PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINE return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE 0015 2'1 I fo'm '2 5; 20M Artxi; "iU V r1119 01 JUL @Mo’bz’gi NOV 12 2002 .DEC 0 652W M APRQf‘ Menorah moo chIfiC/DubOuopGS-p.“ EFFECTIVENESS OF FOCUS GROUP INTERVIEWS AS A PRIMARY DATA GATHERING METHOD FOR RECREATION NEEDS ASSESSMENTS BY Nancy Ellen Knap A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Park, Recreation and Tourism Resources 1999 ABSTRACT EFFECTIVENESS OF FOCUS GROUP INTERVIEWS AS A PRIMARY DATA GATHERING METHOD FOR RECREATION NEEDS ASSESSMENTS BY Nancy Ellen Knap The effectiveness of the focus group interview as the primary data collection method for recreation needs assessments is largely untested. This study evaluated the effectiveness of the method as it was used for the 1999 Ingham County, Michigan recreation needs assessment. Evaluation criteria were based on the goals of the Ingham County Parks Board and generally accepted social science research standards. Findings suggested that the focus group interview method was effective for several reasons. Recreation, a relatively non-controversial topic, was discussed with ease. Purposive and snowball sampling techniques provided a diverse sample, reaching groups underrepresented in previous assessments. Group dynamics, flexibility in questioning, and a diverse sample provided a broad variety of responses. Copyright by NANCY ELLEN KNAP 1999 This thesis is dedicated to my parents, Dr. Florian J. and Delphine Knap who shared with me their lifelong joy of learning. iv ACKNOWLEDGMENTS It is misleading for my name, alone, to appear as the author of this thesis for it clearly would not have been possible without the contributions of others. I extend my thanks to Dr. Dennis Propst for his support and guidance, not only in the writing of this thesis, but also throughout my years of graduate study. My thanks are offered also to Dr. Maureen McDonough and Dr. Daniel Spencer for their work as members of my graduate committee. For their financial support, I thank the people of Ingham County, especially those who also shared with me their hospitality and their thoughts. TABLE OF CONTENTS LIST OF TABLES. . . . . . . . . . . . . . . . . . . . . .vii LIST OF FIGURES. . . . . . . . . . . . . . . . . . . . . xii CHAPTER 1. INTRODUCTION. . . . . . . . . . . . . . . . . . 1 Study Purpose. . . . . . . . . . . . . . . . . . . . .7 CHAPTER 2. LITERATURE REVIEW. . . . . . . . . . . . . . . .9 Traditional Survey Methods. . . . . . . . . . . . . . 9 Focus Group Interview Method. . . . . . . . . . . . .16 Problem. . . . . . . . . . . . . . . . . . . . . . . 26 Objectives. . . . . . . . . . . . . . . . . . . . . .26 CHAPTER 3. PROCEDURES. . . . . . . . . . . . . . . . . . .28 Needs Assessment Procedures. . . . . . . . . . . . . 28 Sampling. . . . . . . . . . . . . . . . . . . . 29 Recruiting Focus Groups. . . . . . . . . . . . .33 Conducting Focus Group Interviews. . . . . . . .35 Coding Data. . . . . . . . . . . . . . . . . . .37 Analyzing Data. . . . . . . . . . . . . . . . . 40 Reporting Data. . . . . . . . . . . . . . . . .42 Evaluation Procedures. . . . . . . . . . . 42 Criteria as Determined by the Goals of the Ingham County Recreation Needs Assessment. 43 Criteria as Determined by Generally Accepted Social Science Research Standards. . . . . 46 CHAPTER 4. RESULTS AND DISCUSSION. . . . . . . .48 Evaluation along Criteria based on the Goals of the Ingham County Recreation Needs Assessment. . . .48 Evaluation along Criteria based on Generally Accepted Social Science Research Standards. . . . . . . .80 CHAPTER 5. SUMMARY AND CONCLUSIONS. . . . . . . . . . . . 96 Summary of Findings. . . . . . . . . . . . . . . . . 96 Conclusions. . . . . . . . . . . . . . . . . 99 Limitations of the Study. . . . . . . . . . .99 Recommendations for the Effective Use of the Focus Group Interview Method in Recreation Needs Assessments. . . . . . . . . . . . . . . . 103 Recommendations for Further Study. . . . . . . . . 107 vi Final APPENDIX A. APPENDIX APPENDIX APPENDIX APPENDIX APPENDIX APPENDIX REFERENCES. B. C. G. Thoughts. FOCUS GROUP RECRUITMENT LETTER. FOCUS GROUP RECRUITMENT TELEPHONE SCRIPT. QUESTIONNAIRE. DISCUSSION QUESTIONS. CODEBOOK. RANKED FREQUENCY COUNTS. MAJOR CONCEPTS 1-8. vfi 108 .112 .115 117 .120 .123 134 .143 .178 LIST OF TABLES Table 1. Leisure behavior survey methods (Gold, 1980). Table 2. Places of residence of nonresident respondents. Table 3. Nonresident distribution in groups. Table 4 Age of sample compared with 1990 U.S. Census for Ingham County. Table 5. Gender of sample compared with 1990 U.S. Census for Ingham County. Table 6. Education of sample compared with 1990 U.S. Census for Ingham County. Table 7. Race/origin of sample compared with 1990 U.S. Census for Ingham County. Table 8. Household type of sample compared with 1990 U.S. Census for Ingham County. Table 9. Residence of sample compared with 1990 U.S. Census for Ingham County. Table 10. Non-facilities related attributes named in response to discussion question 3, reasons for visiting Ingham County parks. Table 11. Non-facilities related attributes named in response to discussion question 3, reasons for not visiting Ingham County parks. Table 12. Non-facilities related attributes named in response to discussion question 4, attributes desired in Ingham County parks. Table 13. Non-facilities related attributes named in :nesponse to discussion.question S, attributes desired in Ingham County parks (discussion aided by list of attributes). vfii .55 .56 .59 .60 60 61 61 62 .67 68 .69 .70 Table 14. Frequency of response in support of a water park, Ingham County recreation needs assessment (1998-1999). Table 15. Frequency of response in opposition to a water .73 park, Ingham County recreation needs assessment(1998-1999) Table 16. Focus group recruitment sources. Table 17. Comparison of responses to Kent County question 5 and Ingham County question 3. Table 18. Comparison of responses to Kent County question 2 and Ingham County question 3. Table 19. Participation in outdoor recreation ranked by frequency of response, Ingham County recreation needs assessment (1999). Table 20. Location of outdoor recreation engaged in ranked by frequency of response, Ingham County recreation needs assessment (1999). Table 21. Identification of Ingham County parks visited ranked by frequency of response, Ingham County recreation needs assessment (1999). Table 22. Reasons for visiting Ingham County parks ranked by frequency of response, Ingham County recreation needs assessment (1999). Table 23. Reasons for not visiting Ingham County parks ranked by frequency of response, Ingham County recreation needs assessment (1999). Table 24. Reasons for visiting noncounty parks, Ingham County parks ranked by frequency of response, Ingham County recreation needs assessment (1999). Table 25. Reasons for not visiting noncounty parks, Ingham County parks ranked by frequency of response, Ingham County recreation needs assessment (1999). ix 73 .78 .83 .86 134 .135 .135 136 .137 138 .138 Table 26. Attributes desired in Ingham County parks ranked by frequency of response, Ingham County recreation needs assessment (1999). Table 27. Attributes desired in Ingham County parks (aided)ranked by frequency of response, Ingham County recreation needs assessment (1999). Table 28. Attributes not desired in Ingham County parks (aided) ranked by frequency of response, Ingham County recreation needs assessment (1999). Table 29. Suggested media for information distribution ranked by frequency of response, Ingham County recreation needs assessment (1999). Table 30. Nonvisitation due to inconvenience & desire for improved public transportation/convenience, Ingham County Recreation Needs Assessment (1998-1999). Table 31. Visitation due to convenience, Ingham County Recreation Needs Assessment (1998—1999). Table 32. In support of land acquisition, Ingham County Recreation Needs Assessment (1998-1999). Table 33. In opposition to land acquisition, Ingham County Recreation Needs Assessment (1998-1999). Table 34. In support of campgrounds (all types), Ingham County Recreation Needs Assessment (1998-1999). Table 35. In opposition to campgrounds (all types), Ingham County Recreation Needs Assessment (1998—1999). Table 36. Desire for information, Ingham County Recreation Needs Assessment (1998»1999). Table 37. In support of interagency cooperation, Ingham County Recreation Needs Assessment (1998-1999). Table 38. In support of connecting parks, Ingham County Recreation Needs Assessment (1998-1999). .139 .140 .141 .141 .144 144 148 .148 .155 155 162 .166 168 Table 39. In support of nominal entrance and rental fees, Ingham County Recreation Needs Assessment (1998-1999). . . . . . . . . . . . . . . . . . . . . . . 171 Table 40. In support of lowcost/free entrance and services, Ingham County Recreation Needs Assessment (1998-1999). . . . . . . . . . . . . . . . . . . . . . . 