AN ABSTRACT OF THE THESIS OF Sara Gress for the degree of Master of Science in Forest Ecosystems and Society presented on June 4, 2015. Title: Diversity in the Outdoors: Student Attitudes About Wilderness in the National Outdoor Leadership School. Abstract approved: ______________________________________________________ Troy E. Hall Outdoor Experiential Education (OEE) programs have been growing in popularity in America for the past few decades, yet participation has continued to be dominated by White, upper-­‐class individuals. These similar disparities exist in general outdoor recreation participation, with individuals from race and/or ethnic minority groups participating in outdoor recreation activities less than Whites. Research in the leisure field has revealed that different meanings and values assigned to wilderness by different ethno-­‐racial groups may explain some of the difference in participation rates, and that wilderness has historically been a concept associated with White, upper-­‐class males. With major demographic shifts occurring in the United States, along with the greater emphasis on social justice in the leisure field, OEE organizations must confront the imbalance in participant demographics. The National Outdoor Leadership School (NOLS) is a worldwide leader in OEE and has made this issue a priority with its Gateway Scholarship Program. This program partners NOLS with community-­‐based organizations and schools to provide low-­‐
income individuals, who also self-­‐identify as people of color, full-­‐tuition scholarships for NOLS courses. The purpose of this mixed methods study was to determine if Gateway scholarship students and non-­‐Gateway scholarship students held different wilderness attitudes and values before and after their NOLS course, if their wilderness attitudes changed during their NOLS experience, and if the predictor variables of previous wilderness experience, ethno-­‐racial identity and urban/rural residence were related to wilderness attitudes. Wilderness attitudes were measured with four constructs (sense of place, environmental ethics, value of wilderness, and environmental awareness) derived from NOLS literature and previous research. All Gateway students and a proportionally representative sample of non-­‐Gateway students from the same NOLS courses during the summer of 2014 were invited via email to participate in the study. A retrospective pre-­‐ and post-­‐test was administered online, and 19 follow-­‐up interviews were conducted via telephone. Results show that while Gateway students (n = 41) entered their NOLS course with significantly less positive wilderness attitudes than non-­‐Gateway students (n = 33), post-­‐course scores between the two groups were not significantly different. Both groups experienced significant positive change in wilderness attitudes from pre-­‐ to post-­‐course scores, with Gateway students experiencing a larger change. Previous wilderness experience was a significant pre-­‐course predictor for all constructs, Gateway status was a significant pre-­‐course predictor for all but sense of place, and community type was entirely non-­‐significant for either pre-­‐ or post-­‐course scores. Post-­‐course, Gateway status was the only variable that remained significant, and only for the value of wilderness construct. Interview data revealed potential reasons for attitude change, including course content (e.g., Leave No Trace, wilderness medicine), a deeper personal connection to wilderness, and heightened awareness of environmental issues. Limitations of the Gateway Scholarship program are discussed and implications of the study’s results are placed in context of social justice goals in OEE. ©Copyright by Sara Gress June 4, 2015 All Rights Reserved
Diversity in the Outdoors: Student Attitudes About Wilderness in the National Outdoor Leadership School by Sara Gress A THESIS submitted to Oregon State University in partial fulfillment of the requirements for the degree of Master of Science Presented June 4, 2015 Commencement June 2015 Master of Science thesis of Sara Gress presented on June 4, 2015 APPROVED: Major Professor, representing Forest Ecosystems and Society Head of the Department of Forest Ecosystems and Society Dean of the Graduate School I understand that my thesis will become part of the permanent collection of Oregon State University libraries. My signature below authorizes release of my thesis to any reader upon request. Sara Gress, Author ACKNOWLEDGEMENTS I would like to thank my major professor, Dr. Troy Hall, for all of her guidance and hard work over the past two years. My graduate school experience would not have been nearly as incredible without her as my major professor. I would also like to thank my committee, Dr. Bo Shelby, Dr. Patti Sakurai, and Dr. Natchee Barnd, for giving me a fresh perspective on my research and being supportive of my graduate work. TABLE OF CONTENTS Page Introduction .......................................................................................................................................... 1 Study context and rationale .............................................................................................. 2 Wilderness attitudes and values ..................................................................................... 8 Research questions and hypotheses .......................................................................... 11 Methods ................................................................................................................................................. 15 Sampling and study design ............................................................................................. 15 Data analysis ......................................................................................................................... 21 Results .................................................................................................................................................... 23 Demographic characteristics ......................................................................................... 23 Reliability and normality of quantitative variables ............................................. 25 Change in wilderness attitudes .................................................................................... 27 Difference in wilderness attitudes between Gateway and non-­‐Gateway students .................................................................................................................................. 32 Personal and cultural factors related to wilderness attitudes ........................ 37 Discussion ............................................................................................................................................ 42 Implications ......................................................................................................................................... 47 Limitations ........................................................................................................................................... 50 Future Research ................................................................................................................................. 51 Conclusion ............................................................................................................................................ 51 References ............................................................................................................................................ 53 Appendices ........................................................................................................................................... 60 TABLE OF CONTENTS (CONTINUED) Page Appendix A IRB approval letter ................................................................................. 60 Appendix B Effect sizes .................................................................................................. 61 Appendix C Survey instrument .................................................................................. 62 Appendix D Cronbach’s alpha results ...................................................................... 69 LIST OF TABLES Table Page 1. Definition of terms…………………………………………………………………………..…….. 7 2. Wilderness attitude constructs and definitions……….…………………..…………..10 3. Interview questions……………………….....…………………………………………………...20 4. Initial Cronbach’s alpha results, dropped items and final alphas..…………… 22 5. Categories and counts of self-­‐reported ethno-­‐racial identities…………………24 6. Gateway and non-­‐Gateway students’ prior experience with overnight wilderness trips………………………………………………………………………………..24 7. Number of students from each community type………………………………………24 8. Codes and definitions from qualitative analysis……………………………………….26 9. Change in attitudes………………………………………………………………………………...29 10. Pre-­‐ and post-­‐course wilderness attitudes reported in interviews…………...30 11. Difference in attitudes between Gateway and non-­‐Gateway students..……..33 12. Personal and cultural factors as predictors of wilderness attitudes………….40 1 Introduction Major demographic shifts are occurring in the United States and the upcoming generations’ ecological perspectives and values will determine the future of America’s wildlands. These shifts will entail a large increase in racial and ethnic diversity (Ortman & Guarneri, 2009). Outdoor recreation and related endeavors are continually growing in popularity, but the typical participant characteristics do not reflect the greater U.S. population, with racial and ethnic minority participation rates not matching population proportions (Johnson, Bowker, Green, & Cordell, 2007). Firsthand experience of the outdoors can be an effective method of creating a positive environmental ethic (Harris, 2002), and outdoor recreation provides that firsthand experience. Also, research shows that positive environmental attitudes are correlated with the extent to which an individual feels a part of nature (Schultz, Shriver, Tabanico, & Khazian, 2004). In line with this, outdoor experiential education (OEE) strives to teach an environmental ethic through firsthand experiences in the outdoors. However, similarly to outdoor recreation, OEE has been historically populated mainly by White, upper class individuals (Holman, Bobilya, & McAvoy, 2008), which -­‐-­‐ when considered in the context of the demographic makeup of the population in the U.S. -­‐-­‐ threatens OEE’s relevance and ethical justification. It is a social obligation of OEE organizations to prepare for and directly engage the people and issues that make up the changing face of the United States (Warren, Roberts, Breunig, & Alvarez, 2014). The National Outdoor Leadership School (NOLS) has made it a goal to reflect the needs and perspectives of diverse communities in its programming (National Outdoor Leadership School, 2013b). To achieve this, and thereby widen the ecologically minded community, it is imperative to understand student attitudes toward the environments where NOLS operates. This goal of understanding and including diverse attitudes, expectations and needs of ethnic and racial minorities extends beyond NOLS; outdoor recreation and park managers must also do the same to ensure the American public continues to benefit from and value our outdoor areas (Roberts & Chitewere, 2011). 2 The purpose of this study was to understand the attitudes and values held by students on NOLS courses and determine if a change in attitudes and values occurred from pre-­‐course to post-­‐course. The study compared students who received a scholarship, based on an interest in the outdoors, financial need and self-­‐
identification as a person of color, to non-­‐scholarship students. While studies have qualitatively explored these perspectives in similar contexts (Roberts & Chitewere, 2011; Roberts & Suren, 2010), and critical analyses of OEE have been published (Rose & Paisley, 2012; Warren, 2005; Warren et al., 2014), a mixed methods analysis, using pre-­‐ and post-­‐tests of the wilderness attitudes and values brought to OEE courses, such as NOLS courses, has not been conducted. This study will help examine the impact of the NOLS diversity initiative and illuminate areas of success and potential places for improvement. Below, after the study context and rationale are established, methods and measurements are explained. Results and implications are then discussed, with attention to potential areas of concern about the NOLS diversity initiative, as well as positive outcomes. Specific definitions of terms used in the context of this study are listed in Table 1 on page 8. These terms will be used throughout this document and, although they are based on previous literature and do not markedly differ from other offered definitions, the definitions in Table 1 only apply to this study context. Study context and rationale It has been empirically established that minority ethno-­‐racial populations do not visit outdoor areas or participate in outdoor recreation activities to the same degree as the dominant ethno-­‐racial group (Whites) in the United States (Floyd, 1998; Johnson et al. 2007; Stanfield, Manning, Budruk, & Floyd, 2005). The differences in outdoor recreation participation levels and behaviors have been a subject of academic discussion and research in the leisure field for many years and continue to be the focus of many scholars (Byrne & Wolch, 2009; Floyd, 1999; Floyd & Johnson, 2002; Johnson, Bowker, & Cordell, 2004; Le, Holmes, & Kulesza, 2012; Roberts & Chitewere, 2011; Warren et al., 2014). Research has covered many 3 recreation contexts, from urban parks to wilderness, to the National Park system and OEE programs. Explanations for the current ethno-­‐racial makeup of outdoor recreationists have come from a variety of sources. Leisure researchers have presented four main theories: marginality, subculture, assimilation and discrimination (Floyd, 1999). These theories assert that socio-­‐economic inequalities, different cultural values, differing levels of cultural integration into the dominant society, and perceived or actual discrimination are the cause of low ethno-­‐racial minority participation in outdoor recreation. These theories have been criticized as being ethnocentric, in that the dominant (White) patterns of outdoor recreation participation and valuation are considered the “normal” against which non-­‐White populations are compared. As such, these theories treat racial and ethnic groups as monolithic, static entities, ignoring intra-­‐group diversity (Floyd, 1998). In light of these criticisms, some scholars have taken a different approach. Geography and landscape/urban planning scholars, as well as government agencies, have attempted to explain differences in perceptions and valuation of natural areas based on cultural differences among recreationists and cultural values embodied in the physical landscapes themselves (Buijs, Elands, & Langers, 2009; Byrne, 2012; Johnson, Bowker, Bergstrom, & Cordell, 2004; Johnson, Horan, & Pepper, 1997). These studies have approached the question about the disparity in racial and ethnic makeup of recreationists with the understanding that wilderness and natural park areas stem from a historically dominant White American identity, and ask instead how the White ideal of nature embodied by these spaces impacts different ethno-­‐
racial groups’ perception and use of these spaces, as opposed to asking what is unique about these groups’ recreation styles. This approach applies the concepts offered by environmental historians, who have established the idea of wilderness in America as ethnocentric, representing a White historical perspective and interpretation of nature, which has led to lingering institutional discrimination embedded in America’s natural park areas, even in the present day (Cronon, 1996; DeLuca & Demo, 2001; Merchant, 2003; Spence, 1999). These studies have offered explanations beyond merely differences in income or preference for different park 4 amenities, as earlier studies offered. Results have shown that, although visitation rates of ethno-­‐racial minorities are lower than that of Whites, this does not mean that non-­‐visitors do not value wilderness and natural park areas; valuation does not necessary translate to visitation, and the two are separate constructs that should not be conflated. Ethno-­‐racial minority participants in many studies report intrinsic, utilitarian, personal and recreation attitudes and values in relation to wilderness and natural areas; reported barriers to visitation range from lack of socio-­‐economic means and perceived or actual discrimination, to feeling unwelcome or perceived differences from other wilderness visitors (Byrne; Johnson, Bowker, Bergstrom, & Cordell; Johnson, Horan, & Pepper). Lack of visitation to park and wilderness areas, and participation in outdoor recreation activities, can be due to the combination of low socio-­‐economic means and differing perceptions of what and for whom wilderness and natural park areas are. Low participation by ethno-­‐racial minorities in outdoor recreation is reflected in the socio-­‐demographic characteristics of OEE participants. Recent research has extended the aforementioned concepts to analyze the lack of ethno-­‐
