1 ASSESSING STAKEHOLDER SUPPORT AND PREFERENCE S FOR MARINE PROTECTED AREA MANAGEMENT ON ANDROS ISLAND, BAHAMAS By DINA LIEBOWITZ A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2007
2 2007 Dina Liebowitz
3 To the Androsian Community
4 ACKNOWLEDGMENTS I would like to thank the many pe ople who made this work possi ble. I owe gratitude to the kindness of the Androsian community as well as th eir willingness to answer my endless queries. This project was greatly aided by the logistical support of th e Bahamas National Trust, The Nature Conservancy, and Forfar Field Station, as well as input from The Bahamas Sportfishing and Conservation Association, the Andros Cons ervancy and Trust, and the College of the Bahamas. I would also like to thank Jacqueli ne Russell for her warmth, acceptance, and fabulous cooking. I am incredibly grateful to my advisor, Dr Susan K. Jacobson, for her guidance, support, and patience, as well as my committee members, Dr Alan Bolten and Dr. Robert Swett, for their generous insights and feedback. I would like to gi ve warm thanks to Jon Dain and the Tropical Conservation and Development program at UF for fostering a collaborative academic environment, as well as for their financial s upport. I would also like to thank the Disney Conservation Foundation for their exceedingly generous funding. This project, as well as my life, was enriched by the energy, generosi ty, and friendship of the graduate student community at the University of Florida. I owe a particular debt of gratitude to Cyndi Langin for her creativit y and eternal good humor. Lastl y, I owe my deepest thanks to my parents, Harold and Judy Liebowitz, my sisters, Esther Glahn, Naomi Maron, and Debbie Kantor, and my extended family and friends fo r being so incredibly supportive, thoughtprovoking, and encouraging throughout this proce ssI could not have done it without them.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........7 LIST OF FIGURES................................................................................................................ .........9 CHAPTER 1 INTRODUCTION..................................................................................................................12 Marine Resource Use and Management.................................................................................12 Classification and Design of MPAs................................................................................13 Biological Efficacy of MPAs..........................................................................................14 Social Efficacy of MPAs.................................................................................................14 Theoretical Context............................................................................................................ ....16 Common Pool Resources................................................................................................16 Stakeholder Involvement and Co-Management..............................................................17 Understanding Factors that Affect Support for Conservation Measures.........................18 Research Context: Fisheries and Ma rine Conservation in the Bahamas................................21 Marine Resource Use......................................................................................................21 Marine Protected Areas in the Bahamas.........................................................................22 Study Site: Andros Island...................................................................................................... .23 Physical Attributes...........................................................................................................23 Community History and Current Stakeholders...............................................................24 History of National Park Establ ishment and Current Management................................25 Research Objectives and Questions........................................................................................26 2 DESIGN AND METHODS....................................................................................................31 Data Collection................................................................................................................ .......31 Surveys........................................................................................................................ ....31 Structure of surveys..................................................................................................31 Sampling method......................................................................................................32 Interviewer effects....................................................................................................32 Group Meeting and Key Informants................................................................................33 Data Analysis.................................................................................................................. ........33 3 RESULTS........................................................................................................................ .......37 Survey Results................................................................................................................. .......37 Socio-demographic and Resource Use Data...................................................................37 Knowledge of National Parks a nd Environmental Organizations...................................39 Perceptions of Resources and Threats.............................................................................39 Preferences for Participation and Outreach.....................................................................41
6 Preferences for Livelihoods, Ma nagement, and Development.......................................41 Support for National Parks, WSNP, and Management Zones.........................................42 Concerns About Parks.....................................................................................................43 Support for National Parks, WSNP, a nd Management Zones: Bivariate Associations.................................................................................................................44 Associations Among Support Variables..........................................................................45 Regression Analysis: Predictors of Support for National Parks, the WSNP, and Management Zones......................................................................................................45 Interviewer Effects............................................................................................................ ......46 Group Meeting Results.......................................................................................................... .48 4 DISCUSSION..................................................................................................................... ....63 Socio-Demographics of the Andros ian Fishers on the West Side..........................................63 Knowledge of National Parks a nd Environmental Organizations..........................................64 Perceptions of Resources and Threats....................................................................................65 Preferences for Livelihoods and Development.......................................................................66 Support for National Parks, the WSNP, and Management Zones: Inconsistencies and Implications for Future Research........................................................................................67 MPAs and Management: The Caribbean Context..................................................................68 MPA Regulations: Compliance and Enforcement Concerns..................................................70 Community Consensus...........................................................................................................71 Limitations of the Study....................................................................................................... ..72 5 RECOMMENDATIONS AND CONCLUSIONS.................................................................74 Implications for Outreach and Participation...........................................................................74 Conclusions.................................................................................................................... .........75 APPENDIX A SURVEY QUESTIONNAIRE...............................................................................................77 B GROUP MEETING DOCUMENTATION............................................................................83 C SPEARMAN'S RANK CORRELATIONS............................................................................85 D CHI-SQUARE FOR RESPONSE AND INDEPENDENT VARIABLES............................90 E CHI-SQUARE AMONG RESPONSE VARIABLES..........................................................106 LIST OF REFERENCES.............................................................................................................108 BIOGRAPHICAL SKETCH.......................................................................................................114
7 LIST OF TABLES Table page 1-1 Agrawals 2002 synthesis of facilitating c onditions for natural resource stewardship.....27 1-2 Stakeholders for the Central Andros National Park System..............................................28 2-1 Locations of samples and percentages of identified resource users interviewed...............35 2-2 Descriptions of variables used fo r bivariate and multivariate analyses.............................36 3-1 Socio-demographics for fishers.........................................................................................50 3-2 Resources collected on the West Side...............................................................................51 3-3 Responses to How do you support yourself if weather is bad for a while or catch is low on Andros?................................................................................................................51 3-4 Knowledge of conservation/ management organizations...................................................51 3-5 Perceptions of threats to the West Side National Park......................................................52 3-6 Preferences for comm unity participation...........................................................................53 3-7 Preferences for outreach met hods and current news sources.............................................53 3-8 Preferences for children's occupati ons, and reasons for that choice..................................54 3-9 Responses to Why do you (support/ not support) general national parks on Andros?"....................................................................................................................... .....54 3-10 Responses to Why do you (support/ not s upport) the protection of the West Side National Park?................................................................................................................ ..55 3-11 Concerns about the West Side National Park....................................................................56 3-12 Significant relationships between support variables a nd independent variables...............57 3-13 Interviewer effects: adjust ed standardized residuals..........................................................58 3-14 Logistic regression for suppor t for national parks on Andros...........................................59 3-15 Logistic regression for support fo r the West Side National Park......................................59 3-16 Logistic regression for support for mana gement zones of the West Side National Park........................................................................................................................... .........60 C-1 Spearman's rank correlation for bivariat e analysis of independent variables....................86
8 D-1 Bivariate associations between res ponse variable (support national parks on Andros) and socio-demographics......................................................................................91 D-2 Bivariate associations between res ponse variable (support national parks on Andros) and knowledge.....................................................................................................93 D-3 Bivariate associations between res ponse variable (support national parks on Andros) and perceptions....................................................................................................94 D-4 Bivariate associations between res ponse variable (support national parks on Andros) and preferences....................................................................................................95 D-5 Bivariate associations between res ponse variable (support WSNP) and sociodemographics................................................................................................................... ..96 D-6 Bivariate associations between response variable (support WSNP) and knowledge....98 D-7 Bivariate associations be tween response variable (suppo rt WSNP) and perceptions...99 D-8 Bivariate associations be tween response variable (suppo rt WSNP) and preferences.100 D-9 Bivariate associations be tween response variable (suppo rt management zones) and socio-demographics.........................................................................................................101 D-10 Bivariate associations be tween response variable (suppo rt management zones) and knowledge...................................................................................................................... ..103 D-11 Bivariate associations be tween response variable (suppo rt management zones) and perceptions.................................................................................................................... ...104 D-12 Bivariate associations be tween response variable (suppo rt management zones) and preferences.................................................................................................................... ...105 E-1 Contingency table of respons e variables: WSNP by NP. ..............................................106 E-2 Contingency table of response va riables: management zones by NP..............................106 E-3 Contingency table of response variab les: management zones by WSNP. .....................106 E-4 Reduced contingency table of re sponse variables: WSNP by NP...................................106 E-5 Reduced contingency table of response variables: management zones by NP. .............107 E-6 Reduced contingency table of response variables: management zones by WSNP. ........107
9 LIST OF FIGURES Figure page 1-1 National park boundaries for the Cent ral Andros National Park system...........................29 1-2 Reference map of MPAs in the Caribbean region.............................................................30 3-1 Knowledge and support of na tional parks and the WSNP.................................................60 3-2 Perceptions of resource trends...........................................................................................61 3-3 Preferences for development on the West Side.................................................................61 3-4 Preferences for types of regulations for the WSNP...........................................................62 3-5 Support for management zones fo r the West Side National Park......................................62
10 Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science ASSESSING STAKEHOLDER SUPPORT AND PREFERENCE S FOR MARINE PROTECTED AREA MANAGEMENT ON ANDROS ISLAND, BAHAMAS By Dina Liebowitz August 2007 Chair: Susan K. Jacobson Major: Wildlife Ecology and Conservation Fisheries are declining globally, and marine biodiversity loss is accelerating. Marine Protected Areas (MPAs) are a form of manage ment proposed to help restore impoverished fisheries, create sustainable resource use syst ems, and conserve biodiversity. However, people rely on the use of marine systems, and more restrictive management may create problems for coastal resource users. Consequent ly, managers are attempting to involve local marine resource users in the MPA planning proce ss, to develop management plan s that facilitate the long-term sustainability of both MPAs and community livelihoods. The Bahamas is currently creating a national network of MPAs. The West Side National Park (WSNP) on Andros Island was established to protect wetlands and coastal environments that support globally significant species and locally signi ficant fisheries, but it does not yet have active management in place. This study surveys local community members on Andros Island, in order to discern what factors are associated w ith local support of natural resource stewardship and incorporate social data into the management of the MPA. Key-informant interviews and a group meeting provided context fo r this study. A 73-item survey of 117 resource users explored the communitys socio-demographic compositi on, knowledge, perceptions, preferences, and support for national parks, the WSNP, and management zones for the WSNP.
11 Results showed that 37% of the responde nts knew about the WSNP, and 20% were familiar with all four of the environmental orga nizations in the area. Responses indicate some general trends in perceptions of changes in resource conditions on the West Side: 54% perceive that bonefish have increased, perceptions of spo nges are mixed, and 48% perceive that crawfish have decreased on the West Side. Perceptions of threats to the West Side were coded into 6 thematic categories: development pressures (n=6 1), poaching (n=47), chemical fishing (n=31), natural disasters (n=23), not hing (n=22), and politics (n= 11). When given options for development, 54% wanted no new development on the West Side, 27% preferred small-scale tourism, and 7% wanted large resorts. Respondents had mixed pr eferences for various types of management practices: the most favored was the volunteer ranger program (84%), and the least favored was limiting the number of people in the national park at a time (41%). The study tested support for na tional parks in general (70% support), the WSNP (57% support), and management zones for the WSNP ( 50% support). Logistic regressions and residual analyses both revealed that each of the three sup port variables is predicted by a different suite of independent variables, indicating a need for extensive outreach effo rts to clarify the implications of neighboring national parks for the local community. Forty-nine percent of respondents reported concern about losing th eir livelihoods, while perceptions of benefits included improvement of fishery resources, protection of wildlife, and the creation of new jobs. Respondents were interested in interactive forms of engagement, with 82 people requesting community meetings and 81 wa nting to aid in community education. These data are a starting point for future researc h, but should be interpre ted cautiously due to interviewer effects.
12 CHAPTER 1 INTRODUCTION Marine Resource Use and Management Marine systems produce an immense amount of resources that people use for consumption and recreation. Human populations rely on almost all coastal marine systems across the globe (National Research Council 2001), but communities that depend on these resources are feeling the pressures of their depleti on. Once thought too vast to be negatively impacted by resource exploitation, research now shows marine production to be severely declining (Worm et al. 2006). World-wide, marine resources are crashing. According to th e National Research Council (2001), 2530% of the worlds fisheries have been overfished and an additional 44% have been fully exploited. Overwhelming human pressure on marine resources also has severely depleted biodiversity and diminished ecological systems, causing some to cease functioning completely (Argady 1997; Jackson et al. 2001). A meta-analysis of worldwide fisheries da ta predicts that if the current trend continues unchecked, all of the current world fisheries will crash by 2048 (Worm et al. 2006). Given that almost 75% of th e worlds fisheries are curren tly overexploited or at full capacity (National Research Council 2001) some recognize that measures must be taken to manage fisheries in a sustainable manner. But marine systems are particularly difficult to manage, as they are common-pool resources that may be susceptible to overexploitation and vulnerable to Hardins (1968) tr agedy of the commons. Historica lly, the cultural mores inherent within some communities have effectively regulated resource exploitation. However, as societies change and grow and resource pressures increa se, these forms of management may no longer adequately protect resources. More recently, many governments st arted regulating resource use with management techniques such as permit syst ems, gear regulations, and catch quotas (Argady
13 1997). In response to severe resource degradati on, Marine Protected Areas (MPAs) were proposed as an additional management technique to conserve biodiversit y as well as restore diminishing fish stocks for neighboring fisherie s. Kelleher et al. (1995) identified 1,306 MPAs globally, though management goals, methods, and e ffectiveness in these sites vary widely. Classification and Design of MPAs The IUCN (1988) defined a marine protected area as Any area of intertid al or subtidal terrain, together with its overl ying water and associated flora, fauna, historical and cultural features, which has been reserved by law or other effective m eans to protect part or all of the enclosed environment. The vagueness of this defi nition can be seen as either a strength or weakness, as it is inclusive, yet potentially mean ingless in its attempt to be all-encompassing. Clear goals for MPAs must be developed in order to design management plans which can balance both biological and soci al needs. Some of the biologi cal objectives of MPAs include protecting specific populations of interest, biodi versity hotspots, and landscape processes. Biologically, managers must balanc e a number of elements, such as size of the protected area (to prevent excessive amounts of the resource lost to spillover effects), protection of multiple habitats for organisms that requir e a diversity of environments fo r their development, replication of habitats, and representation of all distinct habitats (Nowlis & Friedlander 2004). On the social side, the reserve size needs to be small enough to allow at least a limited amount of spillover effect, to enable the local communities to ma ke use of increased abundance of resources. MPA management plans can use a variety of strategi es to accomplish their goals. Common strategies include permanently closing areas to human us e, closing areas seasonally, and establishing spatial zones to regulate ex tractive and non-extractive uses.
14 Biological Efficacy of MPAs Fishers opposed to MPAs often claim that th e biological benefits of closing areas to extraction (no-take zones) ar e yet unproven (Jones 2006). They protest closures, claiming undue hardship for unproven gains, and they cite intensifie d extraction pressure on unprotected areas that are adjacent to no-take zones. Demonstra ting the benefits of notake zones is indeed problematic, since an insufficient number of repl ications for comparable areas exist to provide for statistically robust studies. However, some research shows distinct and rapid gains in abundance, size, and diversity of marine organisms. For exampl e, a study in the Exuma Cays Land and Sea Park demonstrated "increased abun dance, size, and reprodu ctive output of Nassau grouper in the park compared to areas north and south" and a study in Floridas Looe Key Reef found that "after two years protection from sp ear fishing, snapper abundance increased 93% and grunt abundance increased 439%" (BREEF 1998). The benefits of MPAs fall into four main categories: (1) protect ecosystem structure, function, and integrity; (2) improve fisheries [t hrough creating more abunda nt, larger, and more diverse stocks]; (3) expand knowledge and understa nding of marine systems; (4) and enhance nonconsumptive opportunities (Sobel & Dahlgren 2004). While these benefits have not been proven to everyones satisfaction, they are part of the rationa le for supporting MPAs, alongside the precautionary principle, wh ich suggests extreme caution when taking risks in situations where there are many unknown factor s and potentially irreversible losses (National Research Council 2001). Social Efficacy of MPAs MPAs do not exist in a void. Human populations rely on almost all coastal marine systems across the globe (National Resear ch Council 2001), and closing off certain areas for conservation can affect coastal resource users in both posit ive and negative manners. As MPAs are a relatively
15 recent management techniquethe Exuma Cays La nd and Sea Park in The Bahamas, established in 1958, is one of the first MPAs in the Wide r Caribbean regionrecognition of the potential impacts of MPAs on nearby communities is recent as well. There are few long-term socioeconomic studies to track these effects and existing information is largely anecdotal (Gell & Roberts 2003). However, the scientific and management communities acknowledged the social element as a pressing research need (Arg ady 1997; Chuenpagdee et al. 2004; Lundquist & Granek 2005; Mascia 2003), and monitoring pr ograms are being established globally through joint efforts by the World Commission on Prot ected Areas, the Nati onal Oceanographic and Atmospheric Administration, The Centre for Res ource Management and Environmental Studies, the Global Coral Reef Monitoring Network, and others. Benefits of MPAs cited in pr actitioner literature include educational opportunities for local communities, income diversification, reduced conflicts through joint decision making, and increased fisheries catch due to spillover eff ects (Badalamenti et al. 2002; Gell & Roberts 2003). Negative impacts on communities can include loss of income when important fishing grounds are closed, increased conflic t surrounding usufruct rights and parity, and insufficient communication and collaboration among stakeholder groups. For example, local fishermen have stated that they were unfairly punished when they were excluded from fishing grounds while tourism interests were not, despite the fact that heavy tourism can harm the environment (Gell & Roberts 2003). MPAs can also di srupt local settlement patterns and reshape community structure through external pressures, such as occur when switching to less extractiv e occupations such as tourism. Traditional managers and governing bodies ha ve recently recognized the potential for social conflict within comm unities affected by MPAs and ac knowledged that local communities
16 of marine resource users must be involved in the MPA planning process in order to make the protection system valid and practical in the eyes of the stakeholders, a nd ensure the long-term sustainability of the MPA (Dixon 1993; Geoghegan et al. 1999). Theoretical Context Common Pool Resources In general, marine resources ar e not easily delineated, and ther efore they often fall into the open access (an extractive free for all) vs. co mmon property (community management of a commonly owned resource) resource management debate. Hardins tragedy of the commons is often invoked to argue that such common pool re sources will always be treated as open access and are bound to be overexploited. A criticism of Hardins work is the suggestion that he made no distinctions between open access and common pr operty management, and that it is indeed possible to manage common resources in a su stainable manner (Ostrom 1990). In response to this debate, much research has explored the differences between open access and common property resource management to decipher what contributes to sustainable management of common resources. Agrawal (2002) synthesized a list of facilitati ng conditions for sustainable use of common pool resources. These conditions in clude: resource system characteristics, group characteristics, the relationships between th e community and their resources, institutional arrangements, relationships between the resour ce systems and their institutional arrangements, and the exogenous forces (Table 1-1). These facilitating conditions provide an overarching framewor k to explore what factors may be important in advancing successful mana gement of marine reso urces. Possible factors cover a wide range of scales, from the relatively immutable characteristics of the system (such as global currents), to national polic ies, to daily local interactions. Research across all scales is necessary, but to investigate the local issues more specifically, it is important to focus research
17 on group characteristics and their interactions with the resource system (R ichardson et al. 2005), as suggested by Agrawal's 2002 framework. This st udy explores a subset of these conditions by providing baseline data on group characteristic s, and examining relationships among group characteristics and support of resource stewardship through MPAs. Stakeholder Involvement and Co-Management In addition to the importance of biological cr iteria for establishing an MPA, numerous studies stress that social fact ors often determine the MPA's ul timate success or failure (Argady 1997; Chuenpagdee & Pauly 2004; Lundquist & Granek 2005; Masc ia 2003; Salz & Loomis 2005). Many studies highlight the need to invo lve local communities in all aspects of the management of those resources that directly affect them (Dixon 1993; Geoghegan et al. 1999). Friedlander et al. (2003) state th at a successful management plan requires stakeholder input to adequately understand the physical resources and local patterns of use, while Mehta (2001) believes it essential to account for the social el ements of the communities themselves, saying that the long-term survival of protected areas in de veloping nations will be jeopardized if needs, aspirations, and attitudes of local people are not accounted for. A variety of frameworks exist for defining the role of stakeholders in management decisions. They range from complete control by a central government (top-down) to the complete autonomy of local communities, with no intervention from the central government (Christie et al. 2003; Chuenpagd ee et al. 2004; Napier et al. 2005; Pomeroy & Berkes 1997). Currently, a co-management system that include s local communities and the central government is regarded as the best arrangement to ach ieve successful natural resource management (Geoghegan et al. 2001). However, regardless of the management arrangement, information about the community and their needs and attit udes will enable more successful communication
18 and the development of locally acceptable management plans (Chuenpagdee et al. 2004; Lundquist & Granek 2005; Salz & Loomis 2005). Understanding the socio-demographic and attitudinal composition of a community has a number of implications for management, such as enabling more effective outreach programs by better identification of the target audience, as well as providing baseline information to evaluate program effectiveness (Jacobson 1999). In addi tion, such information helps managers account for the livelihood needs of the community and thus create management structures that will not impinge on the ability of the loca l community to sustain itself. Understanding Factors that Affect Support for Conservation Measures Much research focuses on understanding factors that affect stewardship decisions made by local communities. Numerous variables have been tested to determine their effects on community support for both terrestrial and marine pr otected areas (Routhe et al. 2005), as well as their implications on the success of resource management or cons ervation programs. Many of the factors explored fall within the categories of socio-demographics, knowledge, perceptions, and preferences. Marine and terrestrial protect ed area management research often tests demographic and socio-economic variables, such as: age, educatio n, location/proximity to protected area, gender, ethnicity, wealth, diet, pluriact ivity, number of years fished, livelihoods, costs of living, and property ownership. The effects of each vari able are different depending on the context (Alexander 2000; Gelcich et al. 200 5; Infield 1988; McClanahan et al. 2005; Mehta & Heinen 2001; Napier et al. 2005; Pomeroy et al. 1996; Spiteri & Nepalz 2006). For example, a high level of resource dependence may foster greater acceptance of local resource management effortsor it may lead to resource users believing that they have no options but to extract what they can in order to survive (S esabo et al. 2006).