174 xi Figure 1. Figure 2. LIST OF FIGURES Sampling matrix of focus groups conducted. Sample page of coded transcript. xfi .30 .39 CHAPTER 1 Introduction The purpose of an evaluation is to make a judgment about the worth of what is being evaluated so that informed decisions may be made. Henderson (1995) describes an evaluation as, “the systematic collection and analysis of data to address criteria and make judgments about the worth or improvement of something” (p.3). Henderson (1995) identifies five evaluation models, including the professional (expert) judgment model. This model is “criterion referenced”. “Criterion referenced evaluation is not compared to any other organization but simply is used as a standard for measurement” (Henderson, 1995, p.55). This evaluation examines the worth of the focus group interview as the primary data gathering method for recreation needs assessments using comparison with criteria as the standard of measurement. Henderson (1995) describes the purpose recreation needs assessments and how the results are used: Assessment is a process of determining and making specific what a program, a facility, a staff member, a participant’s behavior, or administrative procedure is. In community recreation programs, we often conduct needs assessments. These assessments identify the difference between ‘what is’ and ‘what should be.’ The needs assessment often results in a process of prioritizing results to use in planning programs, places, policies or the use of personnel. (p. 46) In addition to the planning of programs, places, policies and the use of personnel, recreation needs assessments are conducted to avoid the “rubber—stamp replication of standard facilities" and the imposition of the park designer’s view to the exclusion of the user’s need (Molnar & Rutledge, 1992). Molnar and Rutledge (1992) also identified the role of recreation needs assessments in the planning of capital improvements (p. 28). They recognized that recreation needs assessments “do not reveal every facet of need" but that “whatever people say they want springs from a need” (p.23). They conclude that “Realistically, the science of sorting out little—understood human needs is inexact at best. . .” (p. 23). Recreation needs assessments have been conducted using quantitative and qualitative methods, both separately and in combination (Yoder, et al., 1995). The following description of the conventional roles of each approach comes from the field of evaluation research. “The view adopted by most of the research community is that quantitative techniques are the most appropriate source for corroborating findings initially noted from qualitative methods. Likewise, qualitative methods are best used to provide richness or detail to quantitative findings (elaboration), but should precede quantitative ones when clarifying the direction of inquiry (initiation)” (Rossman & Wilson, 1985, p.633). The fullest understanding of social phenomena is achieved through the use of both approaches, the strengths of each compensating for the weaknesses of the other. Reichardt and Rallis summarize more current thinking: The qualitative and quantitative research traditions differ. Qualitative researchers usually seek to explicate the meaning of social reality from the participants' perspectives, while quantitative researchers usually seek to understand relationships, often of a causal nature, without particular emphasis on the participants’ perspectives. Nonetheless, at the most global level, the two traditions have a common goal: to understand and improve the human condition. A defensible understanding of reality can withstand scrutiny from different perspectives and methodologies. Indeed, given its complexities and multiple facets, a complete understanding of human nature is likely to require more than one perspective and methodology. The qualitative and quantitative traditions can provide a binocular vision with which to deepen our understandings. That the qualitative and quantitative perspectives remain partly adversarial in their relationship does not preclude cooperation in working together toward their shared goal. In fact, just the opposite is true. By working together, the two traditions can enhance the practice and utilization of research and evaluation.(Reichardt & Rallis, 1994, p.11) The use of multiple methods in social research study design is recommended to take advantage of the strengths of each (Babbie, 1998, p.108). Yoder et al., (1995) describe the rationale for the use of multiple approaches in research involving parks and recreation agencies. “The understanding of any complex social phenomenon can be enhanced with the use of different points of reference. A social entity, especially one as complex and diverse as a public park and recreation agency, cannot be fully investigated from any single perspective” (p.28). Yet, the decision to use multiple methods will not produce satisfactory results if those methods are employed haphazardly (Yoder et al.,1995, p.27) and few agencies are large enough to support in—house research staffs (Webb & Hatry, 1973, p.59). Limited financial resources and time constraints are major concerns for public agencies conducting citizen surveys and needs assessments (Hudson, 1988, p.21; Webb & Hatry, 1973, pp.51—53; Yoder et al., 1995, p.39). As a result, public agencies tend to choose from methods available to them, rather than attempting to integrate multiple methods into a single study (Hudson, 1988, p.21). In parks and recreation needs assessments and similar research in closely related fields, the most commonly used data collection methods are mail surveys, telephone interviews, and personal interviews (Bannon, 1976; Gold, 1980; Hudson, 1988). Hudson (1988), in her step by step guide to conducting community needs assessments in public parks and recreation, listed the three “major methods utilized by leisure agencies to collect survey information” as: mail questionnaires, telephone interviews, and personal interviews (p.21). Community surveys for leisure planning were described by Bannon (1976). He explained that the term “questionnaire” often includes both personal and self- administered surveys and that, “It [the questionnaire] is probably the most widely used data source in survey research” (p.137). Gold (1980) listed five common survey methods used in the study of leisure behavior of a community. They were: mail questionnaires, self—administered questionnaires, personal interviews, telephone interviews, and field observations (Gold, 1980, p. 188). Hudson and Witt (1984) listed the three types of data collection methods available to leisure agencies as: mailed surveys, telephone interviews, and personal interviews (pp. 18-21). Krueger (1994) contends that such quantitative needs assessment methods as surveys are often incomplete and that, “needs assessment surveys tend to identify concerns that already have achieved some visibility within the community as opposed to the less visible concerns that lie below the surface” (p. 30). He described four different ways that focus groups may be used in relation to quantitative methods. They may be used before quantitative procedures; at the same time, including triangulation; after quantitative procedures; and alone (Krueger, 1994, p. 29). Focus groups used alone, “are helpful when insights, perceptions, and explanations are more important than numbers” (Krueger, 1994, p. 30). Krueger (1994) recommends the use of focus groups for public sector needs assessments. He defined the focus group data collection method through the identification of six characteristics. “These characteristics relate to the ingredients of a focus group: (1) people, (2) assembled in a series of groups, (3) possess certain characteristics, and (4) provide data (5) of a qualitative nature (6) in a focused discussion” (Krueger, 1994, p. 16). Study Purpose The purpose of this study is to provide an evaluation of effectiveness of the focus group interview method in recreation needs assessments. This evaluation is intended to expand the choice of methods available to recreation researchers to include focus group interviews and to improve the understanding of focus group interview data for recreation research consumers. While both a self-administered written questionnaire and focus group interviews were used, the needs assessment in this study neither triangulated nor integrated methods. The dual approach used is best described by Rossman and Wilson (1985) as “situationalist”. There was no attempt to gather the same data using both methods. The questionnaire was used to gather socioeconomic and residency data and estimates of park use. The data gathered using the questionnaire provided an indication of how well the sample represented the study population for socioeconomic and residency characteristics. The data verified the inclusion in the sample of both users and nonusers of the park system under study. Reports of outdoor recreation behavior, reasons for visiting and not visiting parks, and attributes desired were gathered in the focus group interviews. Potential users of