racial minority participation specific to OEE. Social justice, viewed through the lens of critical theory, has been used to reveal the historically created and concurrently reinforced roots of White privilege in the conceptual underpinnings of OEE (McLean, 2013; Rose & Paisley, 2012). This line of research argues that OEE programs can unintentionally reproduce whiteness through the dominant discourse of colonial innocence, uninhabited wilderness and the erasure of indigenous people from these discourses. It contends that merely bringing more ethnically and racially diverse actors into the same system does nothing to change the underlying social structure. The integration of social justice into OEE programs would result in more culturally aware and inclusive teaching methods and outcomes, create a sense of belonging for all students, and foster a more inclusive, versus exclusive, participatory outdoor learning environment (Roberts & Chitewere, 2011; Warren et al., 2014). Scholars employing social justice as a lens have endeavored to discover perspectives differing from the dominant culture to offer an alternative to the prevailing discourses surrounding many OEE programs (Agyeman, 2003; Roberts & 5 Chitewere; Roberts & Suren, 2010). Social justice demands a restructuring of the system, for the dominant culture to recognize its role in creating the current system, and for non-­‐White perceptions of nature and histories to be included in the pedagogy of outdoor education. The National Outdoor Leadership School is directly addressing the need for social justice with its Gateway Scholarship Program. The National Outdoor Leadership School is a world leader in OEE and is addressing the call to become more inclusive of the diverse perspectives of its students, staff and the broader global community (National Outdoor Leadership School, 2013a). A major vehicle for accomplishing this goal is the Gateway Scholarship program. The program partners NOLS with community-­‐based organizations, schools and programs to provide low-­‐income individuals full-­‐tuition scholarships for NOLS courses, as well as post-­‐course service commitments. Gateway Scholarship students must also self-­‐identify as a person of color. This program allows NOLS to reach out to cities and communities that are typically less aware of the NOLS mission, and it is an important effort towards becoming a more diverse organization. The Gateway Scholarship program widens the span of communities and individuals that NOLS, and outdoor recreation in general, can positively impact. For NOLS to be successful in creating a more diverse and inclusive organization, it is imperative to understand the attitudes and values brought to NOLS by students and how NOLS can use knowledge of these attitudes to position itself within a rapidly changing society. Bringing in more students from diverse backgrounds without recognizing the different attitudes about wilderness each individual brings achieves integration, but not inclusion, into the OEE culture. Inclusion requires a better understanding of the diverse perspectives of the students that NOLS instructs; NOLS has made inclusion a priority in its Strategic Plan and seeks to reflect the “perspectives and needs of diverse communities” (NOLS, 2013b, p. 8). The Gateway Scholarship program operated by NOLS presents a unique opportunity to answer the call by Warren and others (2014) for further exploration into how different attitudes held by ethno-­‐racial groups influence how they experience the outdoors by directly comparing individual attitudes, moderated 6 by ethno-­‐racial identity, held by students experiencing the same OEE course and wilderness setting. The specifics of the NOLS approach to wilderness and wilderness attitudes are discussed in the next section. Table 1. Definitions of terms. Definitions used in the context of this study. Term Definition Race Instead of operationalizing race and ethnicity as two separate constructs, this study will Ethnicity operate on the premise that race and ethnicity are both social constructions, attributable to internal and external ascriptions, and function together to form part of an individual’s identity. This study is not looking to draw inferences based on specific and externally assigned racial and ethnic groups; rather, its aim is to discern if self-­‐reported ethno-­‐racial identities contribute to differences in wilderness attitudes. Therefore, the term “ethno-­‐racial group” will be employed to refer to an individual’s self-­‐reported identity in relation to racial and ethnic constructs. Outdoor Experiential Educational programs that use outdoor/wilderness settings and planned activities (“learning Education (OEE) by doing”) to achieve a preconceived and intentional outcome. The National Outdoor Leadership School is a leading OEE organization, and uses wilderness settings and extended stays in those settings to achieve learning outcomes. Wilderness An undeveloped area that is free from substantial human impacts, where nature is dominant and the situations created by this setting have real consequences. Wilderness areas are used by NOLS as a setting for courses. Wilderness in the context of this study will not refer to only legally designated wilderness areas under the Wilderness Act of 1964, although some of the areas utilized by NOLS courses may be legally designated wilderness. Wilderness classroom A wilderness (as defined above) setting that NOLS uses during its courses, in which teaching, learning and interaction between instructors, students and the landscape takes place; the wilderness setting itself can be the teacher by providing real experiences that instructors can help create and interpret. Wilderness attitudes and The meanings and evaluations individual students assign to the wilderness areas used by NOLS values courses. These meanings and evaluations may or may not be stable and can change over time. 7 Supporting Literature (Floyd, 1998; Rose & Paisley, 2012) (Rose & Paisley, 2012; Warren, 2005) (Ewert & McAvoy, 2000; National Outdoor Leadership School, 2014) (Gookin & Leach, 2008; Rose & Paisley, 2012) (Ajzen, 2012; Heberlein, 2012; Steg & de Groot, 2012) 8 Wilderness attitudes and values Wilderness skills, leadership and stewardship play an integral role in the NOLS mission. NOLS has placed wilderness as a central organizational value by striving to be a leader in teaching and developing wilderness skills and leadership, as well as making a commitment to wilderness a priority for the organization and its alumni (National Outdoor Leadership School, 2013b). NOLS seeks ultimately to foster a wildland ethic, which is built with key concepts such as sense of place, wilderness valuation and environmental awareness, as well as through its teachings and use of wilderness classrooms, all of which constitute a wildland ethic (Gookin, 2002a, 2002b; Johnson, 2002; Lamb & Goodrich, 2006). As defined by NOLS, an ethic consists of choice, judgment and values, and its formation is influenced by family, environment, education, society, history, religion and culture; these factors make an ethic unique to every individual (Lamb & Goodrich). In the context of a NOLS course, a wildland ethic defines the relationship between a student and the wilderness classrooms used during the course. NOLS seeks to impart a wildland ethic that students will take beyond their course and integrate into their daily lives (National Outdoor Leadership School, 2013b). NOLS employs a variety of geographic locations and specific skill sets to impart a wildland ethic in its students. NOLS operates all over the world, with courses taught in Alaska, Australia, Africa, India, Mexico, New Zealand, Patagonia and various locations across the American west (NOLS, 2014). Outdoor skills taught range from backpacking, horsepacking and mountaineering, to rock climbing and skiing, to water sports such as sailing, rafting and kayaking. Course length ranges from nine days to an entire academic year. Most NOLS courses are open to anyone who is interested, but some cater to outdoor education professionals looking to further their career, or to youth under 18. The NOLS curriculum of leadership, risk management, wilderness skills and environmental studies is the foundation for every course (selected course descriptions cited as sources are the main four courses taken by Gateway students) (National Outdoor Leadership School, 2015a, 2015b, 2015c, 2015d). The main 9 learning objective for skill transfer is to teach students to be competent and responsible backcountry leaders and travelers. NOLS accomplishes this by slowly giving students more responsibility throughout the course, culminating with a student-­‐led expedition at the end of a course. This exercise allows students to practice what they learned during the course and demonstrate leadership, risk management and outdoor skills, so that these skills can be taken beyond the course and into their personal and professional lives. Environmental studies are also an important component in every NOLS course. Connections between minimum impact backcountry and frontcountry living are taught with Leave No Trace principles, by making students aware of their impact on the environment. NOLS also strives to cultivate a curiosity about the natural world. Environmental issues and potential solutions are a part of each course, with the hope that students will leave NOLS with a better understanding of current environmental challenges. This knowledge, along with Leave No Trace skills, is important to students’ development of a wildland ethic that is transferrable to daily life, which is NOLS’s main objective. The learning objectives described above, operationalized as wilderness-­‐
related attitudes and values, formed the basis of analysis for this study. Four attitude constructs were identified and defined for the purpose of this study: sense of place, environmental awareness, value of wilderness and environmental ethics. The full definition of each construct can be found in Table 2. Referred to collectively as “wilderness attitudes” in this thesis, these four constructs compose a person’s wilderness attitudes, which for the purposes of this study are defined as the meanings and evaluations individual students assign to the wilderness areas used by NOLS courses, while environmental ethics and awarness go beyond those places. These meanings and evaluations can change over time (Ajzen, 2012; Steg & de Groot, 2012), and this study focused on attitude change promoted by NOLS. 10 Table 2. Wilderness attitude constructs and definitions Sense of place An individual’s emotional and interpretative reaction to and (Kyle & Chick, interaction with a specific physical environment, which reflects 2007; MacLean, the meanings assigned to the place by the individual. Sense of 2002) place in the context of this study is relying mainly on NOLS literature, as opposed to the wider literature on sense of place. Environmental A sense of responsibility for the health of the environment, and (Gookin, 2002a; awareness having knowledge of ecological processes and environmental MacLean, 2002) issues. Value of The ecological, recreational and personal evaluations and (Johnson et al., wilderness importance assigned to wilderness by an individual. 1997; Kyle, Bricker, Graefe, & Wickham, 2004) Environmental A sense of membership in an ecological community, recognition (Johnson, 2002; ethic of the intrinsic value of wilderness and commitment to Gookin, 2002b) environmentally responsible behavior. Each individual has his or her own personal version of an environmental ethic. This study attempted to deconstruct certain factors that impact wilderness attitudes before and after a NOLS course. Three predictor variables were chosen, based on the criteria for becoming a Gateway Scholarship student: interest in the outdoors (though not necessarily experience in the outdoors), living in a community less aware of the NOLS mission (typically urban areas), and self-­‐identifying as a person of color. Each of these variables has been examined in previous research dealing with outdoor recreation participation and/or wildland meaning and valuation. Drawing from this previous research, I proposed the following research questions and hypotheses. 11 Research questions and hypotheses This study explored the wilderness attitudes Gateway Scholarship students and non-­‐scholarship students brought to the wilderness classrooms that NOLS courses inhabit, and whether change in these attitudes occurred as a result of a NOLS course. Specifically, I asked: Do students’ attitudes change over the course of a NOLS experience? Why or why not? This question gave rise to the first hypothesis: H1) The NOLS course will have a positive impact on wilderness attitudes. NOLS seeks to foster a wildland ethic and pro-­‐environmental behavior in students through its courses. Attitudes are a product of values and ethics, and attitudes are an antecedent to behavior (Lamb & Goodrich, 2006; Steg & de Groot, 2012). Direct experience can change attitudes, as direct experience can build upon existing beliefs; attitudes formed from direct experience also tend to be more stable (Heberlein, 2012). Individuals who take NOLS courses are most likely already interested in and hold positive beliefs about wilderness, which can predispose them to positive attitude change (i.e., strengthen positive attitudes), provided that the course experience is favorable (Waage, Paisley, & Gookin, 2013). Therefore, a NOLS course would most likely impact wilderness attitudes through direct experience and accessing pre-­‐existing beliefs, leading to positive change in wilderness attitudes. Such a positive impact on attitudes would occur in both Gateway Scholarship and non-­‐scholarship students, because both groups voluntarily self-­‐select to participate. Gateway Scholarship students typically come from different backgrounds and communities than “traditional” NOLS students, and this study examined how their attitudes were similar or dissimilar to those of non-­‐scholarship students, both pre-­‐ and post-­‐course. Therefore, the second research question was: Do Gateway Scholarship students and non-­‐scholarship students have different wilderness attitudes before a NOLS course? Do their attitudes differ after a NOLS course? This study also attempted to explain similarities and differences in wilderness attitudes between the two groups by asking a final question: Are the differences in wilderness attitudes 12 explained by personal and cultural factors (i.e., previous wilderness experience, ethno-­‐
racial group or urban/rural residency)? Given their different backgrounds, I hypothesized that Gateway Scholarship students and non-­‐scholarship students would differ on the three personal and cultural constructs (ethno-­‐racial identity, previous wilderness experience, urban/rural residency), with Gateway Scholars more likely to be non-­‐White, lacking previous wilderness experience, and from urban locations, as these are factors in being considered for a Gateway Scholarship (National Outdoor Leadership School, 2013a). Due to the potential impact of these factors on wilderness attitudes, Gateway Scholars were hypothesized to more likely have less positive pre-­‐course wilderness attitudes. This gave rise to hypotheses 2 and 3: H2) Pre-­‐NOLS course attitudes will be a function of ethno-­‐racial identity, previous wilderness experience and urban/rural residency. H3) Gateway Scholarship students and non-­‐scholarship students will have significantly different wilderness attitudes before the NOLS course, with Gateway Scholarship students having less positive attitudes than non-­‐scholarship students. More specifically, it was hypothesized that non-­‐scholarship students would reflect the historically “typical” demographic of OEE participants in being White and upper-­‐middle class (Holman et al., 2008) and would possess more positive pre-­‐
course wilderness attitudes than Gateway Scholarship students. The potential impact of ethno-­‐racial identity, previous wilderness experience, and community type on wilderness attitudes is discussed below. Race and ethnicity are typically conflated in leisure research, with the terms used interchangeably and measured with the race and ethnic categories given by U.S. Census Bureau. Instead of operationalizing race and ethnicity as two separate constructs, this study operated on the premise that race and ethnicity are both social constructions, attributable to internal and external ascriptions, and function together to form an individual’s identity (Floyd, 1998; Rose & Paisley, 2012). This 13 study was not looking to draw inferences based on specific and externally assigned racial and ethnic groups; rather, its aim was to discern if self-­‐reported ethno-­‐racial identities contribute to differences in wilderness attitudes. Therefore, the term “ethno-­‐racial group” is employed to refer to an individual’s self-­‐reported identity in relation to racial and ethnic constructs. Ethno-­‐racial identity has been shown to relate to attitudes and beliefs about the environment and/or wilderness, as well as outdoor recreation participation, with non-­‐White populations having been shown to have less pro-­‐environmental attitudes and recreation participation rates for reasons ranging from lower socio-­‐
economic status and perceived discrimination to cultural meaning assigned to landscapes and wilderness being rooted in White cultural discourse (Buijs et al., 2009; Johnson et al., 1997; Johnson, Bowker, Bergstrom, et al., 2004; Johnson, Bowker, & Cordell, 2004). Based on this research, and that wilderness is a historically White-­‐dominated concept, I expected that ethno-­‐racial identity might partially explain differences in wilderness attitudes, with White ethno-­‐racial identity corresponding to more positive wilderness attitudes. It must be noted, however, that the research is mixed on whether ethno-­‐
racial identity per se is a significant factor in environmental beliefs and/or recreation participation. When factors such as income, age, gender and place of residence were held constant, Johnson, Bowker, and Cordell (2001) found that race was not a significant factor in outdoor recreation constraints. Similarly, African Americans were shown to hold similar environmental values as Whites, but differ on environmental behaviors and self-­‐reported levels of powerlessness in a study by Parker and McDonough (1999). Given this, the relationship between ethno-­‐racial identity and environmental values/attitudes is a complex phenomenon, and the relationship between ethno-­‐racial identity and wilderness attitudes may be a function of other variables (e.g., income, place of residence, etc.) that are correlated with both identity and attitudes. Prior wilderness experience is included in Hypothesis 2 because previous positive experience with a place has been shown to contribute to more positive attitudes and meanings associated with it (Kyle & Chick, 2007). Additionally, 14 organized group wilderness experiences, such as NOLS, have been shown to result in positive attitudinal change, due to the intense nature of the experience coupled with the self-­‐reflection and critical analysis of that experience by instructors (Ewert & McAvoy, 2000). Therefore, I expected that the level of previous wilderness experience might explain wilderness attitudes, as students with pervious wilderness experience, especially previous NOLS courses or similar type experiences, would have formed some of the positive wilderness attitudes that NOLS promotes. Specifically, I hypothesized that previous wilderness experience would be positively related to wilderness attitudes. This is especially likely in this population, who self-­‐