19 Some literature examines the role of the resource users knowledge in project success (where success is defined differently dependi ng on context and manageme nt goals), looking at issues such as awareness of conservation issues or knowledg e of rules and MPA objectives (Aipanjiguly et al. 2003; Blake 2001; Sah & Heinen 2001; Sesabo et al. 2006; Spiteri & Nepalz 2006). For example, early involvement in conserva tion projects and associated training programs is associated with supportive attitudes (Mehta & Heinen 2001; Sesabo et al. 2006). However, knowledge is not a direct predictor of support for conservation. Blake (2001) discusses how personal experiences with ecologi cal problems may lead to support of environmental policies in some cases but not in others, possibly due to vari ous intervening factors such as wealth. Alessa (2003) found that greater ecological knowledge was actually positively correlated with more depreciative behaviors on the environment. Th e connection between knowl edge and behavior are not straightforward, and may be me diated by a wide variety of elements, such as economics or social norms. Research has also indicated that perceptions are important predictors of support for MPAs. For example, a study of depreciative behaviors by Alessa et al. (2003) found that personal attribution and perception of ecosystem resilience were the main predictive variables, showing that people who believed their act ions can cause change were more likely to be wary of causing damage. Hinds et al. (1997) found that the per ception that fishery reso urces are declining has shown association with support for conservation measures. However, Richardson et al. (2005) found that fishers who perceived that overwhe lming external factor s caused widespread degradation were less likely to try to mediate their own actions, and Mascia (2004) found that some Caribbean fishers did not support MPAs because they believed damage was caused by terrestrial threats or na tural variability, not by their own fishing activitie s. Others have found that
20 supportive attitudes were negatively associated with perceptions of corruption, lack of transparency, or negative inte ractions with conservation offi cials (Lepp & Holland 2006; Mehta & Heinen 2001; Sesabo et al. 2006). However, some argue that support for MPAs may also depend on the perceptions of tangible benefits and costs of conserva tion actions, such as financial gains, infrastructural additions, or cl ear recovery of marine resource populations (Salm et al. 2000). While searching for factors that are associ ated with successful management, Agrawal (2001) suggests that researchers be wary of gath ering data on myriad cond itions that may not be consistent across time and space. Routhe et al. (2005) also voices concern over this approach, claiming that environmental con cern studies are still largely at heoretical and descriptive, lending themselves less to testing hypothese s and furthering theory, and more towards pragmatically informing local po licy. Nevertheless, a great deal of evidence shows that such contextually relevant information is essential to inform local management plans and aid in the sustainable use of resources (Alexander 2000; Fiallo & Jacobson 1995; Ge lcich et al. 2005; Lepp & Holland 2006; Mascia 1999; McClanahan et al. 2005; Sesabo et al. 2006). Therefore, despite the legitimate concern that locally-oriented research approaches should be unified by an overarching framework, these data still contribut e by informing local policy and furthering management needs. This study focuses on one of Agrawals ( 2002) facilitating conditions for resource management mentioned above: the resource user characteristics (socio-demographic as well as psychometric). It tests the associ ations among a number of contextu ally relevant factors that fall into the four categories discussed above: socio-demographics, know ledge, perceptions, and preferences. The research is designed to help managers consider local needs and thereby create
21 viable management plans, as well as identify key target audiences a nd appropriate outreach messages. Research Context: Fisheries and Marine Conservation in the Bahamas Marine Resource Use Fishery resources are an important sector of the Bahamian economy. According to the Fisheries Management Action Plan for the Bahamas developed by the Bahamas Reef Environment Educational Foundation (BREEF), commercial fisheries only generate approximately 2.3% of the GDP, yet commercial fishing provides full employment for 6.8% of the Bahamian workforces and partial employ ment for 8.8%. Additionally, fisheries supply a majority of the nutritional needs for the island inhabitants (BREEF 1998). This dependence lends gravity to the issue that all Bahamian commercial fisheries are severely threatened. According to the BREEF report on the main commercial fisheries: Crawfish fisheries in the Caribbean regi on are mainly at maximum exploitation or overfished. Conch is also considered overfished in the Ca ribbean, and has been placed on Appendix II of CITES [Convention on Internati onal Trade in Endangered Species of Wild Fauna and Flora], as a species which may become threatened with extinction unless trade in specimens is subject to strict regulation. The conservation status of Na ssau grouper in the Caribbean is the most worrying, with overfishing the rule almost everywhere and sp awning aggregations continuing to disappear. Experience in the rest of the Caribbean has s hown that anything other than very controlled fishing pressure on spawning aggregati ons is unsustainable (BREEF 1998). In addition to the importance of the fisher y resources for extrac tion and nutrition, the marine environment is the basis for the touris m industry, which produces approximately 40% of the nations GDP. Therefore, there is a need to pursue management and conservation measures to ensure the sustainability of the marine resources for all uses.
22 Marine Protected Areas in The Bahamas The Commonwealth of the Bahamas created one of the first MPAs in the Caribbean: the Exuma Cays Land and Sea Park (ECLSP). The EC LSP was established in 1958, and is managed by the Bahamas National Trust (BNT), a quasi -governmental organization mandated by the government to develop and manage the national parks system. This MPA has been successful biologically, with documented incr eases in abundance, size, and di versity of organisms within the boundaries (Dahlgren 2004). Socially, the accept ance of regulations has been mixed, as the creation of the ECLSP was not inclusive of co mmunity needs. Poaching has been a problem, which led BNT to work more closely with the community in recent years (BNT meeting, 2006). In 2000, the government began working with BNT and stakeholder groups from around the country to establish a national ne twork of MPAs, some of which are in the process of creating management plans. The Fisheries Management Action Plan for the Bahamas (BREEF 1998) adopted a more inclusive approach than the original establishm ent of the ECLSP, and states that there are various criteria for reserve design and site sele ction, but given the area of relatively unspoiled marine environment that exists within The Bahamas, the main criteria for site selection should be social and economic rather than biological. Accord ing to Dahlgren (2004), the criteria that led to the establishment of the new suite of MPAs in 1998 included the following: Socioeconomic Criteria Fishing Impactthe degree to which fishing w ill be displaced due to the creation of the MPA Community Managementthe ability of the local communities or existing organizations to participate in management of the MPA Community Benefitslikelihood that marine reserve would provide benefits to local communities
23 Ecological Criteria Habitat diversitythe diversity of marine habitats important for supporting Bahamian fisheries and marine biodivers ity. These habitats included se a grass, mangroves, and coral reefs. Regional importancethe potential importan ce of the area for supporting fisheries throughout the Bahamas. Because the prevailing currents in the Bahamas run from SE to NW, larval retention within the Bahamas is likely to be greatest for sites in the SE half of the country. (Dahlgren 2004) Study Site: Andros Island Physical Attributes Andros Island, though sparsely settled, is the largest island in The Bahamas (2300 square miles) and one of the closest to Florida. Andros serves as a major sour ce of drinking water for Nassau and is considered unique due to its numerous blue holes (vertical cave-like habitats, both terrestrial and marine). The major infrastructure on the island includes a US/Bahamian joint naval station (AUTEC), government offices, tourism infrastructure four airports, a clinic, an energy plant, and an agricultural packing plant. Terrestrial ecosystems on Andros consist of mixed hardwoods and Caribbean pine forests, and marine resources include the third largest ba rrier reef in the worl d (along the east coast), tidal creeks, and mangrove wetlands on the west side. Organisms of global and regional interest include the West Indian Flamingo, green, logg erhead, hawksbill turtles, and commercially exploited resources such lobste r, stonecrab, sponge, conch, scal efish (such as Nassau grouper), and sportfish (such as bonefish, tarpon, and permit). A Rapid Ecological Assessment conducted in 2006 provides baseline data on island resources and the physical environment, howev er little information on resource-use and depletion trajectories exist for this region. A 2006 workshop attended by local leaders identified several potential threats to the island resources, including over fishing, improper fishing methods,
24 pollution, tourism, and development pressures (BNT Management Planning Training Meeting, February 2006). Community History and Current Stakeholders The island population consists of approxi mately 7,686 people (2000 census, GoB). The original inhabitants were the Lucayan Indians, wh o were decimated subsequent to the arrival of Europeans. Andros was later se ttled by a colony of freed slaves who were joined by Seminole Indians from southern Florid a (pers. comm. Reverend Newt on 2006). Most of the current settlements, apart from Red Bays and Lowe Soun d, are along the East side adjoining the islands one main highway. The Caribbean Regional Environmental Prog ram (CREP) conducted a project on Andros that identified stakeholders with interests in the national park system, and created a board of representatives. While both the project and stak eholder board are no long er active, the list of stakeholders provides a basis to engage the necessary parties for future efforts. Table 2-1 lists these stakeholders, along with those not include d in the CREP board but with interests in the MPA process and outcomes. Despite the large numb er of stakeholders, only a small number of groups could be surveyed due to the time limitations of the study. There are numerous systems for classifying stakeholders; such as categorizing people according to their potential to influence decisions and project outcomes, or the degree to which a project will affect stakeholde rs. Resource users frequently ha ve minor influence on resource management decisions and projects, but they ar e often highly affected by these decisions and have a large capacity to determine the fate of the project (Garaway & Esteban 2003). Therefore, this study focuses on the direct users of the marine resources under consideration.
25 History of National Park Establ ishment and Current Management In 2002, the government approved the boundaries for 5 national parks on Andros Island, collectively known as the Central Andros National Park system (Figure 2-1), which were created in response to requests by BNT, the Bahama s Sportfishing and Cons ervation Association (BCSA), and the Andros Conservancy and Trus t (ANCAT). The area un der consideration for this study is the largest of the five (68,257 hectares), located on th e west coast and inland (herein referred to as the West Side National Park, or WSNP). The WSNP is considered an MPA as it encompasses tidal creeks, mangrove systems, and th e coastal zone. It is said to include breeding grounds for some of the marine species (such as bonefish), and possibly for the West Indian flamingo. This marine protected area is inaccessible by land, though an old logging road approaches its northern border. The area is access ible by boat, though it is considered remote by locals and expensive to reach due to high gas prices. No management plans exist yet for the MPA, th erefore the current rules remain the same as those in the general fisheries. The Department of Fisheries (DoF ) manages the fisheries using a permit system and catch regulations, and by cl osing seasons for the three big commercial species: Nassau grouper, queen conch, and spiny lobster. The solitary DoF extension agent on Andros conducts spot-checks of catches as th e boats return to the docks. Data collection, research, and coastal enforcement are the respon sibility of the DoF, while the Royal Bahamas Defense Force (RBDF) is responsible fo r enforcement farther out at sea. The Bahamas National Trust held a Manageme nt Planning Training Meeting in February 2006, and is now working with st akeholders to begin the pro cess of developing management plans for each of the national parks. BNT is focu sing much effort on creating viable management plans for the national parks on Andros in concert w ith local groups, and theref ore it is a critical
26 and opportune time to involve stakeholders and u nderstand their interest and needs in reference to the WSNP. Research Objectives and Questions The objective of this research is to explor e which of the factor s (socio-demographics, knowledge, perceptions, and prefer ences) are associated with lo cal community support of MPA management on Andros Island, Bahamas. The research is intended to facil itate effective marine resource management and conservation through informing the management planning process with the attitudes and needs of resource user gr oups. The data from the study will also help to develop outreach messages that address local conc erns and information needs and thereby foster support for sustainable resource management measures. The research questions are: What is the basic socio-demographic profile of users of the West Side National Park resources? What are the current resource users know ledge, perceptions, preferences, and support toward the marine resources and manageme nt of the West Side National Park? What factors are associated with support fo r national parks in general, the West Side National Park, and management zones for the West Side National Park? What are the outreach needs and participatory inte rests of the local mari ne resource users of the West Side on Andros Island?
27 Table 1-1. Synthesis of facil itating conditions for natural re source stewardship by Agrawal (2002). 1. Resource system characteristics i. Small size ii. Well-defined boundaries 2. Group characteristics i. Small size ii. Clearly defined boundaries iii. Shared norms iv. Past successful experiencessocial capital v. Appropriate leadershipyoung, familiar with changing external environments, connected to local traditional elites vi. Interdependence among group members vii. Heterogeneity of endowments, homoge neity of identities and interests (1 and 2). Relationship between resource system characteristics and group characteristics. i. Overlap between user group resident ial location and resource location ii. High level of dependence by group members on resource system iii. Fairness in allocation of bene fits from common resources 3. Institutional arrangements i. Rules are simple and easy to understand ii. Locally devised access and management rules iii. Ease in enforcement of rules iv. Graduated sanctions v. Availability of low cost adjudication (1 and 3). Relationship between resource system and institutional arrangements i. Match restrictions on harvests to regeneration of resources 4. External Environment i. Technology: Low-cost exclusion technology ii. State: a. Central governments should not undermine local authority b. Supportive external sanctioning institutions c. Appropriate levels of external ai d to compensate local users for conservation activities d. Nested levels of appropria tion, provision, enforcement, governance Source: Agrawal, A. 2002. Common resources and in stitutional sustainability. Pages 41-85 in T. Dietz, and E. Ostrom, editors. The Drama of the Commons. National Research Council, Washington, D.C. Agrawals synthesis of ideas draws from Wade (1989), Ostrom (1990), and Balland and Platteau (1998).
28 Table 1-2. Stakeholders for the Cent ral Andros National Park System CREP Stakeholder Board for the Central A ndros National Park System (Disbanded) 1. Bahamas Sportfishing and Conservation International NGO comprised of sportfishers Association (BSCA) working for conservation aims. 2. Department of Agriculture National Government 3. Ministry of Tourism National Government 4. Andros Conservancy and Trust (ANCAT) Local conservation NGO in Fresh Creek 5. The Bahamas Environment, Science and Government commission of representatives Technology (BEST) Commission from environmental oversight departments. 6. Bahamas National Trust (BNT) National NGO mandated by the government to create/manage the National Park system. 7. Crabbers Local land-crab gatherers 8. Fishermen Local marine resource extractors 9. Ministry of Education National Government 10. College of the Bahamas (COB) National level higher education institution 11. Resort and Tour Operators Tourism industries 12. Culture and Craft Private 13. Department of Fisheries (DOF) National Government 14. Caribbean Regional Environmental NGO working with participatory processes Programme (CREP) in management. No longer active on Andros. Additional Stakeholders 15. The Nature Conservancy (TNC) International NGO working with the Andros West Side project. 16. Bahamas Reef Environment Environmental education focused NGO. Educational Foundation (BREEF) 17. Bahamas Environmental Research Bran ch of CoB, research station on Andros Center (BERC)
29 Figure 1-1. National park boundaries for the Centra l Andros National Park system. The largest park is referred to in this document as the West Side National Park (WSNP). Source: The Nature Conservancy, Northern Caribbean Program. West Side National Park
30 Figure 1-2. Reference map of MPAs in th e Caribbean region in 1995. Source: Kelleher, G., Bleakley, C., Wells, S. 1995. A Global Representative System of Marine Protected Areas. Volume 2: Pages 1-42. The World Bank. Washington, D.C Andros Island
31 CHAPTER 2 DESIGN AND METHODS Data Collection Surveys Data was collected MayAugust 2006 and January 2007 using structured, face-to-face surveys. The survey instrument was a 73-item questionnaire that contai ned close-ended, openended, and Likert-type questions (Appendix A). The questionnaire was designed with input from University of Florida faculty and local partners and it was pre-tested with 15 resource users on Andros for clarity and applicability to the lo cal context. After adapting the instrument in response to the feedback from the pre-tests and a group meeting, 117 surveys were conducted by three surveyors: the researcher and two field assistants. Surveys took approximately one hour each to administer, but completion time ranged from 45 minutes to 2.5 hours. Surveys were conducted in a respondents home or at anot her location selected by the respondent. Structure of surveys The survey covered a broad range of topics th at might be associated with resource use, livelihoods, and community engagement in MPAs: Demographics (5 items) Current resource access, use and intensity (8 items) Livelihood strategies and res ource dependence (11 items) Areas that resource users deem im portant for protection (2 items) Knowledge concerning the MPA and cons ervation organizations (5 items) Perception of trends in the st ate of the resources (8 items) Preferences for management and development (15 items) Preferences and barriers for participation in MPA development and management (7 items) Support of the MPA and management practices (8 items) Outreach needs and preferences (4 items) In the open ended questions, respondents were al lowed to give multiple responses, and all were coded thematically for analysis. Variables used fo r quantitative analysis ar e defined in Table 2-2.
