Kollmuss & Agyeman, 2002). obligation or duty to take actions against environmental deterioration in general, or specific environmental concer ns. Perceived environmental responsibility has been identified in several behavioral models as a significant determinant of behavioral intentions (Ajzen, 1991; Hines et al., 1987; Triandis, 1980). A strong relationship has constantly been found between env ironmental responsibility and behaviors that have implications for the well being of the environment (Garling et al., 2003; Van Liere & Dunlap, 1978; Vining & Ebreo, 1992). Garling et al. (2003) emphasized that in order for a behavior to be performed, awar eness of responsibility to perform the behavior stimulates a moral obligation to engage in performing the behavior. The social context has been viewed as shaping ascribed responsibility for pro environmental engagement. Garling et al. (2003) identified pro environmental behavioral intentions as being a reflection of personal norms, ascribed responsibility, and awareness of consequences for oneself, others, and for the biosphere. Therefore, perceived personal

PAGE 145

145 responsibility has been primarily suggested as be ing a strong predictor for behaviors that have collective implications (e.g. participation in public meetings), as compared to behaviors that relate to self interest (e.g. reducing water use). Barr & Gilg (2007) found the social context and alternative pe rceptions of trust and responsibilities within localities as having a mediating role in shaping public understandings of sustainability and environmental issues. In this study, perceived environmental responsibility is primarily relevant, considering the R omanian social context that has been shaped by historical events that primarily emphasized public/governmental responsibility as compared to collective or personal responsibility. Methods Study Area Retezat National Park (RNP) was designated in 1935, bein g the first national park established in Romania. In 1979, RNP was declared an International Biosphere Reserve under the UNESCO Man and Biosphere program and in 2004 RNP received its Protected Area Network (PAN) Certification. Retezat National Park is loca ted in the southwestern Carpathians, and the total surface of the park is 38,138 ha (RNP Management Plan 2008). Within the park, there are more than twenty mountain peaks 2,000 meters or higher, in addition to eighty lakes of glacial origin. There are mor e than 1,100 species of plants, over 50 species of mammals including roe deer, chamois, lynx, bear, and otter and 168 recorded bird species including the golden eagle. RNP was the first park in Romania with a management system in place (van Hal 2006). T he co management framework initiated by RNP is perceived as being a model for other protected areas in Romania. A large portion of the park area (17,564 ha, 46%) is owned by the state, while local associations own the remainder of the land (20,574 ha). Of the 43 villages

PAGE 146

146 adjacent to the park, 26 villages have grazing rights to alpine meadows, with their rights administered either through community based associations or local councils of the five communes to which the villages belong (Kuijs & Bergh 2006). A commune is an administrative division in Romania encompassing one or more villages that share similar economic, socio cultural, geographic and demographic conditions. The total population of these five communes was estimated at 14,006 inhabitants. Commun ities rely on park resources primarily for grazing and the use of other natural resources such as wood, non timber forest products, mushrooms, and medicinal plants. The major management concerns, as it relates to conservation, are related to overgrazing of the pasture areas and illegal wood harvesting (RNP Management Plan 2008). Measurements Participants in this study were asked to express opinions on a series of questions about nature and Retezat National Park. The constructs used for this study included : connection to nature, attachment to the park, conservation attitudes, attitudes towards p ro environmental civic engagement and perceived environmental responsibility. In addition to these constructs, several socio demographic characteristic variables (a ge, gender, education, family status, length or residency, income) were included. A general measure of connection to nature was measured using 18 items previously tested by Pennisi (2007). The items included captured five dimensions of connection to nature : admiration (3 items), spirituality (3 items), identity (4 items), sorrow (2 items), restoration (3 items), and fear (3 items). The eighteen items were measured on a five point Likert scale ranging from 1 to 5, where 1 = Strongly disagree; 3 = Neutral; 5 attachment to Retezat National Park was measured using nine items on a five point Likert scale ranging from 1 to 5, where 1 = Strongly disagree; 3 = Neutral; 5 = Strongly agree. The items

PAGE 147

147 captured two dimensions of attach ment: place dependence and place identity (Kyle et al., 2004; Williams et al. 1992; Williams & Roggenbuck, 1989) 18 items that were adopted from previous studies cond ucted by Infield & Namara (2001), Nguyen (2007) and McFarlane & Boxall (2003). These items capture three dimensions of conservation attitudes: conservation awareness (6 items), conservation benefits (7 items), and management considerations (5 items). The i tems were measured on a five point Likert scale ranging from 1 to 5, where 1 = Strongly disagree; 3 = Neutral; 5 = Strongly agree. Attitudes towards p ro environmental civic engagements were measured using 12 items adapted from previous work done by Garling et al. (2003) and Halpenny (2006). The items were measured on a five point Likert scale ranging from 1 to 5, where 1 = Not at all effective; 3 = Sometimes effective; 5 = Always effective. Perceived environmental responsibility was examined on the basis of environmental responsibility (Garling et al., 2003). The question was assessed using a five point Likert scale ranging from 1 to 5, where 1 = Totally disagree; 3 = Neither agree nor disa gree; 5 = Totally agree. Individual level socio demographics included age (1=18 to 30 years; 2=31 to 50 years; 3=51 to 80 years), gender (1=Male; 2=Female); number of children under 18 years (frequency from 1 to 5 children), educational attainment (1=Eleme ntary school or less; 2=High school graduate; 3=Post high school education), length of residence (open ended) and monthly household income (1=Almost no income to 7=More than 2,000 RON; ~US $661). Data Collection R ural communities adjacent to Retezat Nation al Park belong to five communes encompassing 43 villages with a total population estimated at 14,009 adult residents. Two villages dissimilar in size from each commune were selected, to assure representation of each

PAGE 148

148 commune in the final sample. Nine commun ities (one commune had only one village) were selected for this study using multistage random sampling. The nine villages selected have a population of 4,232 persons residing in 1,159 private households. Multiple strategies to assure face and content valid ity were utilized including a review by a panel of 3 survey experts, translation and retranslation by two Romanian natives and a review of the survey by park staff. Cross sectional data was collected from 260 residents during June October 2009 using face to face interviews (68% response rate) and mail survey (9% response rate). Systematic sampling with a random start was used to select participants for face to face interviews. Due to small community sizes and the challenge of finding people at home, gener ally, every other household was selected. Adults 18 or older were asked to participate, and questions were asked and responses recorded by an interviewer. In addition, 230 mail surveys were sent to four communities where residents were not at home or inter viewer safety was an issue. Data Analysis Data analyses were performed following four stages. First, descriptive statistics were computed for the variables used in the study using the Statistical Package for the Social Sciences (SPSS) version 18.0. Second, the data collected was screened and the critical assumptions underlying the statistical techniques employed were assessed. Third, individual items were examined using Confirmatory Factor Analysis (CFA) and the measurement model for constructs included in the study was estimated using MPLUS version 5.21 to determine how well the indicators captured their specific constructs and the ability of the respondents to differentiate between constructs (Hair et al., 2006). CFA was assessed using the WLSMV (weighted least squares mean and variance adjusted) method of estimation, method recommended for categorical ordinal data (Muthen et al., 1997). Forth, three multivariate analysis of variance (MANOVA) models were assessed.

PAGE 149

149 The fit of the measurement models were ass essed using multiple criteria. T he chi square test of model fit divided by the degrees of freedom was used as a reference criteria supplemented by the Root Mean Square Error of Approximation (RMSEA), Wei ghted Root Mean Square Residual (WRMR) Comparative F it Index (CFI) and Tucker Lewis Index ( TLI ) RMSEA values equal to or less than .06 is indicative of a good model fit values between .08 and .10 indicate acceptable mod el fit, and values higher than .1 0 are considered to be indicative of poor fit (Browne & Cudeck, 199 2; MacCallum, Browne & Sugawara, 1996). CFI and TLI values equal to or greater than 0.95 also indicate good model fit (Hu & Bentler, 1999) The criterion for WRMR is a value less than 1.00 (Yu, 2002 ). After each factor was confirmed by meeting the minimum standards listed above, a score for each factor was calculated using the average of the attributes loaded on the factor. Three multivariate analysis of variance (MANOVA) models were performed to test for any significant age group differences i n the underlying dimensions. In the MANOVA procedure, the dependent variables for the first model capturing an affective response to nature and the park were two factors: connection to nature and attachment to the park. The dependent variables for the seco nd model assessing general conservation attitudes were three factors: conservation awareness, conservation benefits and conservation management. The third model included two dependent variables capturing attitudinal response towards pro environmental civic behavior: attitudes towards p ro environmental civic engagement and perceived environmental responsibility. Age was the independent variable. Taking into consideration that length of residence and education has been previously linked with environmental res ponse the two variables were included as covariates to control for their effects.

PAGE 150

150 Results The average age was 45 years, with one quarter (23.9%) being between 18 and 30 years and about one third (36.0%) being between 31 and 50 years. Respondents who were b etween 51 and 80 years were most represented (40.1%). Fifty four percent were males. The average residency length was 37 years. The majority (65.1%) were married or partnered, and on average, the number of adults per household was three. Almost one third ( 31.6%) indicated high school as the highest level of education attained, 14% had some college or a college degree and 14 % had an elementary school education or less. About one third (33.8%) reported monthly household income between 1,000 and 1,999 RON (abo ut US $330 $660) and 30.6% indicated a monthly household income of more than 2,000 RON (about US $661). The majority of respondents (70.6%) indicated that they did not have any ownership or land use rights in Retezat National Park. To determine convergent validity individual constructs and their measurements were examined using confirmatory factor an alysis (CFA). Marsh, Craven, & Debus (1991) underscored that when a model has been misspecified (poor model fit), the researcher has to respecify the model. O ne way to respecify the model is to delete indicators and the other option is to allow errors to correlate, and decisions should be supported by theory or rationale (Joreskog 1993). Thus, attributes with factor loadings lower than .40 were excluded from f urther analysis (Hatcher, 1994). Attributes correlating with more than one dimension were also alphas above the recommended level of .70 (N unnally & Bernstein 1 994). As a result of initial CFA tests, several items in various factors were dropped due to low factor loadings. Ultim ately, for each construct, the items retained were substantive in size and had significant factor loadings. Connection to nature was not found to have a hierarchical

PAGE 151

151 structure Three factors were eliminated due to low factor loadings and low reliability coefficients. The reliability coefficients for the three factors were below the recommended level of .70 ((Nunnally & Bernstein 1994) ran ging from .42 (nature sorrow) to .67 (nature fear). Furthermore, inter factor correlations between nature restoration and nature identity (.87), and between nature restoration and nature spirituality (.93) were above the suggested criteria of .85 (Kline, 2 005) and suggesting inability of the sample to differentiate between factors Therefore, scale items were collapsed and ultimately seven items were retained as a one dimensional measure of connection to nature. Similarly, the correlation between park ident ity and park depende nce was high (.89) suggesting weak differentiation. Consequently, scale items were collapsed and one measure of park attachment retained including seven items. The fit indices for the measurement model for each construct are provided in Table 5 1. All item (indicator) loadings as well as descriptive statistics are presented in T able 5 2 Evidence of nded level of .70 (Nunnally & Bernstein 1994 ), ranging from .73 (con servation awareness) to .90 (park attachment) and composite reliability (CR) above the reco mmended level of .70 (Fornell & Larker 1981), ranging from .86 (conservation benefits) to .94 (park attac hment). Also included in T able 5 2 is the average variance extracted (AVE) estimate with recommended levels of .50 or higher indicating convergent validity for Larker, 1981). All values exceeded the recommended level ranging from .60 (attitudes pr o environmenta l civi c engagement ) to .71 (park attachment). These findings reveal that the proposed measurement model satisfied all the psychometric requirements, thus the measures were adequate for further analysis.