selected to participate, as presumably, people with negative prior wilderness experiences don’t sign up for a NOLS course. Direct experience with the outdoors is related to higher levels of environmental awareness and stronger pro-­‐environmental ethics (Harris, 2002). Additionally, personal connection to and feeling a part of nature also positively influences environmental awareness and ethics (Schultz et al., 2004). Each of these factors, direct experience and connectedness to nature, may also extend to wilderness attitudes. Individuals from lower socio-­‐economic backgrounds who reside in urban areas tend to participate in outdoor recreation activities at lower rates than Whites (Johnson, Bowker, & Cordell, 2004; Shores, Scott, & Floyd, 2007), and research provides evidence that rural dwellers report fewer constraints to outdoor recreation than urban residents (Johnson et al., 2001). I hypothesized that students from rural areas would have more experience with the outdoors and outdoor recreation than students from urban areas. Therefore, students with rural backgrounds would have more positive wilderness attitudes due to more frequent exposure to the outdoors. If the NOLS course is successful in fostering positive wilderness attitudes, differences between Gateway Scholars students and non-­‐scholarship students, due in part to ethno-­‐racial identity, previous wilderness experience and community type, will be smaller after the NOLS course than before the course. This gave rise to the final hypothesis: 15 H4) Differences in wilderness attitudes between Gateway Scholars students and non-­‐
scholarship students will be significantly smaller after the NOLS course than before the course. Attitude change may not be equal across Gateway and non-­‐Gateway students. I proposed that non-­‐Gateway students might experience a “ceiling effect,” in that these students entered NOLS courses with existing positive attitudes, leaving less room for additional positive change. This phenomenon has been observed in environmental education classes (Eagles & Demare, 1999) and in visitor attitudes towards wilderness (Shin & Jaakson, 1997). In both of these studies, respondents possessed pre-­‐existing positive environmental or wilderness attitudes, clustering near the top of the scale. Gateway students, coming to courses with less positive attitudes, would have a likelihood of larger change. Thus, I expected that all students would experience positive change (Hypothesis 1), but that it would be larger among Gateway students, leading to Hypothesis 4. Answering these questions and testing these hypotheses will help NOLS understand how students experience and perceive the wilderness classrooms where NOLS operates and offer empirical evidence for NOLS to use in the quest for social justice in OEE. Methods Sampling and study design Data were collected using a mixed methods approach with a sequential exploratory strategy consisting of two phases: Phase 1) a quantitative survey, and Phase 2) a qualitative interview component (Creswell, 2003). The quantitative survey was conducted in October and November, 2014, after students had returned home from NOLS courses they took during the summer of 2014. The qualitative interviews were conducted in December, 2014, and January, 2015, after the quantitative survey was closed. Both components were reviewed and approved by 16 Oregon State University’s Institutional Review Board prior to the study (see Appendix A for IRB approval letter). A minimum sample size of 120 returned surveys (60 Gateway Scholarship students and 60 non-­‐scholarship students) was proposed, based on power analysis in Howell (2013) and calculated by G*Power (Faul, Erdfelder, Buchner, & Lang, 2009). The minimum sample size for a two-­‐tailed paired t-­‐test with an alpha error probability of 0.05 and estimated effect size of (d) 0.80 is 23 students. The minimum sample size for an independent samples t-­‐test with an alpha error probability of 0.05 (two-­‐tailed) and estimated effect size of (d) 0.80 is 84 students, 42 Gateway Scholars and 42 non-­‐scholarship students. Following Howell (2013), who suggests sample size should exceed independent variables by at least 50, the minimum sample size for the multiple regression is 53 students. Achieved effect sizes can be found in Appendix B. I expected a response rate of approximately 50%. To achieve the target of 120 surveys would require inviting 240 people. However, after determining that 84 Gateway students were available to sample, the target final sample size was expanded to 152 students. The National Outdoor Leadership School provided a list of students’ contact information, including only those who a) took a course during the summer of 2014, and b) were on a course with at least one Gateway student present. A total of 84 Gateway students took a NOLS course during the summer of 2014; however, only 76 Gateway students had email contact information on file at NOLS. All 76 Gateway students with emails were sampled, along with 84 non-­‐Gateways to match the total number of Gateway students. To sample non-­‐Gateway students, the number of non-­‐
Gateway scholarship students randomly chosen from each course was equal to the number of Gateway students from that course. For example, Whitewater River Expedition had two Gateway students, so two non-­‐Gateway students were randomly selected from the remaining course roster. An online random number generator, (randomizer.org, maintained by the Social Psychology Network) was used to select non-­‐Gateway students from the course rosters. 17 The quantitative survey was conducted with Qualtrics, an online survey tool. After the sample had been selected, students were contacted via email with a link to the survey, inviting them to participate in the study. Three days after the first email was sent, a reminder was sent to those who had not completed the survey. For those students under 18 years of age, an email was also sent to their parents, informing them of their child’s inclusion in the study and allowing them to remove their child from the study. This email included a link to the survey so parents could preview the survey if they wished. No parents requested that their child be removed from the study. A final reminder email was sent out a month after the original invitation, only to those students who had not yet completed the survey. A total of 74 students completed the survey, composed of 33 Gateway students and 41 non-­‐Gateway students, for a response rate of 46%. Wilderness attitudes were measured with 7-­‐point Likert-­‐type scale statements. Respondents were asked how much they agreed or disagreed with each statement using the following scale: (1) Strongly Agree, (2) Agree, (3) Somewhat Agree, (4) Neither Agree or Disagree, (5) Somewhat Disagree, (6) Disagree, and (7) Strongly Disagree. Statements were based on terms used NOLS literature and measures from previous research (Gookin, 2002a, 2002b; Hutson & Bailey, 2008; Johnson, 2002; Johnson et al., 1997; Johnson, Bowker, Bergstrom, & Cordell, 2004; Kyle et al., 2004; Sibthorp, Paisley, Gookin, & Ward, 2007). A copy of the survey with original sources can be found in Appendix C. Often, in studies like this, surveys are administered before and after an experience to measure change in attitudes. However, this may not be the most effective or accurate method to assess differences in pre-­‐ and post-­‐course attitudes. For measurements that rely on a self-­‐reported internal metric and are not stable, such as self-­‐perceptions and attitudes, a retrospective pre-­‐test has been shown to better capture respondents’ self-­‐reported changes (Sibthorp et al., 2007; Miller & Hinshaw, 2012). This approach helps to avoid response-­‐shift bias, which is caused when participants change their internal self-­‐rating system because of their experiences during the program, resulting in pre-­‐ and post-­‐trip self-­‐reported assessments that are difficult to compare (Sibthorp et al.). For example, “average 18 skill level” could mean something different to a person after the course when the participant realized her pre-­‐course skill was in actuality less than average, despite what she previously perceived as average. Additionally, some research suggests that recollections of pre-­‐trip attitudinal states are largely accurate (Dickson & Hall, 2006), and in one study attitude change was more accurately captured using a retrospective pre-­‐test/post-­‐test than a traditional pre-­‐test/post-­‐test for undergraduate education students taking a course aimed at inclusion in the classroom (Miller & Hinshaw, 2012). Therefore, this study used the post-­‐test and retrospective pre-­‐test method, as the focus was on attitudes and values, which are a non-­‐stable and internal metric. Also, this method was more logistically feasible, as the survey was administered only one time, unlike traditional pre-­‐ and post-­‐test surveys. To measure pre-­‐course and post-­‐course attitudes, each statement was presented with two drop-­‐down menus (see Figure 1 for a screen capture of one set of measures). This format was chosen to reduce survey length and respondent burden; the drop down menus allowed the statement and both pre-­‐ and post-­‐course response scales to be displayed on the screen at the same time. Demographics (gender, age, course) and moderating factors (previous wilderness experience, community type, ethno-­‐racial identity) were measured with multiple-­‐choice and open-­‐ended questions. At the end of the survey, a question invited students to be interviewed, asking for their contact information (name, email and phone number) if they were interested. Out of the 74 respondents, 28 students consented to an interview, consisting of 11 Gateway students and 17 non-­‐Gateway students. All 28 students were contacted about an interview through an email reminding them of their participation in the study and asking them when they were available for an interview. A reminder email was sent out a week after the initial invitation to those who had not yet responded. A second reminder was emailed out a week after the first, and a final reminder email, along with a phone call and voicemail, took place ten days later. Interviews were conducted over the phone, and lasted between eight and 30 minutes. Out of the 28 possible interviewees, 19 were interviewed, 19 consisting of ten non-­‐Gateway students and nine Gateway students. The other nine students never responded to any contact attempts. I called the interviewee at the time and date agreed upon during email communication. To begin the interview, I reminded the interviewee that the interview was completely voluntary and asked for permission to record the interview. All interviewees agreed to the interview being recorded. I followed an interview guide, which can be found in Table 3. When necessary, I would ask probing questions, such as, “What do you mean by that?” and “Why do you think that is?” Interviews were conducted with the aid of a guide, not a script, to ensure a narrative interview process, with the stories of interviewees being the main method of explanation and understanding individual perspectives and meanings (Kvale & Brinkmann, 2009). Face-­‐to-­‐face interviews with the respondents would have been ideal, but were not practical in terms of the study time frame or budget. Another advantage of telephone interviews is that respondents were back in their typical daily environments, hopefully leading to better reflection on their NOLS experience and negating any subjective norms or pressures that could exist if interviews were conducted within the NOLS environment. Also, this avoided any immediate post-­‐
course “elation” and measured long-­‐term attitudinal changes more accurately. Figure 1. Example questions from the online survey. These statements compose the “environmental awareness” construct. When clicked, the blank boxes alongside the statements show the 7-­‐point scale, with “Strongly Agree” on top. Screenshot of Qualtrics survey. 20 Table 3. Interview questions. Semi-­‐structured phone interview questions Had you taken a NOLS course, or something similar, before this summer? Had you camped out overnight in the wilderness before your NOLS course? How did you become interested in taking a NOLS course? Was wilderness something you, or your family or friends, were interested in before your course this summer? The word wilderness means a lot of different things to many different people. What did the word “wilderness” mean to you before your NOLS course? What does the word “wilderness” mean to you now, after your NOLS course? Was there something specific that happened on your course that changed your view of wilderness? Did your instructor do or teach you something that changed how you think about wilderness? Did you learn from or have conversations with fellow students that changed how you think about wilderness? Did your feelings about wilderness change in a way that you did not anticipate or are surprised by? After your course, did you talk about what your course was like to family and friends back at home? What was it like? Did they seem to understand what you experienced? Is it important to you that they understand what you experienced? Do you see yourself taking more overnight wilderness trips (or another NOLS course) in the future? 21 Data analysis Quantitative analysis was conducted with SPSS. Three negatively worded items were re-­‐coded to match the direction of the positive items on the survey: “…I didn’t understand why others cared about wilderness,” “…I didn’t feel a commitment to wilderness,” and “…I didn’t understand ecological processes in wilderness.” Indexes for pre-­‐course scores and post-­‐course scores were created for each of the wilderness attitude constructs (sense of place, environmental ethics, value of wilderness, and environmental awareness). Indices were combined overall mean scores from all items in each construct. Cronbach’s alpha was used to ensure internal consistency of the retrospective pre-­‐test and post-­‐test measures. An alpha of 0.7 or higher was considered acceptable, as that is the standard for psychological constructs (Field, 2013). A few items were dropped from indexes after Cronbach’s alpha revealed their inconsistency with the other items in the construct (Table 4). Complete Cronbach’s alpha results can be found in Appendix D. One item from “sense of place” was dropped, leaving a total of four items in the index. No items were dropped from “environmental ethics” or “value of wilderness,” for a total of four and five items, respectively. One item was dropped from “environmental awareness,” leaving a total of three items in the index. Finally, one item was dropped from “course experience,” leaving a total of three items in the index. Course experience items asked about the respondents’ experience during the course, so only one post-­‐course index was created for this construct. Indices created through the steps described above were used as dependent variables in hypothesis tests. Paired t-­‐tests (Howell, 2013) were used to test Hypothesis 1, which states that positive change would occur within students between pre-­‐ and post-­‐ course wilderness attitudes. Independent-­‐samples t-­‐tests (Graziano & Raulin, 2004) were used to test Hypotheses 3 and 4, which state that Gateway Scholarship and non-­‐scholarship students’ pre-­‐course wilderness attitudes would be significantly different from each other, and differences in post-­‐course scores between the two groups would be significantly smaller. Multiple regression (Howell, 2013) was used to test Hypothesis 2, which posited relationships between 22 the predictor variables (ethno-­‐racial identity, previous wilderness experience and urban/rural residency) and pre-­‐course wilderness attitudes. Dummy codes were created for each predictor variable, as all were categorical variables. The baseline category for previous wilderness experience was “zero previous overnight wilderness trips” and was coded “0.” Dummy variables were created for “1 or 2 previous overnight wilderness trips,” “3 to 9 previous overnight wilderness trips” and “10+ overnight wilderness trips.” The Gateway status variable was dichotomous, with Gateway coded as “1” and non-­‐Gateway coded as “0.” The baseline category for community type was “major metropolitan area” and was coded “0.” Dummy variables were created for “suburb of a metropolitan area” and “all other.” Variables were entered into the model with forced entry in SPSS. Table 4. Initial Cronbach’s alpha results, dropped items and final alphas. Construct Pre-­‐course Post-­‐
Item dropped Final pre-­‐
index course course alpha, all index index items alpha, all alpha items Sense of place 0.64 0.49 “…I thought wilderness 0.83 trips would be isolated and lonely” Environmental 0.77 0.66 None 0.77 ethics Value of 0.72 0.49 None 0.72 wilderness Environmental 0.81 0.50 “…I didn’t understand 0.82 awareness ecological processes in wilderness” Course 0.61 “I found the physical experience aspects (i.e. hiking with a heavy pack) of the course easier than I expected” Final post-­‐
course index alpha 0.72 0.66 0.49 0.73 0.82 Interviews were transcribed by a transcription service, Rev.com. Qualitative analysis was conducted using conventional content analysis, with the researchers identifying themes and codes that emerged naturally from the raw interview data, with an emphasis on the initial theoretical wilderness attitude constructs (Hsieh & Shannon, 2005). Reliability was achieved with the use of multiple analysts, the author and another researcher (Patton, 1999). Each began by reading three 23 interviews from beginning to end. They then re-­‐read the interviews while carefully paying attention to and highlighting emergent themes pertaining to the respondents’ conceptualization of wilderness, the impact of the proposed predictor variables, as well as the impact of the course experience, on that conceptualization. Next, the researchers met to go over the themes they identified and compiled the themes and codes agreed upon into a preliminary code tree. This process was repeated twice, with the code tree updated each time and new themes identified and categorized into existing top level codes or new top level codes being created or removed. After a final codebook was established, a final reliability check was done using the method outlined by Kurasaki (2000). Each researcher independently read and coded five interviews using the codebook, and afterwards an inter-­‐reader reliability score, Cohen’s Kappa, was calculated using NVivo. NVivo was also used for the final coding of all interviews. Results Demographic characteristics Descriptive information about ethno-­‐racial identity, previous wilderness experience and community type is presented in Tables 5, 6 and 7. Counts for self-­‐