32 Sampling method Due to a lack of lists of local fishers and the transitory nature of fishing communities, sampling was done in a purposive manner (Bernard 2002) to find fishers who were most likely to use the West Side resources. The basic structure of each fishing commun ity was determined by consulting with local community leaders, which led to a concentrati on of research efforts on fishing communities located in Lowe Sound, Red Bays, Stafford Creek, Fresh Creek, Bowen Sound, Cargill Creek, Behring Point, and Mangrove Cay (Table 2-1) Lowe Sound and Red Bays fishers focus on commercial fishing, while fishers from Cent ral Andros, Bowen Sound, Cargill Creek, and Behring Point focus more on sport fishing. To create groups large enough for analysis, Central Andros, Bowen Sound, Cargill Creek, and Behri ng Point are combined into the Behring Point/Central Andros category, as they are geog raphically close and of a similar composition of types of fishers. Mangrove Cay fishers include a mix of commercial and sport fishers. Due to the non-probability sample, these data ca nnot be interpreted as representative of the entire community; therefor e, these data should be used as a st arting point for further work within these communities. Interviewer effects Three interviewers surveyed the respondents fo r this research: two females from the U.S. and one male from The Bahamas. All surveys with multiple interviewers must deal with potential biases, such as interviewer acquiescence bias, mistrust of interviewers, and differential responses to interviewer gender, age, a nd race (Aquilino 1994, Davis 1997, Finkel et al. 1991, Kane and Macaulay 1993). This research found statis tical differences that ma y be attributable to the interviewer effects; however, the limited sample size, irregular distribution of interviewers by geographic location and type of fisher, and high number of correlations among the variables
33 makes it impossible to quantify these effects. Ch i-square comparisons a nd logistic regressions are methods of exploring the potentia l bias and will be discussed late r; however, in light of this interviewer bias, this work mu st be interpreted with caution. Group Meeting and Key Informants The lead researcher held a group meeti ng in Red Bays on May 25, 2006 to collect qualitative data on the research topics and help finalize the design of the survey instrument (Appendix B). The meeting was attended by appr oximately 25 community members from Red Bays. The meeting format was intended to be a focus group, but community members entered and exited during the meeting, and it was decided not to try to control entry to the meeting due to: (1) the sensitive nature of the topic and (2) pre-existing local apprehensions towards researchers, which would have b een heightened in the absence of complete transparency of the activities to the entire community. Questions we re presented verbally and on flip-charts, and included the following: How would you describe the he alth of the West Side? What fishing areas are most important to you? What areas are important to protect? What type of regulations would you support? How would you like to be involved in management? What is the best way to provide in formation to the Red Bays Community? Answers were recorded on flipchar ts to ensure accuracy of repr esentation of the statements. To obtain contextual information about A ndros Island and resource use issues, key informant interviews were conducted with 17 lo cal merchants, government and NGO officials, and non-fishing community members. Data Analysis Data was entered into an Excel spreadsheet and then imported into both SAS and SPSS for analysis. Chi-square ( ) statistics were calculated in SPSS for all num eric variables and, when
34 assumptions for were violated, Fishers Exact test was calculated using SAS. For significant values, Cramers V was calculated as a measure of association appropriate for categorical data (0 indicates no association a nd 1 perfect association). Cross-tabulations of all signi ficantly associated variables were examined. Adjusted standardized residuals were calcul ated to indicate the direction and strength of association for the values with significant values. For residual analysis, a residual > |2| shows significant departure from independence in that cell at a 95% confidence interv al, and a residu al > |3| shows significance at a 98% confidence interval (Table 3-12). Negative residual values indicate less support for the response variable. Spearman's Rho corre lation coefficients were calculated to test interactions among independent variables. Support measures were compared among the interviewers using tests with collapsed categories to allow for residual analysis with sma ller sample sizes and to explore interview bias issues. The full data set was compared with its tw o subsets: data collected by interviewers 1 and 2 (U.S. females) and data collected by interviewer 3 (Bahamian male). The Resource condition perception scale was created by adding the scores for the three resources (bonefish, sponge, crawfish), then coll apsed to three groups fo r categorical analysis. The full scale was used for regression analysis to enable greater degrees of freedom, so that convergence criteria were met. For multivariate analyses, a l ogistic regression was run on SPSS for the binary response variables: support national parks, support the West Side Nation al Park, and support management zones for the West Side National Park. Due to the large number of poten tial variables and the limitations of the sample size and effective samp le size (due to item non-response), independent variables that showed a high degree of a ssociation with the re sponse variables through tests
35 were chosen and added in a backwards step-wise manner. The same variables were entered in the model for all three response variable s to enable comparison of effects. Qualitative, open-ended answers were entered into Excel, coded for themes, and summarized in tables. Contingency tables were created to examine the associations of the coded data with the response variables. The threats variable was broke n into dummy variables to be included in bivariate an d regression analyses. Table 2-1. Locations of samples and percentage s of identified resource users interviewed, excluding pre-tests. Lowe Sound Red Bays Behring Point/ Central Andros Mangrove Cay Sportfishers 75 % (6/8) N/A 28% (15/53) 67% (14/21) Commercial 34 % (20/59) 55% (34/62) 3/Unknown 72% (18/25)* Mangrove Cay additionally has up to 200 commerci al fisher who are out to sea for several weeks at a time and fish mostly for crawfish on the Great Bahamas Bank, 10+ miles from shore. Though they could not be interviewed at this time and are less directly af fected by the near-shore MPA, they should be consider ed for future studies.
36 Table 2-2. Descriptions of variables used for bivariate and multivariate analyses. Variable Description Dependent Variable 1. Support NP Respondents were asked if they thought that it was a good idea to have National Parks on Andros Island. N=0, Y=1 2. Support WSNP Respondents were asked if it was important to protect the WSNP. No = 1, Do not know (DK) = 2, Yes = 3. For multivariate anal ysis, categories were collapsed: Do not support/DK=0, Support = 1, 3. Support management zones for WSNP Respondents were shown a map of the current boundaries of the park and given four options for spatial zonation of the park, increasing in management intensity. 1= No new regulation up to 4 = half no-entry and half no-take zones. Collapsed to: 0 = do not support zones (previously 1), 1= Support zones (previously 2-4) Independent Variables: Socio-demographics 1. Location Respondents were coded acco rding to where they currently lived. 2. Fisher type This variable refers to their self-categorized main occupation as either: (0) a commercial fisherman (1) a sport-fishing guide. 3. Pluriactivity How many categories of jobs the have. Sportfishing and commercial fishing each represent a category; others include construction, crafts, retail, etc. 4. Years fished How long they engaged in fishing for a living 5. Own boat Whether they own their own vessel. No = 0, Yes = 1. 6. Number in household Number of people living in the same household as the respondent. 7. Age Current age of respondent 8. Education Collapsed into three categories: (1) less than a high school education (2) high school graduate (3) post-high school education, either academic or technical training. Independent VariablesKnowledge, Perceptions, Preferences 9. Know WSNP Respondents were asked if they knew of the existence of the legally declared national park on the West Side. N=0, Y=1. 10. When heard Date they found about the WSNP, coded into 4 periods: 1=Time of survey, 2= During survey research period 3= After park declaration 4= Before park inception 11. Know groups The respondents were whether they had heard of BNT, BSCA, ANCAT, and TNC. (No = 0, Yes = 1). An additive scale (0 to 4) was created of the four. 12. Resource condition perception scale Respondents were asked whether, over the past 10 years, the resources 1=increased,2=same, 3=decreased for bone fish, sponge, and crawfish. The answers created an additive index collapsed into 3 categories for categorical analysis. 13-19. Threats Respondents were asked about the threats to the West Side in an open ended question. Threats were thematically coded and id entified six main categories: nothing, nature/storms, chemical fishing, poaching, politics, and development. 20. Enforcement Respondents were asked if the general fishery rules should be: 1= stronger, 2 = same, or 3 = weaker. 21. Development Respondents were asked how they thought the West Side should be used. 1=No new development, 2=small-scale ecotourism, 3=larg e-scale resorts. 4, 5, and 6 deleted for analysis. 22. Children Asked if they want their children to be fishers as well. 1=Yes, 2=Maybe, 3=No. 23. Interviewer Interviewer 1 an d 2 were U.S. females, interv iewer 3 was a Bahamian male. ____________________________________________________________________________________________
37 CHAPTER 3 RESULTS Survey Results Socio-demographic and Resource Use Data The study identified fishermen who use the West Side resources from four communities: Lowe Sound, Red Bays, Behring Point/Central Andros, and Mangrove Cay (Table 2-1). The number of fishers in each commun ity could not be quantified, as th ere were no official registries and fishermen are a mobile group. However, ke y informants estimated that there were 75 fishers in Lowe Sound, 70 in Red Bays, and up to 200 in Mangrove Cay, though all but approximately 30 of those from Mangrove Cay fish from large commercial vessels 10+ kilometers out to sea for 1 weeks at a time, depending on the weather. Respondents worked in two main sectors: (1) tourism ("catch and release sportfishing guides), and (2) commercial fisheries (resour ce extraction). Sportfis hing guides work with international tourists, fish approximately 40 days a year, and can earn approximately $150 $450 a day. Extractive fishermen harvest a variety of resources from the West Side, such as natural sponge, crawfish/lobster, several species of snapper, and other finfish (barracuda, triggerfish, etc). Fishers were as ked to identify which resource provided their primary source of income (Table 3-2): 49% reported crawfish ing, 39% reported sponge harvesting, and 12% reported snapper. However, many cited difficult y choosing a primary act ivity, as they would engage in all depending on a number of factors such as weather, season, market, and price of gas. Economic data from key informants indicated that crawfish was the more lucrative business at the beginning of the season (opens August 1st), since they could earn up to $11 per pound, and could purportedly bring in up to 200 pounds/pers on/day. However, as the season progresses,
38 catch declines due to fishing pressure, and with high gas prices, it becomes less economical to go the distance necessary to access productive crawfishing grounds. Sponge, on the other hand, may bring in less money per effort ($0.50.00/sponge depending on type, quality, and size, collecti ng up to approximately 200 sponge/day and requiring heavy processing) but is reported to be a dependable in come and has no closed season. However, some community members mentioned a large sponge die-o ff that occurred in 1938, causing the loss of that resource for a time, as we ll as the local extinction of the Velvet Sponge, the most valuable species for sale. The government sponsored a sponge cultivation project in the Bights for a period subsequent to the die-off, but the program was halted several years later. Eighty-two (75%) fishermen claimed to fish for snapper (Table 3-2), however only 9 (12%) of the commercial fishermen claimed it as their most important income, possibly indicating a supplementary role for that resource. In addition, 31 people claimed to gather turtles, but generally in an infrequent, opportunistic ma nner, or when requested by a merchant. Two people reported taking 200 a year, 5 peopl e reported taking 25, and 24 people reported taking fewer than 25 turtles a year or could not recall. Six people specified that they take green turtles and 5 said that they collect loggerhead turtles. Five people volunteered that there were regulations on turtle harvest, mentioning either cl osed seasons on loggerheads or that it is illegal to take hawksbills. Three people volunteered that the populations are being driven away because of motor boat use. Forty-eight percent of the fish ers reported that all their in come is derived solely from marine resources, while 47% cited a secondary source of revenue, and 8% claimed 3 distinct income sources (Table 3-1). When asked how th ey support themselves if weather is bad for a long time or catch is low (Table 3-3), the most frequent answer given wa s savings, both financial
39 as well as food stores and frozen seafood catch. Land based jobs in cluded gatheri ng other natural resources (such as crabbing or hunting), constr uction or mechanical work, food services, and retail, while some people went on credit. Acco rding to key informants, others moved to neighboring islands (such as Chub Cay or New Providence) to look for work. The average age of the fishers was 42 and the average years fished was 19. The average education was 9thth grade and strongly correlated with ag e (Table C-1), partially due to the fact that there were no high school s on the island at that time, so the older fishers did not have the opportunity of further schooli ng. Ninety-four percent of the fi shers interviewed were male, and 6% female, as few women fished. Knowledge of National Parks a nd Environmental Organizations Thirty-seven percent (n =41) of the respondents knew about the West Side National Park (Figure 3-1). Of those 41 people, 39% (n=17) learned about the WSNP within the period of the survey (May 2006January 2007), 48% (n=21) found out after legal declaration of the boundaries, and 11% (n=5) knew about the park during the planni ng phases (Table D-2). There was no statistically significant difference concer ning the percentage of people who knew about the WSNP during the summer or winter research periods. Eighty-eight respondents had h eard of BNT, 66 had heard of BSCA, 34 of ANCAT, and 29 of TNC. The additive index showed that 15 people knew of 0 organi zations, while 22 people knew of all four environmental organizati ons working in the area (Table 3-4). Perceptions of Resources and Threats There was a general trend as to the percepti ons of resource conditions on the West Side (Figure 3-2). Fifty-four percent of fishermen reported that the amount of bonefish on the west side has increased over the past 10 years, whil e 21% thought it was the same, another 21% said the amount of bonefish was decreasing, and 4 % clai med that it depends on the season and year.
40 Fishermen were fairly evenly split about the st ate of sponge, while 48% of fishers thought that crawfish were declining, though 12% claimed that it depended on the weather or that the resource was not decreasing, but was being divi ded among more fishermen, leading to perceived declines per fisherman. However, according to a fisheries official, overall catches have dropped over the past 10 years, though this has not b een quantified. The scale of environmental perceptions was strongly associated with the fishing sector with the sportfishing community viewing the resources as declining more than the commercial fishing community (Spearmans Rho correlation coefficient = 0.400, p < 0.001). Responses to an open ended question asking wh at could harm the West Side were coded into six main categories (Table 3-5): developm ent, poaching, chemical fishing, natural causes, nothing, and political issues. Development issues we re cited most frequently (61 people total), with 22 people citing overfishing as a pr oblem, followed by overpopulation, then habitat destruction (by construction or boat traffic). Reporting that development was a threat was positively correlated with being a sportfisher, having more types of income, owning a boat, and having a higher level of education. It was negativ ely associated with having fished for more years, and greater age (Table C-1). Poaching issues (47 people total) included ge neral problems with poaching, such as taking species out of season and colle cting undersized animals. Thir teen people reported foreign poaching as an issue, specifying commercial poa chers from other island s (Spanish Wells, Abaco) and other nearby countries (Hai ti, Dominical Republic, and Cuba ). Open-ended answers and key informants indicated that poachers came from the U. S. as well, as U.S. sportfishers boated into The Bahamas, did not employ the legally mandated local guides, and illegally extracted fish. Key
41 informants discussed frequent radio broad casts announcing the Defense Force apprehending foreign poachers, as well as newspape r articles that speak of the issue. Thirty-one people reported that chemical fishing is a problem, with activities such as bombing, which entails enclos ing an underwater area contai ning crawfish and pouring in bleach or detergent to kill them all, which aids in rapid collection of the resource but kills everything in the area. Twenty-three people felt that storms were a large threat to the area, while 22 said nothing could hurt the West Side, and 11 people cited various political concerns. Preferences for Participation and Outreach Respondents were proffered 5 ideas for ways to participate in management of the WSNP and were asked to pick the 3 types of communi ty involvement in which they would be most interested (Table 3-6). The most frequently chosen ideas were participating in community meetings (n=82) and educating the Bahamian comm unity about the WSNP with a visitors center and tours (n=81). The least chosen idea was community resource monitoring (n=44). Respondents were queried about their current news sources an d preferences (Table 3-7). The most common news sources were the radio (n=85), television (n=63), and posters/fliers (n=39). They received fisheries news from the Department of Fisheries and the local fisheries officer (n=37), as well as the radio (n=31), and meetings (n=15). When asked their preference for how to best communicate with the community about fisheries resources a nd National parks, the vast majority requested community meetings (n=78), often specifying wanting experts and managers to come speak with them at the meetings. Preferences for Livelihoods, Development, and Management When asked whether they want their childre n to be the same type of fishermen as themselves, 55% said no (Table 3-8), citing a numb er of reasons, the top one being that it is too hard and unstable of a living, and the second bein g that they want thei r kids to get a better
42 education. Thirty three percent sa id maybe, since they want their kids to do whatever they wish, and 12% said yes, claiming that its a good living and they can make good money. Respondents were asked to choos e the best way to use the We st Side (Figure 3-3). Fiftyfour percent preferred leaving th e area as it is with no new development, while 27% wanted small-scale ecotourism, and 7% wanted large scale resorts in the area. Fishermen were split about the general fisher y regulations, with 53% claiming that they should be stronger, 43% saying they should stay the same, and 4% saying they should be weaker. When asked about regulations for the WSNP in specific, fishermen were asked if they thought each of the 5 different types of management were a good idea or bad idea (Figure 3-4). The volunteer warden system is a method suggested by both BSCA and BNT, and the other four are management systems implemented in other Cari bbean MPAs or discussed in MPA management planning literature. There was most support fo r establishing a voluntee r ranger system (83%, n=91), and least support for limiting the number of people who can use the area at a time (38%, n=42). During this section of the intervie w, many people commented about the extreme difficulty of enforcing any of these options. Ther e was a particularly negative attitude towards the option of limiting the number of people in th e park at a time (55 people opposed limiting the number of people), partially due to the potential lack of parity in granting access to the WSNP. Support for National Parks, WSNP, and Management Zones When asked if it was a good idea to have nati onal parks on Andros in general, 70% said yes, 26% said no, and 5% said they did not know (Figure 3-1). For those who supported having national parks on Andros, the most frequently cited reasons were to create tourism jobs (n=13), for future generation (n=11), and general pr otection (n=10). For those who do not support national parks, 21 people cited concern over lo sing their livelihoods, 6 specifically mentioned
43 that national parks close people out of the area, and 3 said there was not hing to protect against (Table 3-9). Of the 4 people who said they did no t know, 3 claimed that it depended on the rules. Fifty-seven percent of the respondents claimed that it is important to protect the area of the WSNP, while 25% claimed that it is not, and 18% said they di d not know (Figure 3-1). When asked to explain their reasons (Table 3-10), s upporters of the WSNP c ited wildlife protection (n=18) and the importance of the area for breed ing grounds (n=11), but some reported concerns with personal agendas and poten tial damage caused by the presence of a private lodge in the area. (n=10). Those who said th ey did not know if the WSNP wa s important claimed that their support of the park depended on the rules (n=7), or reported concerns with political agendas (n=3). Those who said that it is not important to protect the WSNP clai med that the park was protecting the wrong area or there was nothing special in there (n=9), feared that they would lose their livelihoods (n=8) or that there were no threats from whic h to protect the area, therefore a park was unnecessary. Given spatial options for park management zones, 50% chose no new management, while 45% chose one of the three different forms of management zones with increasing intensity of management, and 5% gave no response (Figure 3-5). Concerns About Parks When asked specifically about their concerns for the WSNP (Table 3-11), the most frequent response (49%, n=41) was that they ne ed access and fishing rights for their livelihoods. This was followed by 15% (n=13) concerned about political agendas and the existence of private interests allowing exploitation of the resources within the park. Ten percent (n=8) mentioned the difficulty of enforcement of regulations in an area as remote as the WSNP. The focus group revealed strong attitudes to wards national parks as well, one person voicing that it frightens people, since everyone depends on fish ing for a living and few have