PAGE 152

152 A composite score for each construct was calculated based o n the measurement results. Descriptive statistics for all variables for the three age groups of interest in this study are presented in T able 5 3. Three MANOVA models were assessed in order to capture age d ifferences in terms of affective response to natur e and the park, attitudinal response towards conservation, and a civic behavior attitudinal response (T able 5 3) The assumptions of the MANOVA procedures were assessed and the data was evaluated as being suitable for analysis. The descriptive statistics showed respondents had more neutral responses in terms of their perception of environmental responsibility ( M = 3.39; SD = 1.602) and attitudes towards conservation management ( M = 3.86; SD = .992), while more positive responses were observed for conservat ion awareness ( M = 4.83; SD = .396) and connections with nature ( M = 4.44; SD = .691). The results of the MANOVA procedure finds that residents in different age groups had statistically significant differences in their affective response towards nature and the park (Model 1: p < .05) and attitudes towards pro environmental civic behavior (Model 3: p < .05). No statistically significant group differences were observed as it relates to general attitudes towards conservation (Model 2: p > .05). Furthermore, an assessment of the group differences for each individual construct revealed s ignificant group differences in terms of connection s to nature ( p < .05 ) park attachment ( p < .05), conservation awareness ( p < .05), conservation management ( p < .05), attitudes towards pro environ mental civic engagement ( p < .05 ) and perceived environmental responsibility ( p < .05). The last four constructs were very close to the limit of statistical significant difference (.05). The post hoc tests with Tukey statistics reveal ed that respondents between 18 to 30 years differed significantly in their connections to nature and attachments to the park from those 31 to 50 years ( p < .05) and 51 to 80 years ( p < .05), the younger age group reporting lower levels of

PAGE 153

153 connection to nat ure and park attachment. No significant differences in connections to nature and years ( p years and 51 and 80 years ( p < .05) in terms of conservation awareness was also observed No significant age 18 to 30 years also differed significantly in their atti tudes towards p ro environmental civic engagement ( p < .05) as compared with older adults (age 51 to 80), lower levels of attitudes toward pro environmental civic engagement being reported by the younger group. Furthermore, the young adults (age 18 to 30) d iffered significantly in their perceived personal environmental responsibility ( p < .05) as compared with older adults (age 51 to 80), with the younger generation scoring higher on perceived personal environmental responsibility. To better control for infl uences on the dependent variables, education and length of residence were included as covariates in the models. The group age differences on affective response to nature and the park remained statistically significant even after controlling for education a F = 7.640; p < .05). The age group differences in terms of connection s to nature ( p < .05 ) and park attachment ( p < .05) were statistically significant. The age group differences in terms of conservation attitu des remained = .977; F =. 785; p > .05). Education was found to provide a better explanation for differences in attitudes towards conservation management ( p < .05), and length of residence better capturing differences in terms of conservation benefits ( p < .05). The age group differences on attitudes towards behavior became insignificant when controlling for level of education and length of mbda = .969; F = 1.671; p > .05). Education was found to better explain the

PAGE 154

154 F = 4.991; p < .05). More specifically, level of education providing a better explanation of differences in percepti ons of environmental responsibility ( p = .002). The correlation between age and level of education was significant and suggested young adults had higher levels of education( r = .248), while the correlation between age and length of residence was positive and significant ( r = .616). Discussion This study investigated the implications of age for affective and attitudinal environmental response of adults residing in rural Romania. Age differences were assessed in terms of affective environmental response capt ured by connections with nature and attachments to the park. Attitudinal environmental response included measures of general attitudes towards conservation as well as attitudes towards p ro environmental civic engagement and perceived personal environmental responsibility. There age groups were defined: young adults (age 18 to 30) middle age adults (age 31 to 50) and older adult s (age 51 to 80). The findings of this study are threefold. Firstly, age of the residents was found to be an important factor in even after controlling for length of residence and level of education Young adults reported lower connections to nature and attachments to the park as compared with their older c ounterparts. The affective environmental response did not differ among middle age adults and older adults. Secondly, level of education was found to better explain attitudinal response towards pro environmental behavior. More specifically, differences in a ttitudes towards conservation management were found to be better explained by level of education. Furthermore, level of education was found to play a significant role in explaining differences in perceived gher levels of education reporting higher

PAGE 155

155 levels of personal environmental responsibility. Thirdly, length of residence better captured differences in terms of conservation benefits. Older residents were found to have stronger bonds with nature as well as with the park neighboring their communities. This finding is partially supported by previous studies which underscored that residents highly attached to their place of residence tend to be older (Lawton, 1990 ; Pretty et al., 2003; Rollero & De Piccoli 20 10), but little is known how connections to nature relate to age. Concerns over lack of connections with nature of the younger generation have been raised (Louv, 2006), but less is known about the Romanian context and what factors determined such developme nts. Assertions that length of residence might have a greater predictive power for attachments were not supported in this study where a measure of attachment to the park was employed. This supports prepositions implying that feelings of attachment to place can develop for physical spaces with which individuals have had a recent contact (Bonaiuto et al., 1999; Harris, Brown, & Werner, 1996 ; McCool & Martin, 1994) Furthermore, the age of the residents was not identified as an important factor in explaining t he attitudinal environmental responses of the residents after controlling for education and length of residence McFarlane & Boxall (2003) emphasized that socio demographic characteristics while lacking ability to directly predict behavior, do relate stron ger with values and attitudes. In this study, age was primarily found to provide a better explanation for differences in affective response, connections to nature and park attachments, concepts that have been constantly discussed as shaping personal values and identity ( Brown & Perkins, 1992; Roggenbuck & Driver, 2000). While age was not found as a strong factor in explaining differences in attitudes towards conservation and pro environmental civic behaviors, education had better ability to captur e differen ces in attitudes. Thus, attitudinal environmental responses

PAGE 156

156 towards pro environmental civic behaviors were found to be better explained by an achieved social status, level of education. More specifically, the more highly educated tended to be more critical towards current management approaches towards conservation, but at the same time reported higher levels of perceived personal environmental responsibility. Beggs, Hurlbert, & Haines (1996) concluded that individual s with higher levels of achieved social s tatus (higher levels of education and income) tend ed to have higher expectations for their community and its leaders, and therefore, they may tend to be more critical in their assessment of the community environment. This study supports such assertions, ed ucation being found to better explain differences in attitudes towards conservation management. The results also showed that length of residence better explains differences in conservation benefits, those that lived in the community for a shorter period of time reporting fewer benefits from conservation. The findings of this study were also interpreted in light of comments made by participants in the in depth interviews conducted with 24 community members representing the nine communities adjacent to Reteza t National Park. The qualitative analysis revealed a series of themes that depicted major dynamics that relate to the social and natural environment in the community. The narratives captured a constant emphasis on the amount on changes that occurred over t he years, primarily social changes that ultimately shaped environmental connections. The respondents emphasized that during communism there was more unity and care in the community and also more care for the environment. The respondents emphasized the str onger connection and care for the environment of previous generations, and concerns were raised over the environmental conscious ness of the younger generation. d to have this consciousness to protect the in a forest and look to cut the m ost beautiful tree you should cut the one that is the

PAGE 157

157 one that does look that good the one that is bended, faulty, the one that is dry a we entered this century of high pace, the young generation seems to lack respect for nature, the respect for nature that is used to be is not there anymore, I think. (Miruna) Furthermore, social dynamics are perceived as currently threatening the young generation and their interest in the locale. Concerns were raised over their stability and lack of perspectives in the community. This is what worries me very much, this instability and job insecurity and when you see the young generation worried, when you see them disoriented, you ask yourself, who should we depend on, because you have to depend on th em, no? (Ioan) T he area curre ntly lacks the economic activities that use d to nourish the connections between people and nature. Agriculture and livestock represented for a quite long period of time the main sources of income in these areas. Older generat ions had more interactions with the park through different activities, including agriculture, livestock, and properties that were close to the park, to get to t hem they had to cross the park. Respondents emphasized that since childhood they were exposed to such activities and th ey got to know every trail in the park Currently agriculture and livestock raising is done sporadically and only for family benefit not nec essarily as a source of income. The orientation of the younger generation for more benefici al activities was underscored, agriculture and livestock not being very promising. anymore, people became disinterested, if they take the mountain, let them take it, because anyway is not ours, or because anyway we are not going there anymore. (Verde) will soon be lost.

PAGE 158

158 and sit there with the sheep (Stefan) The quantitative analysis revealed more homogeneous attitudes towards conservation and the respondents tended to be more neutral in their assessment of conservation benefits and conservation management, with level of educa tion capturing differences in attitudes towards conservation management and length of residence differences in perceptions of conservation benefits. The narratives constantly argued the importance of benefiting from the park, and the contribution that the park could have for the well being of the community. Disappointment has been expressed by the respondents due to lack of benefits from lack of employment opportunities for the local residents to the weak benefits from tourism The narratives depicted a cu rrent lack of services for visitors, the tourism infrastructure being rudimentary while prior to 1990, it used to be much more welcoming. Before the revolution I feel that there was more care for Retezat because there were some services, to welcome visi tors, for camping, accommodations, in there anymore, there are not, the population has no more animals, reason for which there are no more sheepfolds, sheepfolds are in my opi nion very interesting and quite pleasa nt for the eye of the tourist. (Veronica) There seems to be an increased awareness of the importance of the park and middleaged and older residents seem to value the fact that the park is protected and taken care of. T he quantitative analysis depicted a high level of conservation awareness in the area. Much of the awareness comes from the admiration and pride of living adjacent to the park, feelings of happiness to see tourists coming and visiting the park and their com munities being expressed. However, less and less tourists are coming due to a lack of infrastructure. Thus, the respondents underscored the ne ed to make the park efficient and relevant for neighboring communities. An overwhelming concern was expressed over the lack of benefits from the park, that ultimately cultivated feelings of disconnect between the population and the park. The respondent s strongly

PAGE 159

159 challeng ed the social role of the park and accentuated desires for stronger connection s and benefit s for lo cal communit ies Some concerns are generated by the restrictions imposed by the park, which seems to lack support from local residents. Much of the current hope for the future d great concerns emphasized that if economic opportunities start to increase (including tourism), there will be much more interest in reconnecting with the local e. The social transition currently occurring in these communities is not unique. The literature on rural communities constantly depicted societal changes associated with a transition period from an agricultural society to a modern society. Pretty (2002) em phasized the weakened links between people and nature due to modern agriculture and industrialization, which has separated people from nature. In the industrial age, communities have been found to experience disconnect from nature and much of the stories, memories and language about land and nature are missing (Pretty, 2002). Furthermore, rapid modernization in both developing and industrialized countries has been associated with weaker sentiments of ownership, an inclination to care, and a desire to take a ction for the collective good. Kellert (2005) underscored that modern societies are characterized by lack of quality and quantity of human interaction with the natural environment. Biodiversity conservation has been found to also play a role in reducing th e amount of interaction with the natural environment. Barrow & Fabricius (2002) discussed the social costs of biodiversity conservation, arguing that people surrounding conservation areas become increasingly alienated from nature and the benefits received from conservation do not always outweigh the costs.