reported ethno-­‐racial identity show a majority of Gateway students identifying as Hispanic, and a majority of non-­‐Gateway students identifying as White. Previous wilderness experience data show that most students, both Gateway and non-­‐
Gateway, had at least one prior wilderness outing, with 28 of 67 (42%) students having one or two previous wilderness overnight trips. Community types were fairly evenly represented amongst non-­‐Gateway students; however, the majority of Gateway students reported living in major metropolitan areas. The gender of respondents for each group was nearly equally split, with 17 male and 12 female Gateway students (four missing/did not answer), and 20 male and 18 female non-­‐Gateway students (three missing/did not answer). The age range for Gateway students was from 16 to 31 years, with 90% of students who gave their 24 age being 19 years of age or younger (four missing/did not answer). The age range for non-­‐Gateway students was from 16 to 28 years, with 95% of students being 19 years of age or younger (three missing/did not answer). Table 5. Categories and counts of self-­‐reported ethno-­‐racial identities (N = 64) Ethno-­‐racial identity Total Gateway (n = 28) Non-­‐Gateway (n = 36) White or Caucasian 33 1 32 Hispanic or Latino 14 13 1 White and Hispanic 4 1 3 Mexican or Mexican-­‐American 3 3 0 Puerto Rican 2 2 0 Asian, Vietnamese, Filipino, Southeast Asian or 6 6 0 Indian Indigenous Mexican and El Salvadorian 1 1 0 Nigerian 1 1 0 *Missing/did not answer: Gateway (n = 5), Non-­‐Gateway (n = 5) Table 6. Gateway and non-­‐Gateway students’ prior experience with overnight wilderness trips (frequency) Number of overnight wilderness trips Gateway (n = 29)* Non-­‐Gateway (n = 38) Zero 10 7 1 or 2 11 17 3 to 9 5 8 10+ 3 6 * Missing/did not answer: Gateway (n = 4), Non-­‐Gateway (n = 3) Table 7. Number of students from each community type Community Type Gateway (n = 29) Major metropolitan area (more than 50,000 residents) 19 4 Suburb of a metropolitan area 6 All other -­‐ Small city (more than 2,500 residents, less than 50,000 residents) -­‐ Small, rural town (less than 2,500 residents) -­‐ Rural area outside of city limits *Missing/did not answer: Gateway (n = 4), Non-­‐Gateway (n = 4) Non-­‐Gateway (n = 37) 10 12 15 25 Reliability and normality of quantitative variables All wilderness attitude and value constructs had a Cronbach’s alpha of .72 or above. Alphas for each item in the individual indexes can be found in Appendix D. Skewness and kurtosis were also examined for each index. All indexes were within acceptable ranges (+2 to -­‐2) (Cameron, 2004) for both measures of normality, except in three cases: the kurtosis for sense of place post-­‐course was 13.37, 5.22 for environmental ethics post-­‐course, and 3.61 for course experience. This must be kept in mind when interpreting results for these constructs. Inter-­‐rater reliability was established for the qualitative results, with an average kappa of .81 and a range of -­‐0.05 to 0.99. Four coded interviews had a negative kappa for a single construct. When these were reviewed, both coders had applied the same code to parts of the same section of text (in this case, a question, answer and probing questions), but did not code overlapping sections. Therefore, there was no major difference in interpretation of the text, just differences in how much context each coder included in the coding. Final themes and definitions are listed in Table 8. Interview respondents were given codes to reflect Gateway status and gender. For example, the code “G-­‐F-­‐3” means the respondent was a Gateway scholarship student, female, and the third female Gateway student to be interviewed. These codes are used as identifiers for direct quotations. 26 Table 8. Codes and definitions from qualitative analysis. Code (sub codes) Definition Experience in wilderness (Yes, No, The level of wilderness experience a student had Access/proximity) before his or her NOLS course over the summer of 2014. Typically refers to whether or not a student had camped out in the wilderness before. “Access/proximity” refers to mention of access or proximity to wilderness areas, and if that impacts level of wilderness experience. Social influence (Family, Friends, College Used when a student explains where his or her Program, Scholarship program, Previous NOLS interest in taking a NOLS course came from. The course, Interest, No interest) sub-­‐codes pertain to the social influences that the student mentions in this context. The last two sub codes, “interest” and “no interest,” refer to whether the student’s social circles, such as family and friends, had an interest in wilderness before the course. Course content/experience impact (Leave No Used when a student refers to a specific Trace, Designated wilderness, self-­‐
activity/content item during the course that discovery/reflection, general, danger/adversity, impacted his or her perception of wilderness. reliance on self/others, wilderness medicine) Also when a student comments on how a specific or general experience impacted his or her view of wilderness. (for example, learning how to deal with dangerous situations makes the wilderness not as frightening, or knowing how to respond in an emergency with wilderness first aid skills). Personal connection (Self, Used when a student mentions his or her Environment/ecological community, Other development of a deeper personal connection people: On the same course, NOLS alumni, with nature/the environment during or post-­‐
Friends back at home) course. Also includes mentions of how students became closer to other people out in the wilderness, more so than back at home. “Self” includes when a student talks about how wilderness helped him or her “think easier,” and act like himself or herself more easily, etc. Concept of wilderness (Pre-­‐course, Post-­‐course, Used when a student explains what wilderness Structural, Emotional, Concept fortified, meant to him or her before and after the course. Appreciation) “Structural” refers to a focus on the physical aspects (no trails, overgrown, etc.) of the students’ wilderness conceptualization, typically mentioned for pre-­‐course attitudes. “Emotional” is used when the student refers to his or her wilderness conceptualization having an emotional element to it, e.g. “happy”, “scary.” “Concept fortified” is used if the student states that the NOLS course reinforced his or her pre-­‐
course concept of wilderness. Appreciation is used when the student talks about how his or her appreciation/respect for wilderness increased because of the NOLS course. 27 Table 8. Codes and definitions from qualitative analysis (Continued) Post-­‐course social (Family and Friends, Direct Used when the student describes what it was like experience: None, Similar) to talk about the NOLS course to family and friends back at home. “Direct experience: none” is used when the student says that his or her family and/or friends didn’t really understand what the NOLS course was like because his or her family and/or friends had never done something like that before. “Direct experience: similar” is used when family and/or friends understood or related to the students’ NOLS experiences because of similar prior experience outdoors. Personal change (Confidence, Gratefulness, Mentions of how the course impacted students Perseverance) personally, in terms of character traits. Repeat trip (Interest, Constraints) Used when a student expresses interest in doing future wilderness trips. “Constraints” refers to when a student expresses interest in doing more wilderness trips or another NOLS course, but needs to find the time/money/etc. Change in wilderness attitudes Results from paired t-­‐tests show that, on average, all students, both non-­‐
Gateway and Gateway, experienced statistically significant positive change in wilderness attitudes between pre-­‐course and post-­‐course scores. Results for each wilderness attitude construct can be found in Table 9. It is important to note that the survey’s Likert-­‐type scale was coded 1 = Strongly Agree to 7 = Strongly Disagree. Therefore, a lower numerical score translates to a more positive wilderness attitude. No pre-­‐course mean scores for any construct were negative (with four corresponding to neutral on the survey). These results support Hypothesis 1, which stated that the NOLS course would have a positive impact on wilderness attitudes. Students do experience attitudinal change during a NOLS course, and the direction of change is positive. The qualitative results reveal the specific nature of the change and potential reasons behind the change. In general, many students talked about how their NOLS course made them more aware of the negative environmental impacts of human society, how their course made them more interested in environmental issues, and 28 how, more generally, they feel more connected with the broader ecological community. Also related to attitude change were the themes capturing students’ concept of wilderness. When asked how they think of wilderness now, after their NOLS course, students typically replied with a more emotionally attached sentiment. Words such as “refreshing,” “appreciate” and “freedom” were used to describe the wilderness after their courses. Students also conveyed that before their course they perceived wilderness as potentially being “scary” or “isolated,” but learning how to travel and live within wilderness made it a place where they felt comfortable. Specific examples are provided in Table 10. Specific aspects of course content were mentioned in relation to their impact on student wilderness attitudes. One aspect directly related to making students feel more comfortable in the wilderness was wilderness medicine. Several students mentioned the wilderness medicine portions of their courses and the impact of that knowledge on their experience. One student summed up how wilderness medicine training impacted his concept of wilderness when he said, “We did one of the wilderness medicine type of things so I guess that was, in my eyes, sort of how dangerous it can be but how if you respect it, you'll be fine. Sort of like a balance of power between the hiker and the wilderness” (G-­‐M-­‐3).
29 Table 9. Change in attitudes. Descriptive statistics and paired t-­‐test results for pre-­‐ and post-­‐course wilderness attitude and value scores* Non-­‐Gateway Students 95% Confidence Interval Pre-­‐course Post-­‐course of the Difference M SD M SD n t df p Lower Upper Sense of place 3.26 1.14 1.60 0.82 41 7.43 40 <0.0005 1.21 2.12 Environmental ethics 2.18 0.79 1.55 0.69 41 6.76 40 <0.0005 0.44 0.82 Value of wilderness 2.62 0.79 1.90 0.69 41 6.63 40 <0.0005 0.50 0.94 Environmental awareness 2.23 0.90 1.68 0.78 40 5.20 39 <0.0005 0.34 0.76 Gateway Students 95% Confidence Interval Pre-­‐course Post-­‐course of the Difference M SD M SD n t df p Lower Upper Sense of place 3.81 1.59 1.72 0.58 33 7.06 32 <0.0005 1.49 2.69 Environmental ethics 2.90 1.17 1.48 0.47 33 6.76 32 <0.0005 0.99 1.85 Value of wilderness 3.68 0.94 2.28 0.77 33 8.49 32 <0.0005 1.06 1.72 Environmental awareness 2.85 1.19 1.49 0.56 31 5.96 30 <0.0005 0.89 1.82 *Likert-­‐type scale was coded as 1 = Strongly Agree to 7 = Strongly Disagree. Where post-­‐course values are smaller than pre-­‐course scores, this indicates attitudes/values became more positive. 30 Table 10. Examples of pre-­‐ and post-­‐course wilderness attitudes reported in interviews Respondent Pre-­‐course wilderness attitude Post-­‐course wilderness attitude code G-­‐F-­‐1 “Wilderness to me meant dirty, scary, “Somewhere that's where I'm at peace. this wild and intense type of Somewhere where I can just be with environment.” myself…I love it.” G-­‐F-­‐2 “It seems like something scary and “I guess it's changed more because now exciting because before my NOLS wilderness isn't something that ... It's course I wasn't really sure whether I something I know I can learn to live in.” would be able to survive in the wilderness for a month.” G-­‐M-­‐2 “Before it meant the wild, quote “It means reality before technology came. unquote, because I did not know It's usually different from what life is anything that I would actually do in the now.” back country and I just did not even know what it actually was.” NG-­‐F-­‐2 “I would think no roads.” “It just means fresh air and freedom and happiness.” NG-­‐F-­‐4 “Wilderness to me just meant the “It's an area of space that is pretty large outdoors.” that it completely free of any man made materials or man made worlds. Anything that's free of society.” NG-­‐M-­‐2 “There were no houses there. There “I guess wilderness more is somewhere was just an empty, natural land.” that is not only untouched by man, but is more overgrown. There are a lot of places that are untouched by man, but they are seen through the realm of a city. There might be a little park in Miami but that doesn't necessarily mean wilderness.” Leave No Trace (LNT) curriculum was also specifically mentioned as being particularly influential on wilderness attitudes. Learning and executing backcountry travel and living techniques made students more acutely aware of their impact on the wilderness and the importance of lessening that impact. One student succinctly summed this up when she said, “overall just learning all the LNT principles made me more aware of what wilderness was and how much of an impact humans can have on it. I guess just overall being super, super careful not to leave a trace, so it can remain wild” (NG-­‐F-­‐3). Interview respondents also reported greater feelings of responsibility for the environment and knowledge of ethical behavior in the wilderness as a result of their NOLS course. When asked about how the course 31 impacted her view of wilderness, one respondent said, “There was a big focus about preserving it, and not ... We didn't make anything out of trees to live in or something. That was actually pretty much frowned upon. I didn't assume that we were going to make shelter for ourselves, but there was a pretty big sense of not destroying anything, and appreciating it, and leaving it for other people” (NG-­‐F-­‐1). Other students also reported the importance of lessening their impact while traveling in the wilderness. These wilderness ethics were translated to life back at home for some students. Many students made statements similar to a Gateway student who said, “it's [wilderness] not something that I feel like I don't have responsibility to, that I think that people ignore and now I like take action to be more wise about how I preserve water at home…to recycle” (G-­‐F-­‐4). Others reported similar sentiments of realizing their place in the larger ecological community and becoming more aware of their responsibility for the environment. Another course component mentioned frequently was the time allowed by instructors for self-­‐discovery and reflection. This reflection time seemed to result in change by giving students time to think about what had happened on their course, though the specifics of change were not always clear from the interview data. Many students made similar statements as this student, when he said, “One of the things our instructors did in the course was have us do refresher time. We would just wander off to our little own area and have us think. Have us sit down for 2 minutes…we'd be completely like miles into the wilderness, away from anything. It was never easier to have a clear mind or just think or relax” (NG-­‐M-­‐3). Other course content and experiences mentioned as being impactful to how students thought about wilderness were learning about designated wilderness, how to deal with adversity and/or dangerous situations, and working with/relying on other students out in the wilderness. 32 Difference in wilderness attitudes between Gateway and non-­‐Gateway students Results from independent t-­‐tests for pre-­‐ and post-­‐course scores for each wilderness construct, split by Gateway status, are reported in Table 11. These results partially support Hypotheses 3 and 4. For the sense of place construct, non-­‐Gateway and Gateway student scores for both pre-­‐ and post-­‐course were not statistically different from each other. Correspondingly, interview data suggest that both non-­‐Gateway and Gateway students entered their course with an interest in wilderness and usually an idea of what wilderness was, and they had similar emotional reactions to the wilderness during the course. One student captured the view of many respondents when asked what wilderness meant to them after the course; he said, “The same thing, but it means a little more. The level of peacefulness and being away from everything, that comes in to play too. I don't even know how to word that. Just serene as well” (NG-­‐
M-­‐4). Many students, from both groups, reported similar post-­‐course sentiments in relation to their emotional reaction to wilderness and also had a difficult time expressing their thoughts and feelings verbally. Non-­‐Gateway and Gateway students were statistically significantly different from each other for both pre-­‐ and post-­‐course value of wilderness scores, with Gateway students reporting less positive scores in both cases. However, post-­‐course group means show a decrease in the magnitude of the difference. The pre-­‐course difference was approximately one full Likert-­‐scale point (1.06). However, the post-­‐
course difference between group means was 0.39, a reduction of more than half. 33 Table 11. Difference in attitudes. Results of independent t-­‐tests and descriptive statistics for pre-­‐ and post-­‐course wilderness attitude scores. Gateway status 95% Confidence Interval of Non-­‐Gateway Gateway the Difference M SD n M SD n t df p Lower Upper Sense of place, pre-­‐