this study are those entrusted with the selection of research methods for recreation needs assessments, including parks governance boards, academic institutions, and private consultants. Other potential users are the consumers of recreation research: park planners, administrators and managers. The benefit to those selecting methods may be an increased understanding of the strengths and weaknesses of the focus group interview method, a method which seems well suited to recreation needs assessments but is largely untested. This understanding may broaden the choice of methods under consideration. The benefit to consumers of recreation research may be an increased familiarity with the data generated using focus group interviews. Familiarity may improve confidence in and reliance on qualitative data for recreation resource decision making. CHAPTER 2 Literature Review Traditional survey methods are used with such frequency that they have been widely accepted in the social sciences, and specifically, in leisure research (Babbie, 1998; Bannon, 1976; Gold, 1980; Hudson, 1988). They are not, however, without weaknesses. These methods continue to be used with some modifications, including their combination with qualitative methods, to compensate for those weaknesses (Wicks & Norman, 1996). This review begins with a discussion of survey methodology, followed by a review of the focus group interview method in recreation research and related fields. The review of the focus group interview method examines its changing role, advantages and disadvantages, and its suitability for recreation needs assessments. ra ' i v ds Babbie (1998) characterizes survey research as, “perhaps the most frequently used mode of observation in the social sciences” (p.255). He identifies the three main methods of administering survey questionnaires as: 1) the self-administered questionnaire, typically the mail survey, 2) the interview survey, typically a “face-to-face encounter”, and 3) the telephone survey (p.276). Although Babbie is referring to social science in general, these classifications closely correspond to those most commonly used in recreation needs assessments (Bannon, 1976; Gold, 1980; Hudson, 1988). As such, Babbie’s discussion of the strengths and weaknesses of survey research as a concept embracing all three methods, and his discussion of the comparative strengths and weaknesses of each method are appropriate. The strengths of survey research include: 1) the ability to describe the characteristics of large populations, 2) flexibility of analysis, and 3) standardized responses produced by asking all respondents exactly the same questions (Babbie, 1998, p.273). The weaknesses of survey research include: 1) the use of standardized questions that represent the least common denominator, 2) the superficial coverage of complex topics, 3) an inability to deal with the context of social life, 4) an inflexibility in the uniformity of the instrument throughout the study, 5) an inability to measure social action (only self-reports of recalled past action or of prospective or hypothetical action) and 6) a general weakness in validity (Babbie, 1998, pp. 273—274). 10 Babbie (1998) compares self-administered questionnaires, face-to-face interview surveys, and telephone surveys, illuminating the strengths of each method within the realm of survey research. He contends that self- administered questionnaires are quicker and cheaper than face-to—face interview surveys. They are cheaper than telephone surveys if the survey is national in scope. Mail surveys may be conducted with a small staff. Anonymous self- administered questionnaires are more effective for sensitive or controversial issues than interview surveys (Babbie, 1998, p.271). Self-administered surveys lack interviewer bias (Babbie, 1998, p.276). Interview surveys have higher completion rates (i.e., fewer incomplete questionnaires), fewer misunderstood questions, and higher return rates than self-administered questionnaires. Interview surveys allow for greater flexibility in sampling in the sense that interviewers may Select the appropriate respondent. Self-administered surveys delivered by mail may be completed by virtually anyone in the household (Babbie, 1998, p.276). Interview surveys are more effective for complicated issues than are self- administered surveys (Babbie, 1998, p.271). In face—to-face interviews, observations made by the interviewer may be 11 recorded in addition to verbal responses to the questions asked, providing more extensive data than either self- administered or telephone interviews (Babbie, 1998, p.272). Telephone interviews are cheaper than face—to—face interviews, may be started and executed quickly, and are safer than door—to-door interviews. Finally, the impact of the interviewer on the response is smaller when the interviewer cannot be seen by the respondent, an advantage of telephone interviews over face-to—face interviews (Babbie, 1998, p.274). Gold (1980) identified many of the same survey methods for use in the study of leisure behavior. He characterizes survey research as, “the best way to study the leisure behavior of a community” and argued that, “Survey research techniques can also be used to assess the recreation needs of special populations, identify major problems and potentials, describe public opinion, and determine the effectiveness of existing facilities or programs” (Gold, 1980, p.118). Gold (1980) advocated the use of probability sampling of both users and nonusers, on-site and in the home. He identified five commonly used survey methods: 1) mail questionnaires, 2)self-administered questionnaires, 3) 12 personal interviews, 4) telephone interviews, and 5) field observations (Table 1). These methods have been employed in recreation needs assessments with such regularity that their limitations are well known. Some of these limitations have been addressed, particularly for population subgroups. Wicks and Norman (1996) address the lack of participation by African Americans in mail survey research and the possibility of nonresponse bias in the results of such surveys. They suggest that the reason for nonparticipation by African Americans in mail surveys is, “the lack of perceived personal connection felt by participants to the subject content being surveyed” (Wicks & Norman, 1996, p.1). They presented a case study in which higher response rates were achieved by modifying the traditional mail survey to achieve a more “personal connection”. Citing the work of Rossman (1994), Wicks and Norman question the validity of responses of African Americans when they do participate in mail surveys. “[M]inorities often offer idealized answers or what they feel might be the ‘correct’ answer” (Wicks & Norman, 1996, p.3). Wicks and Norman (1996) conducted focus groups to identify ways of obtaining information from low income 13 Table 1. Leisure behavior survey methods* (Gold, 1980). Survey Methods Mail Self— Personal Telephone Field question- administered interview interview observa- Major factors naire tion Cost L M H M H Administration L M H M H Privacy/ H M L M L anonymity Sample size/ H M L H L coverage Nonresponse H M L M L bias Follow-up H M M M L required Time required H M M M H Depth/detail L L H H H Personal L M H M L contact Flexibility L L H M L Interviewer L L H M H bias Use of visual L L H L L material Logistics/ L H H L H travel *Summary of relative advantages and disadvantages for typical situation expressed in terms of high (H), medium (M), or low (L), based on criteria described in Survey Research for Community Recreation Services, Michigan State University, Experiment Station Research Report No. 291, February 1976, p. 9. 14 African Americans. The focus groups revealed that nonresponse to mail surveys by this group may be due to the perception that mail surveys: 1) were too complicated, 2) took too much time, 3) were irrelevant to the respondents’ needs, and 4) required reading skills beyond those possessed by the respondent (Wicks & Norman, 1996, p.7). Suggestions for overcoming nonparticipation by this group included: 1) working through “a person or organization that is already established and trusted within their community and then conduct personal interviews or group meetings”, 2) using community center staff as informants, 3) speaking or writing in language appropriate to the respondent, 4) offering incentives, and 5) making respondents more comfortable by explaining the need for information, especially that of a personal nature (Wicks & Norman, 1996, p.9). Wicks and Norman (1996) identified the personal interview as possibly the best alternative to the mail survey for low income African Americans. However, the personal interview was not used for their study due to cost, the need for trained interviewers, and concern for interviewer bias. Instead, a telephone survey which was a modified version of their mail survey was conducted. The 15 wording of the telephone script was changed to achieve a less formal, more conversational tone. Wicks and Norman (1996) concluded that a mixture of both qualitative and quantitative methods may be necessary to reach minority populations and that while qualitative methods, “are likely to be more effective with minorities”, challenges in interpretation will occur (p.13). F0 0 In rvi w M h Social science researchers are not completely unfamiliar with the focus interview method. The use of focus group interviews in the preparation of surveys and other quantitative instruments is common (Minnis et al., 1997; Mitra, 1994; Wicks & Norman, 1996). Their use as a stand— alone method, however, is infrequent outside of marketing (Morgan, 1988, p.10). Support for using focus group interviews as a stand—alone method in social science research, and specifically in public sector needs assessments, seems to be growing (Krueger, 1994; Morgan, 1988). While some of this support predates this thesis by over ten years, the use of focus group interviews in recreation needs assessments as a stand—alone, or even a primary, data gathering method remains largely untested. 