INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand corner and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.
Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.
University Microfilms International A Bell & Howell Information Company 300 North Zeeb Road. Ann Arbor, Ml 48106-1346 USA 313/761-4700 800/521-0600
Order Number 9219046
Perceived structure of the environmental/conservation organization market
Wright, Pamela A., Ph.D.
The Ohio State University, 1992
UMI 300 N. Zeeb Rd. Ann Arbor, MI 48106
PERCEIVED STRUCTURE OF THE
ENVIRONMENTAL/CONSERVATION ORGANIZATION MARKET
DISSERTATION
Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University
By
Pamela A. Wright
*****
The Ohio State University
1992
Dissertation Committee: Approved By
Dr. Robert E. Roth Dr. Roger D. Blackwell j5m ~r Dr. Donald W. Floyd A dviser Dr. Gary W. Mullins Interdisciplinary Program Dr. Emmalou Norland ACKNOWLEDGMENTS
Pursuing a one-of-a-kind Ph.D. program required an extraordinarily cooperative and team- oriented committee from inception through the completion of the dissertation. I was very fortunate to have the five most cooperative and professional faculty members who allowed me to work in an interdisciplinary fashion and pursue a path of my own choosing, at my own pace.
Thanks to Dr. Robert Roth who, as chairman, kept me on track and was the architect for what would have otherwise been a hodge podge of educational explorations. Dr. Roth was able to understand my goals for this program and helped me plan and build an excellent educational opportunity.
Thanks to Dr. Gary Mullins, a true Appalachian intellectual, for bringing me down from the
Great White North, and providing me endless opportunities to experience and learn firsthand about the federal estate in the U.S. Gary has provided me with a host of funding and teaching opportunities, and has been a model of how to get things accomplished in academia.
In addition, he has always been a source of information and guidance but most importantly, moral support and friendship. In the spirit of free trade, I hope we will continue to work and recreate together.
Thanks to Dr. Roger Blackwell my marketing professor, for giving me an orientation sadly needed in our field. The courses I had with him were phenomenal and I appreciate his efforts to focus my interests through individual studies and special group project topics. Dr.
Blackwell continues to challenge me to look at what things mean from both a theoretical and an applied perspective. Thanks to Dr. Emmalou Norland for turning me on to research and for providing me with the skills to do it, from instrument design, to explorations in strange statistical techniques, to 1 educational evaluation practices. Most of all, thank you Emmalou for being my personal role model, and for single-handedly proving that women CAN survive, prosper, and excel within an agricultural college.
Special thanks to Dr. Donald Floyd, for always being there and challenging me to reach out and expand my thinking. Although our areas of focus and interest are somewhat different, he has the ability to think and work across disciplinary areas that I not only needed but enjoyed.
In addition he provided me with a valuable chance to learn how to guide others through the research process. Most importantly, I thank Donald for his friendship and confidence in me as a researcher.
Thanks must go to National Family Opinion, and specifically Dave Boutelle, for providing me with access to their panel of respondents. The instrument design assistance, speedy responses, and excellent quality of data has made this study possible.
Extra special thanks to fellow student Tom Pierre for endless hours of data coding, library searches, and miscellaneous other errands. Most of all, however, thank you, Tom, for five years of wonderful friendship.
Thanks to my fellow graduate students, to numerous to name, within the School of Natural
Resources who generated ideas, reviewed instruments, and provided moral support throughout my education. In particular, thanks to Sue Thomas for the hours of data entry and
Mark McLeod for his insight, proding, and friendship.
Finally, to my husband, Dr. John Hanna, thanks for the day to day menial tasks including data coding, and proofreading; but especially for his guidance, wisdom, support, and love that helped me have the endurance to stick to my studies and have the confidence to pursue my goals.
iii VITA
October 22,1964 ...... Born — Bracebridge, Ontario
1987 ...... B.S. — Lakehead University, Thunder Bay, Ontario
1987 ...... H.B.O.R. — Lakehead University, Thunder Bay, Ontario
1989 ...... M.S. — Parks and Recreation Planning, The Ohio State University
1992—Present ...... Assistant Professor, School of Resource and Environmental Management, Simon Fraser University, British Columbia
PUBLICATIONS
Wright, Pamela A., and Donald W. Floyd. Merging Research and Environmental Education — Environmental Knowledge Among College Students. Tournal of Environmental Education [in review].
W right, Pamela A., and Donald W. Floyd. 1990. Recruiting Natural Resource Students — Some Pertinent Gender and Racial Differences at Ohio State University. Women in N atural Resources. Vol. 12(2):31-34.
Wright, Pamela A., Gary W. Mullins, and Michael Watson. 1989. Market Segmentation of Interpretive Participants to National Park Service Sites. Research in Interpretation. Vol. 4(2):27-34.
Hanna, John W., and Pamela A. Wright. 1987. Self-Guided Trail Analysis. Outdoor Recreation Research journal. Vol. 1(1).
FIELDS OF STUDY
Major Field .Interdisciplinary Program Natural Resource Communications M arketing TABLE OF CONTENTS
ACKNOWLEDGEMENTS...... i i VITA ...... iv LIST OF TABLES...... v ii LIST OF FIGURES ...... xi
CHAPTER
I. INTRODUCTION
Nature of the Problem ...... 1 Theoretical Framework ...... 4 Problem Statement ...... 11 Research Questions ...... 11 Status of Research in the Field ...... 12 Definitions of Terms ...... 13 Limitations of Study ...... 17 Delimitations of Study ...... 17 Basic Assumptions ...... 17
II. LITERATURE REVIEW
Interest Group Theory and Voluntary Associations ...... 19 Characteristics of Environmentalists ...... 27 Consumer Behavior and Decision Making ...... 34 Perceptual Mapping Methods ...... 42 Market Segmentation ...... 47 Sum m ary ...... 49
III. METHODS
Introduction ...... 50 Selection of Organizations ...... 51 Population and Sample ...... 53 Data Collection Techniques and Instrument Development ...... 55 Reliability and Validity ...... 59 Data Analysis ...... 60
v IV. RESULTS
Response Rate and Nonresponse Error ...... 63 Quality of the Data...... 66 Question One ...... 67 Question Two ...... 92 Question Three ...... 128 Question Four ...... 140 Question Five ...... 161
V. DISCUSSION, CONCLUSIONS, & RECOMMENDATIONS
Discussion Question One ...... 187 Question Three ...... 198 Question Four ...... 203 Question Five ...... 207 Question Two ...... 215 Conclusions ...... 227 Recommendations and Implications ...... 232
APPENDICES
A G lossary ...... 239 B Screening Questionnaire ...... 244 C NFO Panel Background Questionnaire ...... 246 D Survey Questionnaire ...... 251
REFERENCES...... 257
vi UST OF TABLES
TABLE PAGE
1 Top Ten Environmental/Conservation Organizations ...... 53 Based on Membership.
2 Sampling and Response Rates ...... 64
3 Frequency Distribution for Demographic V ariables...... 70
4 Environmentally Concerned Citizens Compared to the U.S. Population ...... 73
5 Frequency Distribution for Social Class ...... 75
6 Frequency Distribution for Environmentally Friendly Behaviors ...... 77
7 Frequency Distribution for Recalled Organizations ...... 78
8 Frequency Distribution for Membership in Organizations ...... 79
9 Frequency Distribution for Method of Learning About Organizations ...... 80
10 Frequency Distribution of Environmental Organizational Activities ...... 81
11 Frequency Distribution for Donations ...... 82
12 Frequency Distribution for Group Membership ...... 83
13 Frequency Distribution for Most Preferred Organizations ...... 84
14 Frequency Distribution for Reasons Why Organizations Were Preferred ...... 85
15 Ratings of Importance for Product, Service, and Opportunity Preferences ...... 86
16 Frequency Distribution for Characteristics of an Ideal Organization ...... 87
17 Frequency Distribution for Organizations Respondents Would Consider Joining.. .88
18 Average Rating for Statements About Environmental/Conservation ...... 89 Organizations TABLE PAGE
19 Frequency Distribution for Appropriate Actions for Solving ...... 90 Environmental Problems
20 Frequency Distribution for Self-Description Labels ...... 91
21 Summary of Significant Relationships Between ...... 93 Members and Nonmembers on Selected Independent Variables
22 Mean Score Differences Between Members and Nonmembers o f ...... 96 Environmental/Conservation Organizations.
23 Independent Variables Utilized in the Discriminant Function for Member ship...98
24 Classification Results on Holdout Sample for Membership ...... 99
25 Summary of Significant Relationships Between Extremely ...... 100 Versus Very Environmentally Concerned Respondents and Selected Independent Variables
26 Mean Score Differences Between Extremely and Very Environmentally ...... 102 Concerned Respondents of Environmental/Conservation Organizations.
27 Independent Variables Utilized in the Discriminant Function ...... 104 for Environmental Concern.
28 Classification Results on Holdout Sample for Membership ...... 104
29 Summary of Significant Relationships Between Joiners and ...... 106 Non-Joiners and Selected Independent Variables.
30 Mean Score Differences Between Joiners and Non-Joiners of ...... 107 Environmental/Conservation Organizations.
31 Independent Variables Utilized in the Discriminant Function for Joiners ...... 109
32 Classification Results on Holdout Sample for Joining Behavior...... 109
33 Summary of Significant Relationships Between Donors an d ...... 112 Non-Donors and Selected Independent Variables.
34 Differences Between Donators and Non-Donators o f ...... 113 Environmental/Conservation Organizations.
35 Independent Variables Utilized in the Discriminant Function for Donations 114
36 Classification Results on Holdout Sample for Donations ...... 115
37 Summary of Significant Relationships Between Social Class ...... 118 and Selected IndependentVariables. TABLE PAGE
38 Oneway Analysis of Variance Between Social Class Categories ...... 119 and Selected Independent Variables.
39 Regression of Social Class on Selected Independent Variables ...... 121
40 Summary of Significant Relationships Between Length of Interest ...... 122 in the Environment and Selected Independent Variables.
41 Oneway Analysis of Variance Between Length of Interest Categories ...... 125 and Selected Independent Variables.
42 Regression of Length of Interest on Selected Independent Variables ...... 127
43 Incentive Factors ...... 130
44 Mean Score Differences on Incentives for Participating in ...... 132 Environmental and Conservation Organizations by Factor.
45 Mean Score Differences on Incentives for Participating in ...... 135 Environmental and Conservation Organizations by Segment.
46 Oneway Analysis of Variance Between Social Class and Incentives ...... 137
47 Oneway Analysis of Variance Between Length of Interest and ...... 138 Incentives.
48 Factor Analysis for Product, Service, and Opportunity Items ...... 142
49 Reliability of Product and Opportunity Factors ...... 143
50 Mean Score Differences Between Respondents Product, Service ...... 145 and Opportunity Needs.
51 Oneway Analysis of Variance Between Length of Interest and ...... 147 Product, Service, and Opportunity Factors.
52 Factor Analysis for Characteristics of an Ideal Organization ...... 148
53 Mean Score Differences Between Respondents Ideal Characteristics ...... 150 of Environmental/Conservation Organizations.
54 Summary of Significant Relationships Between Characteristics ...... 151 of an Ideal Environmental/Conservation Organization.
55 Oneway Analysis of Variance Between Social Class an d ...... 152 Characteristics of an Ideal Organization.
56 Oneway Analysis of Variance Between Length of Interest ...... 153 and Characteristics of an Ideal Organization.
ix TABLE PAGE
57 Factor Analysis for Appropriate Actions for Solving ...... 155 Environmental Problems.
58 Mean Score Differences Between Respondents Actions Appropriate ...... 157 for Solving Environmental Problems.
59 Summary of Significant Relationships Between Lobbying as an ...... 158 Appropriate Action by Market Segment.
60 Oneway Analysis of Variance Between Social Class and Appropriate ...... 159 Actions for Solving Environmental Problems.
61 Oneway Analysis of Variance Between Length of Interest and ...... 160 Appropriate Actions for Solving Environmental Problems.
62 Factor Analysis for Organizations Respondents Would Consider Joining ...... 163
63 Mean Score Differences of Organizations Respondents Would ...... 165 Consider Joining by Market Segment.
64 Summary of Significant Relationships of Organizations ...... 166 Respondents Would Consider Joining.
65 Rank Ordered Organizational Preferences by Market Segments ...... 186
66 Rank Ordered Choice Sets for Environmental/Conservation Organizations ...... 192
67 Classification and Prediction of Market Segment Variables ...... 217
68 Variables Included in Discriminant Functions and Regression Equations ...... 218
x UST OF FIGURES
FIGURE PAGE
1 A Model of Purchase and its Outcomes ...... 7
2 Choice Sets Model ...... 9
3 Typology of Voluntary Associations ...... 24
4 Basic Components of the Alternative Evaluation Process ...... 38
5 Preferred Bases for Market Segmentation ...... 48
6 Members Organizational Similarities Maps ...... 170
7 Nonmembers Organizational Similarities Maps ...... 172
8 Joiners Organizational Similarities Maps ...... 174
9 Non-Joiners Organizational Similarities Maps ...... 176
10 Donors Organizational Similarities Maps ...... 178
11 Non-Donors Organizational Similarities Maps ...... 180
12 Extremely Environmentally Concerned Organizational Similarities Maps ...... 182
13 Very Environmentally Concerned Organizational Similarities Maps ...... 184 CHAPTER I
INTRODUCTION
Nature of the Problem
One of the most persistent social movements in American history is the conservation/environmental movement. Like other social movements, its evolution is characterized by periods of growth, entrenchment, stagnation, confrontation, and division.
Throughout its development, an array of voluntary citizen action groups have formed to address a variety of specific issues based on differing political ideologies, practices, and value systems. It is this array of conservation and environmental organizations that were the subject of this study.
Samuel Hays (1987), provides one of the better analyses of the emerging diversity of environmental organizations in Beauty, Health, and Permanence. He divides this social movement into two broad divisions: the conservation era and the environmental era. While many see the latter as a direct outgrowth of the former, Hays suggests that the two eras are fundamentally distinct. The conservation era (1890's — 1940's) was the outgrowth of desires for progress and efficiency. Rather than a movement by the masses, this era was the result of leadership from the applied scientific disciplines who became active in professional and political circles to support rational planning to promote efficient development and use of all natural resources (Hays, 1987). This was a movement from the top down based on a commodity view of society and not the result of widely held assumptions and values about the natural resources.
1 2
Some of the earliest citizen-based organizations that emerged from this era began as sportsmen or recreational clubs. Two such organizations are the Izaak Walton League, that spread to the U.S. from Great Britain in the mid-1800', and the Boone and Crockett Club started by Theodore Roosevelt in 1877.
While the transition from the conservation to environmental era has most frequently been attributed to singular events such as the 1962 publishing of Rachel Carson's Silent Spring
(1962), the 1969 Santa Barbara oil blowout, or the original Earth Day celebration on April 22 in 1970; Hays attributes the 'birth' of the environmental era to earlier changes resulting from postwar recovery in the mid-1940's. Hays views the foundation of the environmental era as a change away from the commodity driven progressive era to the consumption era and to growing concerns about quality of life and amenity values. Unlike the conservation era, the environmental movement was the result of widespread, popular, public values from the middle levels of society that spread outward. It was all a part of the quest for a better life associated with home, community, and leisure. At times, the values associated with this emerging concern for "environmental quality" came in sharp conflict with the "efficient management" views of organizations formed in the conservation era.
Most environmental organizations were organized locally to deal with community problems.
Those who looked at the broader spectrum of environmental quality formed the base of the national organizations. The organizations that emerged gravitated toward specific issues associated with timely concerns. The 1960's saw rapid growth in those organizations concerned with amenity values like the Sierra Club, The Wilderness Society, The National
Parks and Conservation Association, and the National Wildlife Federation. In the 1970's, beginning with the passage of the National Environmental Policy Act (NEPA) and the creation of the Environmental Protection Agency (EPA), a number of organizations formed to react to specific NEPA related assessments. Litigation-focused organizations like the Natural
Resource Defense Council (NRDC), the Environmental Defense Fund (EDF), and the League of
Conservation Voters (LCV) were created to defend environmental interests through lobbying 3 and litigation. Other groups organized around specific issues such as Defenders of W ildlife
(DW) and Friends of Animals (FA) to address growing values for nonconsumptive uses of wildlife, National Toxics Campaign (NTC) for hazardous waste concerns, and Zero
Population Growth (ZPG) over concerns about overpopulation and limits to growth.
The decline in environmental enthusiasm after the original Earth Day brought changes to environmental organizations (Morrison, 1980). Membership growth slowed, and small independent and local groups, especially those based on student membership, faltered. Larger, older, national-based organizations consolidated (Morrison, 1980). While growth slowed, the major environmental groups remained healthy in terms of membership numbers and resources
(Mitchell and Davies, 1978). Orientation of organizations began to change at this time as well. Coalition building and political lobbying emerged as techniques and support of soft, alternative technologies grew after the 1973 oil crisis (Devall, 1980: Morrison, 1980).
The Reagan administration's handling of environmental issues during the early 1980's, major natural resource catastrophe's such as the Exxon Valdez oil spill in 1989, and the celebration of the twentieth anniversary of Earth Day in 1990, fueled the regrowth of national environmental organizations and the creation of a variety of splinter organizations.
Today, there is increasing public awareness of, and concern for, the environment (Americans for the Environment et al, 1989). In 1988, a Gallup Poll found that 84% of Americans were concerned about the pollution of fresh water, 73% were concerned with air pollution, 83% were concerned with ocean dumping, and 60% were concerned with solid waste disposal (Americans for the Environment et al, 1989).
This increased awareness and concern for the environment has continued to fuel the growth of the environmental movement, a movement based largely on voluntary participation in environmental organizations. Unlike most voluntary citizen actions that become social movements, the environmental movement's evolutionary process has not coalesced or focused 4 but has diversified and today encompasses a variety of different viewpoints.
The environmental organizations that have survived this evolutionary process represent this diversity. There are organizations that are wilderness preservation-oriented (The
Wilderness Society), site specific (Save the Bay), education-oriented (National Wildlife
Federation), litigation-oriented (Environmental Defense Fund and Natural Resource Defense
Council), and action-oriented (Earth! First! and Greenpeace). These organizations have memberships ranging from a few hundred (Save the Bay) to over five million (National
Wildlife Federation). As of 1984, there were over 12,000 environmental and conservation groups in the U.S. with 250 new ones being organized each year.
Today, given this proliferation of organizations, choosing to become a member of an environmental group is a challenge. So challenging in fact that Seventh Generation, a corporation that markets environmentally friendly products, has produced the Field Guide to More Than 100 Environmental Groups. Similarly, the National Wildlife Federation's annual Conservation Directory lists over 400 organizations in addition to issue specific citizen's groups and government agencies involved in the environment. Listings like the Field
Guide and Conservation Directory are neither widely available, nor detailed enough to guide the citizen in the decision making process. Thus, the individual citizen is faced with a conflicting array of environmental organizations hawking their causes, products, and needs.
These organizations depend on the support of the environmentally-oriented public and thus must work diligently to market their products to that public.
Theoretical Framework
Marketing principles suggest that for a business to be successful, its products must be positioned distinctly. While positioning a product may involve tangible product differentiation, what is equally more important "is not what you do to a product. Positioning is what you do to the mind of the prospect" (Ries and Trout, 1982). Product positioning involves determining what place the product occupies in the market place and how consumers rank a given product against its competitors on various evaluative dimensions (Wind and Robinson, 1972). This ranking, or development of a perceived market structure, is, however, a mental process and involves the formation of subjective perceptions and preferences. The formation of these perceptions and preferences are part of the consumer decision-making process.
Therefore, understanding how potential consumers make purchase decisions provides the foundation for discovering how environmental organizations can more effectively position themselves in the marketplace. The theoretical framework for understanding this process involves a variety of concepts from marketing, psychology, sociology, and political science.
The foundation for this framework, a marketing-based consumer decision-making model, is enriched with interest group theory, choice sets models, and the influence of perception and preferences. While the literature related to these areas is discussed in detail in Chapter Two, these concept areas are introduced briefly below.
Consumer Decision Making
Consumer decision-making models attempt to portray how consumers make purchase decisions. Models of consumer behavior have been proposed from a variety of orientations including: economic (Brown and Deaton, 1972), attitudinal (Rosenberg, 1953; Fishbein, 1967), stochastic (Massy et al, 1970), perceptual (Green and Rao, 1972), integrative (Nicosia, 1966;
Howard and Sheth, 1969; Engel, Kollat, and Blackwell, 1973), and simulation (Cook and
Mermiter, 1971). While all propose some elements useful to understanding the consumer decision-making process, the integrative model of Engel, Kollat, and Blackwell (EKB) shows the most utility. In its most basic form, the model consists of five primary steps: 6
1. Need Recognition
2. Search
3. Alternative Evaluation
4. Purchase
5. Outcomes
In the first phase, the potential consumer perceives a difference between his/her actual and desired state of affairs. These recognized needs, usually a combination of tangible and hedonic needs, motivate the consumer in the search for information. If the consumer is knowledgeable or experienced, internal search mechanisms may be followed. If the consumer is inexperienced, the search may be external. In the alternative evaluation stage, consumers set a series of evaluative criteria, determine the alternatives they are going to examine, and then assess the performance of these alternatives relative to the criteria. These criteria are determined by the selection of either tangible or hedonic salient dimensions and attributes of products. The consumer then makes a purchase decision for the desired alternative or an acceptable substitute. Finally, the consumer evaluates whether or not the chosen alternative meets needs and expectations.
These five steps are the central decision-making process and are affected, on an individual consumer basis, by environmental differences, individual differences, and psychological processes (see figure 1). Need Recognition EnWronmental Influence* Internal/ External Search
Alternative Evaluation Individual Difference*
Purchase
Outcomes
Source: Engel, Blackwell and Miniard, 1991.
Figure 1. A Model of Purchase and its Outcomes
Interest Group Theory
Theories on the development of voluntary action groups and why people join groups help us to
understand the need recognition portion of the consumer decision-making model within the
context of purchasing an environmental organization membership (Lowi, 1979). There are
three major interest group theories that merit consideration: pluralism, critical (usually
described as elite theory), and interest group liberalism. Pluralism suggests that people group
themselves according to economic interests and that interest groups compete with one another
to "seek programs and policies from government" (Garson, 1978). Critical interest group
theorists suggest that corporation and governments are controlled by the powerful elites and
that interest groups must conform to the predominant values fostered by those elites. Those
who do not conform have little power. The third interest group theory, interest group liberalism (IGL) emerged as a middle ground between the "idealism of pluralism and the pessimism of critical theory" (Grunig, 1989). Proponents of IGL say interest groups arose out of activist traditions that were rooted in community and church and are issue-based because members respond "on the basis of deeply held beliefs about what is right" (Tesh, 1984, p.30).
Depending on the interest group theory, the motivations or incentives for joining groups vary from pursuing common interests, to securing collective goods, to coercive incentives, to a sense of political efficacy (Clay and Wilson, 1961; Wilson, 1973). Classifications of these varying motivations, developed by interest group theorists, can be easily grouped into the tangible and hedonic need classifications from decision-making theory. Tangible needs for joining interest groups include material incentives such as the reward of money, objects, or services.
From the context of environmental organizations this may include a magazine, tax write-off, or access to merchandise. Hedonic needs parallel the interest group motivations based on solidary incentives and purposive incentives. These may include social or status feelings of belonging to a group or the satisfaction of having contributed to a worthwhile cause. The specific type of incentive and combination of incentives varies based on the individual although Moe (1980) found that, in general, material incentives predominated in interest groups formed for economic reasons and solidary and purposive incentives motivated noneconomic groups.
Once needs are recognized, the potential consumer embarks upon a search for products that will fulfill felt needs, and the determination of alternative evaluation strategies for comparing these products. The development of choice sets, groups of products to be considered, becomes an integral part of these two steps.
Choice Set Models
Models of choice sets are based on the assumption that as consumers undergo the process of choosing alternatives, they develop, at each step of the way, a set of possible product or brand choices (Narayana and Markin, 1975; Spiggle and Sewall, 1987). The total set includes
all possible products or brands; those of which the consumer is aware constitute the awareness
set and all others form the unawareness set. The awareness set itself is composed of the
evoked set (those that are actively considered), the inert set (those of which the consumer is
aware but is neutral towards), and the inept set (those of which the consumer is aware but
rejects from purchase consideration because of negative evaluation). An understanding of
consumer's choice sets is useful because it can be used to determine the relative
competitiveness, and the perceptual similarity of products or brands (see figure 2).
Total Set
Unawareness Set Awareness Set
Inert Set Evoked Set Inept Set
Inaction Set Action Set Reject Set
Source: Spiggle and Sewall, 1987.
Figure 2. Choice Sets Models 10
Perceptions and Preferences
The recognition of needs, the construction of choice sets, and the evaluation of alternative products are all influenced by individual perceptions and preferences. Perceptions have been defined as the "complex process by which people select, organize, and interpret sensory stimulation into a meaningful and coherent picture of the world" (Berelson and Steiner, 1963, p. 88). Preferences refer to the degree to which this stimuli is "considered acceptable or adequate to meet individual perceptions" (Bergier, 1981, p.151). For example, during the alternative evaluation process, consumers compare products or brands in their choice sets based on a number of attributes or product characteristics. It is important to realize that the attributes consumers use are not necessarily, and are often not the actual objective and tangible aspects of the product. The attributes that consumers identify are based on their perceptions of a product or group of products. Likewise, research in other fields suggests that perceived attributes used by consumers vary in nature, number, and importance.
Similar attribute-based research into other intangible products such as insurance policies and medical plans have shown that a combination of tangible and hedonic attributes are possible.
For example, two tangible attributes, price and length of time before coverage, may be evaluated alongside an hedonic attribute, such as perceived trustworthiness of an insurance firm. An individual's selection and comparison of attributes takes him /her through some type of perceptual exercise wherein attributes, combinations of attributes, and the fit of specific products/brands to those attributes are mapped and a market structure is perceived.
Any model of buyer behavior needs to recognize that behavior is influenced by a variety of factors including individual and environmental differences. The known gap between attitudes
(i.e. perceptions and preferences), and behaviors suggests that the study of perception and preferences should not be used as a measure or predictor of behavior (Bergier, 1981).
Perceptions and preferences are, however, partial measures of behavior; they are the middle ground between those variables or antecedents of behavior and the behavior itself. This study 11 recognizes that gap between perceptions, preferences, and behavior and does not purport to study it. Rather, this study recognizes the need to examine the perceived market structure that is constructed in the need recognition, search, and alternative evaluation stages of the consumer decision-making process and the relative influence of individual and environmental differences on that market structure.
Problem Statement
The purpose of this study was to determine the perceived structure of the environmental/conservation organization market, and to determine how this perceived market structure varies with individual and environmental differences.
Research Questions
The research questions that will guide this study are:
1. Who are environmentally concerned citizens and what is their involvement and interest in environmental/conservation organizations?
2. How do individual and environmental differences vary by market segment, and what is the best combination of variables to identify these market segments?