0.10 -­‐1.2 0.11 course 3.25 1.14 41 3.81 1.61 33 -­‐1.68 56.27* Sense of place, 1.60 0.83 1.72 0.58 -­‐0.76 72 0.45 -­‐0.47 0.21 41 33 post-­‐course 2.18 0.79 2.90 1.17 -­‐3.05 53.83* 0.004 -­‐1.2 -­‐0.25 Environmental ethics, 41 33 pre-­‐course 1.55 0.69 1.48 0.47 0.51 72 0.61 -­‐0.21 0.35 Environmental ethics, 41 33 post-­‐course 2.62 0.79 3.68 0.94 -­‐5.24 72 <0.0005 -­‐1.46 -­‐0.65 Value of wilderness, 41 33 pre-­‐course 1.90 0.69 2.28 0.78 -­‐2.24 72 0.03 -­‐0.72 -­‐0.04 Value of wilderness, 41 33 post-­‐course Environmental 2.23 0.90 2.85 1.17 -­‐2.53 69 0.01 -­‐1.12 -­‐0.13 40 31 awareness, pre-­‐course Environmental 1.68 0.78 1.49 0.56 1.09 69 0.28 -­‐0.15 0.51 40 31 awareness, post-­‐
course 1.46 0.68 1.66 0.76 -­‐1.16 69 0.25 -­‐0.54 0.14 40 31 Course experience *significant Levene’s test for equality of variances 34 The qualitative results point to students assigning more value to wilderness than before their course, though there was some struggle to articulate specifics. Many students reported appreciating the wilderness more, and expressed similar, somewhat vague, sentiments as a student who said, “I would say there is ... I don't know, before, I just ... Like the wild and being outside. I feel like there's more components to it so it's not just being outside, it has components that relate to things in real life” (G-­‐F-­‐3). Another student said, “It's more than just beauty...it means way more to me now than it used to” (G-­‐F-­‐4). Similar statements about wilderness were made by non-­‐Gateway students, such as, “It just means fresh air and freedom and happiness. Sorry if that wasn't very articulate” (NG-­‐F-­‐2), and “It still means the same thing, like how it's completely untouched by people and there's nothing in it that shows you signs of other people and it's all wild, but still I guess it means more that I appreciate it more” (NG-­‐M-­‐3). The qualitative results point to both Gateway and non-­‐Gateway students appreciating wilderness more and assigning more emotional attachment, as displayed by the previous quotes. Correspondingly, the quantitative results indicate that post-­‐course mean differences are smaller than pre-­‐course mean differences between the two groups. However, the interview data do not provide sufficient explanation for why post-­‐course mean scores for value of wilderness remained significantly different between Gateway and non-­‐Gateway students. Non-­‐Gateway students and Gateway students were statistically different from each other for pre-­‐course environmental ethic, with Gateway students reporting less positive attitude scores. The differences between the two groups became non-­‐statistically significant post-­‐course, with the group means approaching similar values. Gateway students experienced a greater magnitude of change, with the difference between pre-­‐ and post-­‐course scores being 1.42, indicating positive change by more than one Likert-­‐scale point. Non-­‐Gateway students also experienced positive change, although smaller, by 0.62 points. For the final construct, environmental awareness, non-­‐Gateway students and Gateway students were significantly different for pre-­‐course scores, with Gateway students reporting less positive attitudes. These differences were non-­‐
35 significant post-­‐course, with both groups experiencing positive change. Gateway students experienced a larger magnitude of change, with the post-­‐course mean score decreasing by more than one full Likert-­‐scale point (1.36). In contrast, non-­‐
Gateway students’ post-­‐course mean score decreased by only half a Likert-­‐scale point (0.55). Interview data for environmental ethics and awareness constructs are similar and intertwined, and corroborate the quantitative results for both constructs. In general, Gateway students had more to say than non-­‐Gateway students, typically in terms of society’s impact on the environment, when asked, “Did your feelings change about wilderness in any way that you did not anticipate or were surprised by?” Many Gateway students reported a realization of the connection between the urban environment and wilderness. When asked if his change in feelings toward wilderness surprised him, one student said, “we build up a whole lot more trash [at home] than we do in the back country. It's just that there's such a crazy difference out there and back home” (G-­‐M-­‐2). Another said, “Just mostly our effect on the wilderness from in the city. You don't expect it [garbage] to be somewhere halfway across the world” (G-­‐M-­‐4). Many, though not all, non-­‐Gateway students responded similarly to one student who said, “I think I went in to the trip hoping that I'd feel a connection with the outdoors after I'd left it, and I'd say that I did. I guess, nothing unpredictable” (NG-­‐F-­‐1). These students did not exhibit fundamental shifts. Quantitatively, all students reported greater feelings of connection with the outdoors, and increased environmental awareness. However, Gateway students experienced a greater magnitude of change, which is exemplified by the previous quotes and quantitative results. Many students reported becoming more aligned with environmental causes as a result of their NOLS experiences. Two students, one non-­‐Gateway and one Gateway, captured this transition from indifferent to invested well, and illustrate the differing magnitude of attitude change between the two groups. Many non-­‐Gateway students reported that, although they were aware of environmental issues and causes before their course, their NOLS experience highlighted the need for personal interest in environmentalism, as reflected by this non-­‐Gateway student: “My [NOLS] 36 instructor told us that day about…I guess just the whole idea about the global movement, I didn't really think about it in high school and it was mostly looked at as some kind of radical thing that didn't really concern me, but now I've changed a lot” (NG-­‐M-­‐2). Gateway students reported similar sentiments, but some reported heightened environmentalism along with what could be seen as almost a change in identity after their course. One student discussed, in context of how her view of wilderness changed, “those kind of people like tree huggers or granolas, and I wouldn't consider myself that, but after spending one month in Alaska I saw how, why they're so passionate about this. I didn’t expect myself to feel this same, as passionate as them” (G-­‐F-­‐4). This statement gives some insight as to why Gateway students reported greater change, in that many reported feeling surprised by the nature and extent of that change. Further examples of the magnitude of the shift can be found in Table 10. Gateway students were more likely in the interviews to report that they didn’t really know what to expect from a wilderness trip, or they perceived wilderness as “scary.” However, post-­‐course quotes show that Gateway and non-­‐
Gateway students both described wilderness in similar ways, whether describing it as a peaceful, happy place or as simply a place that is “untouched by man.” In general, Gateway students experienced a larger attitude change, and non-­‐
Gateway students experienced a “ceiling effect”. One non-­‐Gateway student in particular exemplifies this difference in magnitude. When asked if his family or friends were interested in wilderness, he said, “my grandmother lives in California and she has 600 acres in Napa Valley, so I spent almost every Christmas there so I've always been exposed to the wilderness and that's always been positive” (NG-­‐M-­‐1). Later in the interview, when asked how his course during the summer of 2014 impacted his idea of wilderness, the same participant said, “I think my idea of wilderness had already been established due to that previous NOLS course.” This student had entered his NOLS course with a fairly established, positive wilderness attitude that stemmed from his previous wilderness experience, leaving little room for further positive change. However, students who had limited or no backcountry experience prior to their summer 2014 NOLS course reported much more substantial change. A student 37 who had never slept in a tent before, when talking about going backpacking with friends after his course, said “that's something that honestly I would have never assumed I would do before NOLS. I wouldn't even have wanted to do it and now I look for it” (NG-­‐M-­‐2). Similarly, another student with limited wilderness experience (one previous backpacking trip) said, “I learned that wilderness is something that can be refreshing and something that isn't as scary as you can learn how to manage yourself and manage it” (G-­‐F-­‐2). These students, along with the previously quoted student who had a strong positive pre-­‐course attitude, display similar feelings and attitudes toward wilderness after the NOLS course, despite having different pre-­‐
course attitudes. Their statements demonstrate why the variation (as measured by standard deviation) in attitude scores, especially among Gateway students, who as a group typically had less previous wilderness experience, was much smaller for post-­‐
course scores, and overall attitude mean scores for both groups approached similar values post-­‐course. Course experience (i.e., feeling a part of the group and that opinions and views were valued) scores were not statistically significant different between non-­‐
Gateway and Gateway students. These results, except for sense of place, support Hypotheses 3 and 4, which state that Gateway students were more likely to have less positive pre-­‐course wilderness attitudes than non-­‐Gateway students, and differences in wilderness attitudes between non-­‐Gateway and Gateway students would be smaller post-­‐
course. Personal and cultural factors related to wilderness attitudes The multiple regression results in Table 12 show the relationship between the predictor variables (previous wilderness experience, community type and Gateway status) and pre-­‐ and post-­‐course wilderness attitude constructs. Initially, ethno-­‐racial identity, not Gateway status, was proposed as a predictor variable. However, self-­‐reported ethno-­‐racial identities were confounded with Gateway status; e.g., White ethno-­‐racial identity was reported by 97% of non-­‐Gateway 38 students and 7% of Gateway students. Due to this confounding factor and small sample sizes for the various ethno-­‐racial identities, the multiple regression analysis used Gateway status as a predictor variable instead of ethno-­‐racial identity. Note that, while 74 respondents completed the survey, ten of these respondents did not report an ethno-­‐racial identity. However, since NOLS provided data on the Gateway status of students, all but eight students were included in the regression models. The eight students not included in the regression models had missing data for one or two predictor variables (community type and/or previous wilderness experience). Pre-­‐course adjusted 𝑅! s ranged from 0.15 to 0.35, with the regression models explaining 15% of the variation in environmental awareness scores, 20% of the variation in environmental ethic scores, 23% of the variation in sense of place scores, and 35% of the variation in value of wilderness scores. Post-­‐course adjusted 𝑅! s were not significantly different from zero. When considered in light of the changes in standard deviation from pre-­‐ to post-­‐course scores found in Table 11, the lack of explained variance in post-­‐course models may in part be due to there being less variance in scores to capture. These results corroborate the independent t-­‐test results, with the differences between Gateway and non-­‐Gateway pre-­‐course scores being statistically significant (except for sense of place), and post-­‐course score means becoming smaller and not statistically different (except for value of wilderness). In general, most pre-­‐course β values for previous wilderness experience for all constructs were negative, and statistically different from zero at the ten or more previous overnight wilderness trips category. This translates to a positive relationship between previous wilderness experience and wilderness attitudes, because smaller values on the attitude scales indicate more positive wilderness attitudes. For example, for pre-­‐course sense of place scores, the dummy variable for 10 or more prior overnight wilderness trips had a β of -­‐.047. So, a change from zero previous overnight wilderness trips to ten or more resulted in a decrease in approximately half of a Likert-­‐scale point (0.47) for the sense of place construct. Post-­‐course β values for all previous wilderness experience categories were not 39 significantly different from zero for all wilderness attitude constructs, indicating previous wilderness experience did not predict post-­‐course wilderness attitude scores. Community type was not a significant contributor to the model for any construct, for either pre-­‐ or post-­‐course scores. This is reflected in the small and not statistically significant β values for each construct, ranging from -­‐0.14 to 0.13 for pre-­‐course scores and -­‐0.09 to 0.19 for post-­‐course scores. As expected from the t-­‐test results, Gateway status was a statistically significant predictor variable in the pre-­‐course models for environmental ethics, value of wilderness and environmental awareness. Similar to the independent samples t-­‐test results, this significance disappeared post-­‐course for environmental ethics and environmental awareness, but remained significant for value of wilderness. However, the magnitude of the significance lessened, moving from β = 0.52 pre-­‐course to β = 0.33 post-­‐course. Non-­‐Gateway was coded “0” and Gateway was coded “1”, so a positive β translates to a less positive wilderness attitude score in the shift from non-­‐Gateway to Gateway status. Although post-­‐course models remained statistically significant, the smaller β values indicate a weaker post-­‐course relationship between Gateway status and wilderness attitudes. Pre-­‐course models including the three variables above significantly improve the predictability of wilderness attitudes over predicting using means alone for each attitude construct. This significance disappears in post-­‐course models, with the predictor variables not significantly contributing to predicted wilderness attitudes, with the exception of Gateway status for value of wilderness. The above results provide partial support for Hypotheses 2, 3 and 4, which stated that ethno-­‐racial identity, previous wilderness experience and community type would impact pre-­‐course wilderness attitudes, Gateway students would have less positive pre-­‐course wilderness attitudes, and differences in attitudes between Gateway and non-­‐Gateway students would be significantly smaller post-­‐course. 40 Table 12. Personal and cultural factors as predictors of wilderness attitudes. Summary of Multiple Regression Analysis for Predictor Variables (N = 74) Variable Sense of place, pre-­‐course Zero prev. wilderness exp. vs. 1 or 2 Zero prev. wilderness exp. vs. 3 to 9 Zero prev. wilderness exp. vs. 10+ Gateway status Major metro. vs. suburb Major metro. vs. all other Sense of place, post-­‐course Zero prev. wilderness exp. vs. 1 or 2 Zero prev. wilderness exp. vs. 3 to 9 Zero prev. wilderness exp. vs. 10+ Gateway status Major metro. vs. suburb Major metro. vs. all other Environmental ethics, pre-­‐course Zero prev. wilderness exp. vs. 1 or 2 Zero prev. wilderness exp. vs. 3 to 9 Zero prev. wilderness exp. vs. 10+ Gateway status Major metro. vs. suburb Major metro. vs. all other Environmental ethics, post-­‐course Zero prev. wilderness exp. vs. 1 or 2 Zero prev. wilderness exp. vs. 3 to 9 Zero prev. wilderness exp. vs. 10+ Gateway status Major metro. vs. suburb Major metro. vs. all other t p -­‐1.69 -­‐3.61 -­‐3.76 0.71 -­‐0.88 -­‐0.92 0.52 0.41 -­‐0.66 1.31 -­‐0.39 -­‐0.57 0.49 -­‐0.56 -­‐2.20 3.48 0.13 0.59 -­‐034 1.09 -­‐0.65 0.59 1.28 0.92 0.10 0.001 0.0001 0.48 0.38 0.36 0.60 0.68 0.51 0.20 0.70 0.57 0.62 0.58 0.03 0.001 0.90 0.56 0.74 0.28 0.52 0.56 0.21 0.36 df adj. 𝑅 ! β F p -­‐0.23 -­‐0.48 -­‐0.47 0.09 -­‐0.11 -­‐0.12 0.08 0.06 -­‐0.10 0.18 -­‐0.06 -­‐0.08 0.07 -­‐0.08 -­‐0.28 0.42 0.02 0.08 -­‐0.05 0.17 -­‐0.09 0.08 0.19 0.13 4.30 6, 59 0.001 0.23 0.75 3.68 0.73 6, 59 0.62 -­‐0.02 0.20 6, 59 6, 59 0.004 0.63 -­‐0.03 41 Table 12. Personal and cultural factor as predictors of wilderness attitudes. (Continued) Variable t p β Value of wilderness, pre-­‐course Zero prev. wilderness exp. vs. 1 or 2 -­‐0.09 0.93 -­‐0.01 Zero prev. wilderness exp. vs. 3 to 9 -­‐1.48 0.15 -­‐0.18 Zero prev. wilderness exp. vs. 10+ -­‐2.04 0.05 -­‐0.24 Gateway status 4.27 0.001 0.52 Major metro. vs. suburb -­‐1.39 0.17 -­‐0.16 Major metro. vs. all other Value of wilderness, post-­‐course Zero prev. wilderness exp. vs. 1 or 2 Zero prev. wilderness exp. vs. 3 to 9 Zero prev. wilderness exp. vs. 10+ Gateway status Major metro. vs. suburb Major metro. vs. all other Environmental awareness, pre-­‐course Zero prev. wilderness exp. vs. 1 or 2 Zero prev. wilderness exp. vs. 3 to 9 Zero prev. wilderness exp. vs. 10+ Gateway status Major metro. vs. suburb Major metro. vs. all other Environmental awareness, post-­‐course Zero prev. wilderness exp. vs. 1 or 2 Zero prev. wilderness exp. vs. 3 to 9 Zero prev. wilderness exp. vs. 10+ Gateway status Major metro. vs. suburb Major metro. vs. all other 0.27 0.57 1.63 0.05 2.49 -­‐0.61 -­‐0.17 0.58 -­‐2.09 -­‐0.26 2.48 -­‐1.02 0.61 0.77 0.46 0.63 -­‐0.44 -­‐0.12 0.82 0.79 0.57 0.12 0.96 0.02 0.55 0.87 0.56 0.04 0.80 0.02 0.31 0.55 0.45 0.65 0.53 0.67 0.92 0.42 0.03 0.09 0.24 0.01 0.33 -­‐0.09 -­‐0.02 0.08 -­‐0.29 -­‐0.03 0.31 -­‐0.14 0.08 0.12 0.07 0.09 -­‐0.06 -­‐0.02 0.12 F 6.79 1.90 2.88 0.37 df 6, 59 adj. 𝑅 ! 