44 alternatives. Some said that they do not want ar eas closed off, but if they can be guaranteed fishing rights, they would agree to the presence of the WSNP. Support for National Parks, WSNP, and Man agement Zones: Bivariate Associations The data from this research contain many si gnificant bivariate associ ations, shown in the residual analysis table (Table 3-12). With the sm all sample size and large number of variables, it is not possible to control for all possible root cau ses of associations, as is often the case with social surveys. Therefore, the following is an expl oration of a number of associations to examine some underlying influences. Support for national parks on Andros in gene ral is associated with 12 independent variables (Table 3-12). More s upport is significantly associat ed (P<0.05, Adjusted Residuals >|2|) with: location of residence (positive fo r Behring Point and Mangrove Cay), being a sportfishing guide, knowledge of the WSNP, knowledge of more environmental groups, perception of environmental condit ions declining, viewing chemi cal fishing and development as a threat, opinion that general en forcement should be stronger, wanting small-scale ecotourism, and being interviewed with interviewer 1. Nega tive views of the park are associated with Location (Red Bays), the perception that ther e are no threats to the WSNP, not wanting any development, not wanting children to be fish ers, and being intervie wed by interviewer 3. Support for the WSNP in specific has only 6 st atistically significant associated variables (Table 3-12), and is positively associated (P <0.05, Adjusted Residuals >|2|) with: being a sportfishing guide, knowing of the WSNP, know ing more environmental groups, wanting stronger fishery enforcement, and wa nting children to be fishers. It is negatively associated with having only one type of income. Support for management zones for the WSNP is significantly associated with 12 variables (P<0.05, Adjusted Residuals >|2| ) (Table 3-12). More support for management zones is
45 associated with: having two income types, knowing about the WSNP, knowing 4 environmental groups, perceiving the resources as declining, perceiving chemical fishing as a threat, wanting stronger enforcement, small-scale development, and being interviewed by interviewer 1. It is negatively associated with loca tion (Red Bays), having only one type of income, hearing about the WSNP only at the time of the survey, wan ting enforcement to be the same, wanting no new development, and being interviewed by interviewer 3. Associations Among Support Variables Contingency tables (Tables E1E6) for the associations among th e response variables showed significant measures between all variables. Cronba ch's alpha, measuring the reliability of these three variables as a scale measuring the same idea, gave a value of 0.585. Though the reliability of the scale is reasonable, there were some interesting patterns in the data: in conducting a cross-tabulation with the variables support WSNP and support national parks in general, 42 respondents appeared in incongruous pairs, claiming that national parks are not a good idea, but WSNP should be protected, and vice versa (Table E-1). Support WSNP and support management zones also had discordant pair s, with 24 people claiming that the park was important to protect, but wanted no new regul ations, while 19 thought th e area was unimportant to protect, then chose management zones rather than no new regulations (Table E-3). Controlling for the interviewer did not change the exis tence of this apparent incongruence. Regression Analysis: Predicto rs of Support for National Parks, the WSNP, and Management Zones The stepwise logistic regression model for s upport for national parks on Andros had a Nagelkerke R of .787 and correctly predicted 90% of responses (Table 3-14). This model identified 6 significant predicto rs of influence: location (par ticularly Red Bays), education, perception of environmental condit ion scale, total number of perc eived threats, and development
46 preferences. While viewing chemi cals as a threat did not meet the p < 0.05 cutoff, it was still included in the model as it cr eated a better model fit. The stepwise logistic regression model for Support for WSNP had a Nagelkerke R of .471 and correctly predicted 78% of responses (Table 3-15). This model identified 2 significant predictors of influence: the number of envir onmental groups they were aware of, and owning their own boat. While location, pluriactivity, and wan ting their kids to be fi shers did not meet the p < 0.05 cutoff, they were important for model inte ractions and included to create a better model fit. The stepwise logistic regression model for support for management zones had a Nagelkerke R of .587 and correctly predicted 81% of responses (Table 3-15). This model identified 2 significant predictors of influence: perception of environmental condition scale, and the perception that development is a threat to the WSNP. The variables th at increased model fit without significance of th eir individual effects were: intervie wer, pluriactivity, when they heard about the WSNP, the perception that there ar e no threats to the WSNP, and enforcement preferences. Interviewer Effects Interviewer effects were examined in a number of ways. Bivariate asso ciations were tested by collapsing variable categories into two by two ta bles to enable chi-square analysis of the smaller sample sizes, then conducti ng a residual analysis for the fu ll data set, the data set from interviewer 1 and 2 (U.S. females), and the data set from interviewer 3 (Bahamian male) (Table 3-13). Only four variables remained significant across all interviewer divisions and all support variables. For the support for national parks variable, support was negatively associated with living in Red Bays, the perception that the resources are increasi ng, and the perception that there are no threats to the West Side. Support was pos itively associated with the perception that
47 resources are decreasing. For the support of management zones variable, support was positively associated with preferring stronger fishery laws and negatively associated with wanting weaker or unchanged laws. Eleven variables remained significant for both the full data set as well as the subset of data from interviewer 1 and 2 combined. For the su pport NP variable, these were: enforcement preference and children's occupation preference. For the support WSNP variable, these were: know about WSNP, when heard about WSNP, enforcement preference, and children's occupation preference. For the support manageme nt zones variable, ther e were: pluriactivity, education level, perception of re source condition, perception of chem ical fishing as a threat, and development preference. Four variables exhibited a patt ern of significance for the full da ta set and interviewer 3, but no significance for interviewers 1 and 2. Two of these were the fish er type variable (for support NP and support WSNP), indicati ng that interviewer effects may have had an influence in creating significance for that variable. Howe ver, there are no instances of significant relationships changing direction due to the interviewer effects. Eleven independent variables which were signi ficant for the full data set lost significance in both reduced data sets, which could potentially be a result of the smaller size of the reduced data sets, as the statistics are sensitive to sample size. All the significant independent variables, in cluding interviewer, were entered into the backward stepwise logistic regre ssions. The interviewer variable did not prove to be a significant factor for support NP or support WSNP, but di d show significance for the support management zones variable. Logistic regressi ons look at the effects of each va riable while holding all others equal, therefore it is a more holistic explorati on of interviewer effects. However, due to item
48 non-response, the regression could only use 62 of the cases, and must also be interpreted with caution. Group Meeting Results The group meeting was held in Red Bays on May 25th, 2006, before the surveys were conducted. Approximately 25 fishers attended the meeting, though an exact count was not possible as people entered and left during th e meeting. Approximately 15 people spoke in response to the 6 main questions presented. Th e meeting yielded contextual information and addressed some of the same issues as the su rvey, but in a group setting, which may present different results than one-on-one surveys (Mar ks 2005). Red Bays was the community with the most negative attitude towards the parks (Table 3-12), therefore greater depth of information is useful to understand their c oncerns. One element that arose frequently was concern for livelihoods, as people claimed that only 3 me mbers of the community had government or salaried positions, and the re st were officially unemploye d, depending on fishing for their livelihoods. Additionally, they mentioned an inte rest in tourism related jobs (Appendix B). One person said that the envi ronment was the same as 20 years ago, while another said production had declined, perhaps due to hurricanes. They explained that fishers can only work during good weather, which has the dual implicati ons that their income is tenuous, and they are not overfishing, since the weather forces them to have closed periods. They also discussed fishing methods, saying that cutting sponge and shif ting the locations of fishing pressures was an adequate management practice already implemente d by the fishermen, but that it was a learning process. They were not willing to discuss locati ons of fishing effort, but claimed to fish the entire Great Bahamas Bank, and from the norther n to southern tips of the island, moving around to spread out fishing pressure and allow resource recovery.
49 When asked for important areas to protect, the first two answers were the ocean and the whole area. One member then cited being closed out of a national park in the Florida Keys, which started a heated debate concerning exclusi on from necessary fishing areas, fear of losing livelihoods, and confusion as to what a nationa l park means in terms of regulations. One fisherman from Exuma Cay cited the case of exclus ion from fishing in the ECLSP, which further generated a host of comments against national parks, claiming that the fisher ies are okay as they are and they do not want a national park. However, if they could be guaranteed that they will not lose fishing rights, then they w ould not have a problem with it. When asked how they would like to be involve d with management, they reiterated that there are no problems in the area, th at they know the fishery rules, a nd that they already take care of management themselves. In terms of outreach they mentioned a number of channels, such as the fisheries officer, local government, and co mmunity meetings, emphasizing that BNT should hold a meeting in Red Bays, and that they want to speak with people in ch arge of the process.
50 Table 3-1. Socio-demographics for fishers (n=117) Variable % (n) Mean SD Location --Lowe Sound 22% (26) Red Bays 30% (35) Central Andros 3% (3) Behring Point 15% (17) Mangrove Cay 31% (36) Primary resource focus --Commercial: (Sponge) 25% (29) (Crawfish) 32% (37) (Snapper/scalefish) 8% (9) Tourism (Sportfishing guide) 30 % (35) Other 6% (7) Pluriactivity/Income di versity 1.67 .633 1 type 44% (48) 2 types 47% (51) 3 types 8% (9) Years fished 19.26 14.45 0-10 31% (32) 11-20 33% (34) 21-30 21% (22) 31+ 15% (15) Own a boat --Yes 61% (70) No 39% (44) Number of people in household 4.12 2.77 1-3 46% (49) 4-6 38% (42) 7-9 8% (8) 10+ 8% (8) Age 42.82 14.12 18-30 19% (22) 31-43 39% (45) 44-56 22% (26) 57+ 15% (18) Gender --Male 94% (110) Female 6% (7) Education 9.64 3.87 0-11th grade 43% (50) 12th grade (HS graduate) 44% (51) 13+ (post-gradate work) 6% (7)
51 Table 3-2. Resources collected on the West Side. Respondents listed all types they collected, and then named the resource that accounted for their primary income and secondary income. Snapper/ Lobster/ Sportfishing Scalefish Crawfish Sponge guide Turtle Total # people collecting 82 81 54 42 31 # Primary income 9 37 29 35 -# Secondary income 25 27 20 2 -Table 3-3. Responses to How do you support yourself if weather is bad for a while or catch is low on Andros? Answers were open-ende d, coded thematically, and divided by sportfishing vs. commercial fishing communities. Note the reliance on savings. Sportfishing Commercial Income Source guide (n=35) fishers (n=75) Land based natural resources: Crabbing (land crab) -11 Hunting (wild boar) 1 5 Straw work -3 Agriculture 1 1 Work/own restaurant/bar -5 Work/own retail 3 2 Mechanic/ Construction (carpentry, elec trician, etc.) 9 11 Government job -2 Musician -2 Unspecified land jobs -4 Savings 14 31 Credit 1 1 Nothing 2 5 Move to other islands for work 1 2 Not a problem 4 2 Table 3-4. Knowledge of conservation/management organizations. Upper sec tion lists number of people aware of each organiza tion. Lower section displays the additive index of how many organizations each person knows. Organization # People aware of it Bahamas National Trust (BNT) 88 Bahamas Sportfishing and Conser vation Association (BSCA) 66 Andros Conservancy and Trust (ANCAT) 34 The Nature Conservancy (TNC) 29 Additive Index of how many organizations pe ople know # people Know 0 organizations 15 Know 1 organization 33 Know 2 organizations 30 Know 3 organizations 12 Know 4 organizations 22
52 Table 3-5. Perceptions of threats to the We st Side National Park. Open ended question, respondents could list unlimited ideas, which were coded for common themes. Threat # People citing theme Development (61 total) Overfishing 22 Overpopulation 13 Habitat destruction/construction 11 Boats disturbing habitat 5 Oil drilling/spills 4 Big commercial vessels 3 Future large scale fishing 1 Increasing fishing technology 1 Increasing demand for product 1 Poaching (47 total) Poaching (out of season, under-sized) 24 Foreign poaching 13 Condos 4 Traps 3 Crabbing 2 Chemical (31 total) Bleach/Joy/Gas fishing 29 Trash 2 Natural (24 total) Storms/Hurricanes 23 Sharks 1 Nothing (22 total) Politics (11 total) Private lodge within WSNP bounds 4 AUTEC 3 Political agendas 2 Lack of education 2
53 Table 3-6. Preferences for community participati on. Each respondent was given the 5 options for participation in management listed below and was asked to choose the 3 out of 5 possible options that they would personally be most interested in doing. Activity type # People Attend community meetings Educate the Bahamian community (visitor center and tours) Educate tourists (visitor center and tours) Act as volunteer park warden (enforcement) Community resource monitoring (biological) 82 81 62 60 44 Table 3-7. Preferences for outreach methods and current news sources. Source/Method Where do you get general news? Where do you get marine resource news What is the best way to share info about NPs? Radio Television Posters/fliers Newspaper Community meetings Word of mouth One-on-one with experts Government (general) Department of Fisheries Local Government Ministry of Tourism Church Conservation orgs Fishery buyers Internet Seminar School Festivals/Fish fry Do not receive news 85 63 39 27 4 15 -37 -1 2 24 3 1 3 ----31 9 4 5 15 13 --37 5 2 -8 3 1 ---6 10 4 9 3 78 9 11 1 3 3 -1 ---4 3 2 -
54 Table 3-8. Fishers' preferences for their childre n's occupations and reasons for that choice. Responses to Do you want your children to be [same fisher type] as well? Response Yes (12%) Maybe (33%) No (55%) Reason n=14 n=36 n=60 Too hard of a living -3 31 Good living/make good $ 9 2 -Whatever they want to be -24 5 Should get a better education -3 8 Should have land jobs --1 Should have office jobs --4 Should have a better living than them -1 7 Girls do not fish -4 6 Fisheries declining --3 Should have multiple jobs -2 -Children should work with parents 2 --Table 3-9. Responses to Why do you (support/ not support) general national parks on Andros?" N = # of people. Open-ended responses were coded for themes. Support NP Do not know Do not support NP Reason 70% (n=79) 4% (n=5) 26% (n=29) Ecological Protect wildlife 9 --Breeding grounds/habitat 8 --No threats --3 Economic/Livelihoods Lose livelihoods/access 8 -21 Improve fishery resources 5 --Create tourism/jobs 13 1 -Future Generations 11 --See new things 5 --Management Protection (unspecified) 10 --Depends on rules 5 3 1 NP excludes local people 1 -6 Need participation/equity 4 --Better management 2 --Need education 1 --
55 Table 3-10. Responses to Why do you (support/ not support) the protection of the West Side National Park? N = # of people. Open-ended responses were coded for themes. Support WSNP Do not know Do not support WSNP Reason 57% (n=65) 18% (n=20) 25% (n=28) Ecological Protect wildlife 18 --Breeding grounds/habitat 11 --No threats --7 Nothing special/ protecting wrong area 2 1 9 If do not harvest sponge, they die --2 Economic/ Livelihoods Improve fishery resources 9 2 1 Lose livelihoods/access 3 -8 Create tourism/jobs 2 --Future Generations 4 --Management Protection (unspecified) 4 1 1 Depends on rules 1 7 -Private interests/lodge in WSNP 10 3 -Hard to enforce 3 -7 NP excludes local people --3 Foreign poaching 6 --Need education 1 --Will not harm anyone 3 --Too big --2 Stay same --1 Want whole West Side 1 --
56 Table 3-11. Concerns about the West Side Natio nal Park. Respondents provided multiple ideas, which were coded for themes. Concern % People citing (n=84) Need access and use for livelihood 49% (41) Private interests/Lodge inside bound aries of WSNP 15% (13) Difficulty of enforcement 10% (8) Need jobs 8% (7) Need for effective conservation, not just boundaries 8% (7) Lack of community awareness and inclusion 7% (6) Depends on rules of the park 5% (4) Potential park fees 2% (2) Need more local participation 2% (2) Tourism can cause problems 2% (2) Unfair to put any restrictions on locals 2% (2) WSNP will not harm people 2% (2) NP closes people out 2% (2) WSNP is too big 2% (2) Need appropriate boundaries 1% (1) Foreign poaching 1% (1)
57 Table 3-12. Significant relationshi ps between support variables and independent variables. Adjusted standard resi duals for significant associations: values higher than | 2 | indicate influence of that category (cel l) in favor (+) or not (-) of support. Variable Support NPs on Andros Support WSNP Support Management Zones Location Lowe Sound -----Red Bays -5.4---3.5 Central Andros/BP 2.4---Mangrove Cay 2.3---Fisher Type Sportfishing guide 2.62.1-Commercial fisher -2.6-2.1-Pluriactivity 1 Type ---2.3-3.1 2 Types ----2.0 3 Types -----Education 0-11th grade -----2.1 12th grade (HS) -----13+ (post-graduate) -----Know about WSNP Yes 18.104.22.168 When heard WSNP 1. Informed at survey -----3.2 2. Survey period -----3. After park creation -----4. Prior to park -----Know groups 0 ---2.6-1 -----2 -2.7---3 -----4 22.214.171.124 Increasing -4.2---3.4 Resource condition perception scale Same -----Decreasing 3.0--2.6 Threat: Nothing Present -2.9---2.0 Threat: Chemical Present 2.1--2.6 Thrt: Development Present 2.9---Enforcement Weaker -----Same -4.6-2.5-4.5 Stronger 126.96.36.199 Development No new development -3.4---3.9 Small-scale tourism 2.9--4.2 Large-scale dev. -----Children Yes --2.2-Maybe --2.1-No -2.5-3.5-Interviewer Interviewer 1 3.3--2.9 Interviewer 2 ---Interviewer 3 -4.7---4.5
58Table 3-13. Interviewer effects: adjusted standardized residuals for si gnificant chi-square associations for support variables. Data are broken into three sections to examine interviewer effects (Full data, interviewer 1&2, and inte rviewer 3. Values greater than |2| indicate a significant infl uence of that category on responses in support (+) or not (-) of the dependent variables. Support NP Support WSNP Support Management Zones Variable Full Data Interviewer 1&2 Interviewer 3 Full data Interviewer 1&2 Interviewer 3 Full data Interviewer 1&2 Interviewer 3 Location Lowe Sound ---------Red Bays -5.4 -3.6 -2.6 ----3.5 --Central Andros/BP 2.4 --------Mangrove Cay 2.3 --------Fisher Type Sportfishing guide 2.6 -2.7 2.1 -3.0 ---Commercial fisher -2.6 --2.7 -2.1 --3.0 ---Pluriactivity 1 Type -2.0 ---2.3 --2.1 -3.1 -2.2 -2+ Types 2.0 --2.3 -2.1 3.1 2.2 -Education 0-11th grade -------2.1 -2.3 -12th+ grade (HS) ------2.1 2.3 -Know about WSNP Yes 2.7 --2.1 2.0 -3.2 --When Heard WSNP Survey period -2.4 ---2.3 -2.2 --2.2 --Prior to park est. 2.4 --2.3 2.2 -2.2 --Know Groups 0 to 1 ----2.5 --2.3 ---2 to 4 ---2.5 -2.3 ---Increasing/same -3.0 -2.1 -2.1 ----2.6 -2.0 -Resource Condition Perception Scale Decreasing 3.0 2.1 2.1 ---2.6 2.0 -Threat: Nothing Present -2.9 -2.2 -2.1 ----2 --Threat: Chemical Present 2.1 -----2.6 2.0 -Thrt: Development Present 2.9 ----2.9 ---Enforcement Weaker/same -4.2 -3.2 --2.3 -2.0 --4.9 -3.3 -2.8 Stronger 4.2 3.2 -2.3 2.0 -4.9 3.3 2.8 Development No new dev. -3.2 ------4.2 -2.7 -Small-scale tourism 3.2 -----4.2 2.7 -Children Yes/ maybe 2.5 2.0 -3.5 2.9 -2.1 --No -2.5 -2.0 --3.5 -2.9 --2.1 --
59 Table 3-14. Logistic regression for support for national parks on Andros. Nagelkerke R Square =.787 (R equivalent for logistic regr ession), indicating the model explains approximately 78% of outcomes. Ninety-p ercent of the responses were predicted accurately from the model, with a Hosmer Lemeshow test of 0.409 (> 0.05 indicates good model fit). Variable B S.E. Wald df Sig. Exp(B) Location (MC) 6.829 3 0.078 Location (LS) -0.143 1.485 0.009 1 0.923 0.867 Location(RB) -3.262 1.398 5.449 1 0.020 0.038 Location (BP) 0.745 1.596 0.218 1 0.641 2.106 Education 0.572 0.202 8.003 1 0.005 1.772 Resource cond. scale 1.314 0.560 5.500 1 0.019 3.721 Threat: Chemicals 2.820 1.497 3.551 1 0.060 16.783 Threat: Total -2.363 1.024 5.326 1 0.021 0.094 Development preference 2.576 1.190 4.687 1 0.030 13.140 Constant -12.179 5.209 5.466 1 0.019 0.000 Table 3-15. Logistic regression for suppor t for the West Side National Park. Nagelkerke R Square =.471 (R equivalent for logistic regression), indicating the model explains approximately 47% of the outcomes. Sevent y eight percent of the responses were predicted accurately from the model, w ith a Hosmer Lemeshow test of 0.238 (>0.05 indicates good model fit). Variable B S.E. Wald Df Sig. Exp(B) Location 2.478 3 0.479 Location (LS) 0.577 0.935 0.380 1 0.537 1.780 Location (RB) 0.495 0.920 0.290 1 0.590 1.641 Location (BP) 2.129 1.355 2.469 1 0.116 8.405 Pluriactivity 0.861 0.566 2.311 1 0.128 2.366 Own boat -1.671 0.804 4.326 1 0.038 0.188 Know groups 0.716 0.329 4.737 1 0.030 2.045 Children -0.980 0.556 3.111 1 0.078 0.375 Constant 0.059 2.033 0.001 1 0.977 1.060
60 Table 3-16.Logistic regression for support for ma nagement zones of the West Side National Park. Nagelkerke R Square =0.587 (R equiva lent for logistic re gression), indicating the model explains approximately 58% of the outcomes. Eighty one percent of the responses were predicted accura tely from the model, with a Hosmer Lemeshow test of 0.551 (>0.05 indicates good model fit). B S.E. Wald df Sig. Exp(B) Interviewer 3.926 2 0.140 Interviewer(1) 1.502 0.850 3.119 1 0.077 4.489 Interviewer(2) 1.920 1.148 2.795 1 0.095 6.819 Pluriactivity 0.967 0.609 2.520 1 0.112 2.631 When hear WSNP 0.026 0.030 .779 1 0.377 1.027 Resource cond. scale 0.603 0.239 6.348 1 0.012 1.828 Threat: Nothing -1.856 1.256 2.182 1 0.140 0.156 Threat: Development -2.162 0.903 5.727 1 0.017 0.115 Enforcement 1.272 0.734 2.997 1 0.083 3.567 Constant -9.061 2.869 9.976 1 0.002 0.000 Knowledge and Attitudes towards National Parks 42 65 79 71 28 29 20 5 0%10%20%30%40%50%60%70%80%90%100% Know of the WSNP? Important to protect WSNP? Good idea to have National Parks on Andros? # People Yes No Don't Know Figure 3-1. Knowledge and support of national parks and the WSNP.