PAGE 160

160 Conclusions T his study finds affective response to nature and the park to be indicative of age differences that relate to environmental access changes in the community. T he amount of interaction as well as benefit from natural resources impacted feelings people hold for their natural environments. From a practical perspective, this study underscores the need to reconnect the younger generation with the natural environment by providing opportunities for i nteraction as well as showing them perceived benefits from conservation. The affective connections with nature and the park of the young adults residing in communities neighboring a national park in Romania are weak as compared with their older counterpart s. Core values that are intrinsic to feelings of connection and attachment to neighboring environments seem to be have been impacted by local social and environmental dynamics and changes. Primarily lack of interaction with the natural environment seems to have generated weaker relationships with nature and the park. Agriculture and livestock have for a long period of time sustained the connection between the residents and the natural environment. The major threat that such developments impose relates to lo ss of the population base, a fact that was constantly mentioned by the residents as a threat to the well being of the community. Disappointment with the current management and lack of benefits has been expressed, with those with higher levels of education being more critical towards the current situation but at the same time reporting higher levels of personal environmental responsibility. Thus, from a management perspective, highly educated individuals, who also tend to be younger, represent a group with p otential for involvement in park management and community development efforts. Connections and attachments to neighboring environments have been viewed as a reservoir for motivation and commitment to places ( Lewicka, 2005 ). Furthermore, l ocal partic ipatio n and civic behaviors have often been connected to an affective link with the environment. Thus, much

PAGE 161

161 of the current lack of local participation might be explained by weak affective connections with the neighboring environment of a generation that lacked t he interactions with the natural environment their counterparts did. Place attachments were found to be nourished by daily interactions with the environment, natural and social environment (neighbors, neighboring parks and other physical areas) (Mesch & Ma nor, 1998; Sampson, 1989; Werner et al., 1993). Thus, there is an emerging need to create programs for young adults that provide opportunities for interaction with neighboring environments and ultimately allow residents to know and truly value the communit y they are part of. Projects targeted at reconnecting the younger generation with the locale could relate to agriculture and traditional activities as well as tourism. Increased importance has to be give n to the local communities. T he natural environment, including the park could ultimately play an increased role in maintaining the social infrastructure in the area. Currently the residents see the park as a precious asset but unfortunately nobody gets to enjoy and truly appreciate it. The lack of benefits from the park was constantly emphasized by the respondents in their narratives. Restrictions imposed by the park are perceived as limiting the quality and quantity of human interaction with the neighboring natural environment as well as impacting current sentiments toward s the park. The respondents tended to be critical in their assessment of the management approach as well as perceived benefits. A wo rd constantly expressed by the respondents in their narratives was restrictions ; the park has increased its area over the last couple of years, increase s that has also resulted in increased restrictions and regulations of the use of resources in the park area and in the buffer zone. Thus, there seems to exist a n emphasi s on increasing benefits and making the pa rk relevant for the local communities. The park management should consider the neighboring communities

PAGE 162

162 in their decisions, otherwise much of what currently represents communities and their culture and traditions of natural connections will be lost. Table 5 1. Goodness of fit indices for each construct CFA Model 2 / df RMSEA WRMR CFI TLI Connection to nature Original 96.73/36 0.083 0.883 0.983 0.988 Respecified 30.09/10 0.091 0.600 0.991 0.994 Park attachment Original 51.479/17 0.091 0.646 0.9 86 0.994 Respecified 24.349/9 0.083 0.460 0.994 0.997 Conservation attitudes Original 257.39/61 0.115 1.355 0.786 0.902 Respecified 60.16/27 0.071 0.780 0.964 0.982 Attitudes pro environmental civic engagement Original 171.71/25 0.156 1.21 9 0.911 0.965 Respecified 7.105/5 0.042 0.289 0.997 0.998

PAGE 163

163 Table 5 2. Reliability and validity of the measurements (CFA m odels) Factors and items Mean SD CR AVE Connection to nature 1 .871 .932 .665 I am connected to nature much like I'm connected to my family 4.40 .974 .88 Nature is a huge part of who I am 4.40 .932 .80 I often feel a sense of oneness with the natural world around me 4.28 1.028 .93 My feelings for nature have influenced my spiritual beliefs 4.12 1.125 .90 When surrounded by nature, I feel at peace 4.91 .327 .68 Listening to the wind go through the trees calms my mind 4.35 1.051 .74 When I'm alone in a n atural area, I have this feeling of complete calm 4.51 .841 .64 Park attachment 1 .903 .944 .712 Retezat National Park means a lot to me 4.54 .784 .91 I am very attached to Retezat National Park 4.18 .992 .94 Retezat National Park is very important to me 4.33 .910 .95 I identify strongly with the Retezat National Park 3.78 1.178 .85 I get many personal benefits out of living near Retezat National Park 3.72 1.280 .62 I enjoy living near Retezat National Park 4.54 .750 .84 I get lots of satisfaction out of living near Retezat National Park 4.17 .996 .75 Conservation awareness 1 .727 .880 .650 It is important to have the Retezat National Park for the survival of various plants and animal species 4.91 .415 0.87 It is necessary to set aside some land for the protection of plants and animals 4.89 .381 0.89 Retezat National Park is our country's pride 4.87 .481 0.84 The illegal cutting of trees in the park should be strictly regulated 4.69 .813 0.63

PAGE 164

164 Table 5 2. Continued Factors and items Mean SD CR AVE Conservation management 1 .834 .881 .651 Retezat National Park is managed successfully for the benefit/enjoyment of future generations 4.16 1.123 0.84 Retezat National Park is managed successfully for a wide range of uses and values, no t just tourism 3.96 1.220 0.79 Retezat National Park management does a good job at protecting the natural resources in the park 4.19 1.066 0.90 The citizens from the communities around the park have enough say in how the park is managed 3.16 1.30 4 0.68 Conservation benefits 1 .767 .859 .606 My community benefits from being near the Retezat National Park 3.95 1.222 0.85 Having the Retezat National Park near my home benefits me and my family 3.97 1.151 0.84 My community is a more b eautiful place to live because we are living near Retezat National Park 4.47 .787 0.76 The tourists who come to the area are useful to we who live in adjacent communitie s 3.82 1.227 0.66 Attitudes pro environmental civic engagement 2 .823 .880 .596 Participating in public meetings related to Retezat National Park 4.00 .983 0.82 Participating in a community project addressing environmental concerns in the area 4.28 .844 0.73 Investing time to learn about the park and environmental prot ection 4.36 .874 0.81 Participating in a worksho p on how to reduce my dependence on park resources 3.87 1.149 0.71 Investing personal time to get involved with the park 3.93 1.034 0.78 Note. = t statistic (> 1.96) at a significance level o f p 1 Measured on a 5 point scale where 1 = Strongly disagree; 2 = Disagree; 3 = Neutral; 4 = Agree; 5 = Strongly agree; 2 Measured on a 5 point scale where 1 = Not at all effective; 2 = Never effective; 3 = Sometimes effective; 4 = Often effective; 5 = Always effective.

PAGE 165

165 Table 5 3. MANOVA of a ge Variable Mean (SD) Test of Between Subjects Effects Age groups 18 30 31 50 51 80 F Value P Val ue Eta 2 Model 1 Affective response N=52 N=82 N=86 Connection to nature 3.95 (.804) 4.51 (.593) 4.67 (.538) 22.549 < .001 .172 Park attachment 3.84 (.765) 4.33 (.694) 4.32 (.835) 8.037 < .001 .069 Model 2 General a ttitu des N=51 N=76 N=84 Cons ervation awareness 4.72 (.400) 4.86 (.451) 4.88 (.319) 3.139 .045 .029 Conservation benefits 3.91 (.655) 4.11 (.880) 4.15 (.855) 1.400 .249 .013 Conservation management 3.67 (.830) 3.76 (1.053) 4.07 (.966) 3.289 .039 .031 Model 3 Attitudes towards beh avior N=57 N=79 N=84 Attitudes p ro environmental civic engagement 3.93 (.660) 4.10 (.785) 4.25 (.767) 3.141 .045 .028 Perceived environmental responsibility 3.86 (1.231) 3.35 (1.664) 3.11 (1.708) 3.875 .022 .034 20, F Value = 11.233, p < .001 Eta 2 =.094 F Value = 1.910, p = .078, Eta 2 =.027 F Value = 3.572, p = .007, Eta 2 =.032

PAGE 166

166 CHAPTER 6 CONCLUSION S Findings The results of this stu dy illustrate the dynamics of the social dimensions of community and their ability to facilitate or hinder pro environmental engagement s First and foremost, this study found the attachment s people hold for the social environment and the park, their connec tion s to nature and attitudes towards conservation as being strongly embedded and define a local environmental identity. Thus, t his study underscores t he importance of attachments, nature connections and conservation attitudes in shaping local environmenta l identities, and the need to sustain such connections through social interactions as well as interaction with natural environments. Furthermore, l ocal environmental identity (as a reflection of attachment, connection to nature, and conservation attitudes) was found to have a significant direct impact on attitudes tow ards pro environmental civic engagement which ultimately were found to influence pro environmental civic behavioral intentions. This suggests t social and n atural environment s the more positive attitudes they have towards engagement that ultimately shapes behavioral intentions. This finding substantiates current understanding of what has been previously described as community identity and its implications fo r environmental values and attitudes (Pol, 2002; Van Vugt, 2001). Perceived collective responsible behavior s to guard natural resources were also found to have a direct impact on attitudes towards pro environmental civic engagement. Furthermore, t his stud y demonstrates the importance of attitudes in the context of pro environmental civic engagement, identity and perceived responsibility influencing attitudes but not behavioral intentions directly. This study provides evidence that with stronger local envir onmental identity

PAGE 167

1 67 environmental engagement will likely increase, ultimately shaping behavioral intentions. In addition attachments to the social and natural environment were fo und to be distinctively predicted by variables commonly associated with community attachment. S ocial interaction was found to have a stronger effect on park attachment, while length of residence had a stronger association with attachment to the social envi ronment. Family status was not found to have a significant eff ect on attachment to the social environment and park attachment This finding suggests that family status does not impact the amount of feelings and affinity people held for the social and natur al environment de fining the place A significant negative effect was found between the number of children under 18 years in the household and the strength of attachment to the social environment. The number of children in the household was found not to hav e an effect on attachment to the park, further reinforcing that attachment to community lay in the multiple facets of the community and various personal conditions distinctly facilitating the development of attachment to different dimensions of the communi ty. Education and income, as a reflection of social position in the community were two other measures assessed as predictors of social and park attachment Education was found to have a significant negative effect on attachment to the social en vironment, while income had a significant influence on attachment to the park. Ultimately, the results suggest that social position does play a role in shaping attachment to the social and natural environment Furthermore, t his study reveals that park attachment, as a reflection of attachment to the natural environment, is strongly correlated with attachment to the social environment emphasizing its role in tangentially supporting the social dynamics that are at the foundation of