3. What incentives motivate people to join environmental/conservation organizations?
4. What are the characteristics of an ideal environmental/conservation organization?
5. How are environmental/conservation organizations perceived by consumers? 12
Status of Research in the Field and Potential Contribution of this Study
Marketing the environmental organization as a product requires an understanding of consumer needs and wants, the consumer decision process, and the perceptions of the products available. Relative to understanding the hedonic and tangible needs behind why people join environmental organizations, there has been little research conducted. At an individual organization level, descriptors of who joins environmental organizations have been collected by organizations such as the National Wildlife Federation, however, much of this information is unavailable due to its proprietary nature. To date, no studies have been published which help explain the alternative evaluation phase of membership purchase decision making in environmental organization interest groups. A number of studies have focused on individual and environmental differences of members of a specific environmental organization (e.g., Sierra Club: Bergier, 1981), but not in the context of consumer decision making and perceptions and preferences.
From an applied perspective, determining the perceived structure of the environmental organization market will assist organizations in:
• understanding their position relative to competitors
• understanding who joins environmental organizations and for what reasons
• determining appropriate re-positioning strategies
• uncovering the possibilities for networking among environmental organizations
• discovering possible opportunities for new product development.
From a theoretical perspective, the study of the perceived structure of the environmental organization market will:
• help define needs more clearly with respect to interest group theory and the need
recognition portion of the decision-making model in a domestic policy setting
• advance the understanding of the influence of perception on the construction of choice
sets for a social cause or "soft' product 13
• help determine what the relative importance of individual and environmental
differences is in determining the perceptual structure of the marketplace.
Definition of Terms
The following terms are defined constitutively and and operationally as they pertain to this study. Definitions of other terms can be found in Appendix A.
A ttribute
Constitutive Definition: A characteristic or property of a product; generally refers to a
characteristic that serves as an evaluative criterion during decision making (Engel,
Blackwell, and Miniard, 1990).
Operational Definition: Attributes were defined as those characteristics, both hedonic
and utilitarian, by which a respondent compared various environmental and conservation
organizations. These include both researcher-generated attributes (Q. 12-14, 18) and
respondent-generated attributes identified from the perceptual maps (Q. 9) and in an
open-ended fashion (Q. 11).
Conservationist
Constitutive Definition: An individual concerned with wise use of natural resources or use
for the greatest good for the greatest number for the longest run (Hay, 1987).
Operational Definition: An individual who is an active member of an organization
whose purpose or activities include wise use of natural resources and not just preservation
(e.g. Ducks Unlimited promotes hunting as well as habitat protection).
Consideration Set
Constitutive Definition: The set of alternatives from which choice is made (Engel,
Blackwell, and Miniard, 1990). 14
Operational Definition: The set of all organizations that respondents indicate they
would be willing to join. Respondents were provided a list of the top twenty organization
to choose from (Q. 15). An "other" option was also included.
Dimensions
Constitutive Definition: Features of a phenomenon (product, service, image, etc.). May be
both perceived (subjective) dimensions and objective dimensions (Hair, Anderson, Tatham,
1987).
Operational Definition: Dimensions are defined as the equivalent or a subset of the
attributes generated. A dimension may not be the direct equivalent of an attribute if the
respondent perceives two or more attributes to vary together along a single dimension.
Environmental/Conservation Organization
Constitutive Definition: A non-governmental social group whose members work to support
some aspect of the environmental movement.
Operational Definition: An organization whose major purpose is for environmental or
conservation causes, who offers membership privileges, is national in scope, and is a non
governmental organization.
Environmentalist
Constitutive Definition: An individual active in preservation of open space, wilderness,
animals or the restoration of clean water, air, etc. (Grolier, 1991).
Operational Definition: An individual who is an active member of an organization whose
purpose and actions promote preservation of a resource.
Evoked Set
Constitutive Definition: The set of all alternative products that the consumer remembers
without guidance (Engel, Blackwell, and Miniard, 1990). 15
Operational Definition: The unguided set of environmental and conservation
organizations that the respondent lists when asked to generate a list of all known
environmental/conservation organizations.
Hedonic Benefits
Constitutive Definition: The symbolic value of a consumption object in terms of emotional
response, sensory pleasure, daydreams, or aesthetic considerations (Engel, Blackwell, and
M iniard, 1990).
Operational Definition: The intangible benefits or an environmental or conservation
organization. These include both researcher-generated hedonic benefits (Q. 13-14),
respondent-generated benefits (Q. 11), and attributes generated from the perceptual maps
(Q. 9).
Ideal Points
Constitutive Definition: The point on a perceptual map which represents the most
preferred combination of perceived attributes (according to respondents) (Hair, Anderson,
Tatham, 1987).
Operational Definition: Respondent ideal points were determined by locating the point
(on the derived perceptual map) that represents the most preferred combination of
perceived attributes. This ideal point is generated using implicit estimation procedures.
Perception
Constitutive Definition: The act of apprehending by means of the senses or o the mind;
cognition; understanding. Immediate or intuitive recognition.
Operational Definition: The way in which a respondent views any pair of organizations
(n=45 pairs of 10 organizations) on a five point scale from one, not at all similar, to five,
extremely similar. 16
Perceptual Map
Constitutive Definition: The visual representation of a respondent's perceptions of two or
more dimensions or features (Hair, Anderson, Tatham, 1987).
Operational Definition: Perceptual maps representing the spatial distance between
environmental/conservation organizations were generated by multi-dimensional scaling
statistical procedures (Q. 9).
Preference
Constitutive Definition: A practical advantage given to one over others. The act of
preferring.
Operational Definition: Preferences for environmental and conservation organizations
were defined as the respondents rank ordered preferences where one was most preferred
and ten was least preferred.
Utilitarian Benefits
Constitutive Definition: Benefits resulting from purchase or other consumer decisions that
are objective, functional product attributes (Engel, Blackwell, and Miniard, 1990).
Operational Definition: Utilitarian benefits were defined as those tangible, material
benefits offered by an environmental/conservation organization. These include both
researcher-generated hedonic benefits (Q. 12-14), respondent-generated benefits (Q. 11),
and attributes generated from the perceptual maps (Q. 9). 17
Limitations of the Study
The following limitations should be considered when reviewing the results of this study:
• The questionnaire was completed by the NFO panel member in each household for
which comparative demographic information was available.
• Instrumentation utilized was a self-report, quantitative measurement instrument.
While the use of qualitative data-gathering techniques may have added to the richness
of the data it was beyond the scope of this study because of limited access to the
population and limited resources.
Delimitations of the Study
This study in conducted under the following known delimitations:
• Data reflect respondents sampled between November and December, 1991.
• Generalization of the study results presumes that those individuals sampled are
representative of the population of environmentally concerned citizens. A nationally-
balanced panel and non-response follow-up techniques were used to help ensure this.
Basic Assumptions
Use of multi-dimensional scaling techniques for data collections suggests the following assumptions:
• Respondents selected were aware and/or knowledgeable enough about environmental
and conservation organizations to evaluate these organizations on a variety of
dimensions.
• Respondents could make similarities comparisons between pairs of organizations.
• Spatial maps based on similarities data represent perceptions.
• All respondents may not perceive an organization to have the same dimensionality
(Hair et al, 1987). 18
• Respondents may not attach the same level of importance to a dimension (Hair et al,
1987).
• Respondents may not judge a stimulus in terms of either dimensions or levels of importance that remain stable over time.
• Instruments designed for this study accurately measured a respondent's perceptions and preferences of environmental and conservation organizations.
• Assumes that respondents are able to recall past purchasing behavior to answer some questions.
• Assumes that respondents are able to accurately describe their ideal organization.
• Assumes that the researcher is able to accurately name the attributes or dimensions of the perceptual maps. CHAPTER II
LITERATURE REVIEW
This chapter prevents a review of the literature pertaining to the theoretical framework for this study. The topics to be discussed in this chapter include: interest group theory and voluntary associations, characteristics of environmentalists, consumer behavior and decision making, perceptual mapping methods, and market segmentation.
Interest Group Theory and Voluntary Associations
Interest group theory developed from the political science literature as an attempt to explain the origin of interest groups and why people join them. A major theory of interest group formation is that people join groups to pursue common interests and because they can more effectively pursue their interest in an organized group format rather than individually.
Olson (1965), however, disagrees with this writing that the concept of a "collective good" is invalid. If interest groups seek collective goods then once a collective good is available to one it is available to all. Therefore, most individuals will find no rational reason to participate in a group to secure a collective good because someone else will do it for them. According to
Olson, four common explanations for the existence of interest groups can be found in group theory:
1. Small latent groups can organize for political activity if some member's share of the
collective good is larger than the cost of organizing the group (Olson).
19 20
2. Political activities are the secondary responsibilities of many organizations that
actively pressure the government (Olson).
3. People may be motivated to join groups by ideological altruism not rational self-
interest.
4. The group founders play a critical role in recruiting members initially to support the
prospect of future collective rewards and later retain members by introducing selective
membership benefits (Salisbury, 1975).
Olson suggests that the concept of a collective good may work minimally in small groups but that in large groups, members must be coerced into joining by providing incentives. Incentives that are often only "marginally" related to group goals. Moe (1980), disagrees, stating that
Olson's theories apply only to economic interest groups where economic motivations supersede political motivations. Interest group liberalism proponents suggest that whether or not a collective good may occur as a result of interest group membership is insignificant. Most members join to gain a "sense of political efficacy."
Since organizations require members to contribute resources such as money, participation, labor, power, information, expertise, access to target groups, and legitimation by listing of names, the organization exchanges a series of benefits or incentives to secure this commitment
(Knoke and Prensky, 1984). The incentive systems serve both to 1) develop high levels of participant commitment, and to 2) acquire sufficient member-contributed resources. In addition to the affective and normative benefits that are an inherent part of participation in an interest group (Knoke and Wright-Isak, 1981; Clark and Wilson, 1961), organizations must also offer utilitarian incentives that are "visible, valued, and utilized" (Burgess and
Conway, 1961). 21
Working from a theory of incentives or motivations, Clay and Wilson (1961), and Wilson
(1973), studied why people join interest groups. Their research also supported three groups of incentives:
Material incentives: money, things, services
Solidary incentives: intangible rewards such as "convivability of coming together,"
prestige of membership, collective status
Purposive incentives: satisfaction of having contributed to a worthwhile cause
Moe (1980), found that material incentives predominated in interest groups formed for economic reasons, and solidary and purposive incentives motivated noneconomic groups.
Many other researchers (Godwin and Mitchell, 1982; Walsh and Warland, 1983; Dennis and
Zube, 1988) separate these incentives into only two groups: expressive and instrumental incentives. Instrumental incentives are those that obtain public goods while expressive incentives can be defined as "personal benefits exclusively reserved for group members"
(Dennis and Zube, 1988).
A number of researchers support a broadening of Olson's perspective based on the strength of the two types of incentives as perceived by members (McFarland, 1976; Berry, 1977; Mitchell,
1979; Moe, 1980). Mitchell (1979), states that Olson's limited view works given the assumption of a rational, economic, perfectly informed individual. Given this never occurs, he states that environmental group members are motivated by more than just expressive incentives but also by the potential increase in instrumental benefits that will occur if they do join thus the members perceived utility of that instrumental benefit becomes an additional incentive.
Following Olson's (1965) incentives theory, voluntary association members should join groups to obtain primarily expressive benefits. Godwin and Mitchell (1982), developed measures of the strength of a number of incentives for group membership finding intrinsic (expressive) 22
rewards as predictors of group membership and activity. Walsh and Warland (1983),
determined that education explained the variance between activists (group members) and
free riders (nonmembers). Dennis and Zube (1988), in a random sample of 1500 California
members of an outdoor recreation cooperative (REI) found that instrumental incentives
appeared to be the strongest incentives for those that held membership in conservation-
oriented voluntary associations. They determined that while expressive incentives were of
value, they were not as highly regarded or as high a predictor of membership as instrumental
incentives.
Gordon and Babchuk (1959) and Jacoby and Babchuk (1963), found a dichotomy of
organizations, not just members, when the concept of instrumental and expressive incentives
were applied to groups. Hendee et al (1969), found a similar relationship and reported that
the expressive and instrumental categories provided close approximation to outdoor
recreation-related clubs and conservation clubs.
Other researchers in the marketing disciplines discuss motivations or incentives for
participation in the light of reference group influences (Engel et al, 1991). They suggest that
there are three main ways that reference groups influence consumer choice, or in this case,
involvement. Utilitarian influences are those that pressure compliance with group norms in
order to maintain group identity through a series of reward and punishments (Homans, 1961).
Value-expressive influences occur when there is a need for psychological association with a
reference group. This may result in an image that is enhanced in the eyes of others or being
identified with those who are admired or respected (Engel et al, 1991). The third form of
reference group influence, informational influences, is where consumers accept the opinions of
others as credible (Calder and Burnkrant, 1977).
Interest group theory literature goes beyond voluntary, nonprofit groups such as environmental and conservation organizations. Understanding the who and why of participation in these
types of organizations, can best be found by reviewing the voluntary association research. 23
Voluntary associations have been defined in a variety of ways including: as "spare-time participatory associations" (Sills, 1968, p. 363), as "nonprofit, nongovernmental, private groups that an individual joins by choice" (Smith and Freedman, 1972, p. 8), and as organizations in that the "majority of whose members are neither paid for participation in the organization nor physically coerced into such participation" (Smith, 1966, p. 483).
Voluntary associations fulfill numerous roles in society including roles such as:
• helping distribute power at the grass roots level
• functioning as service centers
• reinforcing important societal values
• helping implement special personal interests
• providing affectual support for individuals
• providing a vehicle for change or stabilization
• providing a setting to engage in expressive activities
(Amis and Stern, 1974; Babchuk and Booth, 1969).
Gordon and Babchuk (1959), propose a typology of voluntary associations that is based on three dimensions: incentives and purpose, accessibility, and status (see Figure 3). The function or incentive is dichotomized as; for the benefit of the individual, or for the larger good.
Accessibility relates to the qualifications necessary for joining an organization and may be nonexistent or be categorical, achievement, or talent oriented. The third dimension, the status conferring capacity has to do with the capacity of an organization to bestow prestige or to be associated with prestige which in turn accrues to its members. This typology blends well with the incentives theories of organizations. 24
High Access Low Access
High Status Low Status High Status Low Status
Instrumental Young Ku Klux Klan League of Women Lithuanian Republicans Voters Am. Council
Instrumental/Expressive Kiwanis Am. Legion Am. Sociological Alcoholics Society Anonymous
Expressive Scouts YMCA Daughters of the Natl. Negro Am. Revolution Fraternal Org.
Source: Gordon and Babchuk (1969: p. 29)
Figure 3. Typology of Voluntary Associations
Based on this typology, Gordon and Babchuk proposed the following hypotheses:
• organizations with low accessibility as a result of ascriptive criteria which are highly
evaluated will provide high status for their members.
• if membership is highly accessible and organizations espouse and implement widely
held values it will be highly ranked.
• if the means by which organizations implement its goals are controversial, it will rank
lower than one in which this is not the case.
• organizations capable of implementing stated goals will rank higher than ones not
capable.
• organizations with low accessibility of membership as a result of negatively evaluated
ascriptive criteria will be ranked low.
Environmental and conservation organizations can be found that fit into almost all categories within this typology. Organizations range from instrumental, low-access litigative environmental organizations to expressive, high-access recreation-conservation organizations. Understanding who joins voluntary associations in general assists greatly in attempting to understand who joins specific types of environmental and conservation 25 organizations.
In a review of related research, Babchuk and Booth (1969) found that affiliation in voluntary associations is related positively to social class, and to length of time in residence.
Homeowners, married persons, and men were also more likely to be members. The authors also found that while membership numbers tend to remain stable over time membership tenure is greater and turnover is lower in groups that have multiple objectives. Babchuk and Booth
(1969) also found that multiple memberships in categorically similar groups was a common phenomena.
Zimmer and Hawley (1959), found that membership in voluntary associations was higher in the central city (43.1%) than in fringe or suburb areas (24.7%). And that this difference was not due to the demographic composition of the population.
Florin, Jones and Wandersman (1986), studied black participation in voluntary associations and found that during a trend analysis from 1955-1962, blacks participated less than whites although there was a substantial increase in participation as time progressed through that study. By controlling for socioeconomic status, however, Olson (1970), found that black's participatory behavior was even higher than whites although the type of organizations participated in varied. Other researchers (Babchuk and Thompson, 1962; Olson, 1970; Hyman and Wright, 1971), found that within voluntary community organizations, blacks have been identified as participating more than any other racial and ethnic group.
Knoke (1981), and others (Olsen, 1972; Smith and Reddy, 1973), studied less tangible characteristics and found that voluntary group members have higher levels of morale, self esteem, political efficacy, and communication orientation and lower levels of alienation, apathy and social withdrawal. 26
Discrepancies between joiners and active participants in voluntary associations are substantial. Both Stallings (1973) and Turner (1970) suggest that formal membership in groups may not be the most useful way of identifying movement membership. Stallings (1973), discusses active participants in terms of group membership and suggests that membership
"means some form of overt participation" (p. 47). Turner and Killian (1972, p. 324) define membership as "a person who from time to time acts so as to advance the movement and whose action is oriented to the fact that 1) he identifies himself as a member of the group pursuing its common goal, 2) he looks to this group (movement) for direction and support in his action, and 3) he is in communication with others of this group in connection with his own action and in defining the goals and strategies of the movement." Research on formal members of the
Audubon Society and the Isaac Walton League found that only ten to twenty-five percent of local members participated in group programs (Stallings, 1973). Similarly, Faich and Gale
(1971) found that less than thirteen percent of a Sierra Club chapter's membership ever held office or participated in committee work and only eleven percent attended meetings regularly.
Out of their research Faich and Gale identified four "membership" roles: leaders, active members, inactive members, and non-members.
Reddy and Smith (1973), identified a number of predictors to help identify those who were more likely to actively participate in organizations including life cycle stage, socioeconomic status, and gender. Active participation in organizations is relatively high for adolescents and then drops only to peak again when couples have school age children. The authors found that married individuals are proportionately more active and that that participation increases with the number of children. Those with higher socioeconomic status (income, education, social class, occupational prestige, home ownership, etc.) are more likely to participate although the upper class elites tend to be less active at a grassroots level and more active in figurehead roles (Reddy and Smith, 1973). While research suggests that lower class and working class individuals are least likely to join voluntary associations, Reddy and
Smith (1973) determined that those who do are more likely to be active participants. While gender seemed to have little effect according to these researchers on amount of participation, 27 gender divisions do occur with respect to organizational type. The authors suggest that while this trend is certainly changing, traditionally women have participated more actively in charitable, health, and welfare organizations and men in economic and political organizations. Reddy and Smith (1973) also noted that while political party preference seemed to have little effect on active participation, those individuals with more orderly career patterns were more likely to be active participants.
This understanding of interest group theory and voluntary associations provides the background for the investigation of environmental and conservation organization members specifically with respect to who joins these organizations, and why they join organizations.
Characteristics of Environmentalists
Developing an understanding of who belongs to environmental organizations are might best be done by reviewing membership profiles for environmental organizations. Unfortunately, these are largely unavailable due to their proprietary nature. However, there have been a number of public studies conducted on environmentalists specifically in determining the characteristics of those who demonstrate responsible environmental behavior. This literature, and specific studies on environmental organization members is reviewed below.
Studies of specific environmental organization and inter-organizational studies of members and non-members of large, national environmental organizations have provided data useful in describing who joins environmental organizations. One of the oldest environmental organizations, the Sierra Club, has been the subject of many of these studies (Faich and Gale,
1971; Coombs, 1972; Perry, Cleveland, Gillespie, and Lutz, 1975; and Devall, 1970).
Research has suggested a number of factors that influenced participation in environmental organizations. Early experiences (Milbrath, 1984; Tanner, 1979; Peterson, 1982; and Scholl,
1983) are often cited as reasons for joining environmental organizations. Milbrath (1981), also 28 lent support to arguments that environmentally-related threats to self or family may play a significant role in influencing membership.
A great deal of environmental organization research has centered around the belief that members are reflective of the elite in society. Harry, Gale, and Hendee (1969) compared members and nonmembers of an environmental organization and discovered that conservation movement members were largely upper-middle class, members of professional occupations, urban-based, and highly educated. In 1970, Devall replicated elements of the former study and found that 74% of members had four-year degrees, 39% had graduate level work, 49% held professional positions (doctors, lawyers, and university faculty), 21% held lower professional positions (school teachers, writers, artists), 58% had annual incomes of over
$12,000 and 30% had annual incomes of over $18,000 (1970 dollars). Mitchell (1980), also discovered similar figures with respect to education (49% had some graduate level education in contrast with a national level of 7%) (see also Larson, et al. 1982). Other studies (Perry, et al, 1975; Coombs, 1972; and Faich and Gale, 1971) supported these findings with research on other populations of environmental organizational members.
Considerable debate, relative to accusations of "elitism" can be found in the literature
(McEvoy, 1971; Harry, Gale and Hendee, 1971). Tucker (in Andrews, 1980, p. 223), stated that
"[environmentalist's] major direction... has been to work against the interests of the lower middle class and the poor, and for people at the top end of the scale." In contrast, Mitchell expressed that "while the membership of environmental groups is drawn disproportionately from the college-educated, higher income segments of society, the movements supporters are quite broadly based. At all income levels, 60% or more support the movement" (1979). Recent findings from polls conducted by Harris, Tarrance, and Lake (Americans for the Environment et al, 1989), support the findings of broad-based and growing support for the environmental movement. In an investigation of the elitism campaign, a number of researchers (Buttel and
Flinn, 1978; VanLiere and Dunlap, 1980; Mohai, 1985) found that the relationship between environmental concern and socioeconomic status was weak to nonexistent. The distinction 29 comes between differentiating between environmental concern and environmental activism since there is a strong link between activism and elitism (Devall, 1970; Tucker, 1978; Buttel and Flinn, 1978; and Mitchell,1979). Additionally, other research has shown that membership in environmental organizations has never been limited to the elite (Friesema,
1975). Friesema cites cases of Native Americans invoking National Environmental Policy Act legislation to contest power plant locations, and Chicago working-class groups organizing to stop air pollution (Campaign Against Pollution - CAP).
Some organizations appeal specifically to "lower" class segments. The National Wildlife
Federation, has a strong blue collar base because of its affiliations with outdoor recreation- oriented organizations (Friesema, 1975) and has a correspondingly lower percent (22%) of members with education beyond a bachelors degree. Other organizations have made similar appeals to lower-middle class citizens. The Izaak Walton League is made up largely of the small town America, local anglers (Friesema, 1975). So, while research supports arguments for and against the case of class consciousness appeals by environmental organizations, examination of populations studied suggests that class, measured by any combination of the above variables, is a useful and persistent variable in predicting membership in specific environmental organizations. This particularly holds true for the older, larger, nationwide organizations (Devall, 1970).
In 1981, Milbrath gathered evidence that suggested that environmentally-supportive actions are more prevalent among people who are higher on Maslow's hierarchy of needs.
Specifically, environmentalists give less weight to material wealth, higher values to nature, and show greater compassion for other people and species.
With respect to political efficacy, (the feeling of potential power to influence public legislation) Milbrath (1981), found that environmental organizational members believe they can influence environmental policy in their communities more than do the general public.
Active members of the Sierra Club (Manzo and Weinstein, 1987) also scored higher on 30 political efficacy measures than non-members. This data corresponds to voluntary participation literature in sociology that suggests that voluntary group members normally exhibit much higher levels of political efficacy (Knoke, 1981).
There is some evidence to suggest that political efficacy and political orientation may be related. Mitchell (1980), reported that environmentalists are three times more likely to be strongly liberal than the general public, although Mitchell also noted that the degree of left to right orientation was difficult to measure (see also Larson, et al., 1982; and Milbrath,
1981). In contrast, Manzo and Weinstein (1987) found that active members of the Sierra Club were more conservative than less active members. Relative to this relationship between political efficacy and political orientation, Klanderman (1983) suggested that people who have minimal feelings of political efficacy may engage in actions "outside the conventional sociopolitical system." Thus, Manzo and Weinstein (1987) hypothesized that individuals concerned about the environment, who feel that working with traditional environmental organizations is futile, may be prone to join groups advocating civil disobedience or even monkey wrenching. Members of environmental organizations also express very strong attitudes about the seriousness of environmental problems (Milbrath, 1984; Dunlap and Van Liere, 1978; and Maloney and Ward, 1973).
While concern for the environment is a widespread phenomena (Americans for the
Environment, 1987), as of 1980 (Mitchell), less than one percent of the public belonged to an environmental organization. Clearly, being concerned about the environment "is not synonymous with being active" (Manzo and Weinstein, 1987). Voluntary participation research from sociology (Wright and Hyman, 1958) reports that 64% of the general public belong to no voluntary associations, 31% belong to one organization and 21% of those belong to two or more organizations. While multiple memberships are the exception in the general population, this does not hold true for members of environmental organizations. Devall
(1970), noted that 40% of Sierra Club respondents belonged to multiple conservation organizations (see also Faich and Gale, 1971). A number of studies have also examined 31 intensity of participation within environmental organizations. Truman (1951), characterizes the highly involved participants as the "active minority." Milbrath (1981) suggests that there is a substantial gap between passive members and active members. Manzo and
Weinstein (1987) also noted that more active participants were more likely to be involved in other environmental groups. In addition, Devall (1970) found that 37% of Sierra Club leadership and 21% of general membership were active in other organizations in which they held membership.
Stallings (1973), studied other beliefs of members and nonmembers of one environmental group and found that beliefs between leaders and active members were most similar and inactive members and non-members were most alike. Perceptions of membership in environmental organizations also ranges: some members view themselves as subscribers, some as members, some as supporters, some as donors, and some view the organization as a catalog shopping outlet (National Wildlife Federation, 1988).
A number of studies have investigated why people join environmental organizations.
Milbrath (1984), suggests that active members of the Sierra Club joined because of specific environmental conflicts, and because of experience of environmental harm. In contrast, Manzo and Weinstein (1987) who also studied Sierra Club members, determined that general interest in the environment was the primary reason for joining the environmental organization.
O ther research (Harry, et al, 1969; Faich and Gale, 1971; Hoesteray and Bowman, 1976) suggests that members join environmental organizations for the outdoor activities. This reason was also cited by long-term members (13+ years) in a study by Coombs (1972). In contrast,
Coombs found that short-term members joined to support conservation activities. Bartell and
St. George (1974), and Mitchell (1980), found that many new members to the Sierra Club often joined to be financial supporters lending credence to philanthropic motivations cited above. 32
Few studies on search behavior have been conducted on members of environmental organizations. Coombs (1972), discovered that 58% of new members received information that led them to join the environmental organization from a friend. Rogers and Shoemaker (1971) and Leonard-Barton (1981), also indicated that interpersonal sources, particularly friends are important for information acquisition.
Research about the dimensions and attributes of importance when comparing environmental organizations, and the perceptual models constructed out of them within the environmental organizational field are nonexistent. Informational listings of multiple organizations primarily provide statements of mission statements and topical orientation for example, acid rain, ozone protection, or wildlife protection (National Wildlife Federation, 1990; Seventh
Generation). The most common comparisons made about organizations within the environmental movement are the degree to which organizations are labeled as
"environmental" organizations versus, "conservation." Sometimes included as a third division are "outdoor" or "sportsmen's" organizations. Vary rarely, are other major categorizations applied to environmental organizations within the listings. Attribute based, or dimensional characterizations can, however, be found widely in the popular literature.