0.35 p 0.0001 6, 59 0.10 0.08 6, 59 0.02 6, 59 0.90 0.15 -­‐0.06 42 Discussion Overall, the results show that NOLS students experienced positive change in their wilderness attitudes during their course, with Gateway scholarship students experiencing a larger change. The differences between Gateway and non-­‐Gateway students virtually disappeared in post-­‐course scores. The data show that, while Gateway scholarship students came to NOLS with less positive wilderness attitudes than non-­‐Gateway students, both groups left NOLS with similar attitudes and values. Considering these results, there is evidence to answer Research Questions 1 and 2: “Do students’ attitudes change over the course of a NOLS experience? Why or why not?” and “Do Gateway scholarship students and non-­‐scholarship students have different wilderness attitudes before a NOLS course? Do their attitudes differ after a NOLS course?” The answer is almost unequivocally yes, Gateway and non-­‐Gateway students enter NOLS with different wilderness attitudes, both groups experience attitudinal change during their NOLS course, and both come out of their experience with similar wilderness attitudes, although some differences remained. These results echo broader theoretical concepts in attitude research, in that direct experience is one of the most effective vectors of attitudinal change (Heberlein, 2012). In this case, the NOLS course was a positive (on average), direct experience in and of the wilderness, with the specific intent of fostering pro-­‐wilderness attitudes, and was seemingly successful in that pursuit. Additionally, previous research has found that outdoor education is effective in fostering pro-­‐environmental attitudes (Ewert & McAvoy, 2000; Mittelstaedt, Sanker, & VanderVeer, 1999). According to a study investigating the impact of wilderness-­‐based education on environmental ethics in youth, a NOLS course qualifies as a significant life experience and can have a substantial impact on an individual’s environmental ethic (Waage et al., 2013). This study’s results are consistent with this conclusion. The interview data indicated that the NOLS course was an immersive, life -­‐-­‐ and attitude -­‐-­‐ changing experience. Many participants relayed similar sentiments as a student who said, “I didn't expect to have the experience I had and I didn't think that I would look at the way, wilderness I do 43 now…I appreciate a lot more” (NG-­‐M-­‐3). Considering the intensity of a NOLS course, the focus on cultivating pro-­‐environmental attitudes and behavior, and the fact that students voluntarily take NOLS courses, it is not surprising that, on average, both groups of students experienced positive attitude change. The specific reasons that lead to an environmental ethic still remain somewhat unknown, though research has revealed some possibilities. Liefländer, Fröhlich, Bogner, and Schultz (2012) suggest that the length of time spent in nature, the “wildness” of the setting, and the frequency of interactions with nature all influence the strength of connectedness to nature. A typical NOLS course is an extended stay in a wilderness setting, characteristics that would strongly impact connectedness to nature. Correspondingly, interview data pointed to an increased personal connection to nature. Other research has revealed specific course content to be particularly impactful on the formation of an environmental ethic. For instance, Waage et al. (2013) identified Leave No Trace (LNT) and an increase in cognitive awareness of environmental issues as two themes impacting students’ environmental ethics as a result of a NOLS course. My interview data are consistent with these themes, as LNT principles, learning about designated wilderness, and training in wilderness medicine all contributed to respondents’ positive change in wilderness attitudes. The reasons behind the larger change in attitudes experienced by Gateway students are not explicitly clear from the interview data. However, as predicted, non-­‐Gateway students did experience a “ceiling effect,” in that they entered their NOLS courses with strong pre-­‐existing positive wilderness attitudes that did not leave room for large additional change. On average, Gateway students entered their NOLS courses with less positive attitudes, which allowed for a larger increase in positive change. As explained above, NOLS is a significant life experience, and even more so for some students who may have not had similar experiences previously. The answer to the final research question, “Are the differences in wilderness attitudes explained by personal and cultural factors (i.e., previous wilderness experience, ethno-­‐racial group or urban/rural residency)?” is less straightforward. The data reveal the complexity of attitudes and their origins, with both previous 44 wilderness experience and Gateway status being sometimes significant predictors of wilderness attitudes, and community type being non-­‐significant for any of the pre-­‐ or post-­‐course scores. Post-­‐course, Gateway status was the only variable that remained significant, and only for the value of wilderness construct. Previous wilderness experience was hypothesized to have a positive relationship to wilderness attitudes, as previous positive experience with a place has been shown to contribute to more positive attitudes and meanings associated with it (Kyle & Chick, 2007). Additionally, positive wilderness experiences with organized groups have been shown to result in positive attitudinal change (Ewert & McAvoy, 2000). The quantitative results give support to this hypothesis, as prior experience was a significant predictor of pre-­‐course wilderness attitudes for all constructs. Previous wilderness experience was not a significant predictor of post-­‐
course scores, which is most likely because the NOLS course served as a baseline wilderness experience, reducing the experience differences between students. The interview data offer further support for these findings, with most students reporting similar attitudes and values during post-­‐course interviews. The wilderness experience gap that was prevalent in pre-­‐course scores was overridden by the shared intense experience of a NOLS course, with less experienced students becoming similar to more experienced students in wilderness attitudes and values. As previously stated, Gateway status and self-­‐reported ethno-­‐racial identity were highly confounded, with almost all Gateway students identifying as non-­‐White, mainly Hispanic. This is not surprising, considering that to be eligible for a Gateway scholarship, a potential student must self-­‐identify as a person of color, as well as meet certain financial need requirements for a full-­‐tuition scholarship (A. Rajagopal-­‐
Durbin, personal communication, February 4, 2015). With all but one non-­‐Gateway student in the sample self-­‐identifying as White, the non-­‐Gateway student study sample fits the historically “typical” demographic of OEE participants in being White and upper-­‐middle class (Holman et al., 2008). Ethno-­‐racial identity, as opposed to Gateway status, was originally chosen as a predictor variable to address for the possibility that the non-­‐Gateway sample might be composed of students from 45 diverse ethno-­‐racial backgrounds. However, this was not the case, and Gateway status was used in data analysis instead. The inconsistency of Gateway status as a predictor variable reflects the wider literature on the relationship between ethno-­‐racial identity and the meanings and values assigned to wilderness and wildlands. Johnson, Bowker, Bergstrom and Cordell (2004) had similar findings in their study of immigrant status, ethnicity and wilderness values in America. Whereas differences in visitation and type of use were found, there were some shared wilderness values between the various race and ethnic groups. Similarly, African Americans were shown to hold similar environmental values as Whites, but differ on environmental behaviors in a study by Parker and McDonough (1999). When factors such as income, age, gender and place of residence were held constant, Johnson et al. (2001) found that race was not a significant factor in outdoor recreation constraints. In context of those findings and this study's results, it is not surprising that Gateway status was an inconsistent attitude predictor. Adding to the complexity of analysis is the use of Gateway status as a single proxy variable for complex differences among students. Gateway status combines ethno-­‐racial identity and income into one variable, which confounds the independent relationship of each factor to wilderness attitudes. Moreover, simply using ethno-­‐racial identity and/or income as predictors for wilderness attitudes or environmental behavior may not truly reveal the reasons for certain attitudes, and at worst, can paint ethno-­‐racial groups as a single, monolithic entity, erasing intra-­‐
group complexities (Omi & Winant, 2015). Additionally, when drawing conclusions based on Gateway status, it is important to note the small sample size and grouping of many ethno-­‐racial identities (see Table 5) into one variable before painting broader generalizations on ethno-­‐racial identity. Inferences from this study’s results should be made solely in the context of the Gateway Scholarship Program, or very similar programs, as the variables used in this study are specific to the scholarship program and NOLS. The final predictor variable was community type, which was not a statistically significant predictor in any case. This variable was attempting to 46 measure if proximity or access to wilderness was related to wilderness attitudes. I reasoned that more wilderness visitation would lead to more positive attitudes, and that those studetsn who lived closer to wildlands (i.e., rural areas) would have more direct experience. These assumptions are somewhat inconsistent with results of other research related to outdoor recreation and urban/rural residency. In one of the few studies to focus on rural populations, Johnson et al. (1997) found that rural Blacks visit wildland areas less than rural Whites, with visitation related to the more negative impressions of these areas. However, when the wildland meaning variable was included in analysis, the direct impact of race on visitation was significantly reduced. In another study about wilderness values, Johnson et al. (2004) found that while direct use values of wilderness areas differed significantly between some racial and ethnic groups, with Black respondents more often reporting lower direct use (essentially visitation) values, off-­‐site values (values not contingent on visitation) were not considerably different between groups. I am reporting this previous research here to demonstrate that wilderness visitation, which may be related to proximity of residence, does not necessarily indicate whether an individual holds positive atittudes and values about wilderness or environmental issues. As my results and previous research indicate, rural or urban residence does not necessarily determine wilderness attitudes or values. In fact, this study concluded that community type was unrelated to wilderness attitudes. The qualitative data suggested that the ability to travel to wilderness (e.g., visiting family who live near wilderness) may be a more reliable indicator of wilderness attitudes and values. Thus, a more direct question about access to wilderness, whether by geographical proximity or proximity provided by the ability to travel, without using community type as a proxy, might be a more reliable predictor of wilderness attitudes. 47 Implications The stark difference in the self-­‐reported ethno-­‐racial identities of Gateway and non-­‐Gateway students is evidence of the need for the Gateway Scholarship program at NOLS. It is clear that without the program, NOLS courses could conceivably be composed of mostly upper class Whites. This study’s results also suggest that if NOLS wants to continue to impart attitude change in students, the most effective and impactful change is established with the student’s first experience with NOLS. In this way, the Gateway Scholarship program is very important, as it brings in new students from diverse backgrounds who have limited wilderness experience. However, as demonstrated by the qualitative data, non-­‐
Gateway students also positively benefit from NOLS being their first wilderness experience. While there should be no dispute over including a larger diversity of people into NOLS courses and programming, limitations of techniques such as the Gateway Scholarship program must be recognized. The Gateway Scholarship program heavily relies on the assumptions supported by the marginality theory in leisure research. Marginality theory postulates that differences in socio-­‐economic status, mainly income, are the reason for lower participation rates among minorities (Floyd, 1998). There is a good rationale for the use of income as a selection category for increasing ethno-­‐racial diversity on NOLS courses; income is a measurable variable that can be applied objectively to select individuals for scholarships. However, the assumptions of this selection tool should be considered when assessing the effectiveness and inclusiveness of diversity initiatives. When not used carefully, this approach assumes that if minorities were to achieve the socio-­‐economic status of Whites, they would adopt similar leisure preferences and activities, making marginality-­‐based diversity initiatives a potentially ethno-­‐centric approach (Floyd, 1998). This assumption promotes the idea of social stratification, in that individuals with similar incomes will have similar “life chances” and that proceeding up the social hierarchy would result in similar positions (Omi & Winant, 2015). 48 I am not implying that people of minority ethno-­‐racial backgrounds should not be included on NOLS courses and the Gateway Scholarship program be eliminated. Instead, I am cautioning against branding the Gateway Scholarship program as a single solution to diversity issues with the NOLS student population. If the goal is to create an inclusive organization that reflects diverse perspectives on wilderness, then students of all backgrounds should be able to see themselves within the context of a NOLS course. As a leader in wilderness education, NOLS could go beyond simply bringing diverse students into the wilderness and proactively play a influential role in disrupting the White-­‐dominated racialized outdoor leisure identity (Martin, 2004) by continuing the Gateway Scholarship program, supporting Gateway alumni in employment in the wilderness education field, and organizing more expeditions led by people of color, such as Expedition Denali. This should not be difficult, because results from this study suggest that NOLS is already creating an environment of inclusivity, as students’ evaluations of their courses, as measured in course experience scores, were not different between non-­‐Gateway and Gateway students. Moreover, the qualitative data also suggest that students felt that they belonged in wilderness, as discussed further below. It is unclear, however, whether this stemmed from official curriculum content, or certain instructors inserting content pertaining to inclusivity and diversity. Assimilation as the means of addressing “underrepresentation” of minorities in outdoor recreation can be dangerous if not supplemented with social justice oriented staff training and curriculum programming to promote inclusion and diversity of thought. The assimilation trap can re-­‐center whiteness, making it the obvious and final conclusion, as opposed to resisting the patterns of whiteness reflected in outdoor recreation (McLean, 2013; Rose & Paisley, 2012). NOLS should keep this in mind when designing diversity initiatives and course curricula. While NOLS is an organization centered around wilderness, the curriculum should also recognize that outdoor recreation is not the only way of expressing value or connection to wilderness (Martin, 2004), and that there are many ways to experience wilderness beyond the skills that NOLS teaches. In designing diversity initiatives, NOLS (and other similar organizations) should ask: Is the goal of a 49 diversity initiative to add more individuals of diverse backgrounds into the same equation, or to diversify that equation with the inclusion of different perspectives? In other words, NOLS, and the broader OEE community, should be cautious of branding the result of Gateway students becoming more like non-­‐Gateway students as a definitively positive outcome. Every interviewee reported wanting to take future wilderness trips, but some brought up issues of time, proximity or access, and/or financial constraints. For example, when asked if he wanted to do another overnight wilderness trip, one student said, “I can't go that easily and just do the same thing as we did in NOLS, I guess” (G-­‐M-­‐1), and another brought up the expense of a NOLS course, saying, “I would totally do it. I gotta find the money first, but I would do it” (NG-­‐F-­‐4). It appears that, whereas the value of wilderness was effectively conveyed during their NOLS course, the value of “less pristine” non-­‐wilderness areas may have not been highlighted. Students should not feel that the only worthwhile venture into nature is a venture into a distant wilderness (Martin, 2004), or fail to realize that other areas that may be less pristine can be just as important to the overall ecological functioning of the planet. NOLS has done a good job of teaching that wilderness is inclusive of people, as one interviewee eloquently stated when talking about how her instructor brought tools to make ice sculptures: “There is always something out there for someone. For example, those glaciers and when he brought some [tools] and teaching us about them. I just felt that it was really nice to know that there's something out there for me, too” (G-­‐F-­‐2). However, the idea that people must be inclusive of more than just wilderness in their ecological and recreational outlook should be considered a valuable addition the NOLS curriculum. The value of wilderness areas in recreational and ecological terms was effectively conveyed, but the idea that living, ecological systems outside of wilderness, and closer to home, are also vital to human and environmental well-­‐being is also an equally important message if the goal is to foster a wildland ethic applicable to daily life. 50 Limitations The small sample size of this study (N = 74) and use of only one OEE organization (NOLS) must be kept in mind when generalizing these results to the broader OEE field. The use of email and an online survey instrument may have impacted who responded, as these formats assume that all respondents have reliable and consistent access to a computer and the internet. The smaller response rate for Gateway students could reflect a problem with this assumption, considering that as a population, Gateway students are less likely to have these sorts of resources at their disposal. The overall response rate (46%) was standard for social science research (Baruch, 1999). However, a non-­‐response bias check was not conducted and it is possible