PAGE 168

168 the community. The correlation betwee n the two dimensions further supported the shared meanings that the two dimensions capture, a finding strongly supported by the qualitative results T he multidimensionality of the community attachment construct and the intertwined nature of the relationshi p between the emerging dimensions narratives Four distinct dimensions of attachment emerged from the textual analysis, attachment to the natural, social, institutional and cultural environment. While distinct, these dime nsions of attachment do have an intertwined nature that makes them worthy of attention and further assessment. This study also investigated the implications of age for affective and attitudinal environmental response of adults residing in rural Romania. A g e of the residents was found to be ld towards nature and the park, even after controlling for length of residence and level of education Young adults were found to have weaker affective b onds with nature and the park neighboring their communities as compared with their older counterparts. L evel of education was found to better explain attitudinal response towards pro environmental behavior. More specifically, differences in attitudes towar ds conservation management were found to be better explained by level of education with higher levels of education assessing conservation management efforts lower Furthermore, level of education was found to play a significant role in explai ning differences in perceived levels of personal envir onmental responsibility. Lastly length of residence better captured differences in terms of conservation bene fits. In conclusion, this study substantiates the knowledge on rural communities and their interactions with neighboring natural environments, highlighting the interplay of variables that

PAGE 169

169 affect pro environment al civic behavioral intentions. Attachments t o the natural and social environment connect ions to nature conservation attitudes and perceived environmental responsibilities emerging from social contexts were found to ultimately shape pro environmental civic attitudes and behavioral intentions. Impl ications and Future Research This research extends current theoretical understandings of social predictors and inter relationships between attachm ents, connections to nature, conservation attitudes and perceived environmental responsibilities in areas ric h in natural resources and their attitudinal and behavioral implications. Furthermore, this study complements contemporary understandings of the community attachment construct and affective and attitudinal environmental response in rural contexts Primaril y, this study reveals the intertwined relationship between attachments to the social and natural environment, connections to nature, and conservation attitudes. Local environmental identity was proposed as accounting for the hierarchical structure of these constructs. The theory of interpersonal behavior incorporates self identity erception of himself or herself, as one of the major predictors of behavioral intentions (Gagnon et al. 2003; Zhang, Inbakaran & Jackson 2006). This relation ship is based on the argument that our definition of ourselves ultimately defines our actions. This study found local environmental identity influencing attitudes towards the behavior but not behavior al intentions directly. Similarly, perceived collective environmental responsibility was found to impact attitudes but no t behavior intentions directly Attitudes towards the behavior were found to have greater predictive ability for behavior al intentions. F urthe r investigations are needed to b etter understand the environmental identity construct and the nature of the relationships between local environmental identity, perceived

PAGE 170

170 environmental responsibility, and attitudes towards p ro environmental civic engagement and behavioral intentions in areas rich in natu ral resources The conceptualization of local environmental identity could be argued might be a characteristic of the population in this study and the local environment which is rich in natural resources amenities. Thus, this study should be repeated under different conditions, culturally, socially and environmentally to test the environmental identity conceptualization and its implications for attitudes tow ards pro environmental civic engagement and behavioral intentions. Furthermore, this study underscor es the importance of incorporating measures of attachment to the natural environment in the broader assessment of co mmunity attachment In various contexts and settings, the natural environment might be a stronger dimension of attachment having the ability to generate powerful emotional responses. Simply acknowledging that the natural dimension exist s and is most often embedded in the social environment, as suggested by Brehm (2007) is not sufficient There is much more information that can be captured by integrating assessments of various dimension of community attachment. This study brings to attention two more dimensions of attachment, the institutional and cultural environment s T he importance of developing a measurement that focuses on the hierarchica l nature of the construct is emphasized in this study. The multiple dimensions of attachment building on each other to ultimately define an overall attachment to the community, with its foundations being rooted in multiple local aspects. This further sugge sts a need to understand how various dimensions of community attachment are formed, how are they f luctuating and how they separately predict various attitudinal and behavioral patterns. This study found affective environmental response s to nature and the p ark to be indicative of age differences that relate to social and environmental changes that occurred over the years in

PAGE 171

171 the community. Thus, affective environmental responses were found to be more volatile in the face of social changes, as compared with at titudinal environmental response. Attitudinal responses were found to be better explained by an achieved social position (i.e., education) and length of residence in the community An in depth understanding of the affective environmental response of the yo unger generation living in rural communities is warrant ed with a focus on capturing what nature means for the young er generation and what kind of experiences shape emotional response to the natural environment. From a practical perspective, the r esults s how that pro environmental civic behavioral intentions can be improved by strengthening local environmental identities which could be done by encouraging social interaction s as well as interaction with the environment (the park, in this case). Furthermore providing increasing understanding of the benefits of the park, the benefits of conservation, and higher management transparency and cooperation can ultimately generate greater identification of the local population with the local environment. Thus, the findings emphasize the need to create opportunities for local people to utilize and benefit from the park so that ultimately they become an integrated part of the management and conservation stewards. Recommendations include using meetings and informativ e messages to strengthen intentions for local engagement and increase knowledge and awareness of conservation benefits as well the responsibilities derived from living adjacent to the park. Efforts to better inform local residents how current park rules an d management strategies sustain the natural environment to better achieve the environmental goals that the residents expressed, and how they can be involved in making specific decisions facilitating those should enhance synergistic partnerships. Projects that require involvement, interaction, and sharing of knowledge and information (about benefits and management approach) should be supported and constantly encouraged and

PAGE 172

172 implemented throughout the area. This study indicates a need to provide opportunities for engagement and constantly foster ing capacities to be agents of change in their communities and the environments they are attached and connected to. The residents of these communities developed strong attachments and meanings to the surround ing environment and their existence is embedded in a permanent interaction with their environment, and should not be overlooked by management. Furthermore, i nformed programs could be developed aimed at strengthening various dimensions of the community that lack attachment and public care through the establishment of channels of interaction and communication. L istening to the local residents and understanding their attachments within the community context so that informed decision could be made and in the l ong run avoid conflicts is paramount Warren (1987) emphasized the importance of understanding the particular local geographical attributes, due to their implication s for community togetherness that ultimately are crucial element s in t he formation and continuation of community. Thus, considering the multiple facets of the community attachment construct is imperative to provide opportunities for community members to express their opinions and concerns about the things they care about and are attach to. Furthermore this study underscored the need to reconnect the younger generation with the natural environment by providing opportunities for interaction as well as showing them perceived benefits from conservation. The affective connections with nature and the park of the young adults residing in communities neighboring the national park in Romania are weak as compared with their older counterparts. Core values that are intrinsic to feelings of connection and attachment to neighboring enviro nments seem to be have been impacted by local social and environmental dynamics and changes. Primarily a lack of interaction with the natural

PAGE 173

173 environment seems to have generated weaker relationships with nature and the park. Agriculture and livestock have for a long period of time sustained the connection between the residents and the natural environment. The major threat that weaker relationships with neighboring environments impose relates to loss of the population base, a fact that was constantly mentio ned by the residents as a threat to the well being of the community. Disappointment with the current management and lack of benefits has been expressed, with respondents with higher levels of education being more critical towards the current situation but at the same time reporting higher levels of personal environmental responsibility. Thus, from a management perspective, highly educated individuals, who also tend to be younger, represent a group with potential for involvement in park management and commu nity development efforts. Thus, there is an emerging need to create programs for young adults that provide opportunities for interaction with neighboring environments and ultimately allow residents to know and truly value the community they are part of. P rojects targeted at reconnecting the younger generation with the locale should relate to agriculture and traditional activities as well as tourism to facilitate heritage connections Limitations Several measures were taken to reduce potential limitations that could have ultimately impact ed the purity of the study. A major consideration was given to questionnaire development, its face and content validity procedures encompassed a detailed consideration of item wording, ordering effects, social desirability translation and response bias es Measurements previously employed in the literature were used in this study and thus overcoming possible validity limitations. Response rate was a major concern for this study considering that people living in rural commu nities in Romania are not very used to being ask ed to participate in a research study. The response rate was not as high as desired but the major limitation was the result of not finding

PAGE 174

174 people at home and not refusals to participate in the study. Face to face interviews were found to be much more effective than mail survey s sent to community members that could not be found at home even after multiple visits. The response rate was ultimately found to be appropriate for examining the study questions, includi ng the structural equation model (Kline, 2005). In the data analysis stage of the study, several items measuring different constructs employed in the study were found to be weak measures of the underlying constructs and thus, eliminated from further analys is. This situation was primarily attributed to the translation constraints, the meanings of certain items not being very well captured in the Romanian version of the questionnaire. The results of this study are relevant and can be generalized primarily to communities surrounding prote cted areas in Romania and other regions transitioning from a dependent agricultural state to a less dependent agricultural state, yet, communities rich in natural resources. tudes towards the natural environment in these contexts are stimulated by continuous interactions with the natural environment. Thus, relationships between constructs might be a direct response of the local stimuli which incorporate the natural environment as a strong element characterizing the local life in such communities Conclusion s This research sought to provide insight s into the social dimensions of pro environmental civic engagement in the communities surrounding Retezat National Park in Romania. The importance of attachments to the social environment and the park, connections to nature, conservation attitudes, and perceived environmental responsibility in shaping local attitudes towards p ro environmental civic engagement and pro environmental beha vior intentions was assessed ; and most were found to be important contributors The study findings are relevant for

PAGE 175

175 the development and planning of community based programs to support protected area management efforts. This research attempted to complement and advance current understandings of the human park relationships and achieved this goal

PAGE 176

176 APPENDIX A STUDY QUESTIONNAIRE Your Community and Retezat National Park Thank you for agreeing to complete this survey about your community and Retezat National P ark. Please read each question carefully before responding. Answer to the best of your ability and save any additional comments for the end. Your answers will help to better understand your feelings, attitudes, and perceptions about your community and Rete zat National Park. 1. People have different feelings about their community. Please indicate what your level of agreement is with the following statements about your community? Circle one number for each statement on the 5 point scale, where 1=strongly disagree and 5=strongly agree. 2. What are the three words that come to your mind when you think of your community ? A___________________ B___________________ C___________________ 3. Overall, how would you rate the quality of life in your community? P lease one answer. Poor Above poor Average Above average Excellent Strongly Disagree Disagree Neutral Agree Strongly Agree Overall, I am very attached to this community 1 2 3 4 5 I feel like I belong in this community 1 2 3 4 5 The associations that I have with other people in this comm unity mean a lot to me 1 2 3 4 5 If the people in this community were something WE were doing rather than THEY were doing 1 2 3 4 5 If I needed advice about something, I could go to someone in this community 1 2 3 4 5 I agree with most people in this community about what is important in life 1 2 3 4 5 Given the opportunity, I would move out of this community 1 2 3 4 5 I feel loyal to the people in this community 1 2 3 4 5 There are things going on in this communi ty that I am not proud of I plan to remain a resident of this community for a number of years 1 2 3 4 5 There are many things I would like to change about this community 1 2 3 4 5 I like to think of myself as similar to the people who live in this community 1 2 3 4 5 This first section asks about your community and local life conditions.