Media, organizational members, and researchers alike contrast environmental organizations on a continuum that are not always obvious nor supported by research. Hoesterey, and Brown
(1976), contrast organizations from an "outdoor recreation type organization to a recognized environmental advocacy-type organization" while Milbrath (1984) labels another set
"militant environmental organizations."
The first attempt to organize and perceptually differentiate organizations was completed by a news reporter for the Pacific News Service and published in the Rocky Mountain News
(Anderson, 1989). Although based on personal opinion, Anderson identified "four distinct wings" to the environmental movement, a movement that has diversified into "a vast range of organizations, political positions, lifestyles, cults, sects, strategies, faiths, and fanaticisms." 33
Anderson’s four wings consist of:
1. the Pols, the political activists who "see the cause chiefly in terms of public policy"
(e.g. NRDC, EDF)
2. the Greens, who "want to change society deeply, drastically and immediately, through
eco-sabotage, protest, and massive shifts in personal lifestyles" (e.g. Earth First,
Green Party)
3. the NIMBY's (Not In My Backyard), those individuals that "just want to keep their
local communities from being destroyed by development and/or pollution,"and
4. the Globals, who are involved in global scale issues and promoting the ideas of
sustainable development (e.g. World Watch Institute, Global Tomorrow Coalition).
A recent Outside Magazine editorial (Outside, 1991) also attempted to organize and perceptually differentiate twenty-five of the top environmental organizations. These organizations were arrayed on a dimension labeled "Milquetoast to Bombthrower."
On a more academic note, a number of researchers have established dimensions for environmental organizations that include: the degree of wilderness orientation (Hendee,
Catton, Marlow, and Brockman, 1968), political radicalism (Morrison, 1980; Mitchell, 1980), and alternative technology orientation (Mitchell, 1980). Milbrath (1984), developed the first actual perceptual map of environmental thoughts and issues that was comprised of multiple dimensions. Although based on personal reflection, his model consisted of three dimensions: valuation of material wealth versus valuation for a safe and clean environment, advocates versus resisters to social change, dominant social paradigm versus deep ecology/soft technology orientation.
Information on what aspects of environmental organizations contribute to feelings of satisfaction or dissatisfaction may provide insight into attributes of environmental organizations, however, little research has been conducted on the purchase of a membership and its outcomes. Manzo and Weinstein (1987), report that sources for satisfaction in 34 environmental organizations for less active members were based on the knowledge of supporting a worthwhile organization and receiving the magazine. Active members on the other hand report the social aspects of membership, the personal satisfaction from participating in environmental campaigns, and the outings programs as the sources of satisfaction.
In summary, the literature on environmentalists suggests that a number of variables show some promise in identifying members of different environmental organizations. These variables include: political efficacy, political orientation, education, income, Maslowian orientation, participation in other volunteer organizations, and degree of active membership in an organization or intensity of participation.
Consumer Behavior and Decision Making
Literature from consumer behavior and consumer decision making from marketing help us to understand how individuals make purchase decisions and how perception affects the comparison of products. The first three elements of an integrative consumer decision-making model, need recognition, search, and alternative evaluation, provide the basic foundation upon which other techniques and practices of marketing can be built.
Models of consumer behavior and decision-making are intended to replicate how consumers make specific purchase decisions (Engel, Blackwell, and Miniard, 1990). These models are useful because they:
• provide a frame of reference for research
• are useful in theory construction
• can be explanations for the behavior itself.
Most consumer research has used what can be termed "reduced forms" of models to explain specific portions of consumer behavior and decision-making (Kollat, Blackwell, and Engel, 35
1970). These "models" include a number of economic (Brown and Deaton, 1972), attitudinal
(Fishbein, 1967), and perceptual models (Green and Rao, 1972), that while representing significant constructs are limited in that they attempt to explain only one portion of consumer behavior. To truly understand the complexity of consumer buyer behavior and the specific elements of that process, an integrated and comprehensive model of consumer behavior needs to be utilized to structure research (Kollat et al, 1970). The Howard model (1963) is the first such integrative model that use a basic problem-solving process as its underlying structure.
Howard's model was the first to distinguish between variations in buyer-behavior under situations of true problem-solving, limited problem-solving, and automatic response purchases. As the first model, however, many of the linkages between variables were relatively weak and somewhat confusing to follow (Kollat et al, 1970).
The Nicosia model (1966) utilized a flow chart form designed for the advertising sector that included four fields: the firms attributes, consumers attributes, search for and evaluation of means-end relationships, the act of purchase, and the feedback stage. Once again the applicability of this model was limited by unresearched linkages between variables in addition to its relatively narrow scope of application to advertising (Engel et al, 1970). The third comprehensive model, the Engel-Kollat-Blackwell Model of buyer-behavior, moves beyond the simple "black-box" approach of the problem-solving framework and expands to include intervening variables that affect the decision making process (Engel et al, 1970).
Need recognition, the first stage in the problem-solving process, is recognized as a difference in the actual and desired state of affairs. This discrepancy, however, must be of a magnitude that it is noticed (Bruner and Pomazal, 1988). Kotler (1991), suggests that this need can be triggered by either internal or external stimuli (e.g. direct marketing). Engel and others
(Engel et al, 1991) suggests that specifically needs can be activated by: changed circumstances, product acquisition, product consumption, marketing influences, and individual differences.
Bruner and Pomazal (1988), suggest additionally that the recognized need can be broad or narrow, simple or complex. 36
A variety of factors can affect both the actual and desired state of the problem recognition phase. Bruner and Pomazal (1988), suggest that reference groups, novelty, and thinking can affect the desired state while assortment deficiency, arousal of needs, and post-purchase evaluation can affect the actual state. Factors such as finances, previous decisions, family characteristics, culture, social class, individual differences, current situation, and marketing efforts can affect both the actual and desired states. Just because a discrepancy is perceived, however, the consumer may not initiate a consumer action because they may not perceive the problem (e.g., nuclear war) as a consumer problem (Bruner and Pomazal, 1988).
Ragheb (1988), and others have studied leisure and recreation needs and motivations.
Originally, needs were thought of as instinctive however, the behavioral and cognitive literature suggests that needs are more correctly viewed as desires for arousal and stimulation
(Piaget, 1951; Maslow, 1954). Needs are often defined using Maslow's hierarchy of needs
(1943), although Ragheb suggests that too often it is hard to operationalize these needs within the context of leisure and recreation planning. Instead, Ragheb (1988) suggests utilizing Tillman's (1973) ten human needs (e.g., recognition, dominance, service...) since they are often more meaningful. Regardless of how needs are operationalized, they are numerous, expandable, they shift in priority and they vary greatly from person to person (Koontz and
O'Donnell, 1968).
The resulting activated need is expressed ultimately in buying behavior which is designed to provide expected benefits (Engel et al, 1990). These needs may be utilitarian benefits that can be expressed as objective, functional product attribute based characteristics or they may be hedonic or experiential benefits that include sensory pleasures, emotional responses, and esthetic considerations (Engel et al, 1990; Srivastava et al, 1981; Hirschman and Holbrook,
1982).
A recognized discrepancy between the actual and desired state of affairs triggers heightened attention or active searching behavior (Kotler, 1991). Kotler suggests that apart from memory 37 this information comes from three sources: commercial, public, or experiential. This search process is usually a combination of internal and external search that increases in quantity as the consumer moves from limited problem solving to extensive problem solving (Kotler, 1991).
Howard (1963), suggests that whether the search behavior involves extensive or limited problem solving is dependent on the consumers familiarity with the product class and with the alternatives as well as the importance, risk, and frequency of the purchase.
While little is known about the internal search process, most research is founded on the belief that interest and experience leads to knowledge (Bettman, 1979). Green and Morris (1969), found that those consumers interested in music gave similarity ratings of musical groups that resulted in much higher dimensional perceptual spaces than those of consumers that were not interested. They suggest that those individuals who were interested were probably also more experienced and had more information stored in a more complex form than others. Jacoby,
Chestnut, and Silberman (1977) also found that when brand names were present, hence prior knowledge was assumed, less internal search and less total search occurred.
Consumers recognizing an unmet need look for products identified through internal and external search to provide benefits. Each product provides a bundle of attributes that are an attempt to satisfy need and provide benefits (Kotler, 1991). The search and alternative evaluation phases of decision-making are often intertwined during decision making (Engel et al, 1990). While alternative evaluation differs under circumstances of limited and extended problem-solving people generally evaluate alternatives on the basis of the alternatives attributes rather than as a single entity (Lancaster, 1971; Tversky, 1972; Bettman, 1979). The basic components of this stage are portrayed in Figure 4. 38
Determine Determine Evaluative Choice Criteria Alternatives (SEARCH)
Assess Performance of Alternatives
Apply Decision Rule
Source: Engel, Blackwell, and Miniard, 1991
Figure 4. Basic Components of the Alternative Evaluation Process
The first stage of the process entails the development of a set of desired evaluative criteria or attributes to compare product alternatives. The most salient attributes are those that are most important (Kotler, 1991). Myers and Alpert (1976), suggest that the researcher should measure the importance weights that consumers attach to various attributes. Consumers attach a specific utility function to each attribute that in a sense represents a preference for each given level of each attribute (Kotler, 1991). Evaluative criteria or attributes can be both hedonic and utilitarian, however, the most common attributes are price, brand name, and country of origin for most traditional consumer purchases (Engel et al, 1990). Gabor and
Granger (1964), suggest that price may be the most fundamental criteria in traditional consumer purchase situations, however, others (Huber, Holbrook, and Kahn, 1986), suggest that it is often over-rated and that consumers rarely consider price or can even recall it. These evaluative criteria also vary by situation (Dickson, 1982) and the similarity of choice alternatives, motivation, knowledge, and involvement (Gensch and Javalgi, 1987). Just as 39 some attributes are more salient than others, and each has its own particular utility weight for consumers, different consumers utilize different decision rules to select the final attribute- product combination.
Green and Wind (1973), and Myers and Alpert (1968), have detailed a number of different decision rule strategies that consumers may follow including:
• Expectancy-value model—Consumers beliefs about a brand's attributes are multiplied
by the respective importance weights and added to derive a score.
• Ideal-brand model— Consumer compares actual brands to ideal. The closer the actual to
the ideal the more it will be preferred.
• Conjunctive model—Alternatives are evaluated by establishing minimum acceptable
attribute levels.
• Disjunctive model—A noncompensatory model in that products are considered only if
they exceed specified levels on one or a few attributes.
• Lexicographic model— Noncompensatory process where attributes are arranged in
importance and the brand superior on the most important attribute becomes the choice.
• Determinance model— An attribute may be important to the consumer but will not
influence choice if all products possess the same amount of that attribute.
Most analysis models assume that all segments of the population use the same choice model although Gensch and Javalgi (1987), suggest that it is likely that attributes are given different value weights. The authors suggest that instead of assuming all consumers use similar processes, that researchers look for segments a priori based on those consumers assumed to be using a different decision process. Variables that may be useful in this a priori segmentation are those based on research that suggests they have some influence on choice process. Howard and Sheth (1969), suggest current information, involvement, and past experience would be useful while Bettman and Park (1980) and Rothschild (1974) suggest knowledge, involvement, and perceived risk are useful. 40
Throughout the first three steps of the decision making process, the consumer constructs a series of choice sets, or brands/products that are considered at each step of the process. One of the most common choice set models in complex decision making was designed by Narayana and Markin (1975). This model consisted of the following sets:
1. Total set — all those products in the market
2. Awareness set — those that the consumer is aware of
2a. Evoked set — all products the consumer actively considers and
therefore at least initially evaluates favorably
2b. Inert set — those that the consumer is aware of but is initially
neutral to
2c. Inept set — those that the consumer is aware of but negative
tow ards
3. Unawareness set — those the consumer is unaware of.
Spiggle and Sewall (1987) in a model of retail choice elaborated on this model and added the following sets as subsets of the evoked set (for a complete picture see Figure 2 in Chapter 1):
i. Action set — all those that the consumer takes some action towards. Those not included
in the action set become part therefore of the inaction set or the reject set.
ii. Interaction set — Those that the consumer subjects him/herself to personal selling (an
equivalent for nonretail selling could be found). Those not included in the interaction
set include the quiet set and the reject set.
If the search process involves only internal memory search than the choice set for consideration will be that subset of brands recalled from memory - the retrieval set (Engel et al, 1991). If search is external then the consideration set may be larger. Spiggle and Sewall
(1987), suggest that the action and interaction steps require active or external search.
Throughout the need recognition, search, and alternative evaluation phase of the process, perception has a substantial role on influencing decision making. Ritchie (1974), states that 41 there is substantial evidence that individual differences exist in people's perception of a wide range of stimuli. Experimentally, Ritchie suggests that variables that influence perception such as measurement devices, stimulus context, the scenario, and actual stimuli can be controlled. Variables that influence perception that can't be controlled include familiarity, personal values, personality, cultural background and other socioeconomic characteristics.
Schreyer and Beaulieu (1986), call this perceptual awareness of the subjective and objective attributes of products the image. Images are "not static pictures but are processes that begin and end with varying existence" (Bugelski, 1971, p. 35). Images are affected by observations of physical characteristics of products but also by the "perceiver's wants, purposes, preferred attributes, and symbolic values" (Schreyer and Beaulieu, 1986, p. 234). Within categorization theory the authors suggest is an internal structuring system that the respondent uses to organize perceptions. The incoming information is grouped into classes or arrays (Markham,
Horton, and McLanahan, 1980) to make different subca '.egories containing different attributes useful in decision-making. Rosch (1978), suggests that based on cognitive economy the fewest number of sets of attribute categories that relate to the functions of the products will be constructed and maintained.
A number of studies of attributes related to recreation preferences have been conducted with respect to inventory schemes for land classification (Driver and Brown, 1978; Peterson, 1974;
McDonough, 1982). Manfredo (1979), related attribute preference with motives for participation while Cowley and Schreyer (1986) determined interpretive attribute preferences on a continuum of experiences from desire for stress release to desire for learning.
Two of the major criticisms of this type of research to date have been the unidimensional nature of the analysis tools used and the use of fixed lists that may or may not include the most relevant attributes (Schreyer and Beaulieu, 1986). These authors used free elicitation techniques to generate attribute groups and while they used an unsophisticated analysis technique, they were able to conclude that people of different levels of experience and commitment do not differ significantly regarding the types of attributes they identify as 42 important but those that were more experienced were more specific about the attributes and recalled higher numbers of attributes.
Knowledge of the attributes consumers use to compare the products in the environmental/conservation organization market supports Day, Shocker, and Srivastava's
(1979, p. 10) call for consumer defined product markets. They suggest that it is important to define the markets from a consumer, perspective because "people seek the benefits that products provide rather the products per se". Of the three purchaser usage behavior and the four customer judgment techniques they review, perceptual mapping techniques provide the most versatile view of this consumer-defined product-market (Green, and Carmone, 1970).
Perceptual Mapping Methods
Perceptual mapping techniques generate a "geometric representation of customers perceptions
(and preferences) of the qualities possessed by products/brands comprising a previously defined product-market" (Green and Carmone, 1970). Methods for assessing consumer perceptions and preferences center primarily around identifying ahd ranking salient attributes. A number of different methods have originated in the consumer behavior and social psychology fields including:
• conjoint measurement techniques
• direct ratings of importance
• subjective probability models
• open-ended elicitation techniques
• Thurstonian methods
• multi-dimensional scaling
(Green and Carmone, 1970; Jaccard et al, 1986; Fishbein and Ajzen, 1975; Green and Wind, 1973)
These techniques can also be divided into compositional and decompositional techniques.
Compositional techniques, such as multi-dimensional scaling, involve the use of subjective 43 attribute perceptions and values to predict attitudes or preferences (Malhotra, 1988).
Decompositional techniques, such as conjoint analysis, use evaluative judgments of products to infer the values of experimentally manipulated attribute levels (Malhotra, 1988).
Very few comparative studies have been conducted to determine the relative effectiveness of these techniques in assessing the importance of attributes. A recent study (Jaccard et al, 1986), comparing six methods, showed very little convergence between methods suggesting that consumer behavior research that relies on a single measure for inferring attribute importance may be misleading. Jaccard (1986) hypothesized that different methods may tap into different types of attributes.
A number of researchers have studied the applicability of techniques for determining market structure. Green, Tull, and Albaum (1988), suggest that perceptual maps, created by multidimensional scaling, are often too sophisticated and that other simpler techniques such as semantic differential and numerical ratings scales can be used to understand much of the market structure. While these simpler methods have been used to determine how one product compares to another, utilizing these techniques, requires that the respondent compares products on an attribute by attribute basis (Green, Tull, and Albaum, 1988).
Hauser and Koppelman (1979), suggest that research on perceptual mapping techniques have focused almost exclusively on multi-dimensional scaling similarity techniques and that the easier and less expensive techniques of factor analysis and discriminant analysis have been ignored. The researchers compared these three techniques both theoretically and using Monte
Carlo simulations. They determined that factor analysis was superior to the other two techniques based on: theory, goodness of fit, prediction of saved data, interpretation, and ease of use. One of the most important constraints they noted with MDS was that if the respondent's evoked set was relatively small, factor analysis was likely to be superior in identifying more than two dimensions. 44
Fishbein and Ajzen (1975) noted that while factor analysis is the most common method used to examine the multidimensions of a respondent's judgments, it is fundamentally flawed in that the "emerging dimensions are dependent on the particular set of items used to elicit judgments" (Fishbein and Ajzen, 1975, p. 96). In spite, however, of its possible competitive advantage in determining perceptual maps, the a priori requirement, that there is a relatively complete set of known attributes, suggests that it is not appropriate for this study.
Cluster analysis has also been suggested as a possible substitute for multidimensional scaling
(Green, Tull, and Albaum, 1988). Cluster analysis refer to a broad range of techniques often called Q-analysis, typology, classification analysis, and numerical taxonomy whose purpose it is to "separate objects into groups such that each group is more like other objects in its group than like objects outside the group" (Green, Tull, and Albaum, 1988). Similar to factor analysis, however, cluster analysis requires similarities judgments on particular known attributes.
Alpert suggests that while a variety of techniques have been developed to identify product attributes, there is no clear indication of "which ones determine purchases and preferences"
(1971, pp. 184). Instead of presuming that all are equally important, the researcher is really interested in those attributes that are determinant, or lead to product choice. Alpert conducted a study to compare the effectiveness of attribute identification techniques, including: 1) direct questioning, 2) indirect questioning, including motivation research and covariate analysis, and 3) observation and experimentation. He found that direct questioning techniques were generally more efficient than other techniques, however, they were best for straight-forward products and were costly and timely. Alpert suggests that based on the limitations he found with identifying attributes, multidimensional scaling, a technique that works backwards towards those attributes and does not rely on direct questioning, may prove
"superior to more traditional approaches" (p. 190). 45
Petty and Cacioppo (1981), suggest that if an organization discovers the preferences and perceptions for its product it has six management choices: modifying the product (real repositioning), altering the beliefs about the product (psychological repositioning), altering beliefs about a competitors product (competitive depositioning), altering the importance weights for the attributes, calling attention to neglected attributes, or shifting buyers ideals.
Multidimensional scaling determines the basic attribute dimensions consumers use in differentiating products and locates the relative position products occupy on these dimensions
(Consumer/Industrial Research Service, 1987). This provides a profile of how the product in question "measures up" to the competition.
Green and Carmone (1970), in a review of applications of multidimensional scaling in marketing found that perceptual and preference mapping has been utilized for a variety of product classes to answer questions such as:
•What are the major perceptual and evaluative dimensions of the product class?
• What existing brands are perceived as similar to what other existing brands?
• What are the major perceptual points of view among consumers?
• What new brand possibilities are suggested by the configurations of existing brands?
• How are respondent ideal points or preference vectors distributed in the various
perceptual spaces?
• How compatible are communications messages (e.g. advertising) with brand
perceptions.
Green and Carmone (1970), however, noted that multidimensional scaling procedures were
"woefully inadequate for generating, or testing, really new product ideas —those ideas that involve new dimensions outside the consumer's experience" (p. 27).
The only precondition required by multidimensional scaling is the "assumed ability of the respondents to make similarity or dissimilarity judgments" (Bergier, 1981, p. 151). Bergier also suggests that multidimensional scaling can be used not only to study the perceived 46 similarity between products but "to derive clusters of respondents who share similar perceptual structures" (p. 151), thus serving to help study preferences.
Fox suggests that the disadvantage of using multidimensional scaling is of "requiring the researcher to label the dimensions of the resulting (perceptual) map based on the relative positioning of the brands of interest" (1988, p. 47) and he suggests a technique similar to correspondence analysis entitled fundamental matrix decomposition. This technique, however, requires ranking of known attributes.
Ritchie (1975), used the individual differences model (INDSCAL) to classify leisure recreational activities into four dimensions. He used two additional pieces of information to assist in naming the dimensions: the number of times a respondent mentioned the criteria as being the basis for similarity judgments and the average rating of each activity on four bipolar scales. Other researchers (Williams, 1988; Russell and Hultsman, 1987) have used similar techniques to help identify dimensions resulting from an analysis of leisure and outdoor recreation data.
Humphreys (1982), determined that the various methods of obtaining data for multidimensional scaling comparisons do not significantly differ on test-retest reliability procedures. Neidell (1972), compared direct versus indirect measures of obtaining similarities data and found that when the number of stimuli was large it was better to use derived measures.
Malhotra (1988), suggests that metric and nonmetric methods seem to yield very similar results. Other research on validity and structural reliability of multi-dimensional scaling techniques have found that the the order of presentation of stimuli doesn't influence the similarity judgments provided by respondents (Jain and Pinson, 1976). 47
Market Segmentation
While all consumers do not use the same attributes, or ascribe the same salience to attributes when determining the perceptual structure of a market, neither are they alike on other variables. Providing a homogeneous view of the perceived structure of the market and the preferences and perceptions of that market at a macro level covers up the market opportunities that may exist within the larger market (Engel et al, 1991). Market segmentation is a technique that provides the researcher with a way to examine these differences and uncover opportunities (Frank et al, 1972).
Market segmentation is a fundamental market technique based upon the recognition that consumer needs and demands are not alike (Kotler, 1975). It could be defined as "the process of separating a market into groups of customers, prospective customers, or buying situations such that the members of each resulting group are more like the other members of that group then like members of other segments" (Bonomo and Shapiro, 1983, p. 27).
Market segmentation is not a precise science, and the selection of the appropriate variables with which to segment even less precise (MacLachlan and Johansson, 1981). Generally, market segmentation variables can be grouped into four major categories: geographic, sociodemographic, psychographic or lifestyle, and behavioral. Wind (1978), in a review of the issues and advances in market segmentation, suggests that in general, some segmentation techniques are most preferred for specific types of marketing decisions. These suggestions are outlined in Figure 5. 48
For general understanding of a market Benefits sought Product purchase Needs Brand Loyalty For positioning studies Product usage Product preference Benefits sought For new product concepts Reaction to new concepts Benefits sought For pricing decisions Price sensitivity Deal proneness Price sensitivity by purchase/usage patterns For advertising decisions Benefits sought Media usage Psychographic/life style For distribution decisions Store loyalty and patronage Benefits sought in store selection
Source: Wind, 1978, p. 320.
Figure 5. Preferred Bases for Market Segmentation
In addition, to those variable methods proposed by Wind (1978), Dickson (1982) and
Srivastava et al (1981) suggest segmentation based on usage situations. Dickson writes that this type of segmentation is particularily important when the differences in sensitivity to products and their characteristics depends on a person in a specific situation (1982).
While the choice of segmentation approaches varies greatly, Young et al (1978, p. 405), suggest only three situations when segmentation is not useful:
• When the market is so small that marketing to a portion of it is not profitable.
• When heavy users make up such a large proportion of the sales volume that they are
the only relevant targegts.
• The brand is the dominant brand in the market. 49
Wind (1978), suggests that typically most market segmentation research fits into one of two categories: 1) a priori segmentation design in which the basis for segmentation is decided in advance, or 2) clustering-based segmentation design in which segments are determined on the basis of clustering of respondents on a set of "relevant" variables. Regardless of the type of segmentation utilized and the variables selected, fundamental to the chosen segmentation technique is the concept of utility (Kotler, 1975). The resulting segments must have utility given the specific reason for segmentation, specifically segments must be: measurable, accessible, actionable, and substantial (Kotler, 1991). In addition, the variables should divide a market into homogeneous segments that respond differently to different combinations of the marketing mix (Woodside and Motes, 1981; Frank et al, 1972).
Summary
The literature reviewed in this chapter provides two foci. First it provides the reader with a conceptual basis to better explore who joins organizations, specifically environmental and conservation organizations, and why. Secondarily, it provides a foundation from the consumer behavior and decision-making fields to help structure research to determine the perceived structure of the environmental/conservation market and determine variations in that market structure based on individual and environmental differences. CHAPTER III
METHODS
Introduction
The purpose of this study was to determine the perceived structure of the environmental/conservation organization market and to determine how this perceived market structure varies on individual and environmental variables.
To address this research problem, a descriptive, correlational study was conducted using a one-shot case study design with two distinct sections. The first part involved the determination of the market structure e.g., how environmental organizations are positioned by consumers and potential consumers. For this purpose no assumptions were made of the underlying attributes, characteristics, or dimensions that potential consumers used to compare environmental organizations. To determine the market structure, multivariate interdependence methods were needed since the variables in the data set could not be partitioned into criterion or predictor (independent and dependent) variables (Hair,
Anderson, and Tatham, 1987).
The second part of the study sought to determine the relationship between consumer needs and the market structure. Based on the assumption that consumer needs and perceptions would not all be the same, a portion of this analysis also involved the construction of market segments.
While some of the interdependence techniques were useful in determining the market segments, the relationship between variables necessitated the use of dependence methods and the identification of independent and dependent variables.
50 51
This chapter describes the research design and methods used to examine the perceived market structure of environmental/conservation organizations. It addresses various issues related to methods including: selection of organizations, population and sample, data collection technique and instrument development, reliability and validity, and data analysis.
Selection of Organizations
With over 12,000 registered environmental/conservation organizations in the U.S. and limitations based on the response burden of detailed examination of only ten organizations, the selection of the environmental/conservation organizations for study involved the development of a series of criteria. To be selected, an organization had to have: a primary purpose of environmental or conservation protection, a national orientation, offer membership services, and have a large membership size.
The first criterion for selection was whether or not the organization had as its primary purpose environmental or conservation issues. While definitions vary: conservationists are often identified as proponents of wise-use of resources while environmentalists are identified as those who work for purely preservation, not consumption, goals (Hays, 1987). In some cases the major purpose of organizations is less clear with many recreation organizations having as a dual purpose recreation or enjoyment and the preservation of the resource. One organization included in this study, Ducks Unlimited, had a large recreational hunting membership but since they list their primary purpose as "establishes, promotes, assists, and contributes to conservation, restoration, and management of waterfowl habitat" (NWF, 1991) they were included in this study.