61 Perceptions of Resource Trends 63 33 27 38 25 36 2121 25 42 56 50 5 7 14 9 0 10 20 30 40 50 60 70 BonefishSpongeCrawfishGeneral Environment# People Increasing Same Decreasing Depends Figure 3-2. Perceptions of resour ce trends. Respondents were aske d if the amount of [resource] had increased, stayed the same, or decreas ed over the past 10 years. The "General Environment examines the expected trend over the next 10 years. 4 7 2 1 8 32 63010203040506070No Answer Other Industry Housing Large-scale Resorts Small-scale Ecotourism No New Development Figure 3-3. Preferences for development on the West Side. Respondents were asked how they thought the West Side should be used and given the above (closed-ended) options.
62 Opinions concerning regulations for WSNP48 47 42 63 91 21 13 14 10 6 41 51 55 38 13 0%10%20%30%40%50%60%70%80%90%100% No-comercial-fishing zones No-fishing zones Limit # people in park at a time Park permit system Volunteer ranger system Good Idea Don't Know/NO Bad Idea Figure 3-4. Preferences for types of regulations for the WSNP. Respondents were asked if they thought each type of regulation would be a good idea, bad idea, or do not know. 51 26 17 14 0 10 20 30 40 50 60 Zonation 1: No New Regulations Zonation 2: No New Regulations, No-Take Zonation 3: 1/3 No New Rules, 1/3 No-Take, 1/3 No Entry Zonation 4: No-Take, No-Entry # People Figure 3-5. Support for management zones for th e West Side National Park. Respondents were asked to choose from 4 options, ranging from no new regulations to most restrictive.
63 CHAPTER 4 DISCUSSION Socio-Demographics of the Andros ian Fishers on the West Side Approximately half of all fishermen interviewe d work solely with marine resources for a living. For those who reported multiple incomes, fi shing provided the majority of their income for all but 5 people. The issu e of livelihood needs repeatedly emerged in most open-ended questions; for example, 21 people who do not want national parks on Andros cited concern over potential loss of livelihoods. When asked about their general concerns ab out the park, 41 fishers mentioned needing access and fishing rights fo r their livelihoods, so me saying that the government can not expect people to avoid us ing an area unless they provide alternative incomes. The group meeting also emphasized the problem of unemploymen t, and that fishers cannot support management that will detract from their income. There was a positive association between having more than one distinct income and supporting management zones. This suggests that the less dependent people are on the fisher y, the more willing they are to support regulated management practices. The specific livelihood method was associated with support for MPAs in the bivariate analyses (Table 3-12): sportfishing guides were significantly associated with support for both national parks and the WSNP, so they may be an important target group to consider during management planning. However, the fishing type was not associated sign ificantly with support for management zones, implying a more complex set of interactions of the issues. Additionally, that relationship did not appear for the subset of data with interviewer 1 and 2, and in the regression analysis the type of fi sher never featured as a signifi cant predictor (perhaps also due to issues of multicollinearity which were not possi ble to tease apart with the small sample size). Therefore this suggests that while the commer cial fishing and sportfishing communities may
64 have differences, there may be intervening factors at play as well, so focusing on this difference in an attempt to address support for nationa l parks should be approached cautiously. It is also important to reemphasize that many of the fishers work with multiple resources. The geographic communities are comprised of mixe d fishing types, particularly in Lowe Sound and Mangrove Cay. The lack of clear separations among categories may help explain the vast number of significant bivariate as sociations, which could not be t eased out due to the limitations of the sample size. Knowledge of National Parks a nd Environmental Organizations Only 37% of the surveyed resource users knew about the West Side National Park, despite the fact that most of the fi shers know the area. Knowledge of the WSNP was significantly associated with knowledge of the environmen tal groups (Table C-1), suggesting that the organizations may be an important source of in formation on the topic. Prior research also indicates that association membership increase s support for management (Perez-Sanchez & Muir 2003; Pomeroy & Berkes 1997), which may be an important element for these local management organizations to consider. Knowledge of the WSNP is also significantl y and positively associated with all support variables (Table 3-12). Additiona lly, the length of time people have known about the park was strongly associated with preference for manageme nt zones, with less support from the people who found out more recently (Tab le 3-12). This evidence lends cr edence to the suggestions that not only knowledge of the MPA, but early involvement in the projec t, leads to greater support for management. This has been found in other situ ations internationally (Alexander 2000; Infield 1988; McClanahan et al. 2005; Pomeroy & Be rkes 1997) and has been acknowledged by the conservation organizations involved with the WSNP (BNT meeting, February 2006).
65 Education was cited frequently by respondents as an important element for the community, and educating the Bahamian community about the West Side was chosen by approximately 70% of the survey respondents as a method of participation. That quest ion also incited strong verbal support to the effect that education is a necessity. Theref ore, it seems that the community may be receptive to educational efforts from the conservation organizations. Additionally, it is important to note that there is confusion about what the WSNP is and what it means. When asked whether they knew th at the park existed, a number of respondents initially said yes, referenced the privately owned area enclos ed within the WSNP (Flamingo Cay) as the park boundaries, and said they were not allowed in the National Park, which was reported as a source of frustration. While there was no way for this research to verify these claims, it seems important to recognize this conf usion, take it into account when planning an outreach message, and work with the private owner as park management is planned. Perceptions of Resources and Threats The perception that marine resources are dec lining was associated with both support for management zones as well as support for national parks (which is one of the associations which remained consistent across all interviewer data se ts). This association makes intuitive sense, as people who perceive the environment to be de clining or susceptible to degradation may recognize the need to take acti on to halt the decline (Alessa et al. 2003; Cottrell 2003; Sesabo et al. 2006). This is corroborated by the percep tions of threats: people who perceived development as a threat were more likely to support national parks, while those who cited that there are no threats to the area were less likely to support nati onal parks. This information suggests that an important outreach component c ould be to address the perceptions that the environment cannot be depleted. On a cautionary note, this association of increasing resources
66 and lack of support for national parks could conc eivably be a function of people against national parks reporting that the environment is fine in order to avoid external management. Preferences for Livelihoods, Management, and Development The majority of fishers preferred to have no new development, but in terestingly, that was negatively associated with support for manage ment zones, as people who did not want development also tended to be against additi onal regulation. The most favored management method was the volunteer warden system, and according to the group meeting outcome, some fishers wanted to maintain the current status. Th is can be an important element for discussion, as the establishment of parks may be a means of pr eventing environmental da mage from large scale development, not necessarily a method of haltin g local fishing, which may work as a common point of interest for collaboration. The development variable additionally reveal s that people who think national parks are important are also more likely to prefer greater intensity of development, which can make sense in light of the fact that the second most comm on option, after no development, is small scale ecotourism. Table 3-9 shows that th e creation of tourism opportunities and jobs is the most frequent explanation of support for national parks in gene ral. Additionally, lack of livelihood resources are reported most frequently in re sponse to the concerns about pa rks, though a majority of people would prefer for their children not to be fishers. While there is no clear causal direction, the duel preference for parks and small-scale development may be associated with the perception that parks will provide tourism jobs, allowing for alternative occupations and thereby garnering support. This is seen in many MPA situations, wh ere tangible benefits have raised the support of conservation areas. However, it is important to note that only 7% of respondents preferre d large scale resorts on the West Side. Ecological reas ons were most frequently report ed as explanation of support for
67 the WSNP (n=31), and economic reasons were repor ted by less than half of that (n=14). This indicates a broader suit e of local interests, rather than a single concern for the economic situation. Support for National Parks, the WSNP, and Management Zones: Inconsistencies and Implications for Future Work While a majority of people support national parks in general (70%), fewer people support the WSNP in specific (57%), and fewer still su pport the implementation of management zones (50%). This is not surprising in light of livelihood concerns, as an idea may seem feasible in the abstract, but once it is given boundaries and regul ations, potential person al implications may become more concerning to the respondent. In addition to declining support for more spec ific actions, there is a high frequency of apparently discordant pairs when comparing th e response variables. For example, 24 respondents support the WSNP but do not support management zones, while 19 resp ondents do not support the WSNP yet support management zones (Table E-3). This may partially be explained by the fact that the southern section of the current park boundaries covers an area identified as important fishing grounds for some respondents, particularly from the southern communities. Additionally, some respondents who voiced supp ort for the park did so conditionally, for example saying that they support parks, but not if they may lose fishing access or if it is politically motivated. However, as there is not yet clarity as to potential management for the area or what a national park entails, respondents may be reacting to a suite of issues, and there may be a great deal of internal, contextual interpretation concerni ng personal implicat ions of potential management actions. This may also be an indica tion that people are thinking through some of the issues for the first time as they are being surv eyed. This reflects fi ndings concerning general
68 survey research that showed that respondents ma y be inconsistent in matters that they do not contemplate often or for which they are inte rnally conflicted (Convers e 1974, Zaller & Feldman 1992). It is also potentially a tribute to the pow er of survey research as an outreach tool (Chuenpagdee et al. 2004)in respon se to the question of the best way to communicate with the community about resources, some of the respondents who said one-on-one (Table 3-7) followed that comment with just like what youre doing now. In the logistic regressions (Table 3-14 to 3-16), inputting the same independent variables into the models for all three support variables led to a different suite of predictors for each. This may be an indication that in thinking about each question, different concerns are considered, and there is not just one general support thought process. For in stance, the support for national parks in general variable was predicted by location, education, threat and environmental perception variables, and development preferen ces. Support for the WSNP on the other hand was predicted by more tangible, so cio-demographic elements, such as owning a boat, pluriactivity, location, and whether they want th eir children to be fishers (whi ch may possibly be taken as an indication of quality of life, since the most freque nt response was that fish ing is too hard). So it becomes more of a question of specific liveli hoods concerns, versus support for a general principle. The management zone model showed significance for the percep tion of the resource conditions and the perceptions of development as a threat. The frequent inclusion of resource condition and threat perceptions i ndicated an important role of th ese factors, which may be used for an outreach message to promote management effectiveness. MPAs and Management Effectiveness: The Caribbean Context Recent reviews in the Caribbean yield varying accounts of MPA numbers and effectiveness. Kelleher et al.(1995) report that the Wider Caribbean region contains one of the highest concentrations of MPAs globally, with 93 MPAs identified in this region in 1988.
69 However, according to their analysis, two-thirds of these MPAs are not achieving their stated management objectives. A more recent study by Ge oghegan et al. (2001) c ontacted 75 MPAs in the Central Caribbean region, and found that appr oximately 25% have no management at all, 29% have low management, 37% have modera te management, and only 8% have high management. Burke and Maidens (2004) identif ied 285 MPAs in the Wider Caribbean region, reporting good management for only 6%, partial management for 13%, inadequate management for 38%, and the rest unknown. However, all th ree studies rate management effectiveness according to different management goals and ev aluation criteria, illustrating the difficulty in creating an accurate comparison across locations, particularly as the first two surveys relied on self-rating (Alder et al. 2002). Governance may play a large role in the success of MPA management. According to Garaway and Esteban (2003) using the data fr om Geoghegan et al. 2001, 68% of MPAs with a mechanism for local par ticipation had moderate to high ma nagement effectiveness, whereas those MPAs with low stakehol der involvement have only 15% moderate to high management effectiveness. Additionally, 30% of the MPA managers surveyed volunteered information about problems with local compliance end enforcement of MPA regulations. Geoghegan et al. (2001) present the management types of each of the active MPAs they contacted: 15 employ fishery restrictions, 11 us e zoning regulations, 7 use a mixture of zoning and fishery regulations, and 8 prohibit all resour ce extraction in the MP A. The WSNP does not yet have a management plan, but BNT is consid ering the possibility of managing with multi-use zones (BNT community meetings, May 2007), which f its within the realm of experiences of the MPA networks currently in place in the Caribbean.
70 Kelleher et al. (1995) report eight main r easons why MPA management plans have been faltering globally: Insufficient funds and technical resources for management plans Lack of properly trained staff Sparse biological data Insufficient public support and engagement Inadequate enforcement Overuse of resources Impacts from surrounding areas Ambiguous governance It is beyond the scope of this wo rk to provide guidelines for ma nagement effectiveness, other than to reiterate that public support and engagement are among the key issues for MPA management success both globally and within th e Caribbean region, and therefore should be given due consideration. MPA Regulations: Compliance and Enforcement Concerns A related element to be addressed is the question of compliance with new regulations. When questioning respondents about support fo r various types of regulations, a common question raised was: How will they enforce that ? The WSNP is a relatively large area: 68, 257 ha, as compared to the median global MPA area of approximately 1,500 ha. There is currently little traffic through the WSNP, which can be a positiv e factor, as there is a pparently little current pressure on the area. However, this could b ecome a problem of manpower if enforcement becomes an issue. One indication of this issue is that approximately 83% of the respondents claimed that a volunteer warden system is a good idea, but only about 50 % chose that option for participation, many qualifying that it could potentially be a life -threatening task, as there is no one to call for help and poachers may be armed. Lo cal leaders also state that illegal fishermen come from the U.S. and other islands via boat and float-planes, and they claim there is no reasonable way to catch violators, so the la ck of accessibility might become a management
71 constraint. This is in accord with Agrawals ( 2002) framework which states that an overlap between user group residential location and resource location is beneficial. The lack of such an overlap may cause enforcement problems in this scenario. Some research suggests that greater program participation and support will lead to a heightened sense of legitimacy of the management scheme, and will therefore garner greater compliance with regulations (Crawford et al. 200 4; Jentoft et al. 1998; Pollnac et al. 2001), However, Hampshire (2004) writes that poaching is in itself not viewed negatively by all, and many locals see nothing wrong in ta king from what they view as their own waters. This was voiced frequently in terms of needing the right to take a meal and feed their families. Lepp and Holland (2006) found that people associated state level corruption and privat e interests with a conservation program in rural Uganda, which led to little legitimacy, and therefore negative attitudes and lack of compliance. This suggests a need for caution in the Andros case as well (see Crawford et al. 2004 for a thorough treatment of compliance and enforcement issues within the MPA context). Community Consensus While there were some opinions held by a str ong majority of respondents, such as support for national parks in general (70% ) and support for a volunteer rang er system (84%), there were also many questions that received fairly spl it distributions, such as perceptions of sponge resource trends and attitudes towa rds certain regulations. This implies a relatively low state of consensus among fishers, which has been found to be the case in many locations globally (Hampshire et al. 2004; PerezSanchez & Muir 2003; Richards on et al. 2005; Sandersen & Koester 2000). Gelcich (2005) cites that current policies assume that fishers will respond homogeneously and deterministically, which may lead to inappropriate management outcomes. Therefore it is important to cont act a representative sample of st akeholders to ensure that all
72 views are taken into account. Managers need to understand potential attitudinal differences in order to be more participatory and inclusive and hence, effective. Limitations of the Study While this research provides baseline data and preliminary insight into the surveyed community, there are a number of issues which necessitate caution. Fish ing rights and access to livelihood needs are contentious is sues, and there are examples of violent conflict concerning resource use rights in The Bahamas. Therefore, there is naturally some suspicion surrounding such research. This may influence respondents to answer survey questions cautiously or in a manner which they believe will secure their needs, which may bias the results. This couples with the frequently encountered social desirability bias as well as a lack of consistency in answers, possibly due to the newness of the information and lack of practice in thinking through the implications of national parks. Other potential sources of bias incl ude the non-random sample (which precludes interpreting these data as re presentative of the w hole community) and small sample size. Multiple surveyors added a potentially strong ef fect, which was explored in a variety of ways. In the bivariate residual an alysis data (Table 13-12), interv iewers 1 (female from the U.S.) and 3 (male from The Bahamas) had statistically significant associations with both support of national parks and support of the management z ones; respondents were significantly more likely to report supportive attit udes to interviewers 1 and 2, and more likely to report non-supportive attitudes to interviewer 3. When the data se ts were broken down by interviewer, only four variables retained significance ac ross all datasets: location, per ception of the resource conditions, perception of nothing as a threat, and preference s for enforcement (Table 3-13). However, this could partially be due to the reduced sample size upon separating the data, as the chi-square statistic is sensitive to small samples. Furthermore, while there are instances of lost significance
73 when the data are divided, there are no cases in which associations are reversed by the interviewer. The logistic regression analys es (Tables 3-14 to 3-16) addi tionally suggest that, holding all other variables constant, the interviewer variable is not significant fo r support NP and support WSNP. However, it does play a statistically sign ificant role in the support management zones variable, perhaps indicating that as the questions became more sensitive and issues more likely to impact the respondent, the interviewer effect beca me stronger. Unfortunately, due to this small sample size and item non-response, this bias coul d not be fully teased apart, and therefore necessitates caution in the inte rpretation of the results.