that systematic differences existed between respondents and non-­‐respondents. Specifically, it is possible that the majority of respondents had a positive NOLS experience, and if so, potentially less positive wilderness attitudes and values held by the non-­‐respondents would not have been captured. The sampling for the interviews (N = 19) was limited to only those students who agreed to be interviewed. The small number of such students did not allow for the maximum variation approach that was initially planned in the research proposal. Instead, all students who agreed to an interview were contacted, in an effort to achieve a larger interview sample size. Another issue with what was essentially a convenience sample for interviews is that only students who experienced positive attitudinal change during their NOLS course were interviewed. While almost all students reported more positive wilderness attitude scores after their course, a few experienced negative change or no change at all. Not surprisingly, these students did not volunteer to be interviewed. Therefore, the interview data do not reflect these more negative or neutral attitudes and experiences. Finally, there is always the possibility of students reporting “socially acceptable” answers, i.e. saying what they thought I wanted to hear, and reiterating 51 the NOLS conceptualization of wilderness because the interview was about the NOLS course. Although I made efforts to reduce the likelihood of this, the possibility remains. Future Research As discussed above, this study provides evidence that participation in a NOLS course does impact wilderness attitudes and values, and on average, students exit their course with similar positive attitudes and values. As NOLS continues to pursue becoming an organization that reflects diverse perspectives, it remains unclear exactly how this goal can be accomplished in terms of curriculum and course content. Future research could help illuminate way to modify NOLS’s, and other OEE organizations’, curriculum to include a variety of perspectives on wilderness, how to live and travel within it, and how to transfer wilderness attitudes and values to daily life. The literature on social justice and OEE, along with this study, have critically analyzed how certain ideas of wilderness convey a particular cultural viewpoint of nature, but have fallen short of offering practical examples of how to OEE organizations can diversify course content to reflect different cultural and social perspectives. This study also highlighed the limitations of attemping to identify predictors of wilderness attitudes and values, which continue to elude leisure researchers. Other variables to consider could stem from value formation, such as significant social influences’ (i.e., family and friends) interest in wilderness, access (though not necessarily proximity) to wilderness, and individual leisure identity (e.g., “I didn’t see myself as a treehugger”). Conclusion It is clear that a NOLS course positively impacts students’ attitudes about wilderness and broader environmental awareness. While Gateway students entered NOLS courses with less positive wilderness attitudes than non-­‐scholarship students, 52 both groups of students exited NOLS with similar attitudes. The three factors pertaining to Gateway Scholarship selection criteria -­‐-­‐ ethno-­‐racial identity, previous wilderness experience and urban/rural residency -­‐-­‐ proved to be inconsistent and weak predictors of pre-­‐course wilderness attitudes, and entirely non-­‐significant post-­‐course, apart from a single case. These positive post-­‐course attitudes persisted beyond the immediate time period surrounding the course, as the survey and interviews were conducted between two and six months after students returned home. This result echoes previous research in which NOLS alumni reported that knowledge, skills and personal changes (such as increased perseverance) gained during their course were still relevant in their daily lives, even years later (Sibthorp, Furman, Paisley, & Gookin, 2008). This is encouraging, considering that NOLS strives to foster a lifelong pro-­‐environmental ethic in students. However, as Waage et al. (2013) remind us, reaction to these results must be tempered with the possibility that individuals attracted to NOLS courses are self-­‐selecting, in that their prior life experiences have primed them to be especially responsive to the atmosphere of attitude change during a NOLS experience. In terms of this study, it appears that, although students came from a variety of backgrounds, a common interest in wilderness and the environment opened a shared avenue of attitude change, leading students to similar destinations in their attitudes about wilderness. However, if diversity and inclusivity is the goal, diverse ideas about and approaches to wilderness must be cultivated, just as genetic diversity in biota is vital to the future of wilderness. By creating a more diverse discourse surrounding wilderness, more people from a variety of backgrounds will be able to make a personal connection with wilderness and the outdoors. This is increasingly important, as the future of wilderness in America relies on future generations investing their energies and passions in its existence. 53 References Agyeman, J. (2003). Under-­‐participation and ethnocentrism in environmental education research: Developing culturally sensitive research approaches. Canadian Journal of Environmental Education, 8, 80-­‐94. Ajzen, I. (2012). Theory of planned behavior. In P. A. M. Lange, A. W. Kruglanski, & E. T. Higgins (Eds.), Handbook of theories of social psychology (Vol. 1, pp. 438–
459). London, U.K.: Sage. Baruch, Y. (1999). Response rate in academic studies -­‐ A comparative analysis. Human Relations, 52(4), 421–438. Buijs, A. E., Elands, B. H. M., & Langers, F. (2009). No wilderness for immigrants: Cultural differences in images of nature and landscape preferences. Landscape and Urban Planning, 91(3), 113–123. Byrne, J. (2012). When green is White: The cultural politics of race, nature and social exclusion in a Los Angeles urban national park. Geoforum, 43(3), 595–611. Byrne, J., & Wolch, J. (2009). Nature, race, and parks: Past research and future directions for geographic research. Progress in Human Geography, 33(6), 743–
765. Cameron, A. C. (2004). Kurtosis. In M.S. Lewis-­‐Beck, A. Bryman, & T. F. Liao, The SAGE encyclopedia of social science research methods. (Vols. 1-­‐3). Thousand Oaks, CA: SAGE Publications, Inc. Retrieved from http://srmo.sagepub.com/view/the-­‐sage-­‐encyclopedia-­‐of-­‐social-­‐science-­‐
research-­‐methods/n467.xml Creswell, J. W. (2003). Research design: Qualitative, quantitative and mixed methods approaches (2nd ed.). Thousand Oaks, CA: Sage. Cronon, William. (1996). The trouble with wilderness; or, getting back to the wrong nature. In W. Cronon (Ed.), Uncommon ground: Rethinking the human place in nature (pp. 69-­‐90). New York and London: W. W. Norton & Company. DeLuca, K., & Demo, A. (2001). Imagining nature and erasing class and race: Carleton Watkins, John Muir, and the construction of wilderness. Environmental History, 4, 541-­‐560. Dickson, S., & Hall, T. E. (2006). An examination of whitewater boaters’ expectations: Are pre-­‐trip and post-­‐trip measures consistent? Leisure Sciences, 28(1), 1–16. Eagles, P. F. J., & Demare, R. (1999). Factors influencing children’s environmental attitudes. Journal of Environmental Education, 30(4), 33–38. 54 Ewert, A., & McAvoy, L. (2000). The effects of wilderness settings on organized groups: A state-­‐of-­‐knowledge paper. USDA Forest Service Proceedings RMRS-­‐P-­‐
15, 3, 13–26. Faul, F., Erdfelder, E., Buchner, A., & Lang, A.-­‐G. (2009). Statistical power analyses using G*Power 3.1: Tests for correlation and regression analyses. Behavior Research Methods, 41(4), 1149–1160. Field, A. (2013). Discovering statistics using IBM SPSS Statistics (4th ed.). Thousand Oaks, CA: Sage. Floyd, M. (1998). Getting beyond marginality and ethnicity: The challenge for race and ethnic studies in leisure research. Journal of Leisure Research, 30(1), 3–22. Floyd, M. (1999). Race, ethnicity and use of the National Park system. Social Science Research Review, 1(2), 1-­‐24. Floyd, M. F., & Johnson, C. Y. (2002). Coming to terms with environmental justice in outdoor recreation: A conceptual discussion with research implications. Leisure Sciences, 24, 59-­‐77. Gookin, J. (2002a). Defining environmental education: NOLS philosophy of education. In J. Gookin and D. Wells (Eds.), NOLS Environmental Education Notebook. Lander, WY: The National Outdoor Leadership School. Gookin, J. (2002b). Learning ethics: What the research says. NOLS Environmental Education Notebook, eds. J. Gookin and D. Wells. Lander, WY: The National Outdoor Leadership School. Gookin, J., & Leach, S. (Eds.). (2008). 2008 NOLS leadership educator notebook: A toolbox for leadership educators. Lander, WY. Retrieved from http://www.nols.edu/enrolled/admissionforms/instructor_course_curriculum
/2009LENCh1-­‐2.pdf Graziano, A. M., & Raulin, M. L. (2004). Research methods: A process of inquiry (5th ed.). Boston, MA: Pearson Education Group, Inc. Harris, S. (2002). Research in environmental education: What works? In J. Gookin and D. Wells (Eds.), NOLS Environmental Education Notebook. Lander, WY: The National Outdoor Leadership School. Heberlein, T. A. (2012). Navigating environmental attitudes. New York: Oxford University Press. 55 Holman, T., Bobilya, A. J., & McAvoy, L. (2008). Trends and issues in outdoor adventure programming : Perspectives from practitioners and scholars. Taproot, Spring/Summer, 17–24. Howell, D. C. (2013). Statistical methods for psychology (8th ed.). Belmont, CA: Wadsworth. Hsieh, H. F., & Shannon, S. E. (2005). Three approaches to qualitative content analysis. Qualitative Realth Research, 15(9), 1277–1288. Hutson, G., & Bailey, L. (2008). Awakening place awareness during a 30-­‐day wilderness leadership program. International Journal of Wilderness, 14(3), 23-­‐
28. Johnson, M. (2002). NOLS expectations for environmental studies. In J. Gookin and D. Wells (Eds.), NOLS Environmental Education Notebook. Lander, WY: The National Outdoor Leadership School. Johnson, C. Y., Bowker, J. M., Bergstrom, J. C., & Cordell, K. H. (2004). Wilderness values in America: Does immigrant status or ethnicity matter? Society & Natural Resources, 17(7), 611–628. Johnson, C. Y., Bowker, J. M., & Cordell, H. K. (2001). Outdoor recreation constraints: An examination of race, gender, and rural dwelling. Southern Rural Sociology, 17, 111–133. Johnson, C. Y., Bowker, J. M., & Cordell, K. H. (2004). Ethnic variation in environmental belief and behavior: An examination of the New Ecological Paradigm in a social psychological context. Environment & Behavior, 36(2), 157-­‐
186 Johnson, C. Y., Bowker, J., Green, G., & Cordell, H. (2007). “Provide it...but will they come?” A look at African American and Hispanic visits to federal recreation areas. Journal of Forestry, (July/August), 257–265. Johnson, C., Horan, P., & Pepper, W. (1997). Race, rural residence, and wildland visitation: Examining the influence of sociocultural meaning. Rural Sociology, 62(1), 89–110. Kurasaki, K. S. (2000). Intercoder reliability for validating conclusions drawn from open-­‐ended interview data. Field Methods, 12(3), 179–194. Kvale, S., & Brinkmann, S. (2009). Interviews: Learning the craft of qualitative research interviewing (2nd ed.). Thousand Oaks, CA: Sage. 56 Kyle, G., Bricker, K., Graefe, A., & Wickham, T. (2004). An examination of recreationists’ relationships with activities and settings. Leisure Sciences, 26(2), 123–142. Kyle, G., & Chick, G. (2007). The social construction of a sense of place. Leisure Sciences, 29(3), 209–225. Lamb, J., & Goodrich, G. (Eds.). (2006). NOLS wilderness ethics: Valuing and managing wild places. Mechanicsburg, PA: Stackpole books. Le, Y., N. C. Holmes, & C. Kulesza. (2012). Barriers to a backyard national park: Case study of African American communities in Columbia, SC. Natural Resource Report NPS/NRSS/EQD/NRR-­‐2012/604. National Park Service, Fort Collins, Colorado. Liefländer, A. K., Fröhlich, G., Bogner, F. X., & Schultz, P. W. (2012). Promoting connectedness with nature through environmental education. Environmental Education Research, 1–15. Martin, D. C. (2004). Apartheid in the Great Outdoors: American advertising and the reproduction of a racialized outdoor leisure identity. Journal of Leisure Research, 36(4), 513-­‐535. MacLean, R. (2002). Being here now: Teaching a sense of place. In J. Gookin and D. Wells (Eds.), NOLS Environmental Education Notebook. Lander, WY: The National Outdoor Leadership School. McLean, S. (2013). The whiteness of green: Racialization and environmental education. The Canadian Geographer / Le Géographe Canadien, 57(3), 354–362. Merchant, C. (2003). Shades of darkness: Race and environmental history. Environmental History, 8(3), 380-­‐394. Miller, M., & Hinshaw, R. (2012). The Retrospective Pretest as a gauge of change. Journal of Instructional Psychology, 39(4), 251–258. Mittelstaedt, R., Sanker, L., & VanderVeer, B. (1999). Impact of a week-­‐long experiential education program on environmental attitude and awareness. Journal of Experiential Education, 22(3), 138–148. National Outdoor Leadership School. (2013a). Diversity & inclusion in practice. Lander, Wyoming. Retrieved from http://www.nols.edu/about/partnerships/pdf/diversity-­‐and-­‐inclusion.pdf 57 National Outdoor Leadership School. (2013b). Strategic Plan 2008-­‐2013. Lander, Wyoming. Retrieved from http://www.nols.edu/strategic_plan/nols_strategic_plan.pdf National Outdoor Leadership School. (2014). Find your course. Lander, Wyoming. Retrieved from: http://www.nols.edu/courses/ National Outdoor Leadership School. (2015a). Course Description: Absaroka Backpacking. Lander, Wyoming. Retrieved from: https://www.nols.edu/pdf/rocky-­‐mountain/abw_cd.pdf National Outdoor Leadership School. (2015b). Course Description: Alaska Backpacking. Lander, Wyoming. Retrieved from: https://www.nols.edu/pdf/alaska/akw_cd.pdf National Outdoor Leadership School. (2015c). Course Description: Alaska Backpacking and Sea Kayaking. Lander, Wyoming. Retrieved from: https://www.nols.edu/pdf/alaska/akx_cd.pdf National Outdoor Leadership School. (2015d). Course Description: Wind River Wilderness. Lander, Wyoming. Retrieved from: https://www.nols.edu/pdf/rocky-­‐mountain/wrw_cd.pdf Omi, M., & Winant, H. (2015). Racial formation in the United States (3rd ed.). New York: Routledge. Ortman, J. M., & Guarneri, C. E. (2009). United States Population Projections: 2000 to 2050, 1–19. Retrieved from U.S. Census Bureau website: http://www.census.gov/population/projections/files/analytical-­‐
document09.pdf Parker, J. D., & McDonough, M. H. (1999). Environmentalism of African Americans: An analysis of the subculture and barriers theories. Environment and Behavior, 31(2), 155–177. Patton, M. Q. (1999). Enhancing the quality and credibility of qualitative analysis. Health Services Research, 34 (5 Pt. 2), 1189–208. Roberts, N. S., & Chitewere, T. (2011). Speaking of justice: Exploring ethnic minority perspectives of the Golden Gate National Recreation Area. Environmental Practice, 13(4), 354–369. Roberts, N., & Suren, A. (2010). Through the eyes of youth: A qualitative evaluation of outdoor leadership programs. Journal of Park and Recreation Administration, 28(4), 59–80. 58 Rose, J., & Paisley, K. (2012). White privilege in experiential education: A critical reflection. Leisure Sciences, 34(2), 136–154. Schultz, P. W., Shriver, C., Tabanico, J. J., & Khazian, A. M. (2004). Implicit connections with nature. Journal of Environmental Psychology, 24(1), 31–42. Shin, W. S., & Jaakson, R. (1997). Wilderness quality and visitors’ wilderness attitudes: Management implications. Environmental Management, 21(2), 225–
232. Shores, K. A., Scott, D., & Floyd, M. F. (2007). Constraints to outdoor recreation: A multiple hierarchy stratification perspective. Leisure Sciences, 29(3), 227–246. Sibthorp, J., Furman, N., Paisley, K., & Gookin, J. (2008). Long-­‐term impacts of wilderness education. Research in Outdoor Education, 9, 86–103. Sibthorp, J., Paisley, K., Gookin, J., & Ward, P. (2007). Addressing response-­‐shift bias: Retrospective pretests in recreation research and evaluation. Journal of Leisure Research, 39(2), 295–315. Spence, M. D. (1999). Dispossessing the wilderness: Indian removal and the making of the National Parks. New York and Oxford: Oxford University Press. Stanfield, R., Manning, R., Budruk, M., & Floyd, M. (2005). Racial discrimination in parks and outdoor recreation: An empirical study. In Northeastern Recreation Research Symposium (pp. 247–254). Newtown Square, PA: U.S. Forest Service, Northeastern Research Station. Steg, L., & de Groot, J. (2012). Environmental values. In S. Clayton (Ed.), The Oxford handbook of environmental and conservation psychology. New York: Oxford University Press. Waage, T., Paisley, K., & Gookin, J. (2013). Understanding the contribution of wilderness-­‐based experiences to the creation of an environmental ethic in youth. Research in Outdoor Education, 11, 18–27. Warren, K. (2005). A path worth taking: The development of social justice in outdoor experiential education. Equity & Excellence in Education, 38(1), 89-­‐99. Warren, K., Roberts, N. S., Breunig, M., & Alvarez, M. A. G. (2014). Social justice in outdoor experiential education: A state of knowledge review. Journal of Experiential Education, 37(1), 1–15. APPENDICES 59 60 Appendix A. IRB approval letter Institutional Review Board
Office of Research Integrity | Oregon State University
B308 Kerr Administration Building, Corvallis, OR 97331-2140
Telephone (541) 737-8008
irb@oregonstate.edu | http://oregonstate.edu/irb/
Notification Type
Date of Notification
STUDY ID
6384
APPROVED
Study Title
Principal Investigator
Study Team Members
Submission Type
Level
Number of Participants
08/11/2014
Diversity in the Outdoors: Student Attitudes About Wilderness in the
National Outdoor Leadership School
Troy Hall
Sara Gress
Initial Application
Expedited
Category(ies)
6, 7
120
Do not exceed this number without prior IRB approval
Waiver(s)
Risk Level for Children
Funding Source
PI on Grant or Contract
Documentation of Informed Consent
§46.404 minimal risk
None
Proposal #
N/A
N/A
The above referenced study was reviewed and approved by the OSU Institutional Review Board (IRB).