Criterion two for selection was whether or not the organization had as its primary perspective a national orientation. It was presumed that awareness of organizations would be greater for those with a national profile and membership and secondarily, regional and 52 geographic influences would be moderated. Based on the geographic origin of the membership and the environmental problem being addressed, organizations could be subdivided into four broad categories: local, regional, national, and international. A national organization was defined for this study as one which offered membership services and who advertised or recruited membership from the entire U.S. While many "national organizations" could also be classified as international organizations (e.g., World Wildlife Fund, Greenpeace), the organizations had to at least have a national membership before they were included in the study.
Criterion three for selection was that the organization must offer membership services.
While the list of possible membership services varies, for the purposes of this study the payment of annual membership dues in exchange for services such as organizational periodicals, local chapters, educational services, lobbying services, etc., were used to define
"membership" organizations. Membership services from the selected organizations varied with some offering full-color, bimonthly publications, others telephone hotlines for legislative updates and still others with local chapters that offer additional services including volunteer opportunities, local educational services, and social gatherings. All organizations selected, however, required the payment of annual membership dues in exchange for some combination of services.
Criterion four for selection was membership size. Since the original list of 12,000 organizations was not refined sufficiently by the application of the first three selection criteria, the final criteria was based on membership size. The ten largest environmental/conservation organizations were selected in order to ensure that the selection was unbiased and the organizations selected for study were relatively well known. Published statistics on membership numbers vary depending on accounting methods (e.g., some report organizational affiliates as members while others do not) and date of publication. To aid in accuracy membership numbers from the NWF Conservation Directory (the primary listing source for all environmental/conservation organizations) were cross-checked with 53 membership statistics published in annual reports and in the media. The final ten environmental/conservation organizations selected had membership numbers ranging from
250,000 to 5.8 million (NWF, 1991) (see Table 1).
Table 1
Top Ten Environmental/Conservation Organizations Based on Membership
Organization Membership
National Wildlife Federation 5,800,000 National Arbor Day Foundation 1,000,000 Greenpeace, USA* 850,000 World Wildlife Fund 670,000 The Nature Conservancy 550,000 National Audubon Society 550,000 Ducks Unlimited 550,000 Sierra Club 500,000 The Wilderness Society 300,000 Cousteau Society 250,000
* Others sources list Greenpeace with membership of 2.3 million.
Population and Sample
The target population for this study was members and nonmembers of environmental/conservation organizations in the United States during the study period.
Given the relatively low incidence (approximately five to ten percent) of environmental membership in the population (Mitchell, 1980), the target population was further refined to those who identify themselves as "very interested in the environment" and thus would constitute the population likely to purchase a membership in an environmental organization. 54
This population was used because research has shown that membership in environmental/conservation organizations is positively related to environmental concern
(Tucker, 1978; Weigel and Weigel, 1978; Dennis and Zube, 1988).
The determination of sample size was based on a number of factors: the statistical confidence in the survey results, the statistical techniques used in analysis, comparison with similar studies, and budget considerations. Utilizing a confidence-interval approach based on sample proportions to determine sample size (Green, Tull, and Albaum, 1988), a sample size of approximately 400 would have been required given an allowable error of .05, and a confidence level of 95%. For purposes of analysis, minimum cell sizes needed for comparisons of variables was influenced by the possible, although unknown at the time of sampling, representation of membership in varying environmental organizations. Based on estimates regarding cell sizes, numbers of variables, and other similar studies, a returned sample of approximately 400 would be acceptable and a sample size of approximately 1000 would allow greater latitude in analysis.
While a random national sample of individuals would be the statistically ideal way of generating the sample, the relatively low incidence of the phenomena in the population and the inability to obtain an accurate sampling frame suggested an alternate method of accessing the population. National Family Opinion (NFO), a nationally recognized market research firm that maintains a national panel of 400,000 households and nearly one million individuals, provided the sample.
NFO's sample was balanced by national census statistics in total and within each geographic division based on market size, age of head of household, annual household income and size of household. A screening of NFO panelists who have a strong interest in the environment was used to identify the sample frame. This screening questionnaire was mailed on May 15,1991 to a nationally representative sample of 5,000 households (see appendix B). On June 19, 1991, returns were closed for tabulation. At that time, there were 2,978 households returning, 55 representing approximately 60% of the total mailing. Nonrespondents were compared with respondents based on previously collected demographic profiles and no significant differences were found. Of the total returns, approximately half (1400) indicated that they were extremely or very environmentally concerned. These 1400 households constitute the sample for this study. While NFO's panels are non-random in nature, the representative nature of the sample by the act of balancing to census data suggests close proximity to a random sample thus statistics were utilized for analysis and generalizations are made to the population of environmentally concerned citizens at large.
Data Collection Techniques & Instrument Development
Perception and preference data required the completion of relatively complex instruments by a great number of individuals. Possible quantitative data collection techniques, given these constraints, include mail or telephone surveys. Telephone surveys have the advantage of allowing personal interaction and prompting to complete the instrument. Telephone surveys, however, do not allow the use of visual referencing and the number of items respondents can remember are limited, and they generally have high nonresponse rates. Visual requirements for some perception and preference techniques, coupled with the necessity of handling large quantities of data, as was the case in this study, led to the selection of a mail survey as the appropriate data collection device.
Data collection for this study consisted of an 8.5 by 14 inch, four—page booklet with a separate cover letter that was mailed to the NFO respondent in each of the 1400 households
(see Appendix D). Based on NFO's understanding of their panel and the history of NFO panel return rates, there was only one mailing of questionnaires on November 1, 1991 with returns closing on December 9,1991.
While a mail survey was utilized as the primary instrument for this study, data for analysis actually came from three sources: 56
1) Previously collected demographic data on NFO panel members (see Appendix C)
2) Data on environmental behaviors and concern from the sample screening questionnaire
(see Appendix B)
3) The mailed survey, described above, that collected information specific to
environmental/conservation organizations (see Appendix D).
The survey instrument contained a variety of questioning styles and techniques. Most questions were close-ended. Because of the unique nature of the data required for this study, previously developed instruments had limited application in survey design. General instrument construction recommendations from Research for Marketing Decisions (Green, Tull, Albaum,
1988) and Multivariate Data Analysis (Hair, Anderson, Tatham, 1987) for collecting preference and perceptual data were utilized. In addition, a number of technical articles by
Green (e.g. Green, 1970; Green and Rao, 1972; Green and Carmone, 1970; Wind and Robinson,
1972) provided guidance in instrument construction with respect to the methods of collection similarities data.
The first question on the instrument was designed to generate respondent's evoked sets of environmental/conservation organizations through an unaided recall question. This question was included at the beginning of the questionnaire so that respondents would not yet have seen a name of an organization. To assist in truthful responding, no organizational names were included in the cover letter or on the front or back of the questionnaire. The level of measurement was treated as nominal.
The next three questions (Q. 2—4) asked respondents specific about membership in environmental/conservation organizations. Respondents were asked to list the names and years of current memberships and then to indicate from a fixed list, how they decided to become a member of these organizations. Respondents were provided with an opportunity to indicate more than one reason for deciding on a membership since it was thought that some would be members of more than one organization. Thus, more than one response may have been 57 appropriate. All three of these questions were measured at the nominal level. Those respondents who were not members of environmental/conservation organizations were asked to skip to question five.
The next set of four questions (Q. 5—8) asked respondents about their activity and involvement in voluntary organizations. The first question asked respondents to indicate what environmental organizational activities they had participated in. This list was generated from previous survey work by Grunig (1989) and supplemented after consultation with environmental/conservation organization staff persons. Question 6 and 7 were taken from a National Wildlife Federation unpublished survey and asked about donations to environmental/conservation organizations and length of interest in the environment. The last question in this section asked respondents to indicate if they belonged to any other voluntary organizations.
The next section of the questionnaire was designed to examine perceptions and preferences.
Similarities data for determining perceptual differences in environmental/conservation organizations was obtained using an ordinal level of comparison of all possible pairs (45) of the ten study organizations. In this question (Q. 9), respondents were asked to indicate how similar or dissimilar the pairs were. The data were treated as ordinal level for use with non metric multidimensional scaling. Question ten asked respondents to rank, from l=most preferred, to 10=least preferred, the list of ten organizations. Data was assumed to be ordinal.
The last question in this section asked respondents to indicate what they liked best about their most preferred organizations. This was an open-ended question in which respondents were asked to be as specific as possible.
The next two questions (Q. 12 and 13) asked respondents to rate a number items on a five point
Likert-type scale. Respondents were asked to indicate how important each of sixteen product, service, and opportunity items were that could be provided by environmental/conservation organizations. This list was generated based on promotional literature advertising 58 organizational services. An attempt was made to balance the items with half of the items tangible and the other half intangibles. Data was assumed to be ordinal.
Questions 14 and 15 asked respondents to indicate, from fixed lists, what characteristics an ideal environmental/conservation organization would possess and what actions were appropriate for solving environmental problems. In both questions respondents were asked to check all items that applied. Data was assumed to be nominal.
Question fifteen asked respondents to check all organizations that they would consider joining. A list of twenty organizations, plus an "other" category were provided. This list of organizations consisted of the ten study organizations and ten additional organizations selected for high membership levels and representation of subject areas. The Conservation
Directory (NWF, 1991), and an Outside editorial review (1991) of the top twenty-five organizations was used to help in this selection.
Question sixteen asked respondents to indicate on a Likert-type scale how much the respondents felt each of the ten study organizations possessed three characteristics: education-orientation, political-orientation, and protest-orientation. This question was added in an attempt to force respondents into evaluating the similarity of organizations based on specific attributes. Because, this question was added late in the instrument design process, space was limited and it was presented in a matrix style. Data for this question were assumed to be ordinal.
Question seventeen was designed to measure a respondents motivation, based on incentives theory, for participation in environmental/conservation organizations. The question consisted of twelve statements adapted from Dennis and Zube (1988). Respondents were asked to indicate their agreement or disagreement with these items based on a five point Likert-type scale. The question consisted of two subscales designed to be summated, thus, the data was assumed to be interval-level. 59
The final three questions asked respondents to select from fixed lists, the appropriate descriptors for where they grew up, where they currently live, and what self-labels they would ascribe to themselves.
Reliability and Validity
A test of reliability determines the consistency of the measuring instrument in repeated application. Two tests of reliability were used in this study: Cronbach's alpha, and test- retest.
In its original form, Dennis and Zube's (1988) computed reliability coefficients of .79 and .72 for the instrumental and expressive incentives scales. To determine reliability of the remainder of the questions, a test-retest procedure using a one-week delay was conducted on a sample of fifteen individuals representative of the population. The resulting reliability coefficient (coefficient of stability) was .84.
Tests of validity are designed to show that an instrument measures what it purports to measure (internal validity) and the extent to which the findings can be generalized to other situations (external validity).
Internal validity was tested using a review of the instrument by a panel of ten experts for content and face validity. In its original form (Dennis and Zube, 1988), two factors were identified based on the twelve incentive items with factor loadings ranging from .352 to .773.
Non-Response Error and External Validity
Two procedures were used to minimize non-response error. Early and late respondents were compared and demographic characteristics of respondents were compared to non-respondents since data on both had been previously collected. Results are provided in Chapter Four. 60
Data Analysis
The survey data were analyzed using the Statistical Package for the Social Sciences (SPSS)
Macintosh microcomputer program (Norusis, 1990). An alpha level of .05 was established a priori for all statistical analysis. Data analysis utilized the following: 1) descriptive statistics, 2) validity and reliability scale measures, and 3) tests of the research questions.
Descriptive statistics were used to provide frequency distribution of variable, to check the data for abnormal responses, to observe trends, and to determine the assumptions of normality and homogeneity of variance in the distributions. Reliability of the two continuous scaled variables (instrumental and expressive incentives) was investigated using Cronbach's alpha and exploratory factor analysis was used to investigate the scale alignment with theorized factor dimensions. The research questions were analyzed using a variety of statistics including the multivariate techniques of multi-dimensional scaling, factor analysis, discriminant analysis, and multiple regression analysis. Specific techniques are detailed below as they relate to the research questions.
Question two utilized four analytical techniques: chi-square tests, student's t-tests, discriminant analysis, and multiple regression analysis.
Chi-square tests for independence measure significant relationships between categorical level variables. Crosstabulations between two nominal level variables were computed and a significance test with an alpha a priori of .05 was utilized to determine significance. Given large sample sizes and reported similarities under this condition between the Pearson chi square statistic and the likelihood ratio statistic, the Pearson chi-square statistic was utilized for interpretation of independence (Norusis, 1990). Because of the influence of sample size and degrees of freedom on chi square results, chi square summary tables include the sample size, degrees of freedom, significance level, and calculated chi square value. 61
Student's t-tests were used to measure significant differences in group means between continuous variables. Independent t-tests were used in situations where the mean scores of two groups of individuals were being compared and paired sample t-tests were utilized where mean scores for the same respondents were paired. Separate variance estimates were utilized because of differences in group means and standard deviations. All tests were conducted as two-tailed tests with a significance level of .05.
Discriminant analysis was utilized to help determine the best set of variables that distinguish among groups and to help develop a formula for predicting group membership for new cases whose group membership is unknown (Norusis, 1990; Hair, Anderson, and Tatham,
1987). The SPSS procedure, DISCRIMINANT, was used to help identify the categorical dependent variables upon which market segments were divided. A random sample of half of each group was used to calculate the discriminant function coefficient of the linear combination of independent variables. This stepwise discriminant analysis procedure used a variable selection criteria based on minimization of Wilk's lambda. The maximum number of steps in the analysis were twice the number of independent variables. Holdout samples were used to classify unknown cases based on the calculated discriminant function. The overall percentage of cases classified correctly was the sum of cases classified correctly in each group divided by the total number of cases (Norusis, 1990). Prior to computation, dummy variables were created for all categorical (nominal) data, to provide the appropriate level of data for analysis.
Multiple regression analysis was used in the same manner as discriminant analysis when the dependent variable for market segmentation was continuous. Stepwise regression using the
SPSS procedure REGRESSION, was utilized with a significance level of .05 set a priori for entry into the model, and a level of .10 used for extraction from the model. This stepwise procedure identified the best predictor (independent variable) that was regressed on the dependent variables and next, the second best predictor was added to the equation until all independent variables were entered (Norusis, 1990). Prior to computation, dummy variables 62 were created tor all categorical (nominal) data to provide the appropriate level of data for analysis.
Questions three, four, and five utilized factor analysis to help identify the underlying factors or constructs contained within a set of responses. Factor analysis refers to a broad set of interdependence techniques in which all variables are simultaneously studied (Hair,
Anderson, and Tatham, 1987). While factor analysis can perform many functions, it was used in this situation to identify a latent set of dimensions in a larger set of variables to be used in subsequent analysis. The SPSS procedure FACTOR was utilized with the following parameters: common factor analysis with orthogonal varimax rotation (Norusis, 1990).
Eigenvalues of greater than 1.0 were used to select the factors used in interpretation and factor loading scores of .50 or greater were used to determine which items were extracted for each factor (Hair, Anderson, and Tatham, 1987).
Question five examined respondents perceptions of the ten study organizations through the use of non-metric multidimensional scaling (MDS). MDS was utilized to answer the following questions:
• What dimensions (attributes) do respondents use when evaluating
organizations?
• How many dimensions are used?
• What is the relative importance of each dimensions?
• How do environmental organizations relate perceptually?
The MDS procedure, ALSCAL, a Euclidean distance model, was utilized by the SPSS program to analyze the perception data (Norusis, 1990). The data used in this analysis were ordinal
(hence nonmetric MDS), square, symmetric matrix, and conditional data (Young and Lewyckyj,
1979). Three measures of fit were utilized to confirm MDS maps. These include: the S-stress measure ranging from 1 (worst possible fit) to 0 (perfect fit), the KruskaTs stress measure, and the squared correlation coefficient between the data and the distances. CHAPTER IV
RESULTS
Data for this study were analyzed to address the following research questions:
• Who are environmentally concerned citizens?
• How do respondents differ by market segments?
• What incentives motivates people to join environmental conservation organizations?
• What are the characteristics of ideal environmental conservation organizations?
• How are environmental conservation organizations perceived by consumers?
Following a discussion of the response rate and quality of data, results are presented for each of these questions.
Response Rate and Nonresponse Error
Fourteen hundred questionnaires (see Appendix B) were mailed to National Family Opinion
(NFO) panel members screened to be "extremely environmentally concerned," or "very environmentally concerned." The mailing was sent on November 1, 1991, and returns were closed on December 11, 1991. During that six-week period, using only a single mailing, 1017 questionnaires were returned for a seventy-three percent response rate (see Table 2).
63 64
Table 2
Sampling and Response Rates
Group n Percent
Sam ple 1400 100.0 Accepting Sample (Respondents) 1017 72.6 Nonrespondents 383 27.4
While a response rate of seventy-three percent, given the single mailing, and the length of the questionnaire seems satisfactory, nonresponse error can be a major threat to external validity and is a serious concern in survey research. In order to ameliorate concerns of respondent bias and to demonstrate that the responding sample is representative of the population from which the sample was drawn, researchers may choose to: 1) conduct follow- up surveys with a sample of nonrespondents, 2) compare respondents to nonrespondents on the basis of strategic demographic variables, or 3) compare early to late respondents (Sudman and
Bradbum, 1985).
To ensure that nonrespondents did not differ significantly in this study, comparisons of respondents to nonrespondents and early to late respondents were made. Because demographic, and environmental behavior data had been collected earlier by NFO on all panel members, respondents and nonrespondents could be compared on key variables. In addition, research
(Miller and Smith, 1983) suggests, that late respondents are more similar to nonrespondents; thus, a comparison of early respondents (those who responded within the first week and a half of the survey period) and late respondents (those who responded in the final two weeks of the survey period) was conducted. 65
Respondents and nonrespondents were compared on the demographic variable: marital status, geographic location, income, size of household, race, level of education, occupation, employment status, and social group. Respondents and non-respondents were also compared based on environmental behavior variables including individual recycling activities, level of concern about the environment, and affect of a company's environmental record on buying behavior. Each of these tests was conducted using the chi-square test for independence with an alpha level of .05. No significant differences were found.
To determine if early and late respondents differed significantly, the 645 individuals who responded within the first week and a half of the survey were compared to the 105 individuals who responded within the final two weeks of the data collection period. These groups were also compared on those variables listed above in addition to membership in an environmental organization, whether or not they gave money to an organization, whether or not they belonged to any other citizen groups, where they lived, and where they grew up.
Each of these tests was conducted using the chi-square test for independence with an alpha level of .05. No significant differences were found.
The student's t-test was utilized to compare early and late respondents on the basis of the number of organizations they recalled, the number of organizations of which they were a member, length of interest in the environment, motives for joining environmental groups, and total number of environmental activities they participated in. No significant differences were found.
Since respondents and nonrespondents did not differ significantly on demographic or environmental behavior variables, nor were the any differences between early and late respondents, it was assumed that the results for the responding sample were representative of the population from which the sample was drawn. Therefore, we can be relatively confident that the results presented in this study are representative of the environmentally concerned citizen. 66
Quality of the Data
As is the nature of NFO panel surveys, the quality of the data on the questionnaires was excellent. Two exceptions to this generalization should be noted. Question nine was a lengthy question in which respondents were asked to compare, on a Likert-type scale, all possible pairs of ten organizations (n=45 items). Approximately fifteen percent of the sample did not fully complete this question. Given the length and complexity of this question, this proportion of nonresponse was expected. Multidimensional scaling, the analytical tool for this question, not only accepts cases with missing data, but computes similarities matrices based on group averages. Thus, those individuals whose responses were partially or completely absent for this question were represented by the group averaged similarities matrices. As a validity check, those individuals with missing data on this question were compared on the basis of the same variables utilized in the nonresponse comparisons.
Individuals who did not respond to this question were significantly different than those who responded on the basis of two variables, length of years of interest in the environment and membership in an environmental or conservation organization.
Question sixteen asked respondents to compare the ten organizations on the basis of three attributes: educational orientation, political orientation, and protest orientation. This question was added late in the questionnaire design process and it appears that many respondents either did not understand how to complete the question or felt unable to make a comparison among the ten organizations on the basis of these three attributes. Given the poor quality of data for this question, it was excluded from all analysis. 67
Question One: Who are environmentally concerned citizens and what is their involvement and interest in environmental/conservation organizations?
Various researchers suggests that environmentally concerned citizens constitute the population most likely to purchase a membership in an environmental conservation organization (Tucker, 1978; Weigel and Wright, 1978; Dennis and Zube, 1988). Therefore, developing an understanding of who environmentally concerned citizens are, will help the marketer develop specific strategies to reach this population. In addition, a national survey of environmental conservation organizations can provide valuable theoretical information on how this group differs from the population at large and answer specific questions about the issues such as "elitism" within the environmental movement (Harry, Gale, and Hendee, 1960;
McEvoy, 1971; Harry, Gale, and Hendee, 1971).
NFO panel information provided data on the following sociodemographic characteristics: size of household, marital status, social group, race, language, geographic location, market size, income, home ownership, education, occupation, and employment. In addition, respondents on the survey questionnaire were asked to indicate the rural/urban nature of the area in which they grew up and the area where they now live. Table 3 summarizes the results.
Size of Household. The average household size was 2.4 with a quarter (25.6%) of the respondents reporting only one household member and thirty-eight percent (37.8%) of the respondents reporting two household members.
Marital Status. Over half of all respondents (59.1%) reported that they were married and an additional twenty five percent (24.6%) reporting that they were either divorced, widowed, or separated. 68
Social Group. Fifty-nine percent (59.1%) of respondents reported a traditional family grouping, while twelve percent (11.5%) of respondents reported a single parent family.
Twenty-six percent (25.6%) reported that they lived alone.
Race. Ninety-two percent (92.0%) of respondents reported that they were white with the next largest group (5.0%) reporting that they were black.
Origin. Only two percent of respondents (1.7%) reported that they were of Hispanic origin.
Seven percent (7.1%) provided no answer for this question.
Geographic Location. The geographic location of respondents was grouped into nine standard categories. The majority of respondents came from the mid-Atlantic (17.3%), east north- central (17.1%), the south Atlantic (16.7%), or the Pacific (15.8%) regions.
Market Size. Thirty-six (36.4%) of all respondents came from metropolitan areas with a population of two million or greater with the next largest group (24.7%) coming from metropolitan areas under 50,000.
Income. Reported annual family incomes ranged from less than $7,500 to greater than
$175,000. Over half of all respondents reported annual family incomes greater than twenty- five thousand dollars and twenty-four percent (23.8%) reported family incomes between
$12,500 and $24,999.
Home Ownership. Seventy-three percent of respondents (73.3%) reported owning their home while an additional twenty percent (19.7%) reported renting.
Education. Educational information was collected for both the male and female heads of household. On average, the educational level of adult males with at least some college education (60.0%) was greater than those of females with at least some college education 69
(54.9%).
Occupation. The grouped occupation category including retired individuals, students, and armed forces personnel, accounted for the occupation of thirty-seven percent (37.0%) of the males and fifty-one percent (51.5%) of the females. The next largest occupational group for both males (27.9%) and females (23.4%) was in managerial or professional occupations.
Employment. Level of employment was also collected for both the male and female heads of household. Thirty-six percent of females (36.1%) and sixty-three percent (63.4%) of males reported full-time employment. Women were more likely to be employed part-time (16.0%) than men (4.8%).
Age. Ages for both males and females were similar with the largest percent (30.9% and 28.9% respectively) between the ages of 35 and 49. The next largest age group (29.8% and 28.6% respectively) were those over sixty-five.
Rural/Urban Residence. When asked where they grew up, almost forty percent (39.4%) of respondents indicated they grew up in a rural area. The next largest group (34.3%) indicated a suburban area. When asked where they currently lived, forty-four percent (43.8%) classified the area as a suburban area, thirty percent (30.4%) classified the area as a rural area, and the rem ainder (25.8%) lived in an urban area. Frequency Distribution for Demographic Variables
Value Frequency Percent Cumulative Percent*
Size of Household 1 252 25.6 25.6 2 372 37.8 63.5 3 151 15.4 78.8 4 130 13.2 92.1 5 55 5.6 97.7 6+ 23 2.3 100.0 Marital Status m arried 581 59.1 59.1 never married 160 16.3 75.4 divorced/widowed 242 24.6 100.0 Race no answer 15 1.5 1.5 w h ite 904 92.0 93.5 black 49 5.0 98.5 Asian 4 .4 98.9 other 11 1.1 100.0 Language no answer 70 7.1 7.1 Hispanic origin 17 1.7 8.9 not Hispanic 896 91.1 100.0 Geographic Location New England 52 5.3 5.3 Mid-Atlantic 170 17.3 22.6 East North-Central 168 17.1 39.7 West N orth-Central 83 8.4 48.1 South Atlantic 164 16.7 64.8 East South-Central 43 4.4 69.2 West South-Central 87 8.9 78.0 Mountain 61 6.2 84.2 Pacific 155 15.8 100.0 Market Size under 50,000 243 24.7 24.7 50,000—499,999 222 22.6 47.3 500,000—1,999,999 160 16.3 63.6 2 million and over 358 36.4 100.0 71
Table 3 (continued)
Value Frequency Percent Cumulative
Income under $12,500 188 19.1 19.1 $12,500—$24,999 234 23.8 42.9 $25,000—$34,999 172 17.5 60.4 $35,000—$49,999 183 18.6 79.0 $50,000 and over 206 21.0 100.0 Home Ownership no answer 42 4.3 4.3 own 721 73.3 77.6 rent 194 19.7 97.4 live with relatives 16 1.6 99.0 other 10 1.0 100.0 Education—Male grade school 31 4.3 4.3 high school 247 34.9 39.2 attend college 182 25.8 65.0 graduate college 127 17.9 82.9 postgraduate 119 17.1 100.0 Education—Female grade school 17 2.0 2.0 high school 361 42.7 44.7 attend college 240 28.4 73.1 graduate college 125 14.7 87.8 postgraduate 102 12.2 100.0 Occupation—Male m anagerial 199 27.9 27.9 technical 60 8.4 36.3 service 40 5.6 41.9 farming 10 1.4 43.3 craftsman 76 10.6 53.9 laborer 65 9.1 63.0 other/ retired/ student 264 37.0 100.0 Occupation—Female managerial 199 23.4 23.4 technical 126 14.8 38.2 service 71 8.4 46.6 farming 3 .4 46.9 craftswoman 4 .5 47.4 laborer 9 1.1 48.5 other/retired/student 438 51.5 100.0 72
Table 3 (continued)
Value Frequency Percent Cumulative
* Employment—Male no answer 11 1.5 1.5 full-tim e 453 63.4 65.0 part-time 34 4.8 69.7 retired 196 27.5 97.2 not employed 20 2.8 100.0 Employment—Female no answer 18 2.1 2.1 full-tim e 307 36.1 38.2 part-tim e 136 16.0 54.2 retired 223 26.2 80.5 not employed 166 19.5 100.0 Age — Male 21-34 104 14.6 14.6 35-49 220 30.9 45.5 50-64 176 24.7 70.2 65 and over 212 29.8 100.0 Age — Female 21-34 136 16.0 16.0 25-49 254 29.9 45.9 50-64 217 25.5 71.4 65 and over 213 28.6 100.0 Area Where Grew Up rural 379 39.4 39.4 suburban 330 34.3 73.8 urban 252 26.2 100.0 Area Where Currently Live rural 292 30.4 30.4 suburban 420 43.8 74.2 urban 248 25.8 100.0
* Cumulative frequencies are presented to assist in interpretation of continuous level variables. There presentation for categorical variables is simply for table consistency.