PAGE 177

177 4. In general, how would you describe your level of involvement in community activities or events? Please one answer. Not active at all Not very active Somewhat ac tive Very active Extremely active 5. Please name three things from your community that you care about the more, you feel most attached to? A___________________ B___________________ C___________________ 6 How often do you get togeth er or meet the following types of people? Please one answer per statement. 7 People have different feelings about nature. Please indicate your level of agreement with the following statements in terms of th e way you generally feel about nature There are no right or wrong answers. Circle one number for each statement on the 5 point scale, where 1=strongly disagree and 5=strongly agree. Never A few times a year Once a month A few times a month Once a week More than once a week Everyday Immediate family (parents, siblings) Extended family (cousi ns, uncles) Acquaintances Close friends Neighbors Community groups (e.g. church) Public officials Retezat National Park staff Tourists M embers of non governmental organizations Others (please specify _________________ ______) Strongly Disagree Disagree Neutral Agree Strongly Agree I have seen thi ngs in nature that were so amazing; they just filled me with wonder 1 2 3 4 5 When surrounded by nature, I feel at peace 1 2 3 4 5 A lot of nature just scares me 1 2 3 4 5 My love for nature is a big influence in my life 1 2 3 4 5 I feel sorrow because much nature 1 2 3 4 5 I am connected to nature much like I'm connected to my family 1 2 3 4 5 This section asks for your feelings about Retezat National Park and nature in general.

PAGE 178

178 8 Have you ever been inside Retezat National Park? No Yes 9 What are the three words that come to your mind when you think of Retezat Nationa l Park ? A___________________ B___________________ C___________________ 10 During the last 12 months, how often did you go to Retezat National Park? Never A few times a year Once a month A few times a month Once a week More than on ce a week Everyday 11 If you do visit the Retezat National Park, what is the primary purpose of your visit? Please all that apply. Collecting wild resources (fire wood, mushrooms, medicinal plants, etc.) Hiking Camping Work Other (plea se specify___________________) I never go there 12 People living near Retezat National Park have different feelings about the park. Please indicate your level of agreement with each of the following statements about Retezat National Park and what it me ans to you as compared with other places? Circle one number for each statement on the 5 point scale, where 1=strongly disagree and 5=strongly agree. Seeing how much nature is being destroyed affects me emotionally 1 2 3 4 5 Remot e natural areas make me nervous 1 2 3 4 5 Nature is a hu ge part of who I am 1 2 3 4 5 Nature provides me with a spiritual connection 1 2 3 4 5 I often feel a sense of oneness with the natural world around me 1 2 3 4 5 My feelings for nature have influenced my spiritual beliefs 1 2 3 4 5 The power of nature is just incredible 1 2 3 4 5 Listening to the wind go through the trees calms my mind 1 2 3 4 5 The ma gnitude of nature is impressive 1 2 3 4 5 Feeling part of n ature is a spiritual experience 1 2 3 4 5 I have too much fear of nature t o hike in remote natural areas 1 2 3 4 5 When I'm alone in a natural area, I have this feeling of complete calm 1 2 3 4 5 Strongly Disagree Disagree Neutral Agree Strongly Agree Retezat National Park means a lot to me 1 2 3 4 5 I feel no commitment to Retezat National Park 1 2 3 4 5

PAGE 179

179 13 Do you have any property rights (ownership or land use rights) in Retezat National Park? No Yes 14 Retezat National Park is important to people for different reasons. Please indicate for which of the following reasons is the p ark important to you? Please all that apply Use of natural resources (fire wood mushrooms, medicinal plants, fishing, etc.) Livestock grazing Outd oor recreation (hiking, camping etc.) Scenery, its unique landscapes, plants, and animals A gre at place to visit with family and friends Benefits it brings to our community Other, please specify _______________________________ 15 Do you know of any group/forum/or committee that represents your community interests with the Retezat National Pa rk administration? No Yes If yes, please specify the name of the group/forum/ or committee_________________ 16 Below is a list of opinions about conservation in Retezat National Park. Please indicate what of your level of agreement is with each o f the following statements. Circle one number for each statement on the 5 point scale, where 1=strongly disagree and 5=strongly agree. I am very attached to Retezat National Park 1 2 3 4 5 Retezat National Park is very important to me 1 2 3 4 5 I get many personal benefits out of living near Retezat National Park 1 2 3 4 5 I id entify strongly with the Retezat National Park 1 2 3 4 5 living near Retezat National Park 1 2 3 4 5 I enjoy living near Retezat National Park 1 2 3 4 5 I get lots of satisfaction out of living near Retezat Nationa l Park 1 2 3 4 5 Strongly Disagree Disagree Neutral Agree Strongly Agree It is important to have the Retezat National Park for the survi val of various plants and animal species 1 2 3 4 5 It is necessary to set aside some land for the protection of plants and animals 1 2 3 4 5 pride 1 2 3 4 5 Retezat National Park being protected is important for th e benefit of our future generations 1 2 3 4 5 The illegal cutting of trees in the park should be strictly regulated 1 2 3 4 5

PAGE 180

180 If overgrazing continues in the park, all the animals will soon disappear 1 2 3 4 5 What people and their livestock need is mor e important than protecting wild animals and plants 1 2 3 4 5 It is good if some land within the park is allocated to local people 1 2 3 4 5 Retezat National Park is a waste of land 1 2 3 4 5 People should who own/ have property rights in the park shoul d be allowed t o use park resources as they wish Retezat National Park is for outsiders, those who enjoy hiking and wildlife viewing 1 2 3 4 5 The economic stability of communities is more important than protecting park resources 1 2 3 4 5 Retezat N ational Park is managed successfully for the benefit/enjoyment of future generations 1 2 3 4 5 I support the rules and regulations established by the park administration 1 2 3 4 5 Retezat National Park is managed successfully for a wide range of uses and values, not just tourism 1 2 3 4 5 Retezat National Park management does a good job at protecting the natural resources in the park 1 2 3 4 5 The citizens from the communities around the park have enough say in how the park is managed 1 2 3 4 5 The ben efits from the park usually outweigh negative consequences 1 2 3 4 5 My community benefits from being near the Retezat National Park 1 2 3 4 5 Having the Retezat National Park near my home benefits me and my family 1 2 3 4 5 My community is a more beaut iful place to live because we are living near Retezat National Park 1 2 3 4 5 The quality of the air is higher because of living near the Retezat National Park area 1 2 3 4 5 The park resources help local waters stay pure for our community 1 2 3 4 5 The tourists who come to the area are useful to we who live in adjacent communities 1 2 3 4 5

PAGE 181

181 17 There are many ideas on how the park land should be managed. Please indicate who you think should establish the rules and regulations for land managem ent in Retezat National Park? Please all that apply Retezat National Park Administration Local councils Land owners Other, please specify___________________ agement in Retezat National Park 18 People take part in environmental protection in different ways. How effective do you think the following behaviors are in protecting the environment in Retezat National Park? Circle one number for each statement on the 5 point scale, where 1=not at all effective and 5=always effective. 19 In the next 12 mon ths, if opportunities were provided, how likely would you be to perform the following behaviors? Circle one number for each statement on the 5 point scale, where 1=very unlikely and 5=very likely. Not at all Effective Never Effective Sometimes Effective Often Effective Always Effective Attending public presentations about the Retezat National Park 1 2 3 4 5 Participating in publi c meetings related to Retezat National Park 1 2 3 4 5 Participating in a community project addressing environmental concerns in the area 1 2 3 4 5 Learning about the environment 1 2 3 4 5 Investing time to learn about the park and environmental protecti on 1 2 3 4 5 Participating in educational programs about the environment 1 2 3 4 5 Participating in a worksho p on how to reduce my dependence on park resources 1 2 3 4 5 Participation in a local organization that is involved in park protection 1 2 3 4 5 Investing personal time to get involved with the park 1 2 3 4 5 Reducing use of park resources 1 2 3 4 5 Bringing tourists to the park 1 2 3 4 5 Voting for public officials that show interest in environmental issues 1 2 3 4 5 Very Unlikely Somewhat Unlikely Neither Likely nor Unlikel y Somewhat Likely Very Likely Participate in a public meeting related to Retezat National Park 1 2 3 4 5 Attend a public presentation about Retezat National Park 1 2 3 4 5 This section asks about your engagement in environmental protection in Retezat National Par k.

PAGE 182

182 20 People have different perceptions in terms of who should be responsible for protecting the following statements regarding responsibility for environmental protection in Retezat National Park. Circle o ne number for each statement on the 5 point scale, where 1=strongly disagree and 5=strongly agree. 21 How long have you lived in this community? Number of years ________________ 22 How many ad ults and children live in your household? (fill in a number) Number of adults including yourself _______________________ Number of children (under 18) ____________________________ Participate in a community project addressing environmental concerns 1 2 3 4 5 Invest time to learn more about the park and environmental protection 1 2 3 4 5 Give my input into park management decisions 1 2 3 4 5 Be actively involved in an organization that supports park management efforts 1 2 3 4 5 Participate in a workshop on ho w to reduce my dependence on park resources 1 2 3 4 5 Express my concerns about park management to elected officials 1 2 3 4 5 Visit the park at least twice 1 2 3 4 5 Visit the park at least six times 1 2 3 4 5 Totally Disagree Somewhat Disagree Neither Agree nor Disagree Somewhat Agree Totally Agree Personally, I have no responsibility for protecting the environm ent in Retezat National Park 1 2 3 4 5 Every citizen in my community must take responsibility for protecting the environment in Retezat National Park 1 2 3 4 5 Authorities, rather than the citizens, are responsible for protecting the environment in Retez at National Park 1 2 3 4 5 Authorities, together with the citizens, are responsible for protecting the environment in Retezat National Park 1 2 3 4 5 This final section asks about your household and demographic information. This information will be kept confidential and used for statistical purposes only.

PAGE 183

183 23 What is your gender? M ale F ema le 24 In what year were you born? ___________ 25 Are you? Single Married/partnered Divorced/separated Widowed 26 What is the highest level of education you have compl eted? Please one answer. None High school graduate (9 12) Primary school (1 4) Post high school Elementary school (5 8) Some college Professional/ vocational school College degree Some high school (9 10) Advanced degree 27 Which of the following best represents your current employment status? P lease one answer. Retired Working in industry Not working outside of home Working in commerce, tourism, and other services Unemployed Technician, supervisor Student Personnel with higher qualifications Working in agriculture/ owning farming land Business owner, entrepreneur Other (please specify) _______________________ 28 Do you in any way make money from the park or from visitors to the park? No Yes 29 Does anyone in your immediate family make money from the park or from visitors to the park? No Yes 30 What was your approximate household income last month? Almost no income Between 1000 and 1499 RON Less than 250 RON Between 1500 and 2000 RON Between 2 50 and 499 RON More than 2000 RON Between 500 and 999 RON Thank you for completing this survey! In the space provided below, please feel free to include any comments that you think might help us in better understandin g your feelings, attitudes, and perceptions about your community and Retezat National Park.