74 CHAPTER 5 RECOMMENDATIONS AND CONCLUSIONS Implications for Outreach and Participation Outreach may play an important role in cr eating a successful nati onal park. While radio broadcasts were the most common way that res pondents received their news, there was a great emphasis on wanting more personal contact with management staff to discuss National Park issues. Seventy-eight people requested group meetings to discuss the issues and learn more about it, and 11 more wanted one-on-one meetings with officials. Research suggests that early and frequent involvement of the community in mana gement planning is asso ciated with support for national parks. While such intensive interacti on might not always be possible due to time and financial constraints, active involvement with the resource users seems a vital element of an outreach campaign. In reference to the specific message, clarity as to the definition of a National Park appears necessary for success of the park. The documente d concern about losing livelihoods indicates that people assume that a nationa l park means that they will be excluded from the area. However, at recent community meetings (May 2007, January 2007), BNT st aff told attendees that no restrictions will be implemen ted without negotiating acceptabl e rules with the community. Therefore, if management bodies can work with all the stakeholders, these concerns may be addressed, and clarity of pers onal implications for fishers may ease much of the tension. Additionally, clarity of what a pa rk entails may help garner th e support of those who wanted no new development, as BNT has been framing th e West Side National Park as a means of minimizing the potential negative eff ects of large-scale development of the area, rather than as a method to stop locals from fishing. This message has not yet reached many of the fishers at the time of this research, so greater dissemination of this message could pot entially ease tensions.
75 Since support for parks and the WSNP is a ssociated with knowledge of conservation organizations, it may also follow that more public ity of these groups, increased membership, and outreach in general, could help with attitude s towards parks. However, with multiple groups working on the ground, it is important that they collaborate to craft out reach messages that do not have negative or contradictory messages about each other (McClanahan et al. 2005). Another element associated w ith support of parks was the perception of environmental threats and resource declines. Therefore, an outreach message which includes examples of declines in other fisheries and the effects of variou s local threats, as well as the potential benefits of management, may have a positive eff ect on support for the WSNP. Though community resource monitoring was the least often chosen op tion for participation, it was still chosen by 44 people, indicating a potential for a program whic h could both engage the community as well as determine the biological success of the park. This could help fi shers track resource abundance trends and respond to them. Pomeroy (1997) found th at not only actual resource increases, but also the perceptions of impacts of the program are important. Therefore it w ould be beneficial to make sure that the community is aware of even sm all gains or positive outcomes in order to start building trust and a sense of self-efficacy. Due to the strong concern over loss of livelihood s, as well as the majority of people not wanting their children to continue to fish, creating jobs associated with the park could also be a large factor in forging support for the park. While this is not always feasible, this is again an instance in which even small gains may aid in perceptions that could lead to building up the project little by little. Conclusions Common pool resources are cha llenging systems to manage, with insecure tenure and resource availability, multiple levels of mana gement, many stakeholders, and much at stake.
76 Support for conservation endeavors was associated with many factors which were not easily analyzed to reveal clear and unique predictors of national park and management support. The fishers exhibited a great deal of concern about th e parks, frequently speaking of concerns about losing their livelihoods, and of frustrations with politics and personal ag endas interfering with sound management. Yet, despite such fears and concerns, a majo rity of the respondents claimed to support national parks, the West Side National Park, and management zones. This is a good sign for the potential for collaboration and sustainable res ource management, as people concerned for their livelihoods are still willing to c onsider options and en ter into a discussion with managers to potentially support conservation measures. While A ndros is at the beginning of the process to develop both biologically effective and socially acceptable resource management plans, the communities and conservation organizations have taken a first step in moving towards a more collaborative process to involve stakeholders in creating sust ainable resource use systems.
77 APPENDIX A SURVEY QUESTIONNAIRE Note: Survey layout collapsed to shor ten for the purpose of the appendix. Interviewer______________ # Interview_____________ Date _______________ (Do not write the name of the person you are intervi ewingwrite the number, and enter number into XL) Andros Island Marine Resource User Survey INTRODUCTION: (INFORMED CONSENT INFORMATION APPROVED BY U. OF FLORIDA) My name is ____. I am working on a project for the University of Florida. We are doing interviews on what local people think about the management of th e oceans on the West Side of Andros, and how the community uses the ocean life. You w ill not be identified in our report. Your answers will be confidential You do not have to answer any questions you do not wish to answer, and you are free to stop the interview at any time. The interview tak es about an hour. There is no compensation for the interview, but you and your community will receive a copy of our results. Can I include your opinions in this study? ______ (initial if permission granted) There are no right or wrong answers. This project just want to understand the local opinions on the West Side, and how you think the area should be used and managed. This information will be put together into a report, which will be given back to the community, as well as the local government, The Bahamas National Trust, the local organizations, and all the othe r people who help manage the area. Im not here to try to convince you it should be one way or anothe r -this project just wants to get everyones opinions about what they think is the best way to use the area. This can help the managers and the community know what the local people want to happen with the area. This will also hopefully help create management plans that include local opinions and kn owledge about the area, which can help make sure that local needs are met and the environment stays healthy for future generations. A. THREATS FOR MARINE RESOURCES: These first few question will ask about your personal opinion on the condition of the ocean life of the West Side of Andros, and why you think it is that way. 1) Over the past 10 years, do you think the amount of bonefish that lives along the whole West Side has (1) Increased, (2) Stayed the same, or (3) Decreased? 2) Why do you think that is? 3) Over the past 10 years, do you think the amount of sponge in the West Side has (1) Increased, (2) Stayed the same, or (3) Decreased? 4) Why do you think that is? 5) Over the past 10 years, do you think the amount of crawfish in the West Si de has (1) Increased, (2) Stayed the same, or (3) Decreased? 6) Why do you think that is? 7) 10 years in the future, do you think the general marine resources in the West Side will have (1) Increased, (2) Stayed the same, or (3) Decreased? 8) What do you think are the threats to the marine life on the West Side? What might hurt the environment of the West Side?
78 9) What do you think would be the best way to use the West Side: [CIRCLE ONE] 1. Do not develop the areas on the West Side that are not already towns to keep the environment how it is. 2. Develop a few small ecotourism lodges on the West Side that can house up to 20 guests each. 3. Develop large resorts on the West Side that can house more than 50 guests. 4. Develop housing for people to live on the west side. 5. Develop industry on the West Side. 6. Other B. RESOURCE USE AND MAPPING: This project would like to understand where local people go to fish, what animals they use, and how much. Im going to ask you some questions about what animals you fish, and where you go to collect them. 10) What marine resources do you personally collect or use from the West Side? Q13 Animals Taken/Year 1. Sponges ___________________________ ______________ 2. Crawfish/Lobster_______________________________ ______________ 3. Grouper______________________________________ ______________ 4. Snapper______________________________________ ______________ 5. Other finfish _______________________________ __ ______________ 6. Turtles___________________________________ __ ______________ 7. Non-extractivesportfishing_______________ ____ ______________ 8. Tournament fishing ________ ______________ 9. Other_________________ _________________ ______________ 11) Which one is the most important to the way you make a living? 12) Which one is next most important? 13) [FOR ALL BUT THE TWO SPECIES THEY USE THE MOST] Please guess about how many [sponge, snapper, etc.] you take a year. 14) For the first species that you use the most: (write species name) 1. What season do you catch this species? 2. Where does [target species] live as adults and babies? (WRITE NAME OF HABITAT AND DRAW ON MAP) 3. Where do you go to catch [target species]? (WRITE NAME OF HABITAT AND DRAW ON MAP) 4. Where do you put your boat in the water? 5. How many times a week do you catch them? 6. How many hours do you travel on sea to get to your fishing spot? 7. How many hours per trip do you spend actually fishing?
79 8. What gear do you use to get the [TARGET SPECIES]? 9. What method do most of the others fishers use? 10. How much [TARGET SPECIES] do you catch pe r trip on average? [NOTE POUNDS VS NUMBERS] 11. How many years have you fished this resource? 12. How many people do you see out fishing at the same spot, generally? 13. Who do you sell your catch [services] to? 15) For the second species that you use the most: 1. What season do you catch this species? 2. Where does [target species] live as adults and babies? (WRITE NAME OF HABITAT AND DRAW ON MAP) 14. Where do you go to catch [target species]? (WRITE NAME OF HABITAT AND DRAW ON MAP) 15. Where do you put your boat in the water? 16. How many times a week do you catch them? 17. How many hours do you travel on sea to get to your fishing spot? 18. How many hours per trip do you spend actually fishing? 19. What gear do you use to get the [TARGET SPECIES]? 20. What method do most of the others fishers use? 21. How much [TARGET SPECIES] do you catch pe r trip on average? [NOTE POUNDS VS NUMBERS] 22. How many years have you fished this resource? 23. How many people do you see out fishing at the same spot, generally? 24. Who do you sell your catch [services] to? 16) Do you own your own boat? (1) Y / (2) N 17) What type of boat do you fish from? 18) Do you think enforcement of the laws for the areas where you fish should be (1) Stronger, (2) Stay the same, or (3) Weaker? 20) 19) Why? 21) Do you think the actual laws for the areas you fish should be (1) Stronger, (2) Stay the same, or (3) Weaker? 21) Why? 22) Do you see much illegal fishing? (1) Y / (2) N 23) IF YES: What types of activities are they, and are they Bahamian on foreign? 24) Are there any other resources, or ways of using the ocean, that are not being done now, but you think would be a good way to create more jobs? (1) Y / (2) N 25) IF YES, What are they? C. AREAS FOR PROTECTION Later in the survey we will discuss what you think pr otection should mean and how you think it should be done. Protection can mean many things, from simple monitoring to stronger rules. It does not have to mean that you cannot use the area. For now, Im using protection in the broadest way: as in keeping something safe. 26) Thinking of protecting the West Side for future generations, are there any areas of special concern to you, that you would like to see protected from damage? (1) Y / (2) N IF YES: 27A) What areas (name and draw on map)? 27B)Why are they special? 1. __________________________________ ___________________________
80 2. __________________________________ ___________________________ 3. __________________________________ ___________________________ 28) How do you think these areas should be mana ged or protected? What methods do you think should be used to keep the areas and animals safe from harm? 1. ____________________________________________________________ 2. ____________________________________________________________ 3. ____________________________________________________________ D. OUTREACH USE AND NEEDS: These few questions will help us understand the best way to communicate with the community about resources. 29) Where do you usually get your news or find out whats going on? Please check all that you use, and give the station/location/etc. 1. Newspaper______________________________________ 2. Radio __________________________________________ 3. Television_______________________________________ 4. Church announcements____________________________ 5. Posters in public pl aces_____________ _______________ 6. Government officials______________________________ 7. Other__________________________________________ 30) How do you hear about marine resources or fishery rules? 31) What would be the best way share information with the community about ocean life, fisheries management, and parks? E. KNOWLEDGE OF MPA: In 2002, the government approved boundaries for a park by the north bight, but did not manage it or add any regulations. Now a number of organizations, su ch as the Bahamas National Trust, and the local government and organizations, are interested in findi ng better boundaries and the best way to manage the existing area, with the input of the local communities I would like to get some more information about how good the communication about the park was in the past, and what you think about that area. Remember that there are no right or wrong answers. I just want to understand your opinion on the best way to use the area. Your name will not be attached to any of this information. 32) Did you know of the approved National Park area on the West Side? (1) Y / (2) N IF NO, SKIP TO Q. 36 33) IF YES: How did you find out about the West Side National Park? 34) When did you hear about it? 35) Do you know the boundaries? (1) Y / (2) N [IF YES, DRAW ON MAP] 36) [SHOW MAP AND ORIENT PERSON TO LOCATION OF PARK] How important do you think it is to protect this area? Is it (1) Important, (2) Not important, (3) Do you have no opinion bout it, or (9) Do not know? 37) Why? 38) Do you think it is a good idea to have Na tional Parks on Andros? (1) Y / (2) N 39) Why/ Why not?
81 F. MANAGEMENT 40) Do you have any concerns or worries about rul es for the current protected area? (1) Y/ (2) N 41) IF YES: What are they?42) What type of management do you think would be best for this National Park? [SHOW ZONING DRAWINGS] How should the park be used? Some of the options are: 1. Let all the fishing and all activities keep going, but monitor the area. 2. Keep current fishing in some areas, and makes some areas open only for sport-fishing or other activities that do not keep any o cean life, and monitor the area. 3. Have some areas where you can still fish like norma l, some areas where you can only do no-take fishing, and some areas that are completely clo sed to allow the populations to grow back, as a wildlife preserve. Also monitor. 4. Have zones that are open to sport-fishing and u ses that do not take animals, and zones that are closed to everyone to enter, as a wildlife preserve. Also monitor. Please Pick the one you think is best # : 43) And the one you think is second best: # : 44) Are there any other ways to take care of the park that you think would be good? Please tell me if you think the next list of possible re gulations would be a good idea, bad idea, (or you do not know, or have no opinion) for the current National Park on the West Side: 45) Create areas that are closed to commercial fishing. (1) Good Idea / (2) No Opinion / (3) Bad Idea / (9) DK 46) Create small zones in which no fishing at all is allowed. (1) GI / (2) NO / (3) BI / (9) DK 47) Limit the number of people who can use the area at one time. (1)GI / (2) NO / (3) BI / (9)DK 48) Create a permit system, so all users will need to get permits. (1) GI / (2) NO /(3) BI /(9) DK 49) Train willing fishermen to be volunteer rangers (1) GI / (2) NO / (3) BI / (9) DK G. INTEREST IN PARTICIPATION: The next set of questions asks about how much, and in what ways, you would like to have a say about management of protected areas and national parks. 50) From the following list of 5 ideas, please pick the three roles that you think are most important to you, that you would want to be involved with. These pictures each stand for one of the methods [SHOW ILLUSTRATIONS]: 1. Attending community meetings about managing the West Side? 2. Working as volunteer wardens, so you have the auth ority to report or stop people breaking rules. 3. Working with community resource monitoring, to collect information about how healthy the ocean is on the West Side, and how the populat ions of important animals are doing? 4. Educating the Bahamian community with a visitor center and tours of the west side? 5. Educating foreign tourists with a vis itor center and tours of the west side? 51) Do you have any other ideas of ways you woul d like to have the community get involved in managing and caring for the area? 52) Is there any reason you might not be able to participate? (1) Y / (2) N 53) IF YES: Why? 54) How do you feel about the amount of control that you have over ocean regulations now? Are you:(1) Satisfied / (2) No opinion / (3) Unsatisfied / (9) Do not know
82 55) Are you involved in any organized groups? (1) Y / (2) N 56) IF YES: What types? 57) Have you heard of: [Note reactions]: 1. Andros Conservancy and Trust (ANCAT) (1) Y / (2) N 2. Bahamas Sportfishing and Conservation Association (BSCA) (1) Y / (2) N 3. Bahamas National Trust (BNT) (1) Y / (2) N 4. The Nature Conservancy (TNC) (1) Y / (2) N H. RESOURCE DEPENDENCE: Next, Im going to ask you a few questions to help me understand how important the stuff you fish is for the way you make a living. 58) What are the main jobs that you make a liv ing from? What percentage does each add to your income? 1. ___________________________________________ ________% 2. ___________________________________________ ________% 3. ___________________________________________ ________% 59) [FOR COMMERCIAL FISHERMEN] What is the approximate worth of a daily catch? 60) How many people split that? 61) About how much does the aver age [FISHER TYPE] make a week? 62) How many people live in your household, including yourself? 63) How many people do you help support financially? 64) How many times a week do you eat food you catch yourself? 65) How do you support yourself when the weather is bad for a long time or catch is low on Andros? 66) Can you get help from the government in times of need? Y / N 67) What were the occupations of your parents? 1. Mother 2. Father 68) Do you want your children to be [spongers/fishers /crabbers/etc.] as well? (1) Y / (3) Maybe / (2) N 69) Why? I. DEMOGRAPHICS To finish up, Im just going to ask you a few questions about yourself. 70) What year were you born? 71) Where were you born? 72) What is the last grade you finished in school? 73) DO NOT ASK, BUT CIRCLE ONE: Male / Female 74) Would you like a copy of the report? If so, please give me your address: Thank you so much for your time!!! I really appreci ate it, and the information will be written up and sent back to you within the year. Do you have any questions about the project?