Comparisons of the sample of environmentally concerned citizens with U.S. census data can be found in Table 4. 73
Table 4
Environmentally Concerned Citizens Compared to the U.S. Population
Value Environmentally^ U.S. Totals^ Concerned Citizens
Size of Household 1 25.6 24.2 2 37.8 32.1 3 15.4 18.1 4 13.2 15.4 5+ 7.9 10.2 Geographic Location N orth-E ast 22.6 19.2 North-Central 25.5 24.5 South 30.0 34.6 West 22.0 19.9 Market Size under 50,000 24.7 27.5 50,000—499,999 22.6 20.0 500,000—1,999,999 16.3 16.0 2 million and over 36.4 36.5 Income under $12,500 19.1 21.0 $12,500—$24,999 23.8 22.2 $25,000—$34,999 17.5 16.1 $35,000—$49,999 18.6 18.7 $50,000 and over 21.0 22.0 Race W h ite 92.0 84.0 Black, Asian, Indian... 6.5 12.0 Spanish Origin 1.7 8.2 Education—Male < 4 years high school 12.4 15.4 4 years high school 26.5 36.5 1-3 years college 25.5 19.6 4 years college or more 34.4 28.6 74
Table 4 (continued)
Value Environmentally U.S. Totals Concerned Citizens
Education—Female < 4 years high school 7.9 11.8 4 years high school 36.5 42.3 1-3 years college 28.2 21.6 4 years college or more 26.7 24.2 Occupation—Male managerial 44.2 29.2 technical 13.3 19.4 service 8.9 7.8 farming 2.2 4.1 craftsman 16.9 20.3 laborer 14.4 19.1 Occupation—Female m anagerial 48.3 28.9 technical 30.6 42.3 service 17.2 16.2 farming .7 1.1 craftswoman .9 2.4 laborer 2.2 9.1
I 1 Data from this study ^ U.S. Bureau of Census for 1988
Social Class. An ordinal social class index was computed based on Coleman's CSI index
(Coleman, 1983) that was a household calculation based on male education, female education, annual household income, male occupation, or if no male was present, female occupation. This index was ordinally ranked and had a minimum possible score of 3 and a maximum of 32. The midpoint on the scale (18) was used as the starting point from which social classes were constructed. Middle class was identified as that area within one standard deviation (on a normal distribution) from the midpoint, and from there, one standard deviation above was upper middle class while one below was lower middle class, and so on. The largest percentage 75 of respondents (23.1%) were grouped into the middle class segment. The mean score for respondents was 16.3, just within the middle class segment while the mode was 12, just within the lower middle class segment. Table 5 presents a summary of the data.
Table 5
Frequency Distribution for Social Class
Value Frequency Percent Cumulative Percent
lower class 99 10.1 10.1 upper lower class 164 16.7 26.8 lower middle class 215 21.9 48.7 middle class 228 23.1 71.8 upper middle class 157 15.9 87.7 lower upper class 65 6.6 94.3 upper class 55 5.6 100.0
The NFO screening questionnaire used to identify the sample of environmentally concerned citizens provided information on: degree of environmental concern, affect of a company's environmental record on buying behavior, and participation in environmentally responsible activities.
Environmental Concern. Respondents were selected for the survey based on how environmentally conscious or concerned they felt they were. Nineteen percent (18.8%) of those in the NFO survey indicated that they were "extremely concerned" about the environment while eighty percent (81.2%) indicated that they were "very environmentally concerned."
When asked if anyone in the household had changed their buying behavior because of 76 environmental concerns, two-thirds (64.7%) said they had.
Affect of Company's Record. When asked how much a company's environmental record affected likely product purchases, eighteen percent (17.8%) of respondents indicated that they were "extremely affected" by a company's record and an additional forty-six percent
(45.7%) stated that they were" very affected" by a company's record.
Environmentally Friendly Behaviors. Respondents were asked to indicate whether or not they had participated in any of twelve environmentally-responsible behaviors. The activities most frequently participated in were, in descending order, reusing supermarket bags, recycling aluminum, combining errands to save gas, and recycling newspapers. The mean number of activities participated in was 6.12 while the median and mode were both 6.0 (see
Table 6). 77
Table 6
Frequency Distribution for Environmentally Friendly Behaviors
Behavior Frequency Percent^
reusing supermarket bags 808 79.4 recycling aluminum cans 792 77.8 combining errands to save gas 733 72.1 recycling newspaper 697 68.5 consider enviro. impact when buying 649 63.8 buying energy efficient appliances 601 59.1 recycling glass 567 55.7 recycling plastic containers 548 53.9 using water conservation devices 359 35.2 recycling oil and paint 277 27.2 using public transportation 138 13.5 buying organic produce 65 6.4
1 1 More than one category could be indicated.
On the survey questionnaire, respondents were asked a series of questions about their perceptions and preferences of environmental and conservation organizations and their involvement in these organizations. Results follow.
Recalled Organizations. In an unaided recall question, respondents recalled on average of 3.5 organizations. Of the ten focus organizations for the study, the most frequently recalled, in descending order were: Greenpeace (69.3%), Sierra Club (43.7%), the National Audubon
Society (40.4%), and the National Wildlife Federation (32.7%). The least frequently recalled organizations were the Wilderness Society (5.0%), and the Cousteau Society (9.1%)
(see Table 7). 78
Table 7
Frequency Distribution for Recalled Organizations
1 Organization Frequency Percent
Greenpeace, USA 555 69.3 Sierra Club 350 43.7 National Audubon Society 324 40.4 National Wildlife Federation 262 32.7 Ducks Unlimited 137 17.1 World Wildlife Fund 117 14.6 Ntl. Arbor Day Foundation 114 14.2 The Nature Conservancy 82 10.2 Cousteau Society 73 9.1 W ilderness Society 40 5.0
* More than one category could be indicated.
Membership Status. Of those who responded, sixteen percent (15.5%) were current member's of environmental organizations and eighty-five percent (84.5%) were non-members. The top four organizations for membership, in descending order were: National Wildlife Federation,
National Arbor Day Foundation, Greenpeace, and World Wildlife Fund (see Table 8).
Awareness of Organization. Current members of environmental organizations were asked to indicate what influenced their decision to join an organization. Receiving a solicitation in the mail (50.3%) was the most popular method followed by "impressed by their magazine or other publication" (44.2%) (see Table 9). 79
Table 8
Frequency Distribution for Membership in Organizations
1 Organization Frequency Percent
National Wildlife Federation 38 24.1 National Arbor Day Foundation 29 18.1 Greenpeace, USA 28 17.7 World Wildlife Fund 28 17.7 The Nature Conservancy 20 12.7 Sierra Club 20 12.7 Ducks Unlimited 8 5.1 Wilderness Society 4 2.5 Cousteau Society 3 1.9 other national organization 39 24.7 other regional organization 15 9.6 other local organization 20 12.7 n=158 * More than one category could be indicated. 80
Table 9
Frequency Distribution for Method of Learning About Organizations
Method of Learning Frequency Percent^ About the Organization
received a solicitation in the mail 77 50.3 impressed by their magazine 68 44.2 saw/read news reports or articles 52 33.8 recommended by a friend 30 19.5 went to a meeting 21 13.6 saw a T. V. program 18 11.7 saw an advertisement 9 5.8 do not recall 8 5.2 received as a gift 6 3.9 n=158 * More than one category could be indicated.
Environmental Organizational Activities. All respondents were asked to indicate what activities, relative to environmental organizations, they had participated in. While fifty- three percent (53.3%) of all respondents indicated they had not participated in any activities, of those who had, signing a petition (37.7%), and writing a letter on an environmental issue (17.1%) were the two most common (see Table 9). 81
Table 10
Frequency Distribution of Environmental Organizational Activities
A ctivities Frequency Percent
signed a petition 354 37.7 written letters on environmental issues 161 17.1 attended a local meeting 125 13.3 volunteered any time 87 9.3 participated in a sponsored trip 42 4.5 held a position or office 19 2.0 attended national meeting 18 1.9 submitted an article or photo 17 1.8 n=514 1 More than one category could be indicated.
Donations. Twenty-nine percent of respondents (28.9%) indicated that they had donated money in the last year, in addition to any membership dues, to support an environmental or conservation group. Of those who had given money, fifty-nine percent (59.1%) indicated they gave an amount between one and twenty-five dollars. An additional twenty percent (20.3%) reported that they gave between twenty-six and fifty dollars (see Table 11). 82
Table 11
Frequency Distribution for Donations
Donations Frequency Percent
$1—$25 155 59.2 $26—$50 53 20.3 $51—$100 31 11.8 $101—$500 18 6.9 $501—$1000 3 1.1 $10,000+ 1 .4 n=318
Interest in the Environment. The mean number of years respondents indicated they were interested in the environment was 14.5 years. Both the median and the mode were ten.
Group Membership. Fifty-two percent (52.2%) of respondents indicated that they were a member of a civic, fraternal, religious, or neighborhood organization. On average, respondents reported membership in 2.7 organizations (see Table 12). 83
Table 12
Frequency Distribution for Group Membership
Number of Organizations Frequency Percent
1 156 31.5 2 151 30.5 3 94 19.0 4 33 6.7 5 27 5.5 6 14 2.8 7 4 .8 8+ 16 3.2 n=495
Most Preferred Organizations. When asked to rank the ten study organizations from most preferred to least preferred, the three most preferred organizations, in descending order, were:
National Wildlife Federation, National Audubon Society and The Nature Conservancy. The three least preferred organizations, in ascending order, were: Ducks Unlimited, Greenpeace and the Sierra Club (see Table 13). 84
Table 13
Frequency Distribution for Most Preferred Organizations
*1 Rank of Organization 1 Mean Rank
National Wildlife Federation 4.16 National Audubon Society 4.64 The Nature Conservancy 4.94 World Wildlife Fund 5.14 The Wilderness Society 5.55 National Arbor Day Foundation 5.57 Cousteau Society 5.68 Sierra Club 6.13 Greenpeace,USA 6.16 Ducks Unlimited 7.29
1 Descending from most preferred (1) to least preferred (10).
Reasons Preferred. When asked what they liked best in their most preferred organization, the most frequently cited reasons, in descending order, were: interest in wildlife, national recognition, planting trees, preservation of land, birds, active stance, and broad range of issues
(see Table 14). 85
Table 14
Frequency Distribution for Reasons Why Organizations Were Preferred
1 Reason1 Frequency Percent^
w ild life 105 14.0 national recognition 80 10.6 plant trees 72 9.6 preserve land 59 7.8 birds 51 6.8 active stance 46 6.1 broad range of issues 39 5.2 better at communicating 37 4.9 animals, plants... 37 4.9 ocean life 31 4.1 waterfowl/habitat 29 3.9 best magazine 26 3.5 global in focus 17 2.3
1 In descending order. 2 Items do not add to 100 percent because responses less than two percent are not included.
Product. Service, and Opportunity Preferences. Respondents were asked to indicate the importance on a Likert-type scale, of sixteen products, services, and opportunities that an environmental/conservation organization could provide. The most popular items, in descending order, were: the chance to become more aware and knowledgeable, the opportunity to help solve environmental problems, a chance to help educate others, chapter newsletter and bulletins, and a magazine (see Table 15). 86
Table 15
Ratings of Importance for Product, Service, and Opportunity Preferences
Item* Average Rating of Importance
chance to become more aware and knowledgeable 4.12 opportunity to help solve environmental problems 4.06 chance to help educate others 3.89 chapter newsletters/bulletins 3.64 magazine 3.35 opportunity to meet similar people 3.21 local chapter 3.05 conferences or trainings 3.02 opportunity to volunteer 2.92 local chapter meetings 2.85 entertainment, enjoyment, fun 2.80 opportunity to financially support a cause 2.80 local outings 2.79 adventure travel 2.71 gift/catalog services 2.51 chance to get closer to nature 1.29
* In descending order. Mean rating from most important (5) to least important (1).
Characteristics of an Ideal Organization. When asked to describe the characteristics of an ideal environmental/conservation organization, the most popular attributes in descending order, were: educational orientation, environmental problem-solving orientation, preservation orientation, nature appreciation orientation, and a wise-use orientation (see
Table 16). 87
Table 16
Frequency Distribution for Characteristics of an Ideal Organization
Characteristic Frequency Percent^
educational orientation 764 84.9 environmental problem solving 727 81.0 preservation 707 78.6 nature appreciation 637 70.8 wise-use 553 61.4 global 491 54.6 legislative influence 408 45.3 local 382 42.4 multi-issue 340 37.8 conservative 293 32.6 liberal 128 14.2 pacifist 67 7.4 single-issue 53 5.9 radical 34 3.8 militant 24 2.7
1 Items do not sum to one hundred percent since more than one characteristic could be checked.
Organizations Would Consider Toining. When asked what organizations they would consider joining, the most popular organizations, in descending order, were: National Wildlife
Federation, National Audubon Society, Greenpeace, World Wildlife Fund, and the Sierra
Club. The least popular organizations, in ascending order, were: Sea Shepherd Conservation
Society, League of Conservation Voters, and Conservation International. Seventeen percent
(17.2%) of all respondents indicated that they would not join any organization (see Table 17). 88
Table 17
Frequency Distribution for Organizations Respondents Would Consider Joining
1 Organization Frequency Percent
National Wildlife Federation 357 40.0 National Audubon Society 298 33.4 Greenpeace, USA 238 26.7 World Wildlife Fund 236 26.5 Sierra Club 228 25.6 National Arbor Day Foundation 222 24.9 Cousteau Society 221 24.8 Friends of Animals 208 23.3 The Nature Conservancy 176 19.7 Earth! First! 169 19.0 Wilderness Society 169 18.9 Defenders of Wi 1 d 1 i f e 160 17.9 Rainforest Action Network 154 17.3 Ducks Unlimited 121 13.6 Environmental Defense Fund 97 10.9 National Toxics Campaign 84 9.4 Natural Resources Defense Council 59 6.6 Conservation International 57 6.4 League of Conservation Voters 45 5.0 Sea Shepherd Conservation Society 21 2.4 Other organization 21 2.4 No organization 154 17.2
* Items do not sum to 100 percent since more than one organization could be checked.
Statements About Organizations. Respondents were asked to indicate on a Likert-type scale how strongly each of twelve statements reflected their views of environmental and conservation organizations. Of these twelve statements, the most popular, in descending order, were: that organizational achievement of goals would benefit future generations and that important aspects of the respondents life were affected by the environment (see Table 18). 89
Table 18
Average Rating for Statements About Environmental/Conservation Organizations
1 Statement Average Rating
If environmental/conservation organizations achieve their goals, my life and the lives of the next generation will benefit. 4.3
Some important aspects of my life are affected by environmental/ conservation problems. 4.0
Organized groups are very effective in influencing national environmental issues. 3.8
Membership contributions help environmental/conservation organizations influence government action on conservation problems. 3.5
Members gain much from the information they receive from these organizations. 3.5
If people like me contribute to an environmental or conservation organization, the group will be better able to work for improved environmental quality. 3.5
Belonging to these groups makes members feel better about what they are doing with their lives. 3.4
Belonging to these groups is one of the most important things in member's lives. 2.8
Environmental/conservation organizations should actively involve themselves in environmental problems of special groups such as minorities and the economically disadvantaged. 2.8
I think membership in an environmental/conservation organization is important because many knowledgeable and influential people support these groups. 2.7
People become members of environmental/conservation organizations because their friends encourage them to belong. 2.7
Environmental/conservation magazines and other publications arc one of the most important reasons why I would join these organizations. 2.5
1 Average rating on Likert-type scale ranging from agree completely (5) to disagree completely (1). 90
Appropriate Actions. When asked which actions they considered appropriate for solving environmental problems, the most popular actions, in descending order, were: education, supporting research, publicity, and letter writing campaigns (see Table 19).
Table 19
Frequency Distribution for Appropriate Actions for Solving Environmental Problems
Action Frequency Percent^
education 863 91.7 supporting research 728 77.4 publicity 726 77.2 letter writing campaigns 635 67.5 petitions 556 59.1 fundraising 493 52.4 lobbying 452 48.1 testifying at hearings 429 45.6 boycotting 341 36.3 law suits 236 25.1 marching, picketing, or sitting in 167 17.8 strikes 60 6.4 property damage 50 5.3 violent actions against people 14 1.5 other 21 2.1 none of the above 31 3.3
1 Items do not sum to one hundred since more than one action could be checked.
Self-Description. When asked which of six words the respondent would use to identify themselves, the most popular label was concerned citizen, and the second most popular label was conservationist (see Table 20). 91
Table 20
Frequency Distribution for Self-Description Labels
Label Frequency Percent*
concerned citizen 832 86.1 conservationist 293 32.6 environm entalist 246 25.5 educator 173 17.9 naturalist 152 15.7 eco-activist 18 1.9 other 12 1.2 none of the above 61 6.3
* Items do not sum to one hundred since more than one label could be checked. 92
Question Two. How do individual and environmental differences vary by market segment, and what is the best combination of variables to identify these market segments?
While the macro view of environmentally concerned citizens presented for research in question two provides an overall picture of respondents, generalization at this level may hide market opportunities that could exist within specific market segments. Results presented for question two profile select market segments based on significant differences and then attempt to determine the best subset of variables useful for classifying segment members.
Variables useful for market segmentation of the data set included geographic, sociodemographic, psychographic, and behavioral variables. For the purposes of this analysis, psychographic and behavioral variables were used for segmentation. These variables include the psychographic or lifestyle variables of environmental concern and social class. The behavioral variables include membership status, donation behavior, length of interest in the environment, and joining behavior. Because many of the segmenting variables are interconnected, these views do not represent unique market segments but rather overlapping segments. Therefore, each segmenting variable produces a different view of the environmentally concerned citizen market.
The first stage of segmentation involves profiling the segments. The chi-square test for independence was used to determine statistically significant differences between categorical variables. The student's t-test was used to identify relationships between the categorical dependent variable, the segmenting variable, and continuous independent variables. An alpha of .05 was used to determine significance for both tests. Only significant results are presented. The second stage of segmentation involves identifying the variable subset that can be used to assist in correctly classifying respondents in each segment. For segments with a categorical dependent variable, discriminant analysis was used while multiple regression analysis was used for segments with a continuous dependent variable. Members and Nonmembers.
Within the sample, one hundred and fifty-six respondents (15.5%), indicated that they were currently members of an environmental organization. Roughly half (54.2%) indicated that they were a member in only one organization while the rest indicated that they were a member in two (17.6%), three (14.4%), four (10.5%), or five or more organizations (3.3%). Of the ten study organizations, membership was highest in the National Wildlife Federation, the National Arbor Day Foundation, Greenpeace and World Wildlife Fund. Membership was used as the segmenting variable, the dependent variable, in the first market view (see Table
21).
Table 21
Summary of Significant Relationships Between Members and Nonmembers
and Selected Independent Variables
Characteristic n chi-square df P sociodemographics annual family income 972 21.6 4 .000 males occupation 711 24.8 6 .000 females occupation 839 15.2 6 .018 male education 703 32.2 3 .000 female education 835 45.3 3 .000 race 957 4.2 1 .040 social group 972 11.9 3 .007 place where grew up 951 3.2 2 .044 lifestyle characteristics donations 980 138.4 1 .000 joining behavior 975 16.0 1 .000 concern for environment 927 6.6 1 .010 change in buying habits 956 14.4 1 .002 self-labels environmentalist 956 68.3 1 .000 eco-activist 956 6.8 1 .008 conservationist 956 61.9 1 .000 educator 956 15.2 1 .000 n aturalist 956 16.3 1 .000 94
Sociodemographic Characteristics
There was a significant relationship between membership and annual family income
(p=.000). Fifty-six percent of members had an annual family income greater than $35,000 while only thirty-six percent of nonmembers had an annual family income in excess of $35,000.
There was a significant relationship between male and female occupations between members and non-members (p=.000/.018 ). Members were more likely to have white collar occupations than were non-members.
Members were more likely to have more education than were nonmembers (p=.000). Twenty- seven percent of female heads of households and thirty-three percent of male heads of households for members had some postgraduate education compared to non-members with less than fifteen percent for either gender.
There was a significant relationship between race and membership (p=.040). Ninety-seven percent of members were white compared to ninety-two percent of non-members.
Members and nonmembers differed significantly (p=.007) with respect to social group.
Members were more likely to live in non-family households (8.6%) compared to nonmembers
(2.3%).
There was a significant relationship between members and non-members with respect to the area they grew up (p=.044). Members were more likely to have grown up in urban or suburban areas than were non-members. 95
Self-Labels
Members were significantly different than nonmembers with respect to the labels they were willing to ascribe to themselves (p=.000/.008). Members were more likely to use the names: environmentalist, eco-activist, conservationist, educator, and naturalist with respect to themselves than were nonmembers.
Lifestyle Characteristics
Members were more likely to have donated money, beyond any membership dues, to an environmental/conservation organization within the past year (p=.000). Nearly seventy percent (68.6%) of members reported giving some money to an organization while only twenty- two percent (21.6%) of nonmembers reported giving a donation.
Members of environmental/conservation groups were more likely to be members of at least one civic, fraternal, religious, or neighborhood organization (p=.000). Sixty-seven percent (67.3%) of members were joiners while only fifty percent (49.6%) of nonmembers were.
Individuals who expressed that they were extremely concerned about the environment were more likely to be members (p=.010). Twenty-seven percent (27.4%) of members expressed this higher level of concern while only eighteen percent (18.2%) of nonmembers did. Almost eighty percent (79.8%) of members reported that they had changed their buying habits due to environmental concerns compared to sixty-two percent of nonmembers (p=.002).
Student's t-tests were utilized to determine differences between the members and nonmembers on a series of continuous variables (see Table 22). 96
Table 22
Mean Score Differences Between Members and Nonmembers
of Environmental/Conservation Organizations
Variable/Group n mean sd df t-value P
Total Organizations Recalled /member 156 4.1 2.2 / nonmember 849 2.0 1.7 192 11.3 .000 Social Class /member 150 15.0 5.5 /nonmember 822 12.1 5.0 197 5.8 .000 Total Activities Participated In /member 156 2.0 1.7 /nonmember 850 .59 .94 171.8 9.7 .000 Length of Interest in Environment /member 156 16.5 12.8 /nonmember 850 9.5 13.5 223.1 6.1 .000 No. of Environmentally Friendly Behaviors /member 156 6.9 2.5 /nonmember 850 5.9 2.5 219.2 4.73 .000
A student's t-test (p=.000), determined that members participated in, on average, a greater number of environmentally responsible behaviors (mean=7.0) than did non-members
(mean=5.9).
Members were more likely to have been interested in the environment for a longer period of time than nonmembers (p=.000). A student's t-test found a mean number of years of interest for members of 16.5 while nonmembers average length of interest was only 9.5 years.
Social class as measured on a composite index was greater for members than nonmembers
(p=.000). Members scored on average of fifteen on the social class index compared to 97 nonmembers who scored twelve.
Members recalled on average, twice as many environmental/conservation organizations than did nonmembers (p=.000). A t-test found that members recalled on average a total of four organizations while nonmembers recalled only two.
Members were more likely to have participated in environmental activities such as letter writing and signing a petition than had nonmembers (p=.000). Members participated in an average of two activities while nonmembers, on average, participated in less than one (.59) activity.
Discriminating Variables
In order to determine which subset of variables was most useful in identifying members and non-members, stepwise discriminant analysis using the Wilk's Lambda method was used. The dependent variable was membership status and all available continuous independent variables were used, including: total environmental activities, total number of environmentally friendly behaviors, social class, market size, family size, and length of interest in the environment. Six categorical variables were also selected and were coded as dummy variables in order to be compatible with discriminant analysis requirements. These variables included: donator, joiner, type of residence, race, level of concern, and likelihood of changing behavior.
One half of all cases were randomly selected to produce the discriminant function. The remainder of the cases, the holdout sample, was used to test the classification function. The analysis stopped after seven steps with seven variables included (see Table 23). 98
Table 23
Independent Variables Utilized in the Discriminant Function for Membership
V ariable Group* Group Mean Group S.D. Wilks' Lambda p
Total Activities 1 2.07 1.79 .82 .000 2 .60 .96 Donate 1 .66 .47 .76 .000 2 .21 .41 Social Class 1 18.43 7.53 .75 .000 2 14.77 6.94 Household Size 1 2.31 1.17 .75 .000 2 2.47 1.35 Market Size 1 2.65 1.21 .75 .000 2 2.64 1.21 Environmental Concern 1 .27 .44 .74 .000 2 .17 .38 Length of Interest 1 17.08 12.82 .74 .000 2 9.74 13.62 Race 1 1.02 .16 .75 .000 2 1.07 .26
1 Group one are members and group two are nonmembers.
The discriminant function developed from these variables was applied to the holdout sample with a correct classification rate of seventy-nine percent (78.4%). Sixty-nine percent (71.4%) of members were correctly classified and eighty-one (79.8%) of nonmembers (see Table 24). 99
Table 24
Classification Results on Holdout Sample for Membership
Actual Group No. of Cases Predicted Group Membership 1 2
Group 1 63 44 19 71.4% 28.6%
Group 2 331 62 269 20.2% 79.8%
Ungrouped Cases 11 1 10 9.1% 90.9%
Environmental Concern.
Degree of environmental concern was the second variable used to segment markets. Nineteen percent of all respondents indicated that they were "extremely concerned" about the environment while eighty-one percent indicated that they were "very concerned" about the environment. These two groups were compared using the chi-square test for independence on a series of sociodemographic and lifestyle characteristics (see Table 25). 100
Table 25
Summary of Significant Relationships Between
Extremely versus Very Environmentally Concerned Respondents
and Selected Independent Variables
Characteristic n chi-square df P sociodemographics annual family income 972 21.6 4 .000 how heard about organization advertisement 938 7.3 1 .006 news report or article 938 5.0 1 .024 lifestyle characteristics affect of company's record 927 200.1 3 .000 changing behavior 921 34.6 1 .000 self-label environmentalist 889 29.3 1 .000 conservationist 889 13.4 1 .000
Sociodemographics
Annual family income was significantly greater for those individuals who were "extremely environmentally concerned" versus those who were "very environmentally concerned"
(p=.000).
There was a significant relationship between method of hearing about an organization and environmental concern (p=.006/.024). Individuals who were extremely environmentally concerned were more likely to have seen an advertisement or to have read a news report or article on an organization than those who were very environmentally concerned. 101
Self-Labels
There was a significant relationship between environmental concern and self-labels (p=.000).
Those who were extremely environmentally concerned were more likely to call themselves environmentalists or conservationists than were those who were very environmentally concerned.
Lifestyle Characteristics
There was a significant relationship between environmental concern and affect of a company's environmental record. Individuals who were extremely environmentally concerned were more likely to express that a company's environmental record extremely affected their purchase decisions compared to those who were very environmentally concerned (p=.060).
Those who were extremely concerned were also more likely to suggest that they had changed their behavior in response to environmental concerns than those who were very environmentally concerned (p=.000).