PAGE 184

184 APPENDIX B STUDY QUESTIONNAIRE ( IN ROMANIAN ) e Pentru fiecare total Dezacord total Oarecum dezacord Neutru Oarecum de acord Acord total 1 2 3 4 5 1 2 3 4 5 mult pentru mine 1 2 3 4 5 ar organiza ceva ce NOI facem mai curnd dect ceva ce EI fac 1 2 3 4 5 1 2 3 4 5 comun itate 1 2 3 4 5 ar oferi oportunitatea, m 1 2 3 4 5 1 2 3 4 5 lucruri de care nu sunt mndru l ani 1 2 3 4 5 schimb 1 2 3 4 5 1 2 3 4 5

PAGE 185

185 2. Care sunt primele tre A___________________ B___________________ C___________________ ) un Foar te prost Prost Mediu Peste mediu Excelent ( Deloc activ Nu foarte activ Oarecum activ Foarte activ Extrem de activ A___________________ B___________________ C________________ ___ ( ) De cteva ori pe an O pe De cteva ori pe Mai mult pe n fiecare zi surori) Vecini Grupuri n comunitate (spre exemplu, oameni Membrii ai unor guvernamentale _______________ ________)

PAGE 186

186 Pentru fieca 8. Nu Da Retezat? A___________________ B___________________ C___________________ Dezacord total Oarecum dezacord Neutr u Oarecum de acord Acord total extraordinare nct m e 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 Natura este o mare parte a ceea ce sunt 1 2 3 4 5 1 2 3 4 5 meu 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 copaci 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 Mi nepopulate 1 2 3 4 5 Am un sentiment de calm total cnd sunt 1 2 3 4 5 legate

PAGE 187

187 De cteva ori pe an De cteva ori Mai mu lt de o n fiecare zi dvs.? ) toate categoriile care sunt aplicabile n cazul dvs. Colectare/ culegere resurse nat urale (lemne de foc, ciuperci, plante medicinale, etc.) Excursie cu cortul Munca n parc rite opinii/ Nu Da 14 pentru dvs. ) toate categoriile care sunt aplicabile n cazul dvs. Dezacord total Oarecum dezacord Neutru Oarecum de acord Acord total mult pentru mine 1 2 3 4 5 1 2 3 4 5 Retezat 1 2 3 4 5 important pentru mine 1 2 3 4 5 Retezat 1 2 3 4 5 1 2 3 4 5 Retezat 1 2 3 4 5 1 2 3 4 5 Retezat 1 2 3 4 5

PAGE 188

188 Folosirea resurselor din parc (lemn ciuperci, plante medicinale, pescuit, etc.) Este un loc extraordinar ce mer mpreun 15 Nu Da _________________________ ____________________________________________ ________________ 16. Randurile de mai jos con pe o 1= dezacord acord total Dezacord total Oarecum dezacord Neutru Oarecum de acord Acord total 1 2 3 4 5 E animale 1 2 3 4 5 noastre 1 2 3 4 5 vi itoare 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 animalele lor domestice d plantelor 1 2 3 4 5 1 2 3 4 5 teren 1 2 3 4 5 Oamenii care a u drepturi de proprietate/ 1 2 3 4 5 1 2 3 4 5

PAGE 189

189 17 ) toate categoriile care sunt aplicabile n cazul dvs. Consiliile locale P roprietarii de terenuri Nu consider c e nevoie de stabilirea a unor reguli stricte de administrare a terenurilor din cadrul parcului 18 Retezat? 1=total ntotdeauna eficient. parcului 1 2 3 4 5 viitoare 1 2 3 4 5 stabilite de 1 2 3 4 5 success pentru o varietate de nevoi de 1 2 3 4 5 prote jarea resurselor naturale din parc 1 2 3 4 5 parcului le sunt ascultate p rerile referitoare la administrarea parcului 1 2 3 4 5 impactele negative 1 2 3 4 5 Comunitatea mea 1 2 3 4 5 familia mea 1 2 3 4 5 Comunitatea mea este un loc mai frumos cuim 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 parcului 1 2 3 4 5

PAGE 190

190 19 5=foarte probabil. Tot al ineficient Ineficient C eficient Adesea eficient ntotdeauna eficient Retezat 1 2 3 4 5 A participa la ntlniri publice legate 1 2 3 4 5 A parti cipa ntr un proiect al 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 A participa n programe educat ive 1 2 3 4 5 A participa ntr un curs despre resursele din parc 1 2 3 4 5 A participa ntr parcului 1 2 3 4 5 personal pentru a deveni implicat n probleme legate de parc 1 2 3 4 5 A reduce utilizarea resurselor din parc 1 2 3 4 5 1 2 3 4 5 problemele de mediu locale 1 2 3 4 5 Foarte improbabil Oarecum improbabil Nici probabil, nici improbabil Oarecum probabil Foarte probabil Retezat 1 2 3 4 5 parte la o prezentare Retezat 1 2 3 4 5 un proiect al de mediu 1 2 3 4 5 1 2 3 4 5

PAGE 191

191 20 Pentru fieca unde 1= dezacord acord total 21. De ________________ mediului ile n luarea deciziilor administrative ale parcului 1 2 3 4 5 o 1 2 3 4 5 1 2 3 4 5 mele legat de administrarea 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 Dezacord total Oarecum dezacord Nici acord, nici dezacord Oarecum de acord Acord total Personal, nu am nici un fel de responsabilitate pentru protejarea 1 2 3 4 5 1 2 3 4 5 Au Retezat 1 2 3 4 5 responsabili pentru protejarea naturii n 1 2 3 4 5 dem statistice.

PAGE 192

192 __________________________ F emeie 24. n ce an v 19_______ o relatie Divor o? Liceu (9 12 clase) 4 clase) Gimnaziu (5 8 clase) Treapta I de liceu (9 10 clase) Studii postuniversitare (ex: masterat, doctorat) angajare? Pensionar Muncitor Student Personal cu studii superioare ___ Nu Da Nu Da 30. n luna trecut Aproape nici un venit ntre 500 Peste 2000 RON Sub 250 RON ntre 1000 ntre 250 ntre 1500 iile

PAGE 193

193 APPENDIX C SEMI STRUCTURED INTERVIEW GUIDE The purpose of this study is to explore the feelin gs, attitudes, and behaviors people have towards their communities and the neighboring natural areas. I would like to talk to you about your community and also Retezat National Park. Questions: 1. First, please tell me a little bit about yourself what i s your occupation, what is the size and makeup of your family, how long have you lived in this community, and how much do you like it here? 2. Please tell me how would you describe your community to someone who has never been here? 3. Please describe some of the most important things to you about your community; things that you care about and are attached to the most. Concerns regarding your community. 4. Please describe what you think about Retezat National Park as a protected area; your feelings as well as attitudes towards the park. Concerns regarding the park. 5. Please describe some of the things you do or you are willing to do for your community and the park. Is there anything that you would like to add? Do you have any questions or comments? Thank you very much for you participation in this study.

PAGE 194

194 APPENDIX D SEMI STRUCTURED INTERVIEW GUIDE (IN ROMANIAN) Retezat. ia dvs., care este cele mai importante lucruri din comunitate pentru dvs., care sunt Griji legate de comunitate. protejata; Griji legate de parc. studiu.

PAGE 195

195 LIST OF REFERENCES Agrawal, A. (2001). Common property institutions and sustainable governance of resources. World Development 29 (10), 1649 1672. Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes 50 179 211. Altman, I. & Low, S. M. (Eds.). (1992). Place attachment. Human behavior and environ ment New York NY : Plenum Press. Anderson, D., & Gerbing, D. (1988). Structural equation modeling in practice. A review and recommended two step approach. Psychological Bulletin, 103, 411 423. Aoyagi Usui, M., Vinken, H., & Kuribayashi, A. (2003). Pro environm ental attitudes and behaviors: An international c omparison. Human Ecology Review 10 (1), 23 31. Augustyn, M. (1998). National strategies for rural tourism development an d sustainability: T he Polish experience. Journal of Sustainable Tourism 6 (3), 191 209. Bagozzi, R. P. (1994). Structural equation models in marketing research: B asic principles. In R.P. Bagozzi (Ed.), Principles of marketing research (pp. 317 385 ) Oxford UK : Blackwell. Barr, A., & Gilg, A. W. (2007). A conceptual framework for understanding and analyzing attitudes towards environmental behaviour. Geography Annals 89 B (4), 361 379. Barrow, E., & Fabricius, C. (2002). Do rural people really benefit from protected areas rhetoric or reality? Park, 12 (2), 67 79. Bamberg, S., & Sc hmidt, P. (2002). Incentives, morality, or habit? Predicting students care use for university routes with the models of Ajzen, Schwartz, and Triandis. Environment and Behavior 35 264. Beckley, T. M. (2003). The relative importance of sociocultural and e cological factors in attachment to place. In L. E. Kruger (Ed.), Understanding community forest relations. Portland, OR: U.S. Department of Agriculture Forest Service, Pacific Northwest R esearch Station. Beggs, J. J., Hurlbert, J. S., & Haines, V. A. (199 6). Community attachment in a rural setting: A refinement and empirical test of the systemic model. Rural Sociology, 61 (3), 407 236. Bernard, R.H. (2000). Social r esearch m ethods. Thousand Oaks CA: Sage Publications. Bonaiuto, M., Aiello, A., Perugini, M., Bonnes, M., & Ercolani, A. P. (1999). Multidimensional perception of residential environment quality and neighborhood attachment in the urban environment. Journal of Environmental Psychology, 19 331 352.

PAGE 196

196 Bott, S., Cantrill, J. G., & Myers, O. E. (200 3). Place and the promise of conservation psychology. Human Ecology Review 10 (2), 100 112. Brandon, K. E., & Wells, M. (1992). Planning for people and parks: D esign dilemmas. World Development, 20 (4), 557 570. Brehm, J. M. (2007). Community attachment: T he complexity and consequence of the natural environment facet. Human Ecology 35 477 488. Brehm, J. M., Eisenhauer, B. W., & Krannich, R. S. (2006). Community attachments as predictors of local environmental concern. The case of multiple dimensions of attachment. American Behavioral Scientist, 50 (2), 142 165. Brehm, J. M., Eisenhauer, B. W., & Krannich, R. S. (2004). Community attachments as predictors o f local environmental concern. T he case for multiple dimensions of attachment. 67 th Annual Meeting o f the Rural Sociological Society Sacramento, CA. Retrieved from http://www.ruralsociology.org/annual meeting/2004/Brehm,etal.pdf Brennan, M. A. (2007). Community developme nt in the west of Ireland : T wenty years on in the Killala area. Community Development Journal, 42 (3), 330 347. Brennan, M. A. (2006). The development of com munity in the West of Ireland: A return to Killala twenty years on. Community Development Journal, 42 (3), 330 347. Bridger, J. C., & Alter, T. (2008). An interactional approach to place based rural development. Community Development, 39 (1), 99 111. Brown, B. B., & Perkins, D. D. (1992). Disruptions in place attachment. In I. Altman & S. Low (Eds.), Pl ace attachment Human behavior and environment (pp. 279 304). New York, NY: Plenum. Browne, M. W., & Cudeck, R. (1992). Single sample cross validation indices for covariance structures. Multivariate Behavioral Research, 24 445 455. Campbell, K. E., Mar sden, P. V., & Hurlbert, J. S. (1986). Social resources and socioeconomic status. Social Networks, 8 87 117. Chawla, L. (1990). Ecstatic places. 7 (4), 18 23. Chavis, D. M., & Wandersman, A. (1990). Sense of commu nity i n the urban environment: A catalyst for participation and community development. American Journal for Community Psychology 18 (1), 55 81.