83 APPENDIX B GROUP MEETING DOCUMENTATION Focus Group, held In Red Bays, Andros Island, 5/25/06. Documented here are the questions and the bulleted answers as recorded on flipchart paper. Q1: HOW WOULD YOU DESCRIBE THE HEALTH OF THE WEST SIDE? Fairly healthy ame as 20 years ago production lower than 10 years agodue to hurricanes? Quiet and peaceful Not many people out therebonefishing guides from Staniard Creek and Lowe Sound No bonefishing guides in Red Bays More fishermen now than 20 years ago 95% unemployment in terms of government jobs, etc. both men and women. Sponge and fish for a living [sustainability] depends on methodc ut it [sponge], and it grows back [How many people cut, versus hook?] cutting is a learning process Spongers can only work a small timeweather causes muddy water, cant sponge. May go 2 weeks without going out. Then survive on straw work, farming Much farming on a local scale, family gardens, seasonal. 2-3 people in RB farm commercially Q2: WHAT FISHING AREAS ARE MOST IMPORTANT TO YOU? Not going to tell!!! Fish keep moving Great Bahamas Bankfishing grounds Androsians come from Kemps Bay, South Andros Fish different animals at different depths. Shallowbonefish, snapper (mangrove snapper, others), tarpon, permit (game fish, and tasty) Fish from Jolters Cays to Grassy Creek Deep: from Red Bays to Biminiwork Entire Bahamas Bank Work where you can depending on WEATHER Q3: WHAT AREAS ARE IMPORTANT TO PROTECT? The ocean Whole area Spent $33,000 to fight closing area in Florida Keys. If close, sponges start to die/get rotten. Use causes reproduction (squeeze out sperm). Use knife to cut, not pull. Close area for short time, then move on Sponge has seeds. Cant close it! 3 government employees, rest work at sea. Q4: WHAT TYPE OF REGULATIONS WOULD YOU SUPPORT? Will regulations stand? Close area so just fished by people of Red Bays? Already have size limits by law Self regulate Hard to arrest/report peoplemore of them on a boat
84 Any citizen can protect [fishing areas], but DANGEROUS!!! One fisherman almost lost lifebig boats, people have guns Poachers are dangerous In old days, Edwin Bersute [sp?] worked as a government inspector, had a boat and could pull up alongside you, could measure sponges, arrest people, would check boats. Need inspector, but no way to patrol everywhere People doing off season fishing have equipment [???] Protect sponges by cutting, stay away from undersized ones Do not want to be like Exuma! Leave it the way it is Do not want national park Keep laws as they are. Eventually can cause a problem [?] National Park frightens people Doing ok as is They have an inspector and abide by the laws Against closing off areas. If production goes down, they move to another spot. Crawfish, conch, snappergo [fish] in phases [fishermen] spread out fishing, so do not overfish Oceans the same, not much change [fishermen] go to many areas, different Community 100% doesnt want park [comment of one person] Concern is closing off fishing grounds If it can be guaranteed that they can keep on fishing, then theyd be ok with it. Q5: HOW WOULD YOU LIKE TO BE INVOLVED IN MANAGEMENT? Make sure to use different areas [How do you decide where to go/what areas are overused?] Fishermen like family, all decide together, communicate There are no foreigners, not a problem Have seasonal closings: August 1can start crawfish, so let sponges grow back. And when youre sponging, crawfish grow back. There are no threats that fishermen cannot control themselves Everyone knows the limits There are extremely low tides, in which boats are sitting on ground cant fish then. Q6: WHAT IS THE BEST WAY TO PROVIDE INFORMATION TO THE RED BAYS COMMUNITY? AdministratorDr. Christy Community meetings Local government For fishing info, Glenn Gaitor informs. He s ALWAYS around. Appears 5-7 times a week. Enforces laws, doing a good job. Knows every fisherman personally There is interest in tourism jobs BNT should hold a meeting in Red Bays Get people in charge to come and listen
85 APPENDIX C SPEARMAN'S RANK CORRELATIONS
86 Table C-1. Spearman's rank correlation for bivariate analysis of independent variables. Top row (1-23) corresponds with the lab eled variable column. (Ex: 1=Location) 1 2 3 4 5 6 7 8 9 10 11 1. Location Correlation 1.000 Sig. N 114 2.Fisher Type Correlation .309** 1.000 Sig. .001 N 107 110 3.Pluriactivity Correlation .068 .384** 1.000 Sig. .488 .000 N 105 104 108 4.Years Fished Correlation -.004 -.181 -.275** 1.000 Sig. .972 .069 .006 N 100 101 99 103 5. Own Boat Correlation .190* .300** .048 .295** 1.000 Sig. .046 .002 .620 .003 N 111 109 107 102 114 6. # in Hhold Correlation -.080 -.103 -.070 -.100 -.034 1.000 Sig. .421 .299 .472 .325 .730 N 104 103 107 99 106 107 7. Age Correlation .058 -.027 -.095 .633** .236* -.211* 1.000 Sig. .547 .779 .330 .000 .013 .029 N 109 107 107 102 110 107 112 8. Education Correlation -.018 .221* .167 -.410** -.098 -.011 -.451** 1.000 Sig. .858 .025 .091 .000 .316 .915 .000 N 105 103 103 98 106 102 107 108 9. Know NP Correlation -.021 .315** .087 -.039 .198* -.089 .023 .200* 1.000 Sig. .825 .001 .371 .699 .037 .361 .810 .039 N 110 107 108 102 111 107 111 107 113 10. When Hear Correlation -.070 .331** .140 .003 .178 -.087 .051 .238* .922** 1.000 Sig. .456 .000 .149 .973 .058 .371 .591 .013 .000 N 114 110 108 103 114 107 112 108 113 123 11.Know Groups Correlation .133 .572** .249** -.072 .348** -.219* .088 .197* .497** .496** 1.000 Sig. .168 .000 .009 .471 .000 .023 .358 .042 .000 .000 N 109 107 108 102 110 107 111 107 112 112 112 Correlation is significant at the 0.05 le vel (2-tailed) ** Correlation is si gnificant at the 0.01 level (2-tailed).
87 Table C-1. Continued 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 12.Env. Condition Corr. Coeff. .066 .400** .333** -.100 .107 -.028 .034 .090 .273** .239* .290** 1.000 Sig. .528 .000 .001 .363 .306 .794 .744 .401 .008 .018 .005 N 95 91 89 85 94 88 92 89 94 97 93 97 13. Threat: Nothing Corr. Coef. .056 -.183 -.308** .284** .051 .128 .158 -.241* .005 -.059 -.142 -.197 1.000 Sig. .550 .056 .001 .004 .587 .189 .097 .012 .956 .526 .135 .053 N 114 110 108 103 114 107 112 108 113 117 112 97 117 14. Threat: Nature Corr. Coeff. -.009 -.098 .014 .016 -.121 .079 -.051 .105 -.257** -.263** -.112 -.155 -.190* 1.000 Sig. .928 .310 .885 .872 .200 .418 .593 .279 .006 .004 .239 .130 .040 N 114 110 108 103 114 107 112 108 113 117 112 97 117 117 15. Threat: Chemicals Corr. Coeff. -.114 .034 .101 -.071 .008 -.033 -.106 .042 -.007 .034 -.009 .161 -.276** -.243** 1.000 Sig. .225 .723 .297 .476 .933 .735 .264 .670 .943 .718 .929 .114 .003 .008 N 114 110 108 103 114 107 112 108 113 117 112 97 117 117 117 16. Threat: Poaching Corr. Coeff. -.134 .064 -.045 -.034 .106 -.078 -.121 .013 .072 .116 .131 .133 -.289** -.161 .149 1.00 Sig. .156 .508 .643 .735 .264 .422 .203 .894 .447 .214 .169 .195 .002 .083 .109 N 114 110 108 103 114 107 112 108 113 117 112 97 117 117 117 117 17. Threat: Foreigners Corr. Coeff. -.029 .033 -.113 .061 .155 .049 .129 -.129 .098 .127 -.013 .043 -.163 -.172 .197* .116 1.00 Sig. .762 .736 .244 .540 .101 .617 .174 .184 .300 .173 .893 .676 .080 .064 .033 .212 N 114 110 108 103 114 107 112 108 113 117 112 97 117 117 117 117 117 Correlation is significant at the 0.05 level (2-tailed) ** Correlation is significan t at the 0.01 level (2-tailed).
88 Table C-1. Continued 1 2 3 4 5 6 7 8 9 10 18. Threats: Politics Corr .086 .034 -.011 -.023 .032 -.163 .133 .025 .050 .050 Sig. .364 .723 .908 .818 .738 .094 .161 .800 .599 .590 N 114 110 108 103 114 107 112 108 113 117 19. Threats: Development Corr .156 .442** .392** -.233* .208* -.184 -.203* .242* .167 .175 Sig. .097 .000 .000 .018 .026 .057 .032 .012 .078 .059 N 114 110 108 103 114 107 112 108 113 117 20. Enforce Law Corr .042 .355** .222* -.043 .369** -.050 .121 .221* .367** .362** Sig. .671 .000 .025 .671 .000 .620 .217 .026 .000 .000 N 107 104 102 98 108 101 106 102 107 110 21.Development Preference Corr .072 -.217* .070 -.018 -.171 -.073 .078 -.061 .002 .036 Sig. .475 .033 .491 .866 .087 .475 .442 .557 .981 .717 N 100 97 98 91 101 97 99 96 101 103 22.Children Corr -.205* -.288** -.346** -.019 -.108 .154 -.192* .032 -.183 -.168 Sig. .034 .003 .000 .852 .264 .117 .046 .747 .055 .079 N 107 105 106 101 108 105 109 105 110 110 23. Interviewer Corr -.373** -.129 -.209* -.025 -.124 -.011 -.100 -.004 -.308** -.237* Sig. .000 .180 .030 .803 .189 .907 .295 .967 .001 .010 N 114 110 108 103 114 107 112 108 113 117 Correlation is significant at the 0.05 level (2-tailed) ** Correlation is significan t at the 0.01 level (2-tailed).
89 Table C-1. Continued 11 12 13 14 15 16 17 18 19 20 21 22 18. Threats: Politics Corr .178 .054 -.057 -.147 -.017 -.101 .008 1.000 Sig. .060 .598 .543 .115 .854 .280 .931 N 112 97 117 117 117 117 117 117 19. Threats: Development Corr .316** .246* -.374** -.089 .045 .054 -.146 -.025 1.000 Sig. .001 .015 .000 .343 .632 .566 .116 .785 N 112 97 117 117 117 117 117 117 117 20. Enforce Law Corr .426** .307** -.373** .037 .116 .244* .052 .129 .260** 1.000 Sig. .000 .003 .000 .704 .228 .010 .588 .179 .006 N 106 91 110 110 110 110 110 110 110 110 21.Development Preference Corr -.070 .155 -.112 -.072 .186 -.015 .048 .104 -.031 .045 1.000 Sig. .491 .157 .258 .470 .061 .881 .630 .295 .756 .658 N 100 85 103 103 103 103 103 103 103 98 103 22.Children Corr -.334** -.199 .135 .111 .091 -.101 -.138 .009 -.275** -.227* -.100 1.000 Sig. .000 .058 .158 .248 .344 .292 .152 .923 .004 .021 .326 N 110 91 110 110 110 110 110 110 110 104 98 110 23. Interviewer Corr -.187* -.286** -.044 .274** -.132 .057 .092 -.024 -.104 -.230* -.257** .228 Sig. .048 .005 .639 .003 .156 .545 .326 .797 .262 .016 .009 .017 N 112 97 117 117 117 117 117 117 117 110 103 110 Correlation is significant at the 0.05 level (2-tailed) ** Correlation is significant at the 0.01 level (2-tailed).
90 APPENDIX D CHI-SQUARE TABLES FOR RESPONS E AND INDEPENDENT VARIABLES
91 Table D-1. Bivariate associations between re sponse variable (support national parks on Andros) and socio-demographics Total Support NPs on Andros Do not Support NPs on Andros Variable (n) (n) (n) ChiSquare DFp-value Fisher's Exact (p-value) Cramer's V Sociodemographics 1 Location 113793431.8121 3 <.00010.5306 Lowe Sound 25205 Red Bays 351223 Central Andros/ Behring Point 20191 Mangrove Cay 33285 2 Fisher Type 10774336.6838 10.00970.2499 Sportfishing Guide 35305 Commercial Fisher 724428 3 Pluriactivity 10826194.3214 20.1152 1 Type 453913 2 Types 5292 3 Types 11 4 Years Fished 10271315.2168 30.1566 0--10 32248 11--20 34268 21--30 221111 31+ 14104
92Table D-1. Continued Total Support National Parks on Andros Do not Support National Parks on Andros Variable (n) (n) (n) ChiSquare DF p-value Fisher's Exact (p-value)Cramer's V 5 Own Boat 111 77342.6191 1 0.1056 Yes 43 5117 No 68 2617 6 Number of People in Hhold 107 7334 0.05550.2709 1 49 3512 4 42 3111 7 8 26 10 + 8 53 7 Age 111 77340.974 3 0.8075 18 22 157 31 46 3016 44 25 196 57+ 18 135 8 Education 107 7334 0.0616 0-11th grade 49 2920 12th grade (HS) 51 3714 13+ (postgraduate) 7 70
93 Table D-2. Bivariate associations between response variable (s upport national parks on Andros) and knowledge Total Support N Ps on Andros Do not Support NPs on Andros Variable (n) (n) (n) ChiSquare DFp-value Fisher's Exact (p-value)Cramer's V Knowledge 9 Know about WSNP 113 79 34 7.306 1 0.0069 0.2543 Yes 41 35 6 No 72 44 28 10 When Heard WSNP 113 79 34 0.0683 1. Informed during survey 70 43 27 2. Period of research duration 17 13 4 3. After park creation (2002) 21 18 3 4. Prior to Park Declaration 5 5 0 11 Know Groups 112 78 34 17.2635 4 0.0017 0.3962 0 151 8 7 1 33 24 9 2 30 15 15 3 12 9 3 4 22 22 0
94Table D-3. Bivariate associations between response variable (suppo rt national parks on A ndros) and perceptions Total Support National Parks on Andros Do not Support National Parks on Andros Variable (n) (n) (n) ChiSquare DFp-value Fisher's Exact (p-value) Cramer's V 12 Environmental Scale 94 64 30 20.9098 2 <.0001 0.4716 Increasing 22 7 15 Same 51 37 14 Decreasing 21 20 1 13 Threat: Nothing 113 79 34 8.3954 1 0.0038 -0.2726 Present 19 8 11 Absent 94 71 23 14 Threat: Natural 113 79 34 3.5911 1 0.0581 -0.1783 Present 24 13 11 Absent 89 66 23 15 Threat: Chemical 113 79 34 4.4191 1 0.0355 0.1978 Present 28 24 4 Absent 85 55 30 16 Threat: Poaching 113 79 34 2.3398 1 0.1261 Present 31 25 6 Absent 82 54 28 17 Threat: Foreign 113 79 34 1.0000 Present 12 9 3 Absent 101 70 21 18 Threat: Politics 113 79 34 1.0000 Present 9 6 3 Absent 104 73 31 19 Threat: Development 113 79 34 8.5909 1 0.0034 0.2757 Present 43 37 6 Absent 70 42 28
95Table D-4. Bivariate associations between response variable (suppo rt national parks on Andros) and preferences Total Support National Parks on Andros Do not Support NPs on Andros Variable (n) (n) (n) ChiSquareDFp-value Fisher's Exact (p-value) Preferences 20 Fishery Enforcement 1077433 <0.0001 Weaker 330 Same 462125 Stronger 58508 21 Development of West Side 1017130 11.530820.0031 No New Development 623626 Small-scale Ecotourism 31283 Large Scale Development 871 22 Want their children to be fishers too? 1107733 6.413520.0405 Yes 14122 Maybe 36297 No 603624 Interviewer Effects Interviewer 1137934 22.29842<0.0001 Interviewer 1 635211 Interviewer 2 18153 Interviewer 3 321220
96Table D-5. Bivariate associations between response variable (s upport WSNP) and socio-demographics Total Support WSNP Do not Support WSNP Variable (n) (n) (n) ChiSquareDFp-value Fisher's Exact (p-value)Cramer's V Sociodemographics 1 Location 11365486.315330.0972 Lowe Sound 251510 Red Bays 351619 Central Andros/ Behring Point 20164 Mangrove Cay 331815 2 Fisher Type 10761464.412510.0357 0.2031 Sportfishing Guide 352510 Commercial Fisher 723636 3 Pluriactivity 10862466.552820.0378 0.2463 1 Type 452025 2 Types 523319 3 Types 1192 4 Years Fished 10258440.063230.9959 0--10 321814 11--20 341915 21--30 22139 31+ 1486
97 Table D-5. Continued Total Support WSNP Do not Support WSNP Variable (n) (n) (n) ChiSquareDF p-value Fisher's Exact (p-value)Cramer's V 5 Own Boat 11164470.22661 0.634-0.0452 Yes 683830 No 432617 6 Number of People in Household 1076146 0.30880.1884 1--3 493019 4--6 422418 7--9 826 10 + 853 7 Age 11163481.3913 0.70760.1119 18--30 221111 31--43 462917 44--56 251312 57+ 18108 8 Education 1076146 0.49890.1226 0-11th grade 492524 12th grade (HS) 513120 13+ (postgraduate) 753
98Table D-6. Bivariate associations be tween response variable (suppo rt WSNP) and knowledge Total Support WSNP Do not Support WSNP Variable (n) (n) (n) ChiSquareDF p-value Fisher's Exact (p-value)Cramer's V Knowledge 9 Know about WSNP 11365484.59531 0.0321 0.2017 Yes 412912 No 723636 10 When Heard WSNP 11365485.66193 0.1239 1. Informed during survey 703535 2. Period of research duration 17107 3. After park creation (2002) 21165 4. Prior to Park Declaration 541 11 Know Groups 112644813.32564 0.0098 0.3449 0 15411 1 331716 2 301614 3 1293 4 22184
99Table D-7. Bivariate associations betw een response variable (support WSNP) and perceptions Total Support WSNP Do not Support WSNP Variable (n) (n) (n) ChiSquareDF p-value Fisher's Exact (p-value)Cramer's V 12 Environmental Scale 9455391.85962 0.3946 0.1407 Increasing 221210 Same 512823 Decreasing 21156 13 Threat: Nothing 11365480.22361 0.6363 Present 19109 Absent 945539 14 Threat: Natural 11365481.70391 0.1918 Present 241113 Absent 895435 15 Threat: Chemical 11365480.00221 0.9627 Present 281612 Absent 854936 16 Threat: Poaching 11365480.24831 0.6183 Present 311912 Absent 824636 17 Threat: Foreign 11365483.66061 0.0557 0.18 Present 12102 Absent 1015546 18 Threat: Politics 1136548 0.2973 Present 972 Absent 1045846 19 Threat: Development 11365480.78861 0.3745 Present 432716 Absent 703832
100Table D-8. Bivariate associations betw een response variable (support WSNP) and preferences Total Support WSNP Do not Support WSNP Variable (n) (n) (n) ChiSquareDF p-value Fisher's Exact (p-value)Cramer's V Preferences 20 Fishery Enforcement 10761466.02822 0.0491 0.2374 Weaker 321 Same 462026 Stronger 583919 21 Development of West Side 10155460.57352 0.7507 No New Development 623230 Small-scale Ecotourism 311813 Large Scale Development 853 22 Want their children to be fishers too? 110644612.71572 0.0017 0.3400 Yes 14122 Maybe 362610 No 602634 Interviewer Effects Interviewer 11365480.11982 0.9419 Interviewer 1 633627 Interviewer 2 18117 Interviewer 3 321814
101Table D-9. Bivariate associations between response variable (support management zones) and socio-demographics Total Support Management Zones Do not Support Management Zones Variable (n) (n) (n) ChiSquare DFp-value Fisher's Exact (p-value) Cramer's V Sociodemographics 1 Location 109565212.2771 30.00650.3356 Lowe Sound 24168 Red Bays 351025 Central Andros/ Behring Point 19136 Mangrove Cay 311813 2 Fisher Type 10454500.9635 10.3263 Sportfishing Guide 342014 Commercial Fisher 703436 3 Pluriactivity 104535111.035 20.0040.3257 1 Type 451530 2 Types 493019 3 Types 1082 4 Years Fished 9950491.198 30.7535 0--10 311714 11--20 341816 21--30 20812 31+ 1477 _______________________________________________________________________________________________________
102 Table D-9. Continued Total Support Management Zones Do not Support Management Zones Variable (n) (n) (n) ChiSquare DFp-value Fisher's Exact (p-value) Cramer's V 5 Own Boat 10755522.6282 10.1050 Yes 663828 No 411724 6 Number of People in Household 10352512.7101 30.4385 1--3 462323 4--6 412219 7--9 826 10 + 853 7 Age 10756511.8747 30.5988 18--30 221210 31--43 442024 44--56 24159 57+ 1798 8 Education 1035350 0.04730.2456 0-11th grade 471928 12th grade (HS) 492821 13+ (post-graduate) 761