Student's t-tests were utilized to determine differences between environmental concern on a series of continuous variables (see Table 26).
There was a significant relationship between social class and environmental concern (p=.05).
Extremely concerned respondents had an average social class index of 11.9 compared to very concerned respondents whose social class index averaged 12.7.
There was a significant relationship between environmental concern and the number of environmentally friendly behaviors engaged as well as the number of environmental activities participated in (p=. ). Individuals who were extremely concerned participated in an average of 7.2 environmentally friendly behaviors compared to 6.1 for those who were 102 very concerned. Similarly, those who were extremely concerned participated in an average of one environmental activity such as letter writing while those who were very concerned participated in less than one activity (p=.005).
Those who were extremely concerned had been interested in the environment for an average of thirteen years compared to those who were only very concerned who had been interested in the environment for an average of ten years (p=.035).
Table 26
Mean Score Differences Between Extremely and Very Environmentally Concerned Respondents of Environmental/Conservation Organizations
Variable/Group n mean sd df t-value p
Social Class /extremely concerned 185 11.9 5.2 /very concerned 753 12.7 5.2 283.8 -1.96 .050 Total Activities Participated In /extremely concerned 185 1.0 1.4 /very concerned 753 .7 1.1 244.3 2.82 .005 Length of Interest in Environment /extremely concerned 185 13.2 16.5 /very concerned 753 10.4 13.4 246.1 2.1 .035 No. of Environmentally Friendly Behaviors /extremely concerned 185 7.2 2.6 /very concerned 753 6.1 2.3 262.6 5.1 .000 Total Current Memberships /extremely concerned 185 .4 1.0 /very concerned 753 .2 .7 228.4 2.3 .022 103
Discriminating Variables
In order to determine which subset of variables was most useful in identifying "extremely concerned" respondents versus "very environmentally concerned" respondents, Wilk's Lambda- style, stepwise discriminant analysis was used. The dependent variable was environmental concern and a number of continuous, independent variables were selected including: total environmental activities, total number of environmentally friendly behaviors, social class, market size, family size, and length of interest in the environment. Six categorical variables were also selected and were coded as dummy variables in order to be compatible with discriminant analysis requirements. These variables included: donator, joiner, type of residence, race, and likelihood of changing behavior.
One half of all cases were randomly selected to derive the discriminant function. The remainder of the cases, the holdout sample, was used to test the classification function. The analysis stopped after nine steps with nine variables included (see Table 27). The discriminant function developed from these variables was applied to the holdout sample with a correct classification rate of sixty-two percent (61.14%). Sixty-nine percent (69.2%) of extremely concerned respondents were correctly classified and fifty-nine percent (59.3%) of very concerned respondents (see Table 28). 104
Table 27
Independent Variables Utilized in the Discriminant Function for Environmental Concern
Variable Group1 Group Mean Group S.D. Wilks' Lambda p
Change Behavior 1 .8 .3 2 .5 .4 .96 .000 Total Enviro. Behaviors 1 7.2 2.6 2 6.1 2.3 .94 .000 Social Class 1 14.7 7.2 2 15.3 7.1 .93 .000 Length of Interest in Enviro . 1 13.4 16.7 2 10.3 13.0 .92 .000 Total Activities 1 1.1 1.4 2 .7 1.1 .91 .000 Race 1 1.0 .2 2 1.0 .2 .91 .000
1 Group one are extremely environmentally concerned and group two are very environmentally concerned.
Table 28
Classification Results on Holdout Sample for Membership
Actual Group No. of Cases Predicted Group Membership 1 2
Group 1 90 63 26 Extremely Concerned 69.2% 30.8%
Group 2 372 296 448 Very Concerned 40.7% 59.3% 105
Toiners.
The third dependent variable used to segment the environmentally concerned market was joining behavior. Joining behavior refers to the individual's membership in one or more civic, fraternal, religious, or neighborhood organizations. This variable was selected for market segmentation because the voluntary association literature suggests that joiners are more likely to be active in other organizations and more likely to join additional organizations.
Within the sample, slightly over half (52.2%) of respondents were joiners compared to forty- eight percent (47.8%) who were classified as non-joiners.
These two groups were compared using the chi-square test for independence on a series of sociodemographic and lifestyle characteristics (see Table 29).
Sociodemographics
There were significant relationships between all of the elements of the social class indices and joining behavior. Education and occupation for both male and female heads of household and annual family income were significantly higher for joiners than for non-joiners.
Correspondingly a student's t-test between the dependent variable and the social class index found a significant difference between the mean scores for joiners and non-joiners. Joiners scored an average of 16.4 on the composite index compared to non-joiners who scored an average of 14 on the index.
There was a significant relationship between race and joining behavior (p=.029). Forty-nine percent of joiners were white compared to forty-four percent of non-joiners.
There was a significant relationship between method of hearing about an organization and joining behavior (p=.03 or less). Joiners were more likely to have heard about an organization from a magazine, by attending a meeting, from a direct mail piece, or from the newspaper. 106
Table 29
Summary of Significant Relationships Between Joiners and Non-Joiners and Selected Independent Variables
Characteristic n chi-square df P sociodemographics female education 981 36.1 6 .000 male education 981 32.7 6 .000 male occupation 689 14.1 6 .027 female occupation 820 12.5 6 .050 income 949 13.3 4 .009 race 934 4.7 1 .029 how heard about organizations magazine 981 4.5 1 .032 meeting 981 6.3 1 .011 direct mailing 981 6.8 1 .009 newspaper/news report 981 7.9 1 .004 lifestyle affect of a company's record 939 7.8 3 .048 donations 963 13.5 1 .000 self-label conservationist 936 5.6 1 .017 educator 936 14.0 1 .000 concerned citizen 936 10.2 1 .001
Self-Labels
Joiners were significantly more likely to call themselves a conservationist, educator, or concerned citizen than were non-joiners (p=.000/.017).
Lifestyle Characteristics
There was a significant relationship between joining behavior and donations. Joiners were more likely to have given money to an organization within the past year than non-joiners 107
(p=.000) (see Table 30).
Table 30
Mean Score Differences Between Joiners and Non-Joiners
of Environmental/Conservation Organizations
Variable/Group n mean sd df t-value P
Social Class /joiners 491 16.4 7.3 /non-joiners 458 14.0 6.7 946.9 5.4 .000 Total Activities Participated In /joiners 512 1.0 1.4 /non-joiners 469 .6 .9 907.9 5.5 .000 Total Organizations Recalled /joiners 512 2.6 1.9 /non-joiners 469 2.0 1.9 973.0 4.2 .000 Length of Interest in Environment /joiners 512 12.4 14.3 /non-joiners 469 9.2 12.8 987.4 3.70 .000 Total Current Memberships /joiners 512 .3 .8 /non-joiners 469 .1 .7 972.6 3.0 .003 Years of Membership /joiners 90 8.6 8.3 /non-joiners 48 5.6 4.0 135.2 2.87 .005
Student's t-tests found that there were significant relationships between joining behavior and the independent variables, total activities participated in, total organizations recalled, length of interest in the environment, total current environmental memberships and years of membership in organizations. On all five independent variables, joiners mean scores were significantly greater than non-joiners. 108
Discriminating Variables
In order to determine which subset of variables was most useful in identifying joiners and non joiners, stepwise discriminant analysis using Wilk's Lambda method was used. The dependent variable was joining behavior and a number of continuous independent variables were selected, including: total environmental activities, total number of environmentally friendly behaviors, social class, market size, family size, and length of interest in the environment.
Eight categorical variables were also selected and were coded as dummy variables in order to be compatible with discriminant analysis requirements. These variables included: donator, type of residence, race, level of concern, geographic region, social group type, and likelihood of changing behavior.
One half of all cases were randomly selected to derive the discriminant function. The remainder of the cases, the holdout sample, were used to test the classification function. The analysis stopped after five steps with five variables included (see Table 31). The discriminant function developed from these variables was applied to the holdout sample with a correct classification rate of sixty percent (60.4%). Fifty-five percent (54.8%) of joiners and sixty-six percent (66.0%) of non-joiners were correctly classified (see Table 32). 109
Table 31
Independent Variables Utilized in the Discriminant Function for Joiners
V ariable Groupl Group Mean Group S.D. Wilks' Lambda p
Total Activities 1 1.0 1.4 2 .6 .9 .96 .000 Social Class 1 16.5 7.2 2 14.0 6.7 .94 .000 Market Size 1 2.5 1.2 2 2.7 1.2 .92 .000 Length of Interest in Enviro. 1 12.5 14.4 2 9.1 12.3 .92 .000 Donations 1 .3 .5 2 .2 .4 .92 .000
*1 Group one are joiners and group two are non-joiners.
Table 32
Classification Results on Holdout Sample for Joining Behavior
Actual Group No. of Cases Predicted Group Membership 1 2
Group 1 243 139 104 Joiners 54.8% 45.2%
Group2 225 77 148 Non-Joiners 34.0% 66.0% 110
D onations.
Identifying those individuals who are more likely to give money to environmental/conservation organizations is a useful method of segmenting thr> market. If respondents indicated that they had given money, beyond any membership dues, to an environmental or conservation organization within the past year they were classified as a donor. Nearly thirty percent of the sample (28.9%) indicated that they had given money to an organization within the past year.
Donors and non-donors were compared using the chi-square test for independence on a series of sociodemngraphic and lifestyle characteristics (see Table 33).
Sociodemographics
Like those individuals who were joiners, there was a significant relationship between elements of the social class composite index and donating behavior. Education levels for both heads of household, males' occupations, and annual family income were significantly different between donors and non-donors. A t-test between the dependent variable, donations, and the independent variable social class found that on average, donors had a mean score of seventeen on the social class composite index compared to non-donors who had a mean score of fourteen.
Self-Labels
There was a significant relationship between self-labels and donations (p=.000). Those who gave money were more likely to call themselves any one of the six self-labels than were individuals who did not give money. I ll
Lifestyle Characteristics
A series of student's t-tests found significant differences between donation giving and a number of the independent variables (see Table 34). Donors were interested in the environment for an average of 14.5 years compared to non-donors whose interest was on average only 9.2 years.
Donors had participated in an average of 6.7 environmentally friendly behaviors compared to non-donors who participated in an average of 5.8 behaviors. Similarly, donors participated in an average of 1.5 environmental activities such as letter writing compared to non-donors who participated in an average of .5 activities. Donors also recalled significantly more organizations than did non-donors (p=.000). 112
Table 33
Summary of Significant Relationships Betweeen Donors and Non-Donors
and Selected Independent Variables
Characteristic n chi-square df P sociodemographics marital status 954 8.2 2 .016 social group 954 8.4 3 .037 female education 985 32.5 6 .000 male education 985 22.8 6 .000 male occupation 695 18.9 6 .004 income 954 31.6 4 .000 annual family income 972 21.6 4 .000 method heard about organization friend 985 23.2 1 .000 magazine 985 57.3 1 .000 advertisement 985 8.3 1 .003 meeting 985 5.7 1 .016 direct mail 985 84.9 1 .000 television 985 26.3 1 .000 new spaper/new s report 985 69.2 1 .000 life-style affect of a company's record 944 11.4 .009 willingness to change behavior 936 20.6 1 .000 joiner 963 13.5 1 .000 self-label environmentalists 936 34.5 1 .000 eco-activists 936 11.5 1 .000 conservationist 936 23.9 1 .000 educator 936 7.6 1 .005 concerned citizens 936 12.3 1 .000 naturalist 936 15.9 1 .000 113
Table 34
Differences Between Donators and Non-Donators
of Environmental/Conservation Organizations
Variable/Group n mean sd df t-value p
Social Class / donators 277 16.9 7.0 /non-donators 677 14.5 7.0 513.7 4.8 .000 Total Activities Participated In / donators 285 1.5 1.4 / non-donators 700 .5 1.0 394.9 10.19 .000 Length of Interest in Environment /donators 285 14.5 13.0 /non-donators 700 9.2 13.6 551.5 5.6 .000 No. of Environmentally Friendly Behaviors /donators 285 6.7 2.5 /non-donators 700 5.8 2.6 539.3 5.25 .000 Total Organizations Recalled /donators 285 3.3 2.1 /non-donators 700 1.9 1.7 458.6 9.9 .000
Discriminating Variables
In order to determine which subset of variables was most useful in identifying donators and non-donators, stepwise discriminant analysis using Wilk's Lambda method was used. The dependent variable was donation status and a number of continuous independent variables were selected, including: total environmental activities, total number of environmentally friendly behaviors, social class, market size, family size, and length of interest in the environment. Fivet categorical variables were also selected and were coded as dummy variables in order to be compatible with discriminant analysis requirements. These variables included: joiner, type of residence, race, level of concern, and likelihood of changing behavior. One half of all cases were randomly selected to derive the discriminant function. The remainder of the cases, the holdout sample, was used to test the classification function. The analysis stopped after seven steps with seven variables included (see Table 35). The discriminant function developed from these variables was applied to the holdout sample with a correct classification rate of seventy-one percent (71.8%). Sixty percent (60.3%) of donators and seventy-six percent (76.3%) of non-donators were correctly classified (see Table
36).
Table 35
Independent Variables Utilized in the Discriminant Function for Donations
V ariable Group* Group Mean Group S.D. Wilks' Lambda P
Total Activities 1 1.5 1.5 2 .6 1.0 .87 .000 Social Class 1 17.1 6.9 2 14.5 7.0 .86 .000 Enviro. Friendly Behaviors 1 6.9 2.3 2 6.0 2.4 .85 .000 Length of Interest in Enviro. 1 14.6 13.2 2 9.2 13.6 .85 .000 Joining Behavior 1 .6 .5 2 .4 .5 .84 .000 Willingness to Change 1 .7 .4 2 .6 .5 .84 .000 Market Size 1 2.7 1.2 2 2.6 1.2 .84 .000
■1 Group one are donators and group two are non-donators. 115
Table 36
Classification Results on Holdout Sample for Donations
Actual Group No. of Cases Predicted Group Membership 1 2
Group 1 139 88 51 Donators 60.3% 39.7%
Group 2 353 86 267 Non-donators 23.7% 75.3%
Social Class
A social class index constructed on the basis of education, occupation, and income was used as the dependent variable for the fifth segmentation view. This continuous variable, ranging from a minimum of 3 to a maximum of 32, was recoded into five categories for ease in analysis.
These five categories are: lower class, lower middle class, middle class, upper middle class, and upper class. Categorical independent variables were analyzed using chi-square analysis
(see Table 37) while continuous independent variables were analyzed with oneway analysis of variance procedures with a post hoc Scheffe test (see Table 38).
Sociodemographics
There was a significant relationship between marital status and social class (p=.000). Upper class respondents were more likely to be married. 116
Education rates for both male and female heads of household were significantly related to social class (p=.000). As education levels increased, individuals were more likely to be in higher social classes. All respondents in the upper class segment had at least some college and many had post-graduate education.
There was a significant relationship between occupation and social class (p=.000). Those respondents in the upper social classes were more likely to be employed in managerial or professional occupations.
Social class and market size were significantly related (p=.000). As market size increased, so did social class.
Household size was significantly related to social class (p=.000). Single family households were predominantly in the lower classes while households with between three and four members were predominantly upper class.
Race and social class were significantly related (p=.049). Minorities were more likely to be in lower class segments.
Self-Labels
Social class and self-labels were significantly related (p=.000). Upper class segments were more likely to use the self-labels of concerned citizen and educator.
Lifestyle Characteristics
Social class was significantly related to joining behavior (p=.000). Joiners were more likely to be in the upper class segments. 117
Donating behavior was significantly related to social class (p=.000). Lower class segments were more likely to not have given money to an environmental/conservation organization within the past year.
Similarly, membership in organization was significantly related to social class (p=.000).
Upper class segments were more likely to be members of environmental/conservation organizations than lower class segments.
Social class and willingness to change behavior were significantly related (p=.000). Upper class segments were more likely to express that they had changed their behavior as a result of environmental information.
Oneway analysis of variance with a post hoc Scheffe test was used to analyze significant differences between social class categories and continuous independent variables.
There was a significant relationship between social class and environmentally friendly behavior (p=.000). Lower classes performed signigicantly fewer environmentally friendly behaviors than did the middle through upper class segments. Likewise, lower middle class segments participated in significantly fewer environmentally friendly behaviors than did their upper class counterparts.
There was a significant relationship between social class and total organizational activities
(p=.000). Lower class respondents participated in significantly fewer activities that did upper class segments.
There was a significant relationship between social class and organizations recalled (p=.000).
Organizations recalled increased in a positive direction with social class. There were significant differences between a number of the social classes on this variable. 118
There was a significant relationship between social class and total current memberships
(p=.000). Lower class respondents had significantly fewer memberships than upper class respondents and lower middle class respondents had significantly fewer memberships than upper class respondents.
Table 37
Summary of Significant Relationships Between Social Class
and Selected Independent Variables
Characteristic n Chi-Square df P sociodemographics marital status 983 324.82 8 .000 social group 983 284.87 12 .000 female education 983 337.65 24 .000 male education 983 835.59 24 .000 male occupation 714 505.61 24 .000 female occupation 850 123.59 24 .000 market size 983 37.58 12 .000 household size 983 197.69 16 .000 race 968 9.53 4 .049 current residence classification 927 19.74 8 .011 life-style joining behavior 949 28.94 4 .000 donating behavior 954 25.60 4 .000 membership status 972 34.01 4 .000 affect of a company's record 972 25.33 12 .013 likelihood of changing behavior 964 20.31 4 .000 self-label educator 933 32.27 4 .000 concerned citizen 933 10.63 4 .030 119
Table 38
Oneway Analysis of Variance Between Social Class Categories
and Selected Independent Variables
Environmentally Friendly Behaviors Lower L. Middle Middle U. Middle Upper Mean 5.66a 6.05b 6.44ac 718ab 7 37abc Source df. ss ms F ratio Probability Between Groups 4 389 97 17.24 .000 Within Groups 978 5521 6 Total 982 5910
Activities Lower L. Middle M iddle U. Middle Upper Mean .55a .83a .86 .97 1.24a Source dL ss ms F ratio Probabilitv Between Groups 4 51 13 8.47 .000 Within Groups 978 1476 1.5 Total 982 1527
Total Organizations Recalled Lower L. Middle Middle U. Middle Upper Mean 1.80a 2.36ab 2.40ac 2.53ad 3.3 7abcd Source d f_ ss. ms F ratio Probability Between Groups 4 233 58 15.83 .000 Within Groups 978 3599 4 Total 982 3832
Total Current Memberships Lower L. Middle Middle U. Middle Upper Mean .12a .21b .34a .35 .51ab Source dL S. ms F ratio Probability Between Groups 4 16 4 7.20 .000 Within Groups 978 564 .57 Total 982 580
_ i__ J Means with common superscripts are different at a significance level of .05 (Scheffe). 120
Regression Analysis of Social Class
The independent variables were entered into a stepwise multiple regression analysis to determine which variables were best able to account for variability in the dependent variable "social class." Eight variables were determined to be the best set of predictors of social class: donating behavior, family size, race, level of concern for the environment, market size, joining behavior, environmentally friendly behaviors, and membership.
There was a positive relationship between social class and environmental behavior and market size. With an increase in score on these variables there was an increase in the social class level of the respondent. There was a positive relationship between social class and race, donating, joining, level of concern, and membership habits. Respondents who were white, donors, joiners, extremely concerned and members of an environmental conservation organization had correspondingly higher social class indices. The multiple correlation coefficient between social class and these eight variables accounted for twenty-four percent of the variance. Table 39 presents summary information on the variables in the regression equation. 121
Table 39
Regression of Social Class on Selected Independent Variables
V ariable B SEB Beta T SigT
Donate -.84 .48 -.05 -1.75 .008 Family Size 1.84 .15 .34 11.94 .000 Race 2.16 .81 -.07 -2.66 .007 Concern 1.36 .52 -.07 -2.59 .009 M arket Size .88 .16 .14 5.23 .000 Joining 1.93 .40 .13 4.75 .000 Env. Behaviors .52 .08 .18 6.15 .000 Membership 2.69 .60 -.135 -4.45 .000
Regression Statistics M ultiple R .49 R Square .24 Adjusted R Square .24 Standard Error 6.20 F 39.94 Significant F .000
Analysis of Variance DF Sum of Squares Mean Square Regression 8 12308 1538 Residual 959 36935 38 n=969
Length of Interest in the Environment
The final segmentation variable was based on categorization of length of interest in the environment. Seven categories for length of interest were developed from those who did not indicate any length of interest in the environment to those who had been interested for forty or more than years. 122
These length of interest groups were compared using the chi-square test for independence on a series of sociodemographic lifestyle characteristics (see Table 40).
Table 40
Summary of Significant Relationships Between Length of Interest in the Environment
and Selected Independent Variables
Characteristic n Chi-Square df P sociodemographics marital status 983 28.73 12 .004 social group 983 29.30 18 .044 female education 1016 99.85 36 .000 male education 1016 119.57 36 .000 male occupation 714 113.99 36 .000 female occupation 850 69.24 36 .000 income 983 81.13 24 .000 life-style membership 1005 69.87 6 .000 willingness to change behavior 964 46.24 6 .000 donations 985 128.57 6 .000 joining behavior 981 30.13 6 .000 self-label environmentalist 965 114.48 6 .000 eco-activist 965 15.70 6 .015 conservationist 965 89.32 6 .000 educator 965 29.86 6 .000 concerned citizen 965 69.88 6 .000 naturalist 965 52.31 12 .000 123
Sociodemographics
There was a significant relationship between marital status and social group and length of interest in the environment (p=.004/p=.044).
Self-Labels
There were significant relationships between self-labels and length of interest in the environment (p=.000/.015). Those who had been interested in the environment for between ten and twenty-nine years were most likely to identify themselves as environmentalists, eco- activists, conservationists, educators, concerned citizens, and naturalists.
Lifestyle Characteristics
There was a significant relationship between length of interest in the environment and membership in environmental/conservation organizations (p=.000). Members were more likely to have been interested in the environment for between ten and twenty-nine years (see
Table 41).
There was a significant relationship between length of interest in the environment and willingness to change behavior (p=.000). Those who said that they had changed their behavior were more likely to have been interested in the environment for between ten and twenty-nine years.
There was a significant relationship between length of interest in the environment and donating behavior (p=.000). Those who said that they had given money to an environmental/conservation organization were more likely to have been interested in the environment for between ten and twenty-nine years. 124
Those who were interested in the environment for between ten and twenty-nine years were more likely to be joiners in civic organizations (p=.000).
< 125
Table 41
Oneway Analysis of Variance Between Length of Interest Categories
and Selected Independent Variables
Years of Organizational Membership 0 Years 1 —4 5—9 10—19 20—29 30—39 40+Years Mean .19a .23b .66c 1.0d 2.15abe 7.35abcdef 1.92f Source df ss ms F ratio Probability Between Groups 6 2113 352 24.96 .000 Within Groups 1009 14236 14 Total 1015 16350
Total Organizations Recalled 0 Years 1 —4 5—9 10—19 20—29 30—39 40+Years Mean 1.12a 1.92ab 2.44ac 2 g-jabd 3 72abcd 2.65a 2.73a Source df ss. ms F ratio Probabilitv Between Groups 6 786 131 41.43 .000 Within Groups 1009 3193 3 Total 1015 3979
Current Organizational Memberships 0 Years 1 —4 5—9 10—19 20—29 30—39 40+Years Mean .03a .llb .26 14ab .58ab .47 .27 Source df ss ms F ratio Probabilitv Between Groups 6 39 6 10.54 .000 Within Groups 1009 626 .62 Total 1015 666
Environmentally Friendly Behaviors 0 Years 1 —4 5—9 10—19 20—29 30—39 40+Years Mean .5.38a 5.85c 5.48b 6.57a 7.21abc 6.65 6.28 Source df ss ms F ratio Probabilitv Between Groups 6 438 73 10.93 .000 Within Groups 1009 6740 7 Total 1015 7178 126
Table 41 (continued)
Social Class 0 Years 1 —4 5—9 10—19 20—29 30—39 40+Years Mean 12 .57a 14.85b 14.90 17.253 17.88ab 14.92 14.78 Source dL ss. ms F ratio Probabilitv Between Groups 6 3784 630 13.29 .000 Within Groups 976 46297 47 Total 982 50081
abcc* Means with common superscripts are different at a significance level of .05 (Scheffe).
Regression Analysis of Length of Interest
The independent variables were entered into a stepwise multiple regression analysis to determine which variables were best able to predict length of interest in the environment.
Five variables were determined to be the best predictors of length of interest: donating behavior, household size, joining behavior, environmentally friendly behaviors, and total organizational activities. There was a positive relationship between environmental behavior, total activities and the dependent variable, length of interest in the environment.
As the number of environmental behaviors or organizational memberships increased, length of interest in the environment increased (up to 39 years of interest). There was also a positive relationship between donating behavior and joining behavior. Respondents who were joiners were more likely to have been interested in the environment for a longer time (up to 39 years).
The resulting equation accounted for only thirteen percent of the total variance (see Table 42). 127
Table 42
Regression of Length of Interest on Selected Independent Variables
Variable B SEB Beta T SigT
Donate -1.78 .98 -.05 -1.80 .070 Household Size -.65 .31 -.06 -2.05 .039 Joining Behavior 1.77 .84 .06 2.09 .0.6 Env. Behaviors .52 .17 .09 3.01 .002 Total Activities 2.96 .36 .27 8.15 .000
Regression Statistics M ultiple R .35 R Square .13 Adjusted R Square .12 Standard Error 12.84 F 27.74 Significant F .000
Analysis of Variance DF Sum of Squares Mean Square Regression 5 22884 4576 Residual 962 158713 164 n=969 128
Question Three. What incentives motivate people to join environmental/conservation organizations?
Analysis for this question was designed to help determine what motivates people to join environmental/conservation organizations. Specifically, factor analysis was utilized to help determine the underlying incentive factors that motivate membership in environmental/conservation organizations. Then, a series of t-tests and ANOVAs were utilized to help determine which incentives were most important to which market segments.
A twelve-item incentives scale was adapted from Dennis and Zube (1988), to provide insight into how respondents felt about membership in environmental and conservation organizations.
An exploratory factor analysis was employed to group the incentives into categories resulting in ease in data interpretation. Principle components factor analysis with orthogonal varimax rotation produced three factors in six iterations (see Table 43). Items were included in the factor analysis based on two decision rules. The resulting factors all had rotated Eigenvalues greater than one and no item was loaded with a factor loading less than .50. As a result of this second decision rule, two items were excluded from the analysis. Reliability measures for the three factors were computed using Cronbach's alpha. The three factors had alpha values of
.89, .89, and .65 respectively.
Factor one consisted of five items and was labeled "expressive." The items in this factor included elements that describe the personal benefits, or expressive incentives a respondent might get from membership in an environmental or conservation organization.