PAGE 197

197 Clayton, S. (2003). Environ mental identity: A conceptual and an operational definition. In S. Clayton & S. Opotow (Ed s.), Identity and the natural environment: The psychological significance of nature (pp. 45 65). Cambridge, MA: MIT Press. Clayton, S., & Opotow, S. (2003). Introdu ction: I dentity and the natural environment. In S. Clayton & S. Opotow (Eds.), Identity an d the natural environment: The psychological significance of nature (pp. 1 24) Cambridge, MA: MIT Press. Cottrell, S. P., & Cutumisu, N. (2006). Sustainable tourism developmen t strategy in WWF Pan Parks: C ase of a Swedish and Romanian national park. Scan dinavian Journal of Hospitality and Tourism, 6 (2), 150 167. Creswell, J. W. (2007). Qualitative inquiry and research design. Choosing among five approaches ( 2 nd Ed.) Thousand Oaks, CA: Sage Publications, Inc. Cross, J. E. (2003). Conceptualizing communi ty attachment. 66 th Annual Meeting of the Rural Sociological Society Montreal, Canada Retrieved from http://www.ruralsociology.org/annual meeting/2003/CrossJennifer.pdf Cross, J. E. (2001). Protecting our place: E stablishing and maintaining community attachments in the face of population growth and change. Unpublished doctoral dissertation University of California Davis. Cutumisu, N. (2003). A synergy between nature c onservation and development: A framework for analyzing the impact of the sustainable tourism strategy in Natura 2000 sites. Unpublished technical report, PAN Parks de Beer, F., & Marais, M. (2005). Rural communities, the natural environment and developm ent some challenges, some successes. Community Development Journal, 40 (1), 50 61. de Vaus, S. (2001). Research design in social research. London, UK : Sage Public ations. Dolisca, F., McDaniel, J. M., Shannon, D. A., & Jolly, C. M. (2009). A multilevel a nalysis of the determinants of forest conservation behavior among farmers in Haiti. Society and Natural Resources, 22 (5), 433 447. Driver, B. L., & Ajzen, I. (1996). Research needed on hard to define nature based human experiences. In B. L. Driver, D. Du stin, T. Baltic, G. Elsner & G. L. Peterson (Eds.), Nature and the human spirit: Toward an expanded land management ethic State College, PA: Venture Publishing, Inc. Fielding, K. S., Terry, D. J., Masser, B. M., & Hogg, M. A. (2008). Integrating social i dentity theory and the theory of planned behaviour to explain decisions to engage in sustainable agricultural practices. British Journal of Social Psychology 47 23 48.

PAGE 198

198 Fishbein, M. & Ajzen, I. (1975). Belief, attitude, intention and behavior: An introd uction to theory and research Reading, MA: Addison Wesley. Flora, C. B., & Flora, J. L. (1990). Developing entrepreneurial rural communities. Sociological Practice, 8 197 207. Fornell, C., & Larcker, D. (1981). Evaluating structural equation models wit h unobservable variables and measurement error. Journal of Marketing Research, 18 39 50. Fransson, N., & Garling, T. (1999). Environmental concern: Conceptual definitions, measurement methods, and research findings. Journal of Environmental Psychology 1 9 369 382. Gadd, M. E. (2005). Conservation outside of parks: A ttitudes of local people in Laikipia, Kenya. Environmental Conservation, 32 (1), 50 63. Gagnon, M. P., Godin, G., Gagne, C., Fortin, J. P., Lamothe, L., Reinharz, D., & Cloutier, A. (2003). An adaptation of the theory of planned behavior to the study of telemedicine adoption by physicians. International Journal of Medical Informatics 71 103 115. Garling, T., Fujii, S., Garling, A., & Jakobsson, C. (2003). Moderating effects of social value orientation on determinants of proenvironmental behavior intention. Journal of Environmental Psychology, 23 1 9. Goudy, W. J. (1990). Community attachment in a rural region. Rural Sociology, 55 178 198. Granovetter, M. (1973). The strength of weak t ie s. American Journal of Sociology 78 1360 1380. Gursoy, D., & Rutherford, D. G. (2004). Host attitudes towards tourism: An improved structural model. Annals of Tourism Research 31 (3), 495 516. Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E., & Tatham, R. L. (2006). Multivariate data analysis (6 th Ed.). Upper Saddle River, NJ: Person Education, Inc. Hall, D. (2004). Rural tourism development in Southeastern Europe: Transition and the search for sustainability. International Journal of Tourism R esearch 6 165 176. Halpenny, E. A. (2006). Environmental behavior, place attachment, and park visitation: A case study of visitors to Point Pelee National Park. Unpublished doctoral dissertation, University of Waterloo, Ontario, Canada. Harris, P. B., Brown, B. B., & Werner, C. M. (1996). Privacy re gulation and place attachment: P redicting attachments to a student family housing facility. Journal of Environmental Psychology, 16 287 301.

PAGE 199

199 Hatcher, L. (1994). A step by step approach to using the SAS syst em for factor analysis and structural equation modeling Cary, NC: SAS. Hitlin, S. (2003). Values as the core of personal identity: Drawing links between two theories of self. Social Psychology Quarterly, 66 (2), 118 137. Hines, J. M., Hungerford, H. R. & Tomera, A. N. (1987). Analysis and synthesis of research on responsible environmental behavior: A meta analysis. Journal of Environmental Education 18 1 18. Heltberg, R. (2001). Determinants and impacts of local institutions for common resource manage ment. Environment and Development Economics, 6, 183 208. Hu, L., & Bentler, P. (1999). Fit indices in covariance structure modeling: Sensitivity under parameterized model misspecification. Psychological Methods, 3, 424 453. Hummon, D. M. (1990). Commonpl aces: Community ide ology and identity in American c ulture. Albany, NY: State University of New York Press. Ioras, F. (2003). Trends in Romanian biodiversity conservation policy. Biodiversity and Conservation, 12 9 23. Infield, M., & Namara, A. (2001). C ommunity attitudes and behavior towards conservation: an assessment of a community conservation program around Lake Mburo National Park, Uganda. Oryx, 35 (1), 48 60. Jennings, G., & Nickerson, N. P. (2006). Quality tourism experiences. Burlington, MA: Else vier Butterworth Heinemann. Joreskog, K. G. (1993). Testing structural equation models. In K.A. Bollen and J.S. Long (Eds.), Testing structural equation models ( pp. 294 316 ) Newbury, CA: Sage. Kasarda, J. D., and Janowitz, M. (1974). Community attachmen t in mass society. American Sociological Review 39 328 339. Kappelle, R. J. (2001). Relationships between local people and protected natural areas: A case ew Zealand Unpublished m thesis, Lincol n University, Caterbury, New Zealand. Kellert, S. R. (2005). Building for life. Designing and understanding the human nature connection. Washington, DC: Island Press. Kihlstrom, J. F., Cantor, N., Albright, J. S., Chew, B. R., Klein, S. B., & Neidenthal, P. M. (1988). Information processing and the study of the self. In L. Berkowitz (Ed.), Advances in experimental social psychology ( pp.145 180 ) San Diego, CA: Academic Press.

PAGE 200

200 Kline, R. B. ( 2005 ) Principles and practice of structural equation modeling New York, NY: Guilford. Kollmuss, A., & Agyeman, J. (2002). Mind the gap: Why do people act environmentally and what are the barriers to pro environmental behavior? Environmental Education Research 8 (3), 239 260. Kuijs, J., & Bergh, J. van den. (2006). Tourism and nature c onservation in Retezat National Park : A case study of sustainable financing mechanisms and tourism potentials of Wageningen Netherlands. Kyle, G., Bricker, K Graefe, A., & Wickham, T. (2004). relationships with activities and settings. Leisure Sciences, 26, 123 142. Lawton, M. P. (1990). Knowledge resources and gaps in housing the aged. In D. Tilson (Ed.), Aging in place (pp. 287 309). Glenview, IL: Scott, Fo resman. Tourism Management 28, 876 885. Lewicka, M. (2005 ). Ways to make people active: T he role of place attachment, cultural capital, and neighborhood ties. Journal of En vironmental Psychology, 25 381 395. Louv, R. (2006). Last child in the woods. Saving our children from nature deficit disorder. Chapel Hill, NC: Algonquin Books of Chapel Hill. Luloff, A. E. (1990). Community and social change: How do small communities act? In A. E. Luloff and L. E. Swanson (E d s .), American Rural Communities (pp. 214 227). Boulder, CO: Westv iew Press Luloff, A. E. & Swanson, L. E. (1995). Community agency and disaffection: Enhancing collective resources. In L. J. Beaulieu & D. Mulkey ( E ds. ) Investing in people: The human capital needs of rural America (pp. 351 372). Boulder, CO : Westview Press MacC allum, R. C., Browne, M. W., & Sugawara, H. M. (1996). Power analysis and determination of sample size for covariance structure modeling. Psychological Methods 1 130 149. Mannell, R. C. (1996). Approaches in the social and behavioral sciences to the systematic study of hard to define human values and experiences. In B. L. Driver, D. Dustin, T. Baltic, G. Elsner & G. L. Peterson (Eds.), Na ture and the human spirit: Toward an expanded land management ethic (pp. 405 416). State College, PA: Venture Publishing, Inc. Manfredo, M. J., Vaske, J. J., Bruyere, B. L., Field, D. R., & Brown, P. J. (2004). Society and natural resources: A summary of knowledge. Jefferson MO : Modern Litho.

PAGE 201

201 Manzo, L. C. (2005). For better or worse: E xploring multiple dimensions of place meaning. Journal of Environmental Psychology, 25 67 86. Marsh, H.W., Craven, R. G., & Debus, R. (1991). Self concepts of young child ren 5 to 8 years of age: Measurement and multidimensional structure. Journal of Educational Psychology 83 (3), 377 392. Mayer, F. S., & Frantz, C. M. (2004). The connectedness to nature scale: A measure of individuals' feeling in community with nature. Jo urnal of Environmental Psychology, 24 (4), 503 515. McCool, S. F., & Martin, S. R. (1994). Community attachment and attitudes toward tourism development. Journal of Travel Research, 32, 29 34. McFarlane, B., & Boxall, P. C. (2003). The role of social psyc hological and social structural variables in e nvironmental activism: A n example of the forest sector. Journal of Environmental Psychology 23 79 87. Mesch, G. S., & Manor, O. (1998). Social ties, environmental perception, and local attachment. Environmen t and Behavior 30 (4), 504 519. Muthn, B., C du Toit S. H., & Spisic D ( 1997 ) Robust inference using weighted least squares and quadratic estimating equations in latent variable modeling with categorical and continuous outcomes. Unpublished technical report. Nguyen, T. N. (2007). Role of social capital in natural resources conservation: A case study of Cat Tien National Park in Vietnam. Unpublished doctoral dissertation, University of Florida. Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric theory (3rd e d .) New York, NY: McGraw Hill. Oostenbrink, W. & Kosterink, P. (2005). Public participation in sustainable regional development in EU accession countries Bulgaria and Romania Retrieved from www. milieukontact.nl Ormsby, A., & Kaplin, B. A. (2005). A framework for understanding community resident perceptions of Masoala National Park, Madagascar. Environmental Conservation 32 (2), 156 164. Pagdee, A., Kim, Y. S., & Daugherty, P. J. (2006). What makes community forest management successful: A meta study from community forests throughout the world. Society and Natural Resources 19 33 52. PAN Parks (2009). Vision Retrieved from http://w ww.panparks.org/Introduction/Vision

PAGE 202

202 Payne, L. L., & Schaumleffel, N. A. (2008). Relationship between attitudes toward rural community parks and recreation and rural community satisfaction. Journal of Park and Recreation Administration, 26 (3), 116 135. Pennisi, L. (2007). Connection to nature: Developing a measurement scale. Unpublished doctoral dissertation, University of Florida Perrin, J. L. & Benassi, V. A. (2009). The connectedness to nature scale: A measure of emotional connection to nature? Jo urnal of Environmental Psychology, 29 434 440. PJB Associates. (2006). New perspectives for learning: Engaging people in active c itizenship in Central and Eastern Europe Retrieved from http://www.pjb.co. uk/npl/bp64.htm Pol, E. (2002). The theoretical background of the city identity sustainability network. Environment and Behavior, 34, 8 25. Popescu, C. (1993). Romanian industry in transition. GeoJournal, 29 (1), 41 48. Pretty, J. (2002). Agri culture. Reconnecting people, land, and nature Sterling, VA: Earthscan Publications Ltd. Pretty, G., Chipuer, H., & Bramston, P. (2003). Sense of place amongst adolescents and adults in two rural Australian towns: T he discriminating features of place attachment, sense of community and place dependence in relation to place identity. Journal of Environmental Psychology, 23 273 287. Pretty, J., & Smith, D. (2004). Social capital in biodiversity conservation and management. Conservation Biology 18, 631 638. Retez at National Park. (2008). About the park. Retrieved from http://retezat.ro/index.php/romana/despre parc.html Retezat National Park (2008). Management plan. Retr ieved from http://retezat.ro/index.php/romana/proiect life/119 plan management pnr sit natura2000.html Riger, S., & Lavrakas, P. J. (1981). Community ties: P atterns of attachment and social interaction in urban neighborhoods. American Journal of Community Psychology, 9 55 66. Roberts, L., & Simpson, F. (1999). Developing partnership approaches to tourism in Central and Eastern Europe. UK Journal of Sustainable Tourism, 7 (3, 4), 35 47. Robertson, J., & Lawes, M. J. (2005). User perceptions of conservation and participatory management of iGxalingenwa forest, South Africa. Environmental Conservation 32 (1), 64 75.