103 Table D-10. Bivariate associations betw een response variable (support management zones) and knowledge Total Support Management Zones Do not Support Management Zones Variable (n) (n) (n) ChiSquare DFp-value Fisher's Exact (p-value) Cramer's V Knowledge 9 Know about WSNP 109575210.3417 10.00130.3080 Yes 402911 No 692841 10 When Heard WSNP 1095752 0.00770.3253 1. Informed during survey 672740 2. Period of research duration 17125 3. After park creation (2002) 21147 4. Prior to park declaration 440 11 Know groups 10857519.6673 40.04640.2992 0 1569 1 311615 2 301218 3 1165 4 21174
104 Table D-11. Bivariate associations between response variab le (support management zones) and perceptions Total Support Management Zones Do Not Support Management Zones Variable (n) (n) (n) ChiSquare DF p-value Fisher's Exact (p-value) Cramer's V 12 Resource Condition Percep. Scale 91 4744 14.2363 2 0.00080.3955 Increasing 21 417 Same 49 2722 Decreasing 21 165 13 Threats: Nothing 109 5752 3.9579 1 0.0467-0.1906 Present 19 613 Absent 90 5139 14 Threat: Natural 109 5752 3.583 1 0.0584-0.1813 Present 23 815 Absent 86 4937 15 Threat: Chemical 109 5752 6.8247 1 0.00900.2502 Present 27 207 Absent 82 3745 16 Threat: Poaching 109 5752 0.3173 1 0.3173 Present 30 1713 Absent 79 4039 17 Threat: Foreign 109 5752 0.0249 1 0.8747 Present 11 65 Absent 98 5147 18 Threat: Politics 109 5752 0.4933 Present 9 63 Absent 100 5149 19 Threat: Development 109 5752 1.0267 1 0.3109 Present 41 2417 Absent 68 3335
105Table D-12. Bivariate associations between response variable (s upport management zones) and preferences Total Support Management Zones Do not Support Management Zones Variable (n) (n) (n) ChiSquareDF pvalue Fisher's Exact (p-value) Cramer's V Preferences 20 Fishery Enforcement 1045450 <0.00010.4851 Weaker 202 Same 451233 Stronger 574215 21 Development of West Side 995148 <.00010.4264 No New Development 612239 Small-scale Ecotourism 30255 Large Scale Development 844 22 Want their children to be fishers too? 10657495.73932 0.05520.2338 Yes 1266 Maybe 342410 No 602733 Interviewer Effects Interviewer 109575220.59632 <.00010.4347 Interviewer 1 603921 Interviewer 2 17125 Interviewer 3 32626
106 APPENDIX E CHI-SQUARE TABLES FOR RESPONSE VARIABLE Table E-1. Contingency table of response variables: WSNP by NP. Note the high number of discordant pairs (in bold), showing support for the WSNP but not national parks in general, or vice versa ( =5.318, p=.021 for association of the two variables). Support WSNP Total Not important/ do not know Important Support NP (general) No 20 14 34 Yes 28 51 79 Total 48 65 113 Table E-2. Contingency table of response variables: management zones by NP. Note the high number of discordant pair s (in bold), showing support fo r the management zonation but not national parks in general, or vice versa ( =35.410, p<0.0001 for association of the two variables). Management Zones Total No zonation Zonation Support NP (general) No 30 3 33 Yes 22 54 76 Total 52 57 109 Table E-3. Contingency table of response variables: manageme nt zones by WSNP. Note the high number of discordant pairs (in bold ), showing support for the management zonation but not national parks in general, or vice versa ( =4.665, p=0.031 for association of the two variables). Management Zones Total No zonation Zonation Support WSNP Not important/do not know 28 19 47 Important 24 38 62 Total 52 57 109 Table E-4. Reduced contingency table of response variables: WSNP by NP. Note the high number of discordant pair s (in bold), showing support fo r the WSNP but not national parks in general, or vice versa ( =4.855, p=.028 for association of the two variables). Support WSNP Total Not important/ do not know Important Support NP (general) No 8 3 11 Yes 19 33 52 Total 27 36 63
107 Table E-5. Reduced contingency ta ble of response variables: management zones by NP. Note the high number of discordant pairs (in bold), showing support for the management zonation but not national parks in general, or vice versa (Fisher's Exact Test yields p=0.002 for association of the two variables). Management Zones Total No zonation Zonation Support NP (general) No 8 2 10 Yes 13 37 50 Total 21 39 60 Table E-6. Reduced contingency ta ble of response variables: management zones by WSNP. Note the high number of discordant pairs (in bold), showing support for the management zonation but not National Parks in general, or vice versa ( =7.163, p=0.007 for association of the two variables). Management Zones Total No zonation Zonation Support WSNP Not important/Do not know 14 12 26 Important 7 27 34 Total 21 39 60
108 LIST OF REFERENCES Agrawal, A. 2001. Common Property Institutions and Sustainable Governance of Resources. World Development 29 :1649-1672. Agrawal, A. 2002. Common resources and instituti onal sustainability. Pages 41-85 in T. Dietz, and E. Ostrom, editors. The Drama of the Commons. National Research Council, Washington, D.C. Aipanjiguly, S., S. K. Jacobson, and R. Fl amm. 2003. Conserving Manatees: Knowledge, Attitudes, and Intentions of Boaters in Tampa Bay, Florida. Conservation Biology 17 :1098. Alder, J., D. Zeller, T. Pitcher, and R. Suma ila. 2002. A Method for Eval uating Marine Protected Area Management. Coastal Management 30 :121-131. Alessa, L., S. M. Bennett, and A. D. Kliske y. 2003. Effects of knowledge personal attribution and perception of ecosystem health on deprecia tive behaviors in the intertidal zone of Pacific Rim National Park and Reserve. Journal of Environmental Management 68 :207218. Alexander, S. 2000. Resident attitudes toward s conservation and black howler monkeys in Balize: the Community Baboon Sanctu ary. Environmental Conservation 27 :341-350. Allendorf, T. D. 1999. Local residents perception s of protected areas in Nepal: Beyond conflicts and economics. Page 98. The Univer sity of Minnesota, St. Paul. Aquilino, W. 1994. Interviewer Mode Effects in Surveys of Drug and Alcohol Use. Public Opinion Quarterly. 58: pages 210-240. Argady, T. S. 1997. Marine Protected Areas and Ocean Conservation. R.G Landes Company, Austin, TX. Badalamenti, F., A. A. Ramos, E. Voultsiadou, Aacute, J. L. Nchez Lizaso, G. D'Anna, C. Pipitone, J. Mas, J. A. R. Fernandez, D. Whitmarsh, and S. Riggio. 2002. Cultural and socio-economic impacts of Mediterranean marine protected areas. Environmental Conservation 27 :110-125. Bernard, R. 2002. Research Methods in Anthropol ogy: Qualitative and Quantitative Approaches. AltaMira Press, Walnut Creek, CA. Blake, D. 2001. Contextual Effects on Environmen tal Attitudes and Beha vior. Environment and Behavior 33 :708-725. BREEF, MacAlister Elliott & partners Ltd. 1998. Fisheries Management Action Plan for the Bahamas. 80 Pages. A report to the Bahamas Department of Fisheries.
109 Burke, L., and J. Maidens, editors. 2004. Reef s at risk in the Cari bbean. World Resources Institute. Washington, D.C. Christie, P., B. McCay, M. Miller, C. Lowe, A. White, R. Stoffle, D. Fluharty, L. McManus, R. Chuenpagdee, C. Pomeroy, D. Csuman, B. Blount, D. Huppert, and R.-L. Eisma. 2003. Toward developing a complete understanding: A social science research agenda for marine protected areas. Fisheries 28 :22-25. Chuenpagdee, R., J. Fraga, and J. I. Eun -Avila. 2004. Progressing Toward Comanagement Through Participatory Research. Society & Natural Resources 17 :147-161. Chuenpagdee, R., and D. Pauly. 2004. Improving the State of Coastal Areas in the Asia-Pacific Region. Coastal Management 32 :3-15. Converse, P. 1974. Nonattitudes and American Public Opinion: Comment: The Status of Nonattitudes. The American Political Science Review 68 : 650-660. Cottrell, S. P. 2003. Influence of Sociodemogra phics and Environmental Attitudes on General Responsible Environmental Behavior among Recreational Boaters. Environment and Behavior 35 :347-375. Crawford, B. R., A. Siahainenia, C. Ro tinsulu, and A. Sukmara. 2004. Compliance and Enforcement of Community-Based Coastal Re source Management Regulations in North Sulawesi, Indonesia. Coastal Management 32 :39-50. Dahlgren, C. P. 2004. Bahamian Marine Reserves -Past Experience and Future Plans. Pages 268-286 in J. A. Sobel, and C. P. Dahlgren, editors. Marine Reserves: A Guide to Science, Design, and Use. Island Press, Washington. Davis, D. 1997. The Direction of Race of Interviewer Effects Among African-Americans. American Journal o Political Science. 41: 309-22 Dixon, J. A. 1993. Economic benefits of marine protected areas. Oceanus 36 :35-36. Fiallo, E., and S. Jacobson. 1995. Local communiti es and protected areas : attitudes of rural residents towards conservation and Machalil la National Park, Ec uador. Environmental conservation 22 :241-249. Finkel, Guterbock, and Borg. 1991. Race-of-Intervie wer Effects in a Pre-election Poll. Public Opinion Quarterly. 55:313-30. Friedlander, A., J. Sladek Nowlis, J. A. Sanch ez, R. Appeldoorn, P. Usseglio, C. McCormick, S. Bejarano, and A. Mitchell-Chui. 2003. Designi ng Effective Marine Protected Areas in Seaflower Biosphere Reserve, Colombia Based on Biological and Sociological Information. Conservation Biology 17 :1769-1784. Garaway, C., and N. Esteban. 2003. Increasing MPA effectiveness through working with local communities: Guidelines for the Caribbean. Pages 1-44. MRAG Ltd., London, UK.
110 Gelcich, S., G. Edwards-Jones, and M. J. Ka iser. 2005. Importance of Attitudinal Differences among Artisanal Fishers toward Co-Managemen t and Conservation of Marine Resources. Conservation Biology 19 :865-875. Gell, F. R., and C. M. Roberts. 2003. Benefits beyond boundaries: the fisher y effects of marine reserves. Trends in Ecology & Evolution 18 :448-455. Geoghegan, T., Y. Renard, N. Brown, and V. Krishnarayan. 1999. Evaluation of Caribbean experiences in participatory planning and management of marine and coastal resources. Pages 1-52. The Caribbean Natural Resour ces Institute. Laventille, Trinidad. Geoghegan, T., A. Smith, and K. Thacker. 2001. Char acterization of Caribbean marine protected areas: An analysis of ecologica l, organizational, and socio-ec onomic factors. Pages 1-20. Caribbean Natural Resources Institute. Hampshire, K., S. Bell, G. Wallace, and F. Stepukonis. 2004. "Real" Poachers and Predators: Shades of Meaning in Local Understandings of Threats to Fisheries. Society & Natural Resources 17 :305-318. Hardin, G. 1968. The Tragedy of the Commons. Science 162 :1243-1248. Himes, A. H. 2003. Small-Scale Sicilian Fish eries: Opinions of Artisanal Fishers and Sociocultural Effects in Two MPA Case Studies. Coastal Management 31 :389. Hinds, M. J., M. S. Chinnan, L. R. Beuchat, R. S. Pomeroy, R. B. Pollnac, B. M. Katon, and C. D. Predo. 1997. Evaluating factors contributing to the success of commun ity-based coastal resource management: the Central Visayas Regional Project-1, Philippines. Ocean and Coastal Management 36 :97-120. Infield, M. 1988. Attitudes of a rural community towards conser vation and a local conservation area in Natal, South Afri ca. Biological Conservation 45 :21-46. Ite, U. 1996. Community percepti ons of the Cross River National Park, Nigeria. Environmental Conservation 23 :351-357. IUCN. 1988. Proceedings of the 17th Session of the General Assembly of IUCN and the 17th Technical Meeting. San Jose, Costa Rica. Jackson, J. B. C., M. X. Kirby, W. H. Berger, K. A. Bjorndal, L. W. Botsford, B. J. Bourque, R. H. Bradbury, R. Cooke, J. Earlandson, J. A. Este s, T. P. Hughes, S. Kidwell, C. B. Lange, H. S. Lenihan, J. M. Pandolfi, C. H. Peters on, R. S. Steneck, M. J. Tegner, and R. R. Warner. 2001. Historical Overfishing and th e Recent Collapse of Coastal Ecosystems. Science 293 :629. Jacobson, S. 1999. Communication Skills for Co nservation Professionals. Island Press Washington, D.C.
111 Jentoft, S., B. J. McCay, and D. C. Wilson. 199 8. Social theory and fisheries co-management. Marine Policy 22 :423-436. Jones, P. J. S. 2006. Collective action problem s posed by no-take zones. Marine Policy 30 :143156. Kane, E and Macaulay, L. 1993. Interviewer Gender and Gender Attitudes. Public Opinion Quarterly. 57: pages 1-28. Kelleher, G., Bleakley, C., Wells, S. 1995. A Global Representative System of Marine Protected Areas. Volume 2: Pages 1-40. Th e World Bank. Washington, D.C. Lepp, A., and S. Holland. 2006. A Comparison of A ttitudes Toward State-Led Conservation and Community-Based Conservation in the Village of Bigodi, Uganda. Society and Natural Resources 19 :609-623. Lundquist, C. J., and E. F. Granek. 2005. Strate gies for Successful Marine Conservation: Integrating Socioeconomic, Political, and Sc ientific Factors. Conservation Biology 19 :1771-1778. Marks, L. C. 2005. Participatory Planning for R ecreation Management in Abaco National Park, Bahamas. Wildlife Ecology and Conservati on. University of Florida, Gainesville. Mascia, M. B. 1999. Governance of Marine Protected Areas in the Wider Caribbean: Preliminary Results of an International Ma il Survey. Coastal Management 27 :391-402. Mascia, M. B. 2003. The Human Di mension of Coral Reef Mari ne Protected Areas: Recent Social Science Research and Its Poli cy Implications. Conservation Biology 17 :630. Mascia, M. B. 2004. Social Dimens ions of Marine Reserves. Pages 164-186 in J. A. Sobel, and C. P. Dahlgren, editors. Marine Reserves: A Guide to Science, Design, and Use. Island Press, Washington. McClanahan, T., J. Davies, and J. Maina. 2005. F actors influencing resour ce users and managers' perceptions towards marine protected ar ea management in Kenya. Environmental Conservation 32 :42-49. Mehta, J., and J. Heinen. 2001. Does Community-B ased Conservation Shape Favorable Attitudes Among Locals? An Empirical Study from Nepal. Environmental Management 28 :165-177. Napier, V., G. Branch, and J. Harris. 2005. Evalua ting conditions for successful co-management of subsistence fisheries in KwaZulu-Natal, South Africa. Environmental Conservation 32 :1-13. National Research Council. 2001. Marine Pr otected Areas: Tools for Sustaining Ocean Ecosystems. National Academy Press, Washington,D.C.
112 Nowlis, J. S., and Friedlander, A. 2004. Design and Designation of Mari ne Reserves. Pages 128163 in J. A. Sobel, and C. P. Dahlgren, ed itors. Marine Reserves: A Guide to Science, Design, and Use. Island Press, Washington. Ostrom, E. 1990. Governing the Commons: The E volution of Institutions for Collective Action. Cambridge University Press. Perez-Sanchez, E., and J. F. Muir. 2003. Fish ermen perception on resources management and aquaculture development in the Mecoacan estuary, Tabasco, Mexico. Ocean & Coastal Management 46 :681-700. Pollnac, R. B., B. R. Crawford, and M. L. G. Gorospe. 2001. Discovering factors that influence the success of community-based marine protec ted areas in the Visayas, Philippines. Ocean and Coastal Management 44 :683-710. Pomeroy, R. S., and F. Berkes. 1997. Two to tang o: The role of government in fisheries comanagement. Marine Policy 21 :465-480. Pomeroy, R. S., R. Pollnac, C. Predo, and B. Katon. 1996. Impact evaluation of communitybased coastal resources management projects in Philippines. International Centre for Living Aquatic Resources Management, Metro, Manila. Richardson, E. A., M. J. Kaiser, and G. EdwardsJones. 2005. Variations in fishers' attitudes within an inshore fishery: implications for management. Environmental Conservation 32 :213-225. Richardson, E. A., M. J. Kaiser, G. Edwards-J ones, and H. P. Possingham. 2006. Sensitivity of Marine-Reserve Design to the Spatial Resolu tion of Socioeconomic Data. Conservation Biology 20 :1191-1202. Routhe, A. S., R. E. Jones, and D. L. Feld man. 2005. Using theory to understand public support for collective actions that im pact the environment: allevia ting water supply problems in a nonarid biome. 86 :874(824). Sah, J., and J. Heinen. 2001. Wetland resource use and conservation attitudes among indigenous and migrant peoples in Ghodaghodi Lake ar ea, Nepal. Environmental Conservation 28 :345-356. Salm, R. V., J. R. Clarke, and E. Siirila 2000. Ma rine and Coastal Protected Areas: a Guideline for Planners and Managers. World Conservation Union, Washington, D.C. Salz, R. J., and D. K. Loomis. 2005. Recreat ion Specialization and Anglers Attitudes Towards Restricted Fishing Areas Human Dimensions of Wildlife 10 :187-199. Sandersen, H. T., and S. Koester. 2000. Co-manag ement of Tropical Coasta l Zones: The Case of the Soufriere Marine Management Area, St. Lucia, WI. Coastal Management 28 :87-97.
113 Sesabo, J., H. Lang, and R. Tol. 2006. Perceived Attitude and Marine Protected Areas (MPAs) establishment: Why households characteristics matters in Coastal resources conservation initiatives in Tanzania. Pages 1-39. Ham burg University, Centre for Marine and Atmospheric Science. Sobel, J. A., and C. P. Dahlgren 2004. Marine Reserves: A Guide to Science, Design, and Use. Island Press. Washington, D.C. Spiteri, A., and S. Nepalz. 2006. IncentiveBased Conservation Programs in Developing Countries: A Review of Some Key Issu es and Suggestions for Improvements. Environmental Management 37 :1-14. Worm, B., E. B. Barbier, N. Beaumont, J. E. Duffy, C. Folke, B. S. Halpern, J. B. C. Jackson, H. K. Lotze, F. Micheli, S. R. Palumbi, E. Sa la, K. A. Selkoe, J. J. Stachowicz, and R. Watson. 2006. Impacts of Biodiversity Loss on Ocean Ecosystem Services. Science 314 :787-790. Zaller, J. and Fledman, S. 1992. A Simple Theory of the Survey Response: Answering Questions versus Revealing Preferences. American Journal of Political Science 36 : 579-616.
114 BIOGRAPHICAL SKETCH Dina Liebowitz was born in Israel and raised in Austin, TX and Isra el. She earned a BA in biology and environmental studies from the Universi ty of Pennsylvania in 2000 then spent a year working with environmental and co-existence organi zations in Jerusalem. Dina worked the next four years leading ecology field trips for Harvard' s Department of Organismic and Evolutionary Biology, before coming to the University of Florid a for her MS. After attaining this degree, she plans to continue her educa tion by pursuing a PhD with th e NFS IGERT ProgramAdaptive Management: Water, Wetlands, and Watersheds.