16 Focus groups are not new to social science research. Merton et al.(1956) used focus groups in studies of the effectiveness of mass communications. This early work followed experimental procedures. Subjects were exposed to the same stimulus, typically viewing a film, and subsequently participated in a focus group interview. Merton et a1. (1956) identified the advantages of focus group interviews as compared with individual focused interviews. They included: 1) release of respondents’ inhibitions, 2) greater variations in responses, and 3) activation of forgotten details via group discussions (Merton et al., 1956, p. 146). Disadvantages stemmed from responses that “may be more nearly related to this interplay of personalities and status claims in the group than to the subject matter” (Merton et al., 1956, p.147). These included the “leader effect”, irrelevant discussions, interruptions in continuity, and inhibitions caused by fears of “public speaking” (Merton et al., 1956, pp. 149-153). More recently, Morgan (1988) used focus group interviews employing a less experimental approach than Merton et al.(1956). Morgan (1988) identified the strengths and weaknesses of focus group interviews and their role as a 17 —"'F 8]: A”, “I v0.5 ~Y‘A\ ‘v. u.. "- - ‘ ha ‘4 H) \a p ‘4 ‘x‘. “‘ N ‘\ ’ . ‘Q ~ F... I . ‘.‘v ‘F ‘. ‘1 - .V ,- NAN?" ~‘\ self-contained method3f He identified pairs of strengths and their corresponding weaknesses in the use of focus groups. He found focus groups: 1) are easy to conduct, but not based on natural settings, leaving “residual uncertainty about the accuracy of what participants say”, 2) are effective in exploring topics and generating hypotheses, but researchers have less control over data, and 3) create opportunities for group interaction without preconceived questions, but may not “mirror” individual behavior (Morgan, 1988, pp. 20-21). Morgan (1988) stated, “From a social science point of view, focus groups are useful either as a self—contained means of collecting data or as a supplement to both quantitative and other qualitative methods” (p. 10). He pointed out that in asserting the importance of using focus group interviews as a self—contained method, researchers should not overlook their value in combination with other methods. In combination with other methods, he found focus groups suited to: 1) orientation to a field new to the researcher, 2) hypothesis generation, 3) evaluation of study 'TTHe terms “self-contained” and “stand—alone” are used iI"Iterchangeably herein to describe the focus group interview metlnod when it is not used in conjunction with other methods or ;in.the preparation of survey instruments. This is not to Suggest that it cannot be used with other methods. 18 sites or populations, 4) development of interview schedules and questionnaires, and 5) interpretation of the results of previous studies (Morgan, 1988, p. 11). Morgan (1988) suggested that the decision to use focus groups depends on how “actively and easily” participants can discuss the topic under study (p.23). Focus groups were appropriate as a self-contained method when new areas are being explored or the participants’ point of view is sought in well-known areas (p. 24). Morgan (1988) reiterates, “The key distinguishing feature of a self—contained focus group is that the results can stand on their own” (p.25). Krueger (1994) also recognized this stand-alone feature and suggested a broader range of uses. Krueger’s (1994) practical guide to focus groups was intended specifically for use by public and nonprofit sectors (p.vii), suggesting applications for public sector recreation agencies. Krueger (1994) contends that focus groups “work” because they: 1) tap into human tendencies; that as products of our environment, we are influenced by the others around us, and 2) provide a nonjudgmental environment where people terki toward self—disclosure (pp. 10—11). He stated the Purjpose of focus groups is, “to determine the perceptions, ieetlings, and manner of thinking of consumers regarding 19 products, services, or opportunities” (Krueger, 1994, p.19). He organized the uses of focus groups according to the stage of the process being studied from planning to “postmortems”, specifically identifying “needs assessments” (Krueger, 1994, pp. 21—25). Krueger (1994) identified the following limitations of focus groups: 1) compared with individual interviews, there is less researcher control, 2) group interaction makes the data more difficult to analyze, 3) trained interviewers are required, 4) variations in group demeanor make a sufficient number of groups necessary, 5) groups are difficult to assemble, and 6) logistical problems may require participant incentives (p.36). Focus group interviews have been used in natural resource planning. Minnis et a1. (1997) discussed the pre- quantitative uses of focus groups in natural resource policy research. In the Minnis et al. study, focus groups provided insights into the level of hunters’ understanding of state bear hunting policies. These insights led to the use of hypothetical scenarios in the subsequent mail questionnaire. Minnis et a1. (1997) compared the use of focus groups in (questionnaire design with questionnaire development excfilusively from the researcher perspective. They concluded: 20 1) focus groups enabled researchers to better develop and address research questions, 2) focus groups exposed factors unanticipated by researchers, but critical to the topic, 3) focus groups provided a participant perspective, and 4) focus groups are useful, and perhaps underutilized, in human dimensions research (pp. 46—47). Henderson (1991), in her discussion of qualitative methods for recreation, parks, and leisure research states: Indepth interviewing can be expensive, time consuming, biased, and sometimes inefficient, but the method also provides some of the richest data that we can find. Indepth interviews are hard to pretest, have unpredictable results because different people tend to respond differently, and are difficult to standardize and replicate. On the other hand, they also offer many advantages in allowing the researcher to have a greater understanding of the complexities of social reality from a number of perspectives. (p.72) She identifies focus group interviews as a variation of qualitative interviewing (p.82). Her discussion extensively cites a 1988 work by Krueger, linking the focus group interview method as defined by Krueger, with recreation, parks, and leisure research. She concurs with Krueger that focus group interviews may be used alone or in combination Imith.other methods adding that they are appropriate for PrCXgram evaluation (p. 82). Henderson’s (1991) idethtification of focus group interviews for recreation 21 program evaluation suggests their suitability for recreation needs assessments. Mitra (1994) used focus group interviews in connection with a public sector recreation needs assessment, not as a stand—alone or even a primary data collection method, but in a pre-quantitative role. She discussed the use of focus groups in developing a written questionnaire for a public sector recreation needs assessment. She found focus groups, “help to make a questionnaire specific to a community” by: 1) providing community specific input, and 2) helping to design a final instrument for specific programs. She noted that by conducting preliminary focus groups, it was possible to increase awareness of recreation opportunities and to publicize the study which may increase response rates to the subsequent survey. She identified an additional value to focus groups, that of creating a “public relations channel” for the recreation provider (Mitra, 1994, p.134). Mitra (1994) established two criteria for focus group success: 1) sufficient interest in providing input, and 2) moderator awareness of recreation issues to keep special :hlterests from turning focus groups into “forums for airing COTnplaints and grievances” or digressing to irrelevant issues (p. 138). 