Factor two consisted of three elements and was labeled "political efficacy." The elements included in this factor expressed statements such that individual membership in an environmental or conservation organization was a way to work as a part of a larger whole towards political action and environmental problem solving. Factor three consisted of two elements and was labeled "Normative Compliance." The elements included in this factor included "joining because friends encourage respondents to belong" and" joining to benefit current and future generations." 130
Table 43
Incentive Factors
Statem ent Expressive Political Normative
Members gain much from the information they receive from these organizations. .56 .28 .38
Belonging to these groups makes members feel better about what they are doing with their lives. .65 .17 .14
Belonging to these groups is one of the most important things in member's lives. .66 .04 .12
I think membership in an environmental/conservation organization is important because many knowledgeable and influential people support these groups. .74 .07 -.10
Environmental/conservation magazines and other publications are one of the most important reasons why I would join these organizations. .63 .09 -.06
Organized groups are very effective in influencing national environmental issues. .09 .74 .06
Membership contributions help environmental/ conservation organizations influence government action on conservation problems. .07 .76 -.04
If people like me contribute to an environmental or conservation organization, the group will be better able to work for improved environmental quality. .37 .53 .26
If environmental/conservation organizations achieve their goals, my life and the lives of the next generation will benefit. .32 .33 .62
People become members of environmental/ conservation organizations because their friends encourage them to belong. .18 .15 .75 131
Relationships between the three factors and the market segments identified in question two were examined. Mean scores for each market segment were compared using the students t-test for independent groups. Since the n's were substantially different, separate variance estimates were utilized to determine significance. Table 44 presents the results for each segment by factor.
Membership
Members rated all three incentives for participating in environmental and conservation organizations higher than did nonmembers (p=.000). Members rated expressive incentives, those personal gain benefits received from organizations as sixteen out of a maximum of twenty-five. This was significantly different than nonmembers who rated expressive incentives as 14.8 out of twenty-five. On the political efficacy scale, members rated the importance of those incentives as almost twelve (11.9) out of fifteen while nonmembers rated those items as 10.5 out of fifteen. Similarly, the third factor, normative compliance was also rated higher by members than by nonmembers.
Donations
Significant differences were also found on all three incentive factors between those who gave money to an environmental/conservation organization than to those who did not. Donors scored significantly (p=.001) higher on the rating of the expressive incentives scale than did non-donors. Donors similarly rated the political efficacy and the normative compliance factors significantly higher than did non-donors.
Environmental Concern
Those who were extremely environmentally concerned rated the political efficacy factor as significantly (p=.004) more important than did those who were very environmentally concerned.
Table 44
Mean Score Differences of Incentives for Participating in
Environmental and Conservation Organizations by Factor.
Incentive/Group n mean sd df t-value P
EXPRESS /member 154 16.0 3.4 /nonmember 701 14.8 3.69 237 3.82 .000
/donor 274 15.6 3.63 /nondonor 569 14.7 3.66 542 3.50 .001
EFFICACY /member 155 11.9 2.06 /nonmember 713 10.5 2.25 240 7.19 .000
/donor 275 11.5 2.01 /nondonor 580 10.4 2.27 585 6.72 .000
/ extremely concerned 155 11.3 2.33 /very concerned 653 10.6 2.27 228 2.94 .004
NORMATIVE COMPLIANCE /member 154 7.2 1.03 /nonmember 718 6.9 1.29 267 3.04 .003
/donor 274 7.2 1.08 /nondonor 586 6.9 1.32 643 2.95 .003 133
To further examine the differences between the market segments of interest, scores on the three scales were normalized by dividing the total score for the factor into the number of items in that score. Using a paired sample t-test all possible combinations of the three factors were examined to see if a particular market segment scored higher on one incentive than another (see Table 45).
Members
Significant differences were found between the normalized scores for all possible combinations of the three factors (p=.000). The most important incentive for participating in environmental/conservation organizations for members, in descending order, were the political efficacy factor, the normative compliance factor, and the expressive factor. Those individuals who were non-members ranked political efficacy as significantly more important than expressive incentives (p=.000) and normative compliance significantly more important than expressive incentives (p=.000). For non-members there was, however, no significant difference between the normative compliance incentive and the political efficacy incentive.
Environmental Concern
Those individuals who were extremely environmentally concerned also ranked the three factors in the same order as did members. The differences between all possible pairs of the three factors were significant (p=.000). Those individuals who were very environmentally concerned ranked political efficacy as significantly more important than expressive incentives (p=.000) and normative compliance as significantly more important than expressive incentives (p=.000). There was, however, no significant difference between the normative compliance incentive and the political efficacy incentive for very environmentally concerned citizens. 134
Donors
Donors also ranked the incentives the same way as the other three segments. In descending order, the political efficacy factor, the normative compliance factor and the expressive factor were the most influential incentives. Those individuals who were non-donors ranked political efficacy as significantly more important than expressive incentives (p=.000) and normative compliance significantly more important than expressive incentives (p=.000). There was, however, no significant difference between the normative compliance incentive and the political efficacy incentive for non-donors. 135
Table 45
Mean Score Differences of Incentives for Participating in
Environmental and Conservation Organizations by Segment.
Segment/Factor n mean sd df t-value p
MEMBERS /expressive 154 3.2 .68 /efficacy 154 3.9 .69 153 -11.21 .000
/expressive 153 3.2 .68 /normative compliance 153 3.6 .51 152 -7.74 .000
/ efficacy 154 3.9 .68 / normative compliance 154 3.6 .51 153 5.24 .000
NONMEMBERS
/expressive 695 2.9 .74 /efficacy 695 3.5 .74 694 -19.08 .000
/expressive 698 2.9 .74 / normative compliance 698 3.4 .64 697 -16.66 .000
EXTREMELY CONCERNED
/expressive 153 3.1 .83 /efficacy 153 3.7 .76 152 . -9.90 .000
/expressive 153 3.0 .82 /normative compliance 153 3.5 .68 152 -6.47 .000
/efficacy 153 3.7 .77 / normative compliance 153 3.5 .68 152 2.53 .013
VERY CONCERNED
/expressive 636 2.9 .72 /efficacy 636 3.5 .74 635 -18.13 .000
/expressive 637 2.9 .72 /normative compliance 637 3.4 .62 636 -16.17 .000 136
Table 45 (continued)
Segment/Factor n mean sd df t-value P
DONORS /expressive 274 3.1 .73 / efficacy 274 3.8 .69 273 -15.90 .000
/expressive 273 3.1 .72 /normative compliance 273 3.6 .54 272 -9.91 .000
/ efficacy 274 3.8 .69 / normative compliance 274 3.6 .54 273 5.14 .000
NON-DONORS
/ expressive 563 2.9 .73 / efficacy 563 3.5 .74 562 -16.37 .000
/ expressive 567 2.9 .73 /normative compliance 567 3.4 .663 566 -15.31 .000 137
Social Class
Social class was significantly correlated with expressive incentives (p=.004). Specifically, there was a significant difference between lower class respondents ranking of expressive incentives (x=15.71) and middle class respondents ranking of expressive incentives (x=14.19)
(see Table 46).
Table 46
Oneway Analysis of Variance Between Social Class and Incentives
Expressive Incentives Lower L. Middle Middle U. Middle Upper Mean 15,71a 15.22 14.493 14.62 14.67
Source dL ss ms F ratio Probability Between Groups 4 209 52 3.86 .004 Within Groups 830 11252 13 Total 834 11461 a Means joined by a common superscript are significantly different at p=.05 (Scheffe)
Length of Interest in the Environment
There was a significant relationship between length of interest in the environment and incentives for joining organizations. There was a significant relationship between political efficacy and length of interest in the environment. Specifically there was a relationship between those who indicated that they were not interested in the environment and those who were interested in the environment for between ten and thirty-nine years. While there was an overall significant relationship between the length of interest in the environment and expessive and normative compliance incentives, a post hoc analysis found no significant 138 differences between means (see Table 47).
Table 47
Oneway Analysis of Variance Between Length of Interest and Incentives
Political Efficacy 0 Years 1 —4 5—9 10—19 20—29 30—39 40+Years Mean 9.97a 10.70 10.52 l l .l l 3 11.253 11.55a 10.97 Source df_ ss ms F ratio Probabilitv Between Groups 6 196 32 6.51 .000 Within Groups 870 4377 5 Total 876 4574 a Means joined by a common superscript are significant at the .05 level (Scheffe)
The three incentive factors were added to the list of independent variables used to discriminate between market segments to see if they assisted in the power of that discrimination.
Without the incentive factors, twelve iterations of Wilk's style stepwise discriminant analysis resulted in twelve variables that were loaded that resulted in a seventy-nine percent (79.4%) correct classification rate of members and nonmembers. With the addition of the three incentive factors to the variable list, fourteen iterations resulted in fourteen variables including the incentive factor, political efficacy. The correct classification rate of members and nonmembers remained at seventy-nine percent (79.0%) but the chance of correctly classifying a member improved (69.3% to 74.0%) while the chance of correctly classifying a nonmember and the correct classification of ungrouped cases decreased. A comparison of "extremely concerned" versus "very environmentally concerned" respondents yielded similar results. Nine iterations resulted in a correct classification of sixty-two
(62.2%) of the respondents without the use of the incentive factors. With incentive factors included in the discriminant analysis, that classification rate increased to sixty-six percent
(65.7%) and the political efficacy variable was included within twelve iterations. 140
Question Four. What are the characteristics of an ideal environmental/conservation organization?
Understanding the characteristics of an ideal environmental/conservation organization from
the perspective on the respondents may help organizations design better products and better
position and market their offerings. Underlying characteristics of the ideal organization
based on products, service, and opportunity needs, ideal characteristics, and ideal activities
were studied using factor analysis. Then the differences between segments based on the
derived factors were studied.
Data to support this research question came from a number of different survey questions.
Questions twelve and thirteen provided information, in an ordinal format, on the relative
importance of products, services, and opportunities that could be offered by
environmental/conservation organizations. In question fourteen, respondents were asked to
identify the elements of an ideal organization and in question eighteen they were asked to
indicate which actions they considered appropriate for solving environmental problems.
Without specifying any optional "other" characteristics, respondents were presented with a
list of forty-one characteristics. In order for more meaningful interpretation of the results,
factor analysis was utilized on the results of these questions to assist in the identification of
smaller groups of characteristics.
Products, Services, and Opportunities. Respondents were asked to rate, on a Likert-type scale,
the importance of twelve products, services, and opportunities they would like from an
environmental/conservation organization. An exploratory factor analysis was utilized to
group these items into a smaller set of like services. Principal components factor analysis
with orthogonal varimax rotation extracted four factors in eight iterations. Table 48 displays
the results. 141
Factor one contained four items and was labeled "local structure." Each of these elements related to products or services that were consistent with a local chapter structure to an environmental/conservation organization.
Factor two contained five items and was labeled "hedonic opportunities." These five items describe a package of membership opportunities that, with the exception of financial contributions, are more hedonic in nature. Within this context, however, financial contributions could be seen as providing hedonic benefits to the giver.
Factor three consisted of four items and was labeled "social benefits." Each of these elements had in common the provision of socializing benefits ranging from meeting similar people to adventure travel.
Factor four consisted of three items and was labeled "tangible needs." These three items were tangible products that were all tangible physical benefits of membership. 142
Table 48
Factor Analysis for
Product, Service and Opportunity Items
Item Mean Factor 1 Factor 2 Factor 3 Factor 4 [LOCAL] [HEDONIC] [SOCIAL] [TANGIBLE]
local chapter 3.1 .88 .16 .03 .13 local outings 2.8 .77 .08 .38 .06 local chapter meetings 2.8 .89 .16 .03 .13 conferences or training 3.0 .65 .34 .14 .14
chance to become more 4.1 .26 .73 .06 .25 knowledgeable opportunity to financially 2.8 .00 .64 .26 .00 support a cause chance to get closer to nature 1.2 .18 .64 .36 .07 chance to help educate 3.9 .22 .81 .01 .13 others opportunity to help solve 4.1 .19 .83 .01 .15 problems
' adventure travel 2.7 .20 .08 .72 .23 opportunity to meet similar 3.2 .46 .29 .50 .02 people entertainment, enjoyment, fun 2.8 .14 .18 .76 .03 gift/catalog services 2.5 -.09 .05 .56 .56
magazine 3.3 .10 .12 .16 .79 chapter newsletters/bulletins 3.6 .38 .38 .00 .60 gift/catalog services 2.5 -.09 .05 .56 .56
Cronbach's alpha was used to compute reliability indexes for each of these four factors. Each of the alpha's was greater .70 (see Table 49). 143
Table 49
Reliability of Product and Opportunity Factors
Factor No. of Items Cronbach's Items A lp h a
1. Local Structure 4 .92 2. Hedonic Benefits 5 .94 3. Social Benefits 4 .84 4. Tangible Needs 2 .79
Mean scores on the four factors for each market segment were compared using the students t- test for independent groups. Separate variance estimates were utilized to determine significance given the relative different n's. Table 50 presents the results by factor.
Members
Members and nonmembers did not differ significantly with respect to social benefits and local structures. Members, however, rated tangible needs more highly than did nonmembers
(p=.008). Members also rated hedonic needs significantly higher than did nonmembers
(p=.000).
Donors
Donors differed significantly to non-donors on all four need factors. Donors' mean scores were significantly higher on local structure (p=.028), social benefits (p=.030), tangible needs
(p=.000), and hedonic needs (p=.000). 144
Joiners
Joiners and nonjoiners did not differ significantly on any of the four need factors.
Environmental Concern
Those individuals who were extremely environmentally concerned rated only the tangible needs significantly higher (p=.05) than did those individuals who were very environmentally concerned. 145
Table 50
Mean Score Differences Between Respondents
Product, Service, and Opportunity Needs
Need/Group n mean sd df t-value p
LOCAL STRUCTURE
/donor 271 12.2 4.19 /nondonor 568 11.4 4.60 578 2.21 .028 SOCIAL
/donor 273 11.6 3.52 /nondonor 564 11.0 3.72 565 2.17 .030 TANGIBLE
/member 153 9.3 1.91 /nonmember 718 8.9 2.19 244 2.69 .008
/donor 272 9.4 1.89 /nondonor 585 8.7 2.23 615 4.85 .000
/ extremely concerned 158 9.3 2.09 / very concerned 651 8.9 2.17 245 1.93 .055 HEDONIC /member 155 19.6 3.69 /nonmember 715 18.2 4.52 264 4.07 .000
/donor 278 20.0 3.39 /nondonor 578 17.6 4.66 722 8.36 .000
Social Class
Utilizing an ANOVA analysis with a post hoc Scheffe test, there were no significant differences discovered between social class and product, service, and opportunity factors. 146
Length of Interest in the Environment
There is a significant difference between length of interest in the environment and product, service, and opportunity factors. While there was a significant difference between length of interest in the environment and local structure a post hoc Scheffe test found no significant differences between means. There was a significant difference between tangible needs and length of interest in the environment. Specifically there was a difference between those who were not interested in the environment (x=8.41) and those who were interested in the environment for between five and nine years (x=9.33). There was a significant difference between hedonic benefits and length of interest in the environment. Specifically, there was a significant difference between those who were not interested in the environment (x=15.87) and those who were interested in the environment from between one and thirty-nine years (see
Table 51). 147
Table 51
Oneway Analysis of Variance Between Length of Interest
and Product, Service, and Opportunity Factors
Tangible Needs 0 Years 1 —4 5—9 10—19 20—29 30—39 40+Years
Mean 8.41a 9.00 9.33a 9.18 9.00 9.25 9.25
Spurce d i §§. ms F ratio Probability Between Groups 6 80 13 2.90 .0083 Within Groups 873 4021 4 Total 879 4101
Hedonic Benefits 0 Years 1 —4 5—9 10—19 20—29 30—39 40+Years Mean 15.873 18.74a 19.24a 19.14a 19.40a 19.48a 17.52 Source dL ss ms F ratio Probability Between Groups 6 1531 255 14.24 .000 Within Groups 872 15631 17 Total 878 17162 a Means joined by a common superscript are significantly different at p=.05 (Scheffe).
Characteristics of an Ideal Environmental Organization. Principal component analysis using orthogonal varimax rotation was used to determine common clusters of characteristics of an ideal environmental/conservation organization. Fifteen presence and absence attributes were coded as ones or twos and after ten iterations, four factors with Eigenvalues greater than one were extracted. Eleven of the fifteen characteristics were accounted for in these four factors
(see Table 52).
Factor one consisted of six items: global, educational, nature appreciation, preservation, wise- use, and problem-solving, and was labeled "traditional." Factor two consisted of two items: liberal and pacifist and was labeled "liberal." Factor three also contained two items: radical and militant, and was labeled "radical." The fourth factor contained only one item, single issue. Reliability measures for all four factors were computed using Cronbach's alpha. Factor one obtained an alpha of .80 while the second and third factors had lower alphas (r=.48 and r=.64 respectively). An alpha could not be computed on the fourth factor since it contained only a single item.
Table 52
Factor Analysis for
Characteristics of an Ideal Organization
Characteristic Factor 1 Factor 2 Factor 3 Factor 4 TRADITIONAL LIBERAL RADICAL SINGLE
global .52 .17 .20 -.05 educational .74 .02 .04 -.04 nature appreciation .70 -.03 -.01 .13 preservation .74 .08 -.03 -.04 wise-use .68 .14 -.02 -.01 environmental problem-solving .72 .07 -.03 -.04 liberal .03 .74 .10 -.01 pacifist .00 .64 .17 .06 radical -.05 .10 .81 .00 m ilitan t -.00 .08 .85 -.05 single issue -.09 .16 -.06 .78
Factors not extracted in analysis, local conservative multi-issue legislative influence 149
The first three factors for the characteristics of an ideal organization were analyzed by market segment using a t-test for independent groups (see Table 53). The single element factor, and those factors that did not load were analyzed separately using the chi-square statistic
(see Table 54).
Member
Members rated traditional ideal characteristics significantly higher than did nonmembers
(p=.000). Members had a mean score of 7.7 compared to nonmembers whose mean score was 3.6.
Members also rated the liberal ideal characteristics significantly higher than did nonmembers (p=.000). Members mean score was .4 compared to nonmembers mean score of .1. In a chi-square test to determine differences between members and nonmembers on the single characteristic, multi-issue, members tended to prefer the multi-issue characteristic (p=.000).
Donor
Donors rated the first three ideal characteristic factors significantly higher than did non donors. Donors mean score on traditional characteristics was 4.6 compared to 3.5 for non-donors
(p=.005). Donors mean score for liberal characteristics was .3 compared to non-donors mean score of .1 (p=.000). The mean score for donors on radical characteristics was .1 compared to .0 for nonmembers (p=.009). In a chi-square test to determine the difference between characteristics that did not load as factors, donors tended to prefer the multi-issue characteristic (p=.000).
Joiner
Joiners rated only the traditional characteristics significantly higher than did non-joiners
(p=.005). Joiners mean score was 4.0 compared to 3.6 for non-joiners. Environmental Concern
No differences were found with respect to ideal characteristic factors for those who were
"extremely" versus "very environmentally concerned."
Table 53
Mean Score Differences Between Respondents
Ideal Characteristics of an Environmental/Conservation Organization
Characteristic/Group n mean sd df t-value P
TRADITIONAL /member 156 7.7 1.36 / nonmember 849 3.6 2.05 301 8.40 .000
/donor 285 4.6 1.38 /nondonor 700 3.5 2.10 786 9.75 .000
/joiner 512 4.0 1.84 /nonjoiner 469 3.6 2.1 934 2.80 .005 LIBERAL /member 156 .4 .67 /nonmember 849 .1 .42 177 4.41 .000
/donor 285 .3 .57 / nondonor 700 .1 .42 418 4.68 .000 RADICAL /donor 285 .1 .37 /nondonor 700 .0 .233 376 2.64 .009 151
Table 54
Summary of Significant Relationships Between Characteristics of an Ideal
Environmental/Conservation Organization and Market Segments
Characteristic/Group n Chi-Square df P
MULTI-ISSUE member/nonmember 890 39.52 1 .000 donor/nondonor 878 31.14 1 .000
Social Class
There was a significant difference between social class and traditional characteristics of an ideal organization. Specifically, there was a significant difference between the lower social class rankings of traditional characteristics and the middle to upper class segments.
Additionally, there was a significant difference between lower middle class segments
(x=3.71) and upper class segments (x=4.50) (see Table 55). 152
Table 55
Oneway Analysis of Variance Between Social Class
and Characteristics of an Ideal Organization
Traditional Organizations Lower L. Middle Middle U. Middle Upper
Mean 3.39a 3.71b 3.98a 4.21a 4.50ab Source df_ ss ms F ratio Probability Between Groups 4 144 36 9.23 .000 Within Groups 978 3834 4 Total 982 3979 a Means joined by a common superscript are significantly different at p=.05 (Scheffe)
Length of Interest in the Environment
There was a significant difference between traditional characteristics and length of interest in the environment. Specifically, there is a significant difference between those who indicated they were not interested in the environment and those who were interested in the environment for any number of years. There is a significant difference between liberal characteristics of an ideal organization and length of interest in the environment.
Specifically, there is a significant difference between those who did not indicate a length of time for interest in the environment (x=.05) and those who were interested in the environment for between ten and twenty-nine years (x=.28 and x=.38). There was a significant difference between those who were interested in the environment for between one and four years (x=.10) and those who were interested for between ten and twenty-nine years (x=.28 and x=.38).
Additionally, there was a significant difference between those who were interested in the environment for five and nine years (x=.18) and those who were interested in the environment 153 for between twenty and twenty-nine years (x=.38) (see Table 56).
Table 56
Oneway Analysis of Variance Between Length of Interest
and Characteristics of an Ideal Organization
Traditional Characteristics 0 Years 1 —4 5—9 10—19 20—29 30—39 40+Years Mean :2.23a 4.05a 4.23a 4.42a 4.63a 4.403 4.71a Source df_ ss. ms F ratio Probabilitv Between Groups 6 930 155 49.71 .000 Within Groups 1009 3147 3 Total 1015 4077
Liberal Characteristics 0 Years 1 —4 5—9 10—19 20—29 30—39 40+Years Mean .05a .10b ,18c .28ab ,38abc .20 .15 Source df_ ss ms F ratio Probabilitv Between Groups 6 13 2 10.73 .000 Within Groups 1009 215 .21 Total 1015 228 a Means joined by a common superscript are significantly different at the .05 level (Scheffe).
Appropriate Actions. Fourteen items were included on the instrument to determine which actions respondents felt were appropriate for solving environmental problems. In order to reduce these items to a manageable number and to find common groupings, principal component factor analysis using orthogonal varimax rotation was used. Within five iterations, three factors were extracted that included all but one (lobbying) of the fourteen items (see Table 57). 154
Factor one contained six items including: letter writing, petitions, supporting research, publicity, fundraising, and education. Since all of these items were action oriented but were political moderate this factor was named "low risk actions." Cronbach's alpha was used as the reliability factor for this factor and an alpha of .80 was determined.
Factor two contained five items including: marching, lawsuits, testifying at hearings, boycotting, and strikes. Once again all of these items had an action orientation but were more politically oriented thus this factors was named "demonstration actions." Cronbach's alpha for this factor was .73.
The third factor contained only two elements, property damage and violent actions against people and was name "eco-terrorism actions." Only a few individuals (less than five percent for each item) reported doing either of these activities which may help explain the relatively small alpha (r=.38) that was determined using the Cronbach's reliability procedure. 155
Table 57
Factor Analysis for
Appropriate Actions for Solving Environmental Problems
Action Factor 1 Factor 2 Factor 3 LOW RISK DEMONSTRATION ECO-TERRORIST
letter writing campaigns .68 .15 .06 petitions .59 .32 -.03 supporting research .77 -.01 .09 publicity .69 .14 -.01 fundraising .63 .20 .05 education .73 .16 .02 marching, picketing .07 .78 .04 law suits .10 .72 .10 testifying at hearings .45 .51 .07 boycotting products .24 .60 -.06 strikes .10 .68 .03 property damage .01 .09 .77 violence against people .04 .08 .79 * lobbying
* Factor not extracted in analysis.
To determine differences between market segments based on these three factors, a t-test for independent groups was used between each market segment (see Table 58). No significant differences were found between segments with respect to factor three, the eco-terrorist actions, probably because of the low numbers of individuals who did either of these activities.
Lobbying, the action that did not load as a factor was analyzed separately using chi-square analysis. 156
Members
Members rated both low risk and demonstration actions considerably more than did nonmembers. Members supported an average of 4.9 low risk actions compared to nonmembers who supported an average of 3.7 low risk actions (p=.000). Members preferred an average of
1.9 demonstration actions compared to nonmembers who preferred an average of only 1.0 demonstration actions (p=.000). In a chi-square analysis of lobbying as an independent action, members were more likely to support lobbying than were nonmembers (p=.000).
Joiners
Joiners supported significantly more low risk actions than did non-joiners (p=.020). Joiners supported an average of 4.1 low risk actions compared to non-joiners who supported on average, 3.8 actions.
Donors
Donors selected significantly more low risk and demonstration actions than did non-donors.
Donors selected an average of 4.7 low risk actions compared to non-donors average of 3.6
(p=.000). Donors selected an average of 1.7 demonstration actions compared to non-donors who averaged only 1.0 demonstration actions (p=.000). In a chi-square analysis of lobbying as an independent action, donors were more likely to select lobbying than were non-donors (p=.000)
(see Table 59).
Environmental Concern
No significant differences were found between those who were extremely versus very environmentally concerned with respect to actions appropriate for solving environmental problems. Mean Score Differences Between Respondents
Actions Appropriate for Solving Environmental Problems
Action/Group n mean sd df t-value P
LOW RISK /member 156 4.9 1.31 /nonmember 849 3.7 1.99 304 9.76 .000
/donor 285 4.7 1.54 /nondonor 700 3.6 2.01 684 8.89 .000
/joiner 512 4.1 1.81 /nonjoiner 469 3.8 2.06 934 2.34 .020
DEMONSTRATION /member 156 1.9 1.61 /nonmember 849 1.0 1.34 196 6.03 .000
/donor 285 1.7 1.53 /nondonor 700 1.0 1.33 468 6.60 .000 158
Table 59
Summary of Significant Relationships Between Lobbying as an
Appropriate Action by Market Segment
Group n Chi-Square df P
member/nonmember 930 20.89 1 .000 donor/nondonor 913 16.22 2 .000
Social Class
There was a significant relationship between social classes and low risk activities and demonstration activities (see Table 60). There was a significant relationship between lower class respondents and the middle to upper class segments on low risk activities. There was a significant relationship between lower class respondents and upper class segments on demonstration activities. 159
Table 60
Oneway Analysis of Variance Between Social Class
and Appropriate Actions for Solving Environmental Problems
Low Risk Activities Lower L. Middle M iddle U. Middle Upper Mean 3.48a 3.95 4.03a 4.10a 4.57a Source df ss ms F ratio Probabilitv Between Groups 4 146 36 9.77 .000 Within Groups 978 3669 3 Total 982 3816
Low Risk Activities Lower L. Middle Middle U. M iddle Upper Mean 1.00a 1.13 1.32 1.37 1.54a Source df ss ms F ratio Probabilitv Between Groups 4 34 8 4.30 .001 Within Groups 978 1961 2 Total 982 1995 a Means joined by a common superscript are significantly different at p=.05 (Scheffe)
Length of Interest in the Environment
There was a significant relationship between appropriate actions for solving environmental problems and length of interest in the environment (see Table 61). There was a significant difference between low risk activities and length of interest in the environment. Specifically, there was a relationship between those who did not indicate any years of interest in the environment and those who were interested in the environment for any length of time.