PAGE 203

203 Roggenbuck, J., & Driver, B. L. (2000). Benefits of nonfa cilitated uses of wilderness Wilderness Science in a Time of Change Conference Missoula, MT Rollero, C., & De Piccoli, N. (2010). Place attachment, identification and environment perception: An empirical study. Journal of Environmental Psychology, 30 198 205. Roman, M. M., & Roman, M. D. (2003). Time use for leisure in Romania. Retrieved from www.petra1.istat.it/timeuse/paper/paper55.ppt Romanian National Institute of Statistics. (2007). Statistical y earbook 20 07. Retrieved from http://www.insse.ro/cms/rw/pages/index.en.do Sampson, R. J. (1989). Local friendship ties and community attachment in mass society: A multilev el systemic model. American Sociological Review, 53 766 779. Saris, W. E., Satorra, A., & Srbom, D. (1987). The detection and correction of specification errors in structural equation models. In C. C. Clogg (Ed.), Sociological methodology (pp. 105 129). San Francisco: Jossey Bass. Schultz, P. W. (2002). Inclusion with nature: The psychology of human relationships. In P. Schmuck & P. W. Schultz (Eds.), Psychology of sustainable development (pp. 61 78) Norwell, MA: Kluwer Academic Publishers Group. Schu ltz, P. W., Shriver, C., Tabanico, J. J., & Khazian, A. M. (2004). Implicit connections with nature. Journal of Environmental Psychology, 24 (1), 31 42. Sheldrake, R. (1999). Dogs that know when their owners are coming home. New York, NY: Crown. Snyder, M ., & Cantor, N. (1998). Understanding personality and social behavior: A functionalist strategy. In D.T. Gilbert, S. T. Fiske & G. Lindzey (Eds.) The handbook of social psychology ( pp. 635 679 ) Boston, MA: McGraw Hill. Stanciu, E. (2002). Developing co llaborative management in Retezat National Park, Romania Retrieved from http://www.earthlore.ca/clients/WPC/English/grfx/sessions/PDFs/session_2/Stanciu.p df Stedman, R. C. (2003). Is it really just a social construction? The contribution of the physical environment to sense of place. Society and Natural Resources, 16, 671 686. Stedman, R. C. (2002). Towards a social psychology of place: Predicting behav ior from place based cognitions, attitude, and identity. Environment and Behavior, 34 (5), 405 425.

PAGE 204

204 Steele, J., Bourke, L., Luloff, A. E., Liao, P. S., Theodori, G. L., & Krannich, R. S. (2001). The drop off/ pick up method for household survey research. J ournal of the Community Development Society 32 (2), 230 250. Stern, P. C. (2000). Toward a coherent theory of environmentally significant behavior. Journal of Social Issues 56 (3), 407 424. Stern, P. C., & Dietz, T. (1994). The value basis of environment al concern. Journal of Social Issues, 50 65 84. Stern, P. C. & Oskamp, S. (1987). Managing scarce environmental resources. In D. Stokols & I. Altman (Eds ), Handbook of environmental psychology Vol. 2, (pp 1043 1088) New York NY: Wiley. Stinner, W. F ., Van Loon, M., Chung, S. W., & Byun, Y. (1990). Community size, individual social position, and community attachment. Rural Sociology, 55 (4), 494 521. Strauss, A., & Corbin, J. (1998). Basics of qualitative research: Grounded theory procedures and techn iques (2 nd ed.). Newbury Park, CA: Sage. Strauss, A., & Corbin, J. (1990). Basic of qualitative research: Grounded theory procedures and techniques. Newbury Park, CA: Sage. Stokols, D., & Shumaker, S. A. (1981). People in places: A transactional view of settings. In J. Harvey (Ed.), Cognition, social behavior, and the environment (pp. 441 488). Hillsdale, NJ: Erlbaum. Summers, G. (1986). Rural community development. Annual Review of Sociology 12 237 371. Theodori, G. L. (2004). Examining the associati on between length of residence and community attachment: A research note. Southern Rural Sociology, 20 (1), 107 122. Theodori, G. L. (2000) Levels of analysis and conceptual clarification in community attachm ent and satisfaction research: C onne ctions to co mmunity development. Journal of the Community Development Society, 31 (1), 35 58. Theodori, G. L., & Luloff, A. E. (2002). Position on environmental issues and engagement in proenvironmental behaviors. Society and Natural Resources, 15, 471 482. Theodori, G. L., & Mayfield, C. A. (2008). Attendance at religious meetings and community involvement. Southern Rural Sociology, 28 (1), 266 276. Triandis, H. C. (1980). Values, attitu des and interpersonal behaviour. I n: M. M. Page (Ed.) Nebraska symposium on motiv ation: Beliefs, attitudes and v alues ( pp. 195 259 ) Lincoln NB : University of Nebraska Press.

PAGE 205

205 Tuan, Y. (1974). Topophilia: A study of environmental perception, attitudes, and values Englewood Cliffs, NJ: Prentice Hall. United Nations Development Progra m. (2001). The integration of biodiversity into National Environmental Assessment Procedures. National Case Studies. Romania. Retrieved from http://www.unep.org/bpsp/EIA/Case%2 0Studies/Romania%20(EIA).pdf V an Hal, M. (2006). Evaluation of c o management in national parks: T he case of Retezat National Park, Romania. of Wageningen Netherlands. Van Liere, K. D., & Dunlap, R. E. (1978). M oral norms and environmental behavior: An application of Schwartz's norm activation model to yard burning. Journal of Applied Social Psychology 8 174 188. Van Vugt, M. (2001). Community identification moderating the impact of financial incentive s in a n atural social dilemma: W ater conservation. Personality and Social Psychology Bulletin, 27 1440 1449. Vaske, J. J., & Donnely, M. (1999). A value attitude behavior model predicting wildland preservation voting intentions. Society and Natural Resources, 12 523 537. Vaske, J. J., & Kobrin, K. C. (2001). Place attachment and environmentally responsible behavior. Journal of Environmental Education 32 (4), 16 21. Vining, J. (2003). The connection to other animals and caring for nature. Human Ecology Review, 10 (2), 87 99. Vining, J., & Ebreo, A. (1992). Predicting recycling behavior from global and specific environmental attitudes and changes in recycling opportunities. Journal of Applied Social Psychology 22 1580 1607. Vorkinn, M., & Riese, H. (2001). Env ironmental concern in a local context: The significance of place attachment. Environment and Behavior, 33, 249 263. Walpole, M. J., & Goodwin, H. J. (2001). Local attitudes towards conservation and tourism around Komodo National Park, Indonesia. Environme ntal Conservation, 28 (2), 160 166. Warren, R. (1987). The Community in America (3rd ed.). Lanham, MD: University Press of America. Weladji, R. B., Moe, S. R., & Vedeld, P. (2003). Stakeholder attitudes toward wildlife policy and the Benoue Wildlife Conse rvation Area, North Cameroon. Environmental Conservation, 30 (4), 334 343. Werner, C. W., Altman, I., Brown, B. B., & Ginat, J. (1993). Celebrations in personal

PAGE 206

206 relationships: A transactional/dialectic perspective. In S. Duck (Ed.), Social context and rela tionships: Understandi ng relationship processes series, Vol. 3 (pp. 109 138). Newbury Park, CA: Sage. White, A. T., & Runge, F. C. (1994). Common property and collective action: Lessons from cooperative watershed management in Haiti. Economic Development and Cultural Change 43, 1 41. Williams, D. R., Patterson, M. E., Roggenbuck, J. W., & Watson, A. E. (1992). Beyond the commodity metaphor. Leisure Sciences, 14, 29 46. Williams, D. R., & Roggenbuck, J. W. (1989). Measuring place attachment: Some prelim inary results In L. H. McAvoy & D. Howard (E ds.), Abstracts: 1989 Leisure Research Symposium (p. 32). Arlington, VA: National Recreation and Park Association. Williams, D. R., & Vaske, J. J. (2003). The measurement of place attachment: Validity and gener alizability of a psychometric approach. Forest Science, 49 830 840. Wilkinson, K. P. (1986). In search of the community in the changing country side. Rural Sociology, 51, 1 17. Wilkinson, K. P. (1991). The community in rural America. Middleton WI : Soci al Ecology Press. Woldoff, R. A. (2002). The effects of local stressors on neighborhood attachment. Social Forces, 81 87 116. Yu, C. Y. (2002). Evaluating cutoff criteria of model fit indices for latent variable models with binary and continuous outcome s. Unpublished d octoral dissertation, University of California. Zhang, J., Inbakaran, R. J., & Jackson, M. S. (2006). Understanding community attitudes towards tourism and host guest interaction in the urban rural border region. Tourism Geographies 8 (2), 182 204.

PAGE 207

207 BIOGRAPHICAL SKETCH Natalia Buta studied Marketing at University of Oradea, Romania. After graduation, she worked for two years as a human resources assistant for a company in Romania providing accounting services. Her passion for the outdoors and continues involvement in developing outdoor recreation programs for youth provoked and inspired her to continue her studies Consequently, in 2004 she joined the graduate program in Recreation and Parks Administration at Central Michigan University w here she received her Master of Arts in Recreation and Parks Administration. I n 2006 she began her Ph.D. program at the University of Florida with a concentration in natural resources recreation. Over the years, she has been involved in several research p rojects in cooperation with faculty. She worked on research projects that explored the role of the leisure experience in life satisfaction, impact of outdoor recreation participation on quality of life, as well as image formation process of natural landsca pes in the context of outdoor recreation participation. Furthermore, she conducted research examining online representation of ecotourism operations, primarily focusing on socio cultural, economic, educational, and environmental sustainability messages. He r accumulated research knowledge has formed the building blocks for designing her dissertation research, which is a strong representation of her research interest. She intends on continuing her work to better understand the role of outdoor recreation and natural and cultural resources in shaping community and ecological well being.