22 Mitra (1994) concluded that, “a very good understanding of the leisure needs of the community” may be obtained solely through focus groups, but rejected them as a single data gathering method for recreation needs assessments. She stated two reasons: 1) focus groups most often represent frequent users of the recreation provider, not necessarily a random sample of the communityz, and 2) recreation planners and administrators tend to prefer “hard numbers”. She stated this tendency can be “corrected” by increasing focus group use and improving qualitative data coding methods (Mitra, 1994, p.139). The effectiveness of focus groups in leisure research was studied by Siegenthaler et al.(1998), but as a secondary objective. They also found focus group interviews activated forgotten details, a feature identified by Merton et al.(1956). Siegenthaler et al.(1998) used focus groups to examine the leisure perceptions of older women in retirement communities. A second purpose of the study was, “to examine the effectiveness of focus groups for stimulating thought 2Groups in her study were made up of recreation staff, interest groups, and volunteers responding to press releases. It seems that the sampling method, not the use of fixzus groups, was the cause of this limitation. 23 and discussion about leisure and recreation with older women” (Siegenthaler et al., 1998, p.55). The authors found focus groups were an “effective means of eliciting discussion” and “the interactions with other group members appeared to stimulate their own thoughts about their experiences and perceptions” (Siegenthaler et al., 1998, p. 65). Finally, “the women clearly enjoyed the discussion and seemed to appreciate the social interaction it provided” (Siegenthaler et al., 1998, p. 65). In summary, the uses of focus groups largely have been pre-quantitative (e.g., to prepare self-administered questionnaires) or in combination with other methods, both quantitative and qualitative. Uses have been determined by the purpose of the research and the ease with which the topic is likely to be discussed by participants. Strengths of focus groups stem from group interaction and gaining the participant’s perspective, although the degree to which they provide a “natural” setting conducive to discussion appears to be debatable. Weaknesses of focus groups lie in the logistical difficulties of recruiting and assembling groups, the lack of researcher control, and in the interpretation and analysis of results. 24 “Assessment involves getting potential baseline information about available inputs, what needs and interests people have, current involvement, attitudes and reactions to leisure or a particular situation, and an assessment of what knowledge, skills, aspirations, and attitudes now exist” (Henderson, 1995, p. 52). The focus group interview method is appropriate for recreation needs assessments because focus group interviews: 1) access participant perspectives, 2) activate forgotten details through group interaction, 3) avoid forcing participants to answer preconceived questions and choose from limited options, 4) mirror the natural decision making process which involves outside influences, 5) provide the opportunity for indepth explanations and details, 6) are well suited to noncontroversial topics such as recreation, 7) allow for the modification of questions as unanticipated topics arise, 8) provide personal contact that improves response for typically underrepresented groups such as African Americans, 9) open public relations channels, and 10) may be used in subsequent design of a quantitative instrument. In light of the challenges presented to traditional methods by diverse populations, the foundations of focus group interviews in social sciences, the support for focus group interviews as a stand-alone method, and its 25 prior inclusion in leisure research, consideration of the focus group interview as the primary data gathering method for public recreation needs assessments is merited. Problem Traditionally, parks and recreation agencies have selected from among mailed surveys, telephone interviews and personal interviews in collecting data for recreation needs assessments. The focus group interview has been identified as a useful, self—contained data collection method for social science research (Morgan, 1988) and appropriate for public sector needs assessments (Krueger, 1994). Yet, the method is rarely used as the primary data collection method in recreation needs assessments. Its effectiveness in this role remains largely untested. Objectives The objectives of this study are: 1.) to conduct a recreation needs assessment using focus group interviews as the primary data gathering method. 2.) to identify criteria for evaluating the focus group interview method based on the goals of the parks agency using the findings for planning purposes and generally accepted social science research standards. 26 3.) to evaluate the focus group interview method along the criteria identified. 4.) to provide recommendations for the effective use of the focus group interview method in recreation needs assessments. 27 CHAPTER 3 Procedures Two sets of procedures are explained in this chapter. First, the procedures used to conduct the needs assessment are described including sampling, recruiting focus groups, conducting focus group interviews, coding data, analyzing data, and reporting data. Second, the procedures used to evaluate the focus group interview method used in the needs assessment are described including criteria as determined by the goals of the agency sponsoring the study, and criteria as determined by generally accepted social science research standards. Needs Assessment Procedures The needs assessment was conducted to determine the recreation needs of the residents of Ingham County for the Ingham County Parks Department. Ingham County is located in the center of Michigan’s lower peninsula. The county may be characterized as the home of Michigan’s state capital in Lansing, Michigan State University-—the state’s largest institution of higher learning, and General Motors Corporation manufacturing (Huisjen, 1999). A majority (65%) of the county’s 281,912 residents live in East Lansing and Lansing (U.S. Department of Commerce, Economics and 28 Statistics Administration, Bureau of the Census, [U.S. Census], 1992). The Ingham County parks system is comprised of over 1,000 acres of parkland in five parks located primarily in the northwest corner of the county near the county’s population centers. The size of parks ranges from 40 acre Baldwin Park to 540 acre William Burchfield Park and Riverbend Nature Center. System facilities include softball fields, volleyball courts, basketball courts, playgrounds, picnic shelters, horseshoe pits, swimming beaches, snack bars, fishing ponds, canoe rentals, pedal boat rentals, sledding hills, ice skating rinks, cross-country ski trails, hiking trails, and a nature day camp. A soccer complex is scheduled to open in 2000 (Ingham County Parks Department, 1999). Sampling Purposive and snowball sampling techniques were employed in the selection of pre—existing and constructed groups along a sampling matrix (Figure 1). The vertical axis of the matrix comprised a list of communities located throughout the county. The horizontal axis comprised a selection of socioeconomic and special interest characteristics. There was no attempt made to fill all of 29 Altar Society Loeation/ Recreation Age Ethnicity Gender OtherJ Population Distribution Lansing Big Brothers/ Big Big Brothers/ Big Sisters Sisters Friends of Fenner St. Mary's Cathedral Police Athletic Police Athletic Police Athletic League League League Greater Lansing Labor Council NAACP Center for Independent Living East Lansing MSU Outing Club MSU Outing Club ASMSU Outdoor retailers East Lansing Seniors‘ Commission Julian Samora Julian Samora Institute Institute Stockbridge Stockbridge Township Supervisors Webberville Webberville Parks Advisory Committee Leslie Leslie High School students Leslie Lions Club Mason Mason High School students Mason Rotary Okemos Meridian Senior Center Williamston Williamston City Council Dansville Dansville Board of Dansvillc Board of Education Education Countywide Park professionals bicyclists Boy Scout leaders Boy Scout leaders Boy Scout leaders Ingham County Commissioners Retired and Senior Volunteer Program Figure 1. Sampling matrix of focus groups conducted. 3Consists of environmental interests, units of government, park professionals and other community leadership groups. 