Annual continuing review applications are
due at least 30 days prior to expiration date
Approval Date: 08/11/2014
Expiration Date: 08/10/2015
Documents included in this review:
Protocol
Consent forms
Assent forms
Alternative consent
Letters of support
Recruiting tools
Test instruments
Attachment A: Radiation
Alternative assent
Project revision(s)
External IRB approvals
Translated documents
Attachment B: Human materials
Grant/contract
Other:
Comments:
Principal Investigator responsibilities for fulfilling the requirements of approval:
All study team members should be kept informed of the status of the research.
Any changes to the research must be submitted to the IRB for review and approval prior to the
activation of the changes. This includes, but is not limited to, increasing the number of subjects to
be enrolled.
Reports of unanticipated problems involving risks to participants or others must be submitted to the
IRB within three calendar days.
Only consent forms with a valid approval stamp may be presented to participants.
Submit a continuing review application or final report to the IRB for review at least four weeks prior
to the expiration date. Failure to submit a continuing review application prior to the expiration date
will result in termination of the research, discontinuation of enrolled participants, and the
submission of a new application to the IRB.
OSU IRB FWA00003920
1
IRB Form | v. date February 2014
Appendix B. Effect sizes Variable Sense of place Pre-­‐course index Post-­‐course index Environmental ethics Pre-­‐course index Post-­‐course index Value of wilderness Pre-­‐course index Post-­‐course index Environmental awareness Pre-­‐course index Post-­‐course index Course experience n 33 33 33 33 33 33 31 31 31 Gateway M SD 3.81 1.60 1.72 0.58 2.90 1.17 1.48 0.47 3.68 0.94 2.28 0.77 2.85 1.19 1.49 0.56 1.66 0.76 61 n 41 41 41 41 41 41 40 40 40 Non-­‐Gateway M SD 3.26 1.14 1.60 0.82 2.18 0.79 1.55 0.69 2.62 0.79 1.90 0.69 2.23 0.90 1.68 0.78 1.46 0.68 Cohen’s d -­‐0.40 0.17 0.72 -­‐0.12 1.22 0.52 0.59 -­‐0.28 0.28 62 Appendix C. Survey instrument Oregon State University is conducting a research study titled "Diversity in the Outdoors: Student Attitudes About Wilderness in the National Outdoor Leadership School". We would like to invite you to participate in this study about your experiences and thoughts regarding your NOLS course this summer (2014). Your answers will provide NOLS with insight as to the diverse realities and perspectives about wilderness held by students, and will allow NOLS to better reflect these attitudes in its courses. The survey should take about 10 minutes to complete. Your participation is voluntary, and you are free to stop completing the survey at any time. Your answers will be kept strictly confidential. Your name will not be connected to any of your responses. You will have the opportunity to provide contact information at the end of the survey if you wish to participate in a follow up interview in the coming weeks. We thank you in advance for your participation. This study has been reviewed and approved by the Oregon State University Institutional Review Board. If you have any questions, concerns or comments about this study, please do not hesitate to contact: Institutional Review Board B308 Kerr Administration Building Corvallis, Oregon 97331-­‐2140 Telephone: (541) 737-­‐8008 Fax: (541) 737-­‐3093 E-­‐mail: irb@oregonstate.edu Please be aware that data collected online cannot be guaranteed; confidentiality will be maintained to the extent permitted by the technology being used. Information collected online can be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain viruses. Please click the >> below to advance the survey. By clicking this button, you are agreeing to take our survey. 63 The following statements ask about your feelings and opinions about wilderness and the environment before and after your NOLS course. The term wilderness in this survey refers to an undeveloped area that is free from substantial human impacts, where nature is dominant and the situations created by this setting have real consequences. When responding to these statements, please try to remember how you felt before your course and how you now feel after your course. It is most accurate to describe your AFTER responses first because that memory is freshest. For each statement, please use the drop down list to choose the best answer that describes how much you agree or disagree with that statement. Sense of place1 Before/After my NOLS course… Strongly Agree Agree a) …I expect to feel safe while in the wilderness. b) …I am very attached to wilderness areas.
c) …I feel like I have a good idea of what wilderness is. d) …I feel wilderness trips would be isolated and lonely. e) …I feel confident in my wilderness skills, e.g. navigation, finding water. 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 Somewhat Agree Neither Agree or Disagree Somewhat Disagree Disagree (a, b) adapted from Johnson, Horan, & Pepper (1997). (e) adapted from Sibthorp, Paisley, Gookin, & Ward (2007). 1 Wilderness attitude construct headers were not displayed on the online survey instrument. Also, numerical values were not assigned to the Likert-­‐type scale responses on the online survey Strongly Disagree 64 Environmental ethics Before/After my NOLS course… Strongly Agree Agree a) …I feel that the trees, wildlife, free flowing water, rock formations, and meadows that wilderness protects have value themselves, whether or not humans benefit from them. b) …I support protecting wilderness so it will always exist in its natural condition, even if no people were to ever visit or otherwise benefit from it. c) …I make an effort to make my daily life as environmentally friendly as possible. d) …I understand that I am a part of the larger ecological community. 1 2 3 4 5 6 Somewhat Agree Neither Agree or Disagree Somewhat Disagree Disagree Strongly Disagree 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 (a, b) Adapted from Johnson, Bowker, Bergstrom, & Cordell (2004) (d) Adapted from Hutson & Bailey (2008) Value of wilderness Before/After my NOLS course… Strongly Agree Agree a) …I feel that wilderness areas are important because they help to preserve plant and animal species that could have important scientific or human health values, such as new medicines. b) …A lot of my life is centered around recreating in wilderness. c) …I get more satisfaction from visiting 1 2 3 4 5 6 wilderness areas than any other type of recreation area. d) …I don’t understand why others care about wilderness. e) …I don’t feel a commitment to wilderness. Somewhat Agree Neither Agree or Disagree Somewhat Disagree Disagree Strongly Disagree 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 (a) Adapted from Johnson, Bowker, Bergstrom, & Cordell (2004) (b, c) Adapted from Johnson, Horan, & Pepper (1997) (e) Adapted from Kyle, Bricker, Graefe, & Wickham (2004) 65 Environmental Awareness Before/After my NOLS course… Strongly Agree Agree a) …I feel that wilderness areas were important to protect because they contribute to air and water quality. 1 2 3 4 5 6 b) …I don’t understand ecological processes in wilderness. c) …It is important to me to understand and be aware of current environmental issues. d) …I feel a sense of responsibility for the environment. Somewhat Agree Neither Agree or Disagree Somewhat Disagree Disagree Strongly Disagree 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 (a) Adapted from Johnson, Bowker, Bergstrom, & Cordell (2004) (d) Adapted from Johnson, Horan, & Pepper (1997). Course Experience The following statements ask about your experience on your NOLS course. During my NOLS course… Strongly Agree a) …I felt like I was a part of the group. b) …I found the physical aspects (i.e. hiking with a heavy pack) of the course easier than I expected. c) …I felt that my opinions and views were valued. d) …I learned a lot about wilderness and how to travel in it. Agree Somewhat Agree Neither Agree or Disagree Somewhat Disagree Disagree Strongly Disagree 1 1 2 3 4 5 6 2 3 4 5 6 7 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 66 Socio-­‐demographic measures Now we would like to know a bit more about you. 1) What year were you born? Year: 2) What is your gender? o
o
o
Male Female Transgender 3) Were you a NOLS Gateway Scholarship Student? o
o
Yes No 4) Which course did you take? Please provide course name, dates and location. Course name: Course dates: Course location: 5) How long was your stay in the wilderness during your course? Days: 6) Were you evacuated from your course? 2 o
o
Yes No 2 Some NOLS students are evacuated from their courses, for a variety of reasons, such as a medical emergency. 67 7) Which race and/or ethnicity do you most closely identify with? Please write your answer in the space provided. 3 6) What type of community do you reside in? Please choose one. o
o
o
o
o
Major city or metropolitan area (more than 50,000 residents) Suburb of a metropolitan area Small city (more than 2,500 residents, less than 50,000 residents) Small, rural town (less than 2,500 residents) Rural area outside of city limits 7) How many overnight wilderness trips had you gone on before your NOLS course? o
o
o
o
Zero 1 or 2 3 to 9 10+ 8) If you would be interested in participating in a follow-­‐up telephone interview about your NOLS experience in the coming weeks, please provide your contact information. If you do not wish to be contacted, you may leave this question blank. Name: Email: Phone number: 3 Question 7 is left open-­‐ended to measure the self-­‐ascribed ethno-­‐racial identity of the student, as opposed to an externally assigned identity. 68 Thank you for taking our survey. Your answers will be helpful for future NOLS programming. If you have any questions, concerns or comments, please contact: Sara Gress Department of Forest Ecosystems and Society Oregon State University 321 Richardson Hall, Corvallis, OR 97331 Tel: 541-­‐737-­‐2244 sara.gress@oregonstate.edu 69 Appendix D. Cronbach’s alpha results Variable Sense of place I expect to feel safe while in the wilderness. I am very attached to wilderness areas. I feel like I have a good idea of what wilderness is. I feel wilderness trips would be isolated and lonely. I feel confident in my wilderness skills, e.g. navigation, finding water. Environmental ethics I feel that the trees, wildlife, free flowing water, rock formations, and meadows that wilderness protects have value themselves, whether or not humans benefit from them. I support protecting wilderness so it will always exist in its natural condition, even if no people were to ever visit or otherwise benefit from it. I make an effort to make my daily life as environmentally friendly as possible. I understand that I am a part of the larger ecological community. Value of wilderness I feel that wilderness areas are important because they help to preserve plant and animal species that could have important scientific or human health values, such as new medicines. A lot of my life is centered around recreating in wilderness. I get more satisfaction from visiting wilderness areas than any other type of recreation area. I don’t understand why others care about wilderness. I don’t feel a commitment to wilderness. Environmental awareness I feel that wilderness areas were important to protect because they contribute to air and water quality. I don’t understand ecological processes in wilderness. It is important to me to understand and be aware of current environmental issues. I feel a sense of responsibility for the environment. Course experience I felt like I was a part of the group. I found the physical aspects (i.e. hiking with a heavy pack) of the course easier than I expected. I felt that my opinions and views were valued. I learned a lot about wilderness and how to travel in it. * item deleted Pre-­‐course α α if deleted 0.64 0.48 0.50 0.46 0.83* 0.51 Post-­‐course α α if deleted 0.49 0.26 0.38 0.40 0.72* 0.38 0.77 0.73 0.68 0.73 0.63 0.65 0.54 0.73 0.49 0.72 0.71 0.65 0.69 0.81 0.68 0.60 0.74 0.50 0.31 0.82* 0.75 0.73* 0.33 α 0.39 0.74 α if deleted 0.61 0.44 0.82* 0.37 0.60