Additionally, there was a significant relationship between demonstration activities and 160 length of interest in the environment. Specifically, there was a significant difference between those who did not indicate any length of interest in the environment and those who had between interested in the environment for between one and twenty-nine years.
Table 61
Oneway Analysis of Variance Between Length of Interest
and Appropriate Actions for Solving Environmental Problems
Low Risk 0 Years 1 —4 5—9 10—19 20—29 30—39 40+Years
Mean 2.49a 4.31a 4.34a 4.45a 4.83a 3.87a 4.40a Source df ss. ms F ratio Probabilitv Between Groups 6 787 131 42.28 .000 Within Groups 1009 3131 3 Total 1015 3918
Demonstration Activities 0 Years 1 —4 5—9 10—19 20—29 30—39 40+Years Mean .51a 1.12ab 1.36ae 1.52a 2.00abcde 1.12c 1.17^ Source df ss ms F ratio Probabilitv Between Groups 6 246 41 22.86 .000 Within Groups 1009 1812 1.7 Total 1015 2058 a Means joined by a common superscript are significantly different at the .05 level (Scheffe). 161
Question Five. How are environmental/conservation organizations perceived by consumers?
Understanding consumer preferences and perceptions of environmental/conservation organizations can be used to help determine market structure from the consumer's point of view.
Three questions provided most of the information for this section of the analysis. In question fifteen, respondents were asked to indicate which of a list of twenty organizations, including the ten study organizations, they would consider joining. In question nine, respondents were asked to indicate how similar all possible pairs of ten organizations were. This similarities data was collected for analysis utilizing multidimensional scaling, a perceptual mapping technique. In question ten, respondents ranked the ten study organizations from most preferred to least preferred. Finally, to assist in interpretation of the results, respondents were also asked, in an open-ended format, to indicate what they liked most about their number one choice.
Organizations Respondents Would Consider joining.
Principal components factor analysis was used to attempt to group together the twenty possible organizations that respondents would consider joining. Using the orthogonal varimax rotation procedure with nine iterations, six factors were extracted. Only two organizations
(The Nature Conservancy and Conservation International) were not included in these factors, however, low Cronbach's alphas on four of the six scales suggests caution in the interpretation of the resulting factors (see Table 62).
Factor one included five organizations that with one exception have in common a focus on wildlife, thus, these organizations were grouped together in a factor named "wildlife organizations." The exception, the Wilderness Society, certainly has wildlife as one of its concerns but instead of a species orientation it tends to focus more at a habitat level. Since, 162 however, the two orientations are compatible this common name will be used. A reliability indicator, using the Cronbach's procedure resulted in an alpha of .77.
The second factor grouped together five very action oriented organizations and thus the factor was named "action organizations." Additionally, these organizations have in common a fairly single or dominant issue focus. A somewhat lower Cronbach's alpha, r=.66, was calculated for this factor.
The third factor consisted of two organizations, the League of Conservation Voters and the
Natural Resource Defense Council. These organizations both work within the political and legal spheres and thus the factor was named "political/legal organizations." A low
Cronbach's alpha of .40 was computed for this factor suggesting caution in the interpretation of the reliability of this factor.
The fourth factor, also with an extremely low Cronbach's alpha (.34) also contained just two organizations, the Cousteau Society and the Sierra Club. While these organizations are both fairly large and traditional no single common element between the two seemed obvious.
Because of the low reliability indicator, it is not recommended that much credence be placed in this factor grouping.
The fifth and sixth factors have similar problems. Three organizations, the National Arbor
Day Foundation, The Audubon Society, and Ducks Unlimited are included in the fifth factor.
An alpha of .34 was obtained using the Cronbach's reliability procedure. The final factor includes only one item, the Sea Shepherd Conservation Society, which is a relatively small, animal-oriented, activist organization. The low frequency with which this organization was reported (2.4%) may account in large part for this factors extraction. The alpha for this last factor could not be computed with only one item in the scale. 163
Table 62
Factor Analysis for
Organizations Respondents Would Consider Joining
Organization Fact. 1 Fact. 2 Fact. 3 Fact. 4 Fact. 5 Fact. 6 WILDLIFE ACTION POLITICFOUNDER
Friends of Animals .66 .10 -.10 -.05 .06 .41 W orld Wildlife Fund .73 .13 .04 .18 -.04 -.00 Defenders of Wildlife .69 .20 .00 -.18 .03 .30 W ilderness Society .66 .14 .26 .11 .04 .00 National Wldlife Federation .71 .06 .04 .09 .13 -.06 Ntl. Toxics Campaign .02 .62 .16 -.03 .05 .21 Environmental Defense Fund .23 .52 .37 .09 -.04 -.12 Greenpeace .15 .54 -.14 .41 -.12 .13 Rainforest Action Network .16 .62 .11 .21 .04 .01 Earth! First! .16 .66 .01 -.12 .02 .06 Nat. Res. Defense Council .01 .43 .56 -.06 .02 -.04 League of Conservation Voters .03 .02 .69 .09 -.01 .18 Cousteau Society .02 .12 -.02 .64 .12 .35 Sierra Club .11 -.01 .10 .71 -.02 -.10 Ntl. Arbor Day Foundation .00 .23 .01 -.03 .73 -.06 Ntl. A udubon Society .16 -.12 .13 .26 .64 -.11 Ducks Unlimited .06 -.10 -.03 -.14 .55 .32 Sea Shepherd Con. Society .10 .14 .10 .15 -.08 .67 * Conservation International * The Nature Conservancy
* Factors not extracted in analysis.
Given the low reliability values for the last four factors, only the first two factors will be summated and analyzed using t-tests (see Table 63). Analysis for the remaining organizations by market segment will utilize Chi Square analysis (see Table 64). 164
Members/Nonmembers
Members of environmental/conservation organizations indicated that they would be more likely to join any of the wildlife organizations than did nonmembers (p=.000). They also indicated that they would be more likely to join action oriented organizations than did non members (p=.001). Members would also be more likely to join the following organizations:
Natural Resource Defense Council, League of Conservation Voters, Sierra Club, Audubon
Society, National Arbor Day Foundation, and The Nature Conservancy.
Donors/Non-Donors
Donors would be more likely to join both wildlife organizations (p=.000) and action organizations (p=.000) than would individuals who did not give money to an organization.
Donors were also more likely to join all eight other organizations not included in the first two sets of organizations.
Joiners/Non-Joiners
No significant differences were noted between joiners and non-joiners with respect to organizations they would consider joining.
Environmental Concern
Individuals who were extremely concerned about the environment, were more likely to consider joining action organization than were nonmembers (p=.002). They were also more likely to consider joining the League of Conservation Voters, the Natural Resource Defense
Council, and the Sea Shepherd Conservation Society. 165
Table 63
Mean Score Differences of Organizations Respondents
Would Consider Joining by Market Segment
Organization/Group n mean sd df t-value p
WILDLIFE ORGANIZATIONS
/member 156 1.7 1.58 /nonmember 849 1.0 1.44 205 5.15 .000
/donators 285 1.7 1.61 /non-donators 700 .9 1.38 462 7.54 .000
ACTION ORGANIZATIONS
/member 156 1.0 1.42 /nonmember 849 .66 1.07 188 3.49 .001
/donators 285 1.2 1.38 /non-donators 700 .54 .97 402 7.48 .000
/ extremely concerned 185 1.0 1.38 ' /very concerned 753 .67 1.06 239 3.13 .002 166
Table 64
Summary of Significant Relationships of Organizations Respondents Would Consider Joining
Organization/Group n Chi-Square df P
Sea Shepherd Conservation Society extr. concerned/very concerned 823 12.06 1 .001 donor/non-donors 869 7.38 1 .006
Natural Resource Defense Council extr. concerned/very concerned 823 4.44 1 .034 member/nonmembers 882 7.56 1 .005 donor/ non-donor 869 17.71 1 .000
League of Conservation Voters extr. concerned/very concerned 823 4.57 1 .032 member/nonmember 882 10.39 1 .001 donor/non-donor 869 10.06 1 .001
Sierra Club member/nonmember 882 14.47 1 .001 donor/non-donors 869 19.59 1 .001
Audubon Society member/nonmember 882 6.86 1 .008 donor/non-donor 869 3.56 1 .059
Arbor Day Foundation member/nonmember 881 24.90 1 .000 donor/non-donor 868 12.27 1 .000
The Nature Conservancy member/nonmember 882 46.38 1 .000 donor/non-donor 869 16.02 1 .000
National Toxics Campaign donor/ non-donor 869 24.18 1 .000
Cousteau Society donor/non-donor 869 20.49 1 .000 167
Organizational Similarities.
Respondents similarity ratings on the forty-five pairs of organizations were averaged by market segment and a similarities matrix constructed. Analysis was conducted using a classical nonmetric MDS model with the SPSS procedure ALSCAL. One Euclidian matrix was constructed for each segment that resulted in a square symmetric data set. Three measures of fit are referred to in the results of each plot: S-stress, Kruskal's stress measure and the squared correlation coefficient. These three measures are used to help determine how well the
Euclidian model or map describes the relationships between the pairs of organizations. S- stress is a measure of fit ranging from one, the worst possible fit, to zero, a perfect fit (Young,
1987). Like goodness of fit measures in factor analysis, S-stress changes with the number of iterations. Kruskal's stress measure is a measure of the proportion of variance or disparity which is not accounted for by the MDS model and helps us determine the appropriate number of dimensions to include in the analysis. The final goodness of fit measure is r-squared measure that is a squared correlation between data and distances.
From a purely statistical sense, maximizing all three goodness of fit measures would be most desirable. However, to minimize stress additional dimensions must be added to the MDS solution and thus interpretability can be clouded. Different MDS procedures maximize different goodness of fit measures. Hair, Anderson, and Tatham (1987) in addition to Young
(??) suggest that when using the ALSCAL procedure the r-square measure of fit is the technique to rely upon since ALSCAL is designed not to maximize Kruskal's stress measure but the ALSCAL S-stress measure of fit. The r-square statistic ranges, as it does in normal correlations, between -1 and +1. Hair, Anderson, and Tatham suggest that an r-square value of
.60 or greater in nonmetric (ordinal level) MDS is considered acceptable. For this analysis, r- square will be maximized at the expense of s-stress or stress measures in order to keep solutions to a minimum number of dimensions. 168
In addition to the goodness of fit measures, some guidelines regarding the shape of the solution can also assist in determining the appropriate number of dimensions. If the data appear to be either clumped into two separate groups or circular in nature, the solution may be somewhat degenerate and may be the result of too low dimensionality (SPSS). 169
Members
Similarities data from members were plotted initially in two dimensions. Given a two dimensional solution within eight iterations an S-stress level of .09, a stress measure of .09 an r-square value of .95 was achieved (see Figure 6).
The horizontal dimension was named the "habitat" dimension because it seemed that the organizations varied from left to right from avian habitats to aquatic habitats. The vertical dimension was named the "organism" dimension because it seemed that the organizations varied from bottom to top from zoological to botanical species. The two most similar organizations were the Wilderness Society and The Nature Conservancy. The Cousteau
Society, Ducks Unlimited, and the National Arbor Day Foundation represented the three organizational extremes that were plotted. Greenpeace received similarities ratings so different than the other organizations that it was not plotted on the map. 2.1 - + : + : 1 : :
1.0 - + : + : : 3 :
: 85 : : 0.0 - + ------+ : 9 7 :
: 6 4 : - 1.0 - + : + : 2 : :
- 2.1 - + : +
-+ ------+------+----+------+-----+------+------+------+------+------+- - 2.5 - 1.5 - 0.5 0.5 1.5 2.5
Key 1 Arbor Day Foundation 2 Ducks Unlimited 3 Sierra Club 4 Cousteau Society 5 Wilderness Society 6 World Wildlife Fund 7 National Wildlife Federation 8 The N ature Conservancy 9 Audubon Society A Greenpeace
Figure 6. Members Organizational Similarities Maps 171
Nonmembers
Similarities data from nonmembers was plotted initially in two dimensions. Given a two dimensional solution within eight iterations an S-stress level of .12, a stress measure of .10 an r-square value of .93 was achieved (see Figure 7).
The horizontal dimension was named the "habitat" dimension because it seemed that the organizations varied from left to right from avian habitats to aquatic habitats. The vertical dimension was named the "organism" dimension because it seemed that the organizations varied from bottom to top from zoological to botanical species. The most similar organizations were the Wilderness Society, The Nature Conservancy, and the Audubon Society. Ducks
Unlimited and the National Arbor Day Foundation represented the most dissimilar organizations. Greenpeace and the National Wildlife Federation received similarities ratings so different than the other organizations that they were not plotted on the map. -+ + + +— -+- -+ + + + -
2.1 -+ +
1.0 - +
0.0 - + - - 9—- 8-
- 1.0 - +
+ - 2.1 - +
- + + -
- 2.5 - 1.5 -0.5 0.5 1.5 2.5
Key 1 Arbor Day Foundation 2 Ducks Unlimited 3 Sierra Club 4 Cou«feau Society 5 W ilderness Society 6 World Wildlife Fund 7 National Wildlife Federation 8 The Nature Conservancy 9 Audubon Society A Greenpeace
Figure 7. Non-Members Organizational Similarities Maps 173
Joiners
Similarities data from joiners was plotted initially in two dimensions. Given a two dimensional solution within six iterations an S-stress level of .12, a stress measure of .10 an r- square value of .94 was achieved (see Figure 8).
The horizontal dimension was named the "habitat" dimension because it seemed that the organizations varied from left to right from avian habitats to aquatic habitats. The vertical dimension was named the "organism" dimension because it seemed that the organizations varied from bottom to top from zoological to botanical species. A number of organizations were relatively similar to each other in terms of distances including the Wilderness Society, The
Nature Conservancy and World Wildlife Fund and National Wildlife Federation. The
Nature Conservancy, the Wilderness Society were plotted on the exact midpoint of the horizontal dimension, with the National Arbor Day Foundation only slightly off the midpoint. Greenpeace, National Arbor Day Foundation, Cousteau Society and Ducks
Unlimited were plotted as the outliers around the cluster of more similar organizations. — I t------+ ——— H I V— ———+— I (-
2.1 - + ' 1:
1.0 - +
- 1.0 - +
4
- 2.1 - +
_+ + + + + + + + + + +
- 2.5 - 1.5 - 0.5 0.5 1.5 2.
1 Arbor Day Foundation 2 Ducks Unlimited 3 Sierra Club 4 Cousteau Society 5 Wilderness Society 6 World Wildlife Fund 7 National Wildlife Federation 8 The Nature Conservancy 9 Audubon Society A Greenpeace
Figure 8. Joiners Organizational Similarities Maps 175
Non-Joiners
Similarities data from non-joiners was plotted initially in two dimensions. Given a two dimensional solution within seven iterations an S-stress level of .11, a stress measure of .09 an r-square value of .94 was achieved (see Figure 9).
The horizontal dimension was named the "habitat" dimension because it seemed that the organizations varied from left to right from avian habitats to aquatic habitats. The vertical dimension was named the "organism" dimension because it seemed that the organizations varied from bottom to top from zoological to botanical species. Non-joiners perceptual map had a number of different relationships in it. The two most similar organizations were the
World Wildlife Fund and the National Wildlife Federation and the two least similar the
Cousteau Society and the National Arbor Day Foundation. There was an almost linear pattern of organizations ranging from the Audubon Society up through the National Arbor
Day Foundation. Compared to joiners, non-joiners ranked most organizations much higher on the vertical dimension. -+ + + + + +- -+ + + + -
2.1 -+
1.0 - +
- 1.0 - +
- 2.1 - +
-+ + + + + + + + + + +- -2.5 - 1.5 - 0.5 0.5 1.5 2.5
Key 1 Arbor Day Foundation 2 Ducks Unlimited 3 Sierra Club 4 Cousteau Society 5 W ilderness Society 6 World Wildlife Fund 7 National Wildlife Federation 8 The Nature Conservancy 9 Audubon Society A Greenpeace
Figure 9. Non-Joiners Organizational Similarities Maps 177
Donors
Similarities data from donors was plotted initially in two dimensions. Given a two dimensional solution within eight iterations an S-stress level of .11, a stress measure of .09 an r-square value of .95 was achieved (see Figure 10).
The horizontal dimension was named the "habitat" dimension because it seemed that the organizations varied from left to right from avian habitats to aquatic habitats. The vertical dimension was named the "organism" dimension because it seemed that the organizations varied from bottom to top from zoological to botanical species. The placement of the organizations on the donors map had the most dissimilar organizations more dispersed than any of the other organizations. The Wilderness Society and the Nature Conservancy were once again the most similar organizations. Five of the ten organizations clustered around the midpoint of the horizontal dimension and five also clustered around the midpoint of the vertical dimension. Thus, in spite of the overall appearance of dispersion, most of the organizations were tightly clustered. 178
-+ + + + + + + + + + +- : 1 2.1 -+
1.0 - +
- 1.0 - +
- 2.1 - +
_+ + + + + + + + + + +-
- 2.5 - 1.5 - 0.5 0.5 1.5 2.5
K< 1 Arbor Day Foundation 2 Ducks Unlimited 3 Sierra Club 4 Cousteau Society 5 Wilderness Society 6 World Wildlife Fund 7 National Wildlife Federation 8 The Nature Conservancy 9 Audubon Society A Greenpeace
Figure 10. Donors Organizational Similarities Maps 179
Non-Donors
Similarities data from non-donors was plotted initially in two dimensions. Given a two dimensional solution within eight iterations an S-stress level of .14, a stress measure of .12 an r-square value of .90 was achieved (see Figure 11).
The horizontal dimension was named the "habitat" dimension because it seemed that the organizations varied from left to right from avian habitats to aquatic habitats. The vertical dimension was named the "organism" dimension because it seemed that the organizations varied from bottom to top from zoological to botanical species. Unlike the donors map, there was very little clustering of organizations for non-donors. Once again the Wilderness Society and The Nature Conservancy were the most similar organizations the the four organizations on the extremes remained the same but overall there was much more spread, particularity on the vertical dimension. - + + + - - + + + + -
2.1 -+
1.0 - +
- 1.0 - +
- 2.1 - +
-+ + + + + + + + + + +- - 2.5 - 1.5 - 0.5 0.5 1.5 2.5
1 Arbor Day Foundation 2 Ducks Unlimited 3 Sierra Club 4 Cousteau Society 5 Wilderness Society 6 World Wildlife Fund 7 National Wildlife Federation 8 The Nature Conservancy 9 Audubon Society A Greenpeace
Figure 11. Non-Donors Organizational Similarities Maps 181
Extremely Environmentally Concerned
Similarities data from respondents who were extremely environmentally concerned was plotted initially in two dimensions. Given a two dimensional solution within six iterations an
S-stress level of .15, a stress measure of .12 an r-square value of .91 was achieved (see Figure
12).
The horizontal dimension was named the "habitat" dimension because it seemed that the organizations varied from left to right from avian habitats to aquatic habitats. The vertical dimension was named the "organism" dimension because it seemed that the organizations varied from bottom to top from zoological to botanical species. The two most similar organizations were the World Wildlife Fund and the National Wildlife Federation. On the horizontal dimension, the most dissimilar organizations were Greenpeace versus the Audubon
Society and Ducks Unlimited. On the vertical dimension, the most dissimilar organizations were the Cousteau Society and the National Arbor Day Foundation. 182
- + + + - - + + + + -
2.1 - + +
1.0 - +
- A - +
- 1.0 - +
- 2.1 - +
-+ + + + +— - - + ------+ + ------+ + - - 2.5 - 1.5 - 0.5 0.5 1.5 2.5
Key 1 Arbor Day Foundation 2 Ducks Unlimited 3 Sierra Club 4 Cousteau Society 5 Wilderness Society 6 World Wildlife Fund 7 National Wildlife Federation 8 The Nature Conservancy 9 Audubon Society A Greenpeace
Figure 12. Extremely Environmentally Concerned Organizational Similarities Maps 183
Very Environmentally Concerned
Similarities data from individuals who were very environmentally concerned were plotted initially in two dimensions. Given a two dimensional solution within six iterations an S-stress level of .13, a stress measure of .10 an r-square value of .94 was achieved (see Figure 13).
Unlike all of the previous perceptual maps, the map for the very environmentally concerned segment is somewhat different. The horizontal dimension in this map has become the
"organism" dimension with organizations ranging from zoological species on the left side to botanical species on the right. The vertical dimension then has become the "habitat" dimension ranging from aquatic habitats on the top to avian habitats on the bottom. This rotation in dimensions may simply have been caused by the way in which the computer numerically solved the Euclidean distances plotted to produce the map. A number of organizations are perceived as fairly similar to each other including: the World W ildlife
Federation and the Wilderness Society and The Nature Conservancy. 184
- + ------+ — ■— + - - + + + -
2.1 -+
1.0 - +
0.0 - + -
- 1.0 - +
- 2.1 - +
— (------h------h------h------h------1------V------y—------+ ------+ + — - 2.5 - 1.5 - 0.5 0.5 1.5 2.5
K ey 1 Arbor Day Foundation 2 Ducks Unlimited 3 Sierra Club 4 Cousteau Society 5 Wilderness Society 6 World Wildlife Fund 7 National Wildlife Federation 8 The Nature Conservancy 9 Audubon Society A Greenpeace
Figure 13. Very Environmentally Concerned Organizational Similarities Maps 185
Organizational Preferences.
Respondents rank ordered the ten study organizations from one — most preferred to ten —least preferred. This rank ordered preference was compared by market segment using the Chi
Square test for independence. Segments that demonstrated significant differences in their rank ordering are displayed in Table 65.
Members and Nonmembers
There was a significant difference between the rank ordering of most preferred organizations by members and nonmembers (p=.03).
Donor/Non-Donors
There was a significant difference between the rank ordering of donors and non-donors (p=.01).
Donors ranked World Wildlife Fund and The Nature Conservancy higher than did non donors.
Joiners/Non-Joiners
There was a significant difference between joiners and nonjoiners (p=.01). While both segments preferred the National Wildlife Federation, Joiners second choice was National
Audubon Society followed by The Nature Conservancy. Non-joiners ranked World W ildlife
Fund second followed by The Nature Conservancy.
Environmental Concern
There was a significant difference between extremely versus very environmentally concerned citizens with respect to rank order preference. While the top five choices changed order only 186 by one place, a number of organizations in the last five places slipped two ranks.
Table 65
Rank Ordered Organizational Preferences by Market Segment
Segment Organizations Member/Non Donor/Non Joiner/Non Extreme/Very
National Wildlife Fed. 1 1 1 1 1 1 1 1 National Audubon Society 3 2 4 2 2 4 3 2 The Nature Conservancy 2 3 3 3 3 3 2 3 World Wildlife Fund 4 4 2 5 4 2 4 4 The Wilderness Society 5 6 6 7 7 5 5 5 National Arbor Day Fndn. 8 7 7 6 7 4 8 7 Cousteau Society 6 6 8 7 5 6 6 6 Sierra Club 7 9 9 8 8 9 9 8 Greenpeace 9 8 7 9 9 7 7 9 Ducks Unlimited 10 10 10 10 10 10 10 10 CHAPTER V
DISCUSSION, CONCLUSIONS, AND RECOMMENDATIONS
The purpose of this study was to determine the perceived structure of the environmental/conservation organization market, and to determine how this perceived market structure varies with individual and environmental differences. General discussion is presented for research questions one, three, four and five followed by a discussion of the findings by market segment. Conclusions and recommendations are provided in the latter part of the chapter.
DISCUSSION
Question One. Who are environmentally concerned citizens and what is their involvement and interest in environmental/conservation organizations?
This study identified selected characteristics of environmentally concerned citizens and their level of involvement and interest with environmental and conservation organizations. Based on these results, a profile of the "typical" environmentally concerned citizen would be as follows: married with few children, white, well educated, and with a median family income of greater than $25,000. Male heads of household were most likely to be employed fulltime in managerial or professional occupations while women heads of household were most likely to be employed parttime. The "typical" respondent was a home owner who lived in a metropolitan area of over 50,000.
187 188
While these characteristics describe "typical" environmentally concerned citizens, the actual sample of respondents was much more diveise. Selection of respondents based on environmental concern was based on a sample balanced to national U.S. household characteristics, thus, an examination of those who met the screening criteria, environmental concern, aids our understanding of what differentiates an environmentally concerned citizen.
Examination of Table 4 (p. 73) shows that on the basis of geographic location, market size, and annual family income, environmentally concerned citizens are very similar to the U.S. population at large. Environmentally concerned citizens are, however, substantially different on the basis of a number of other variables. Environmentally concerned citizens are more likely to be white, to be of non-Spanish origin and to come from smaller families. In addition, education levels for both male and female heads of household are substantially higher for environmentally concerned citizens than the population at large. Likewise, male and female environmentally concerned citizens are more likely to hold higher status occupations, specifically managerial and professional occupations.
In spite of the similarity between annual family income for the environmentally concerned citizen versus the general population, the marked differences shown in both education and occupation between environmentally concerned citizens and the general population supports an often divided literature base that argues that environmentalists are reflective of the elite in society (Harry, Gale, and Hendee, 1969; Devall, 1970; Mitchell, 1980; Perry et al, 1975;
Coombs, 1972; Faich and Gale, 1971). Interestingly, while there appears to be no difference between the sample and the population at large based on income, there does appear to be a substantial racial bias in the data. Environmental elitism arguments have most often found the variables of race and income, in combination, and not race, occupation, and education as shown in this study.
Thus, these findings suggest an interesting refinement to the conservation elitism argument.
With the removal of income as an access barrier and a prerequisite to environmental concern, there is a much greater possible market for environmental interest and for products that meet 189 these interested individual's needs.
As was indicated by 1988 census data, size of household was somewhat smaller for environmentally concerned respondents. When the results for the variables, marital status and social group are examined, this difference is clarified. Seventy-five percent of respondents indicated that they were married, divorced, or widowed and sixty percent indicated their social group as a traditional nuclear family (e.g., husband and wife). Given the large percent of traditional family groupings yet the relatively small household/family sizes two conclusions can be made: 1) that environmentally concerned citizens who have traditional family groupings have children who are older and who have left the home, or 2) that environmentally concerned citizens have fewer children than the population at large.
To further examine the possibility that environmentally concerned citizens were in later stages of family lifecycle with no children at home, i.e., empty nesters, age of both male and female heads of household were examined. Significant relationships between those over fifty, small household sizes, and traditional family groupings confirmed this supposition.
This finding is supported by Babchuk and Booth (1969). In contrast, Reddy and Smith (1973) suggest that participation increases with the number of children. The rival hypothesis, that environmentally concerned citizens have fewer children than the general population appears to have little grounding in this study.
Correlations between stage in family life cycle and environmental concern present some unique marketing opportunities. Empty-nesters make a particularly attractive target to business because of their increased disposable income and available time (Engel et al, 1991). In addition, the fact that few families with children at home express environmental concern provides not only a challenge for environmental educators but also an opportunity for an organization that can identify and meet the specific latent environmental concern needs of families with young children. 190
While involvement and concern for the environment also varied, "typical" respondents were those who expressed that they