30 the resulting cells of the matrix, but to insure an even distribution. The geographic locations and characteristics that formed the basis of the matrix were determined by the Ingham County Parks Board, a nine—member appointed body. These locations and characteristics were chosen by the parks board to include persons in the study that were ‘unrepresented in previous attempts to obtain citizen input such as public hearings (R. Moore, personal communication, July 28, 1999). Ingham County Parks Board members and park staff were encouraged to suggest existing groups and constructed groups of individuals that were homogeneous along the sampling characteristics of the matrix. Published lists of organizations were also utilized. Groups declining participation were asked to recommend groups with similar characteristics. Twenty-six groups were interviewed (Figure 1). Twenty-three of the groups were pre-existing, three were constructed (outdoor retailers, park professionals, and Boy Scout leaders). Pre—existing groups were selected for several reasons in addition to homogeneity along sampling characteristics. First, pre—existing groups tend to have a pre—established 31 comfort level among members that is not present among strangers. Second, focus group interviews were conducted in conjunction with that group’s regularly scheduled meeting at their regular meeting place which reduced logistical difficulties for participants. Third, the use of pre- existing groups provided an opportunity for agency outreach. Traditionally, focus groups are intentionally made up of persons unknown to each other but sharing a given characteristic (i.e., constructed groups). According to Morgan (1988), the risks in using groups composed of friends include: 1) “taken-for-granted assumptions” among friends that can hide information, and 2) topics tacitly deemed unsuitable for discussion (p.48). Krueger (1994) warned that: 1) people who interact socially on a regular basis may be responding to past experiences instead of the current topic, 2)”familiarity tends to inhibit disclosure” (p.18), 3) superior-subordinate relationships, such as supervisor- employee, inhibit discussion, and 4) “pre-established lines of communications" among colleagues or friends makes interpretation difficult (p.87). Yet, he reported the uncertainty among researchers of the necessity and practicality of recruiting groups of strangers, particularly in community-based research where it may be virtually 32 impossible for participants to be unacquainted with each other. He suggested that the effects of the familiarity of participants is an issue for analysis (Krueger, 1994). Wicks and Norman (1996) conducted a pre—quantitative focus group study to determine more effective ways to survey low-income African Americans. They selected pre—existing groups, “to ensure that each group would be homogeneous and compatible” (Stewart & Shamdasani, 1990 cited in Wicks & Norman, 1996, p.5). Wicks and Norman (1996) also contend that less energy needs to be devoted to “group maintenance” in homogeneous groups, particularly if the topic under study is a sensitive issue (p.5). A list of six groups was assembled according to the sampling matrix for the first wave of recruiting. Recruiting was done in several waves so that an even distribution of groups in the matrix would result. . . E 3 Initial contact was via telephone, when necessary, to determine or confirm the name, address, and telephone number of the contact person for the selected group. Both the Ingham County Parks Department and Michigan State University were identified as the sponsor and researcher, respectively, in all calls and correspondence associated with the study. A 33 personalized letter on University letterhead was sent explaining the purpose of the study, questions to be asked, the amount of time required for the discussion, and the approximate date the contact person could expect a follow—up call from a member of the research team (Appendix A). Each letter was signed by each researcher. Follow-up calls were placed within two to four business days after expected receipt of the letter (Appendix B). Confirmation calls were placed to groups agreeing to participate one to two days prior to the scheduled meeting. Groups were not offered, nor did they receive, compensation. Sixty-one groups were contacted. Groups declined to participate for a variety of reasons. Some believed their meetings were already too long so that the addition of the focus group interview would reduce participation below the ten person minimum suggested in the recruitment letter. Some declined to participate because they believed their membership did not visit Ingham County parks. Recruiters assured group leaders that the views of nonusers of parks were sought which persuaded some groups to reconsider. Some groups did not give reasons for not participating, stating simply that they were not interested. None of the groups 34 stated that they declined because incentives were not offered. c r i w Focus group interviews took place immediately before, after, or within a group’s regular meeting or event at the group’s meeting location, with the exception of constructed groups. Constructed groups met in various conference room settings that were convenient to participants. Each participant was asked to complete a self— administered written questionnaire designed to gather socioeconomic data and self—reported county park use estimates (Appendix C). Participants also were asked if they wished to receive further county park information via mail and if they wanted to be contacted for an informational county park presentation to any groups to which they belonged. The socioeconomic and residency data were used to determine the representativeness of the sample. Information request data were gathered to enhance outreach efforts by the Ingham County Parks Department. Participant names and addresses of those requesting information were recorded separately to retain individual anonymity (Appendix C). Focus group interview questions were designed to gather data according to the goals of the Ingham County Parks 35 Board, Michigan Department of Natural Resources funding guidelines, and generally accepted focus group methods (Appendix D). Five main discussion questions were asked regarding outdoor recreation participation, where that recreation took place, reasons for visiting or not visiting Ingham County parks, attributes desired in Ingham County parks, and support for or opposition to a list of suggested attributes. In addition, probe questions were asked as needed to clarify a response or encourage further, more indepth discussion. The research team arranged the sequence of the discussion questions to guide the participants from general, easy to answer questions to specific, thought provoking issues. This approach eased groups into more difficult discussions only after rapport had been established and participants became fully engaged in the topic. Interviews were limited to one hour in length. Audio recordings were made and transcribed. Transcripts were prepared by the Michigan State University Office Services Department using word processing computer software. A moderator team approach was used for the majority of interviews. An experienced lead moderator identified county parks for participants using photographs and a county map 36 prior to questioning. The lead moderator and one or two apprentice moderators asked questions. The lead moderator asked the majority of questions and directed the flow of discussion. Apprentice moderators took notes, distributed and collected questionnaires, arranged seating and handled other logistical details. Informational brochures were distributed after the discussion. Moderators were provided with an on-site tour of the Ingham County parks by the parks director. They familiarized themselves with the park literature and local recreation issues. 0 ' D ta Socioeconomic, residency, park usage, and information request data, gathered via the self—administered written questionnaire were coded and entered into Statistical Package for the Social Sciences (SPSS), release 8.0.0, computer software (SPSS, Inc., 1998). Because a fairly structured interview guide was used, the focus group interview data coding system was based on the guide. A global code was established for each question in the guide. Four person codes (self, family, client, and others) were established to distinguish self—reports from informant reports. Place codes indicated if the response 37 referred to county parks or parks, in general. Response codes indicated the substance of the idea expressed in the response. Responses that clearly and distinctly expressed recurring themes were assigned a quote code (Appendix E). Responses were coded using combinations of the codes described above. Combinations differed based on the nature of the question (Appendix E). For example, if in response to question 1, a participant said, “I play golf”, the response would receive three codes: the global code for question 1 “recreation participation”; the person code “self” because it is a self-report; and the response code “golf” because that is the substance of the idea expressed. A place code would not have been assigned because a place was not indicated by the participant. Each transcript was coded twice. The first coding was performed independently. The second coding was an evaluation of the first coding. Additions and changes identified in the second coding were incorporated into the final coding. Differences in interpretation were resolved by the two coders working together to achieve a consensus. Coded transcripts were entered into Ethnograph v5.02 computer software for the analysis of text-based data (Qualis Research, 1998) (Figure 2). 38 Focus Group 18 Julian Samora Research 1 Institute January 22, 1999 2 F1: Okay, the first question that we 4 have is, what kinds of outdoor 5 recreation do you engage in, and it 6 doesn't have to be just in these parks 7 or really in any parks but what kinds 8 of things do you do for recreation 9 outdoors? 10 !—RBCPART l-SELF !-BIKE !-WALK R: Bike riding, walking, swimming. 12 I F1: Riding, would that be... 14 R: Bike riding. 16 F1: Okay. 18 #-RECPART #-SBLF #-CONCERT R: We attend outdoor concerts at times 20 -# the summer. 21 —# #-RBCPART #-SELF #-SOCIALIZE #-PLAYGROUND R: Take kids to the park. Play on the 23 -# swings. 24 -# #-RECPART #-SELF #-SLED R: Do tobogganing some. I've got kids 26 —# I get into a lot of that. 27 —# #-RECPART #-SBLP #-SLED R: Yeah, the wintertime, sledding and 29 —# stuff. 30 —# F1: Okay. What about you guys? 32 i-RRCPART #-SELP #-FISH #-SPORTS BASKETBALL R: Yeah, fishing, pretty much, a little 34 -# bit, sports, baseball, basketball. 35 —# Figure 2. Page of coded transcript. 39 l-SWIM #-BASBBALL' #- W Using SPSS, frequencies and percentages were calculated for the socioeconomic, residency, park usage, and information request data. Percentages for socioeconomic and residency data were compared with respective categories of the 1990 U.S. Census for Ingham County. As discussed in the literature review, data generated using focus group interviews present challenges in interpretation of results. Morgan (1988) describes two approaches to focus group data analysis: “ethnographic", relying on direct quotation, and “content analysis”, producing numerical descriptions of data. Recognizing the strength in combining the two approaches, he stated, “Thus a largely ethnographic approach may benefit from a systematic tallying of one or two key topics, while a basically quantitative summary of the data is improved immensely by including quotes that demonstrate the points being made” (Morgan, 1988, p.64). Krueger (1994) cautioned against the use of numbers in analyzing focus group results. “Numbers sometimes convey the impression that results can be projected to a population, and this is not within the capabilities of qualitative research procedures” (Krueger, 1994, pp.154—155). He listed 40 seven considerations in analysis, including the frequency and extensiveness of comments and warned against equating frequency and extensiveness with importance. He suggested finding the “big ideas” that “emerge from an accumulation of the evidence” (Krueger, 1994, pp.150—151). As recommended by Morgan (1988), qualitative data from focus group interviews were analyzed using content analysis. Frequencies of response were counted and ranked for each discussion question (Appendix F). For the top five ranked responses, frequencies by focus group and percentage across groups were calculated to identify the origin and distribution of the most frequently given responses. As suggested by both Morgan (1988) and Krueger (1994), major concepts also were identified by the research team using the ethnographic approach. Major concepts were determined by their tendency to surface in response to different discussion questions. A narrative summary of each major concept was illustrated with quotations from focus group interview participants. In addition, frequencies were recalculated along response codes regardless of question code providing content analysis for major concepts (Appendix G). 41 R ' ta Reports to the parks board were made on a monthly basis throughout the course of the study. The parks board’s input was sought including suggestions for discussion and questionnaire questions, recommendations of groups to be recruited, decisions on topics for analysis and analysis formats, and the content and format of the final report. As suggested by Morgan (1988), decisions about how to report findings were largely already made in the choice of an approach to analysis. Two approaches were used to report the results. First, ranked frequency counts were reported, organized by discussion question (Appendix F). Second, major concepts as identified by the research team and requested by parks board members were described in short narratives, verified by frequency counts and illustrated by direct quotations from focus group interview transcripts (Knap & Propst, 1999 and Appendix G). v i r r Objective 2 of this study is to identify criteria for evaluating the focus group interview method. In a discussion of evaluation research, Babbie (1998) stated, “[Olne of the biggest problems faced by researchers is getting people to agree on criteria of success and failure" (p.25). The 42 criteria used in this evaluation had two origins: (1) the goals of the Ingham County recreation needs assessment and (2) generally accepted social science research standards. Several sources were consulted to determine the goals of the assessment. Yoder et al., (1995) recommended that the agency’s “expressed and unexpressed desires for planning should be gathered. This will come from a variety of sources including minutes of meetings, letters to other agencies, notes from within the agency, or personal conversations inside and outside the agency" (p.31). The first set of criteria was compiled from the Ingham County Parks Department’s Request for Proposal #30-98 (Ingham County Parks Department, 1998b), Michigan Department of Natural Resources guidelines, Community Recreation Planning (Michigan Department of Natural Resources [MDNR], 1999), and the Ingham County Parks Board (Ingham County Parks Board, 1998) . ‘ . e.‘ .‘ ‘ 11.!‘0. .0 ‘ Cal 0f he I'Qell 0- Criterion 1. To provide “a barometer of current and future demands for recreational facilities and services" (Ingham County Parks Department, 1998b). 43 Criterion 2. To fulfill the requirements of the state funding application process (Ingham County Parks Department, 1998b). 1 er' n To limit the assessment to the residents of Ingham County (Ingham County Parks Department, 1998b). Criterio 4. “To conduct a needs assessment in such a way that the diversity of the county, in terms of socioeconomic characteristics and geographical distribution, will be represented" (Ingham County Parks Department, 1998b). Criterion 5. To “hold costs and time requirements to a minimum” (Ingham County Parks Department, 1998b). Criterion 6. To provide “more than just a wish list” of facilities (Ingham County Parks Board meeting, 1998). 44 Criterion 7. To establish contacts with constituency groups to further “outreach” efforts (Ingham County Parks Board, 1998). The criteria identified by the Ingham County Parks Department are among the criteria of a well conducted community needs assessment according to recreation research literature. In generalized terms these criteria are: identification of current and future desires (Hudson & Witt, 1984, p. 15), meeting funding requirements (Yoder et al., 1995, p.31), identification of the population under study (Howard & Crompton, 1984, pp.35—36), representativeness of the population by the sample (Howard & Crompton, 1984, p.35; Hudson & Witt, 1984, p.22), cost effectiveness (Hudson, 1988, p.4; Hudson & Witt, 1984, p.14), and expansion of community outreach (Hudson & Witt, 1984, p. 14). The second set of criteria was based on generally accepted social science research standards. These criteria are important in performing credible social science research. The combination of both sets of criteria resulted in an evaluation that was both pragmatic and rigorous. Babbie (1998) offered a list of questions consumers of 45 social science research should ask. The following criteria were selected and adapted from that list. - f2 -_ I- -.:i -. o en-r.l A -9ted ocial S i-n f Subjects and observations should be selected in such a way as to provide a broad overview of the phenomenon under study to determine if the researcher has paid special attention to deviant or disconfirming cases (Babbie, 1998, p.462). Criterio 9. Measures obtained should be reliable and valid (Babbie, 1998, p.463). Tests for reliability of focus group data were not found in the literature on focus group methods. However, Babbie (1998) recommended certain practices may be followed to guard against researcher unreliability. These practices include: 1) a coding system agreed upon by coders, 2) multiple independent coding, and 3) resolution and re- evaluation of disagreements in coding (Babbie, 1998, p.133). Tests for validity were similarly absent from focus group literature. Krueger (1994) argued that because, “Validity is the degree to which the procedure really 46 measures what it proposes to measure” that “Focus groups are valid if they are used carefully for a problem that is suitable for focus group inquiry” (Italics is in the original.) (Krueger, 1994, p.31). “Typically focus groups have high face validity, which is due in large part to the believability of comments from participants” and because “The decision maker, when confronted with focus group results, may find explanations that seem infinitely reasonable, explanations that have come directly from the clients and not from secondhand summaries” (Krueger, 1994, p.32). Criterion 10. Conclusions reported should be based on actual findings (Babbie, 1998, p. 464). In Chapter 4, Results and Discussion, the focus group interview method is evaluated along each criterion (Objective 3). The results of the recreation needs assessment (Objective 1) are presented within the context of the criteria. For a fuller discussion of the recreation needs assessment, see Knap and Propst, 1999. 47 CHAPTER 4 Results and Discussion In this chapter, each criterion described in Chapter 3 is restated along with an operational definition. Then the method is evaluated along each criterion. The data used in the evaluation were gathered using both the self— administered questionnaire and the focus group interview. Evaloetion along Criteria besed on the Goale of the Ingham n R r ion N d Assessmen ; . . E To provide “a barometer of current and future demands‘iku' recreational facilities and services” (Ingham County Parks Department, 1998b). Jubenville (1976) described the importance of identifying needs in the outdoor recreation planning process. “It is imperative that we properly allocate our limited fiscal and natural resources for recreation, using ‘ The word "needs" has been used as a substitute for the word "demands" to conform to “needs assessment” terminology and because "demands" may have unintended economic connotations for some readers. The word "demands" appears in the original request for proposal as a goal for the recreation needs assessment. The word “needs” is not intended to infer basic human needs commonly referred to in psychology literature. 48 some priority system. Generally, the priority system should conform to identified leisure needs of the potential user population. . .” (Italics is in the original.)(p.65). To provide measures of current and future needs for recreational facilities and services, responses to three questions were examined (Appendix D). Discussion question 3 asked participants why they did or did not recreate in Ingham County parks at present. Responses to this question as they related to facilities and services were used to determine if a measure of current needs was provided. Discussion question 4 asked participants to name facilities and services they thought the county should provide in the future. Question 5 also was concerned with the future, asking participants which facilities they would support from a list of suggestions. Responses to questions 4 and 5, as they related to facilities and services, were used to determine if a measure of future needs was provided. Only some reasons for visiting Ingham County parks were related to facilities and services. The following reasons were facilities and services related and were given five or more times: playground play, picnicking, availability of picnic pavilions, canoeing, cross—country skiing, well- maintained parks, swimming, walking, observing nature, 49 enjoying the beach (excluding swimming), trails, boating (excluding canoeing), and the availability of varied activities in the parks (Table 22, Appendix F). Only some reasons given for not visiting Ingham County parks were related to facilities and services. The following reasons were facilities and services related and were given five or more times: poor water quality of swimming lakes, a lack of activities available in parks, and the presence of jetskis on park waters (Table 23, Appendix F). Thirty—three facilities and services needed in the future in Ingham County parks were named. For brevity, only the top ten ranked by frequency of response appear here. They were: land acquisition, connecting parks with trails/greenways, bicycle trails (not including mountain bike trails), nature areas, youth programming, hiking trails, skateboard park, clean lake water for swimming, minimal development in parks, and more access to natural waterways (e.g., lakes, rivers)(Table 26, Appendix F). Participants had the opportunity to express their support for or opposition to facilities presented to them on the list of suggestions (Appendix D, Question 5). Respondents expressed support for twenty-one facilities and services in response to a list of suggestions. The list 50 contained ten items, three of which were open—ended (additional trails, please specify trail type; other athletic facilities, please describe; other, please describe) accounting for more than ten types of responses (Appendix D). For brevity, only the top ten ranked by frequency of response appear here. They were: paved paths, land acquisition, water park, campgrounds, skateboard park, soccer fields, softball/baseball fields, hiking trails (not including paved paths), well—maintained sports facilities/groomed trails and tent campgrounds (Table 27, Appendix F). Facilities opposed by participants and named five or more times were: water park, soccer fields, skateboard park, campgrounds, paved paths, and land acquisition (Table 28, Appendix F). Evalaation of Criterion l. A “barometer” of current and future needs for recreation facilities and services was provided using the focus group interview method. Two indicators were identified. First, the variety of responses given indicated thorough coverage of the topic. One hundred twenty different facilities and services were named in response to discussion questions 3, 4 and 5. A recreation needs assessment for Kent 51 County, using a self—administrated questionnaire, provided forty responses for participants to choose from in response to similar questions (Nelson & Johnson, 1997). Second, the range of frequencies provided a numerical description of the data which may be used to prioritize management and planning actions (The number of times a given facility or service was named ranged from one to forty—six.). So, while there is no objective standard for the variety and number of responses that determines the effectiveness of a method, the depth and breadth of response achieved seemed to provide an appropriate measure of current and future needs. . . g To fulfill the requirements of the state funding application process (Ingham County Parks Department, 1998b). The Michigan Department of Natural Resources (Michigan Department of Natural Resources [MDNR], 1999) requires a community recreation plan for grant eligibility. The planning process must include citizen input. Solicitation of Citizen input may take many forms. MDNR (1999) suggests public hearings, surveys, workshops, informational letters and flyers, and telephone calls. Whatever form citizen input takes, two requirements must be satisfied. To satisfy the 52 firsst requirement, the method must involve the public in the prcxzess of creating a community recreation plan including: in diversity in representation, 2) inclusion of underrepresented groups, and 3) use of multiple media for public notification. To satisfy the second requirement, the data collected must provide insights into the “recreation priorities” of the community to maintain and improve recreation opportunities (MDNR, 1999, p.7). W. The MDNR (1999) requires representation by “diverse segments” of the population and recommends involving “segments of the population whose concerns are often overlooked” (MDNR, 1999, p.7 and p.3). Previous efforts by Ingham County solicited public input through of public hearings. These efforts tended to be dominated by persons living near the parks or special interest groups (R. Moore, personal communication, July 28, 1999). The purposive sampling method used in the Ingham County recreation needs assessment employed a sampling matrix for selection and recruitment of groups. The representativeness of the sample was monitored and adjusted throughout the course of the study to insure diversity along those variables listed in the matrix (Figure 1, p.30). This sampling method met the 53 i “1' 1". .111. . I I—h YEA 8“ .4. “A '5‘- “v ‘ MDNR requirement of diversity and followed its recommendation to involve “overlooked” segments. Focus group interviews are not listed among MDNR suggested methods for public notification (MDNR, 1999, p.7) probably due to their infrequent use in recreation needs assessments. The data collected in the Ingham County recreation needs assessment satisfied and exceeded MDNR requirements for determining community recreation priorities. Discussion questions gathered data on participant priorities as required. In addition, data 'were gathered on outdoor recreation participation, reasons for visiting parks, and suggested media for distribution of park information. Written questionnaire items collected data on frequency of park use and socioeconomic characteristics, also in excess of MDNR requirements. This recreation needs assessment met and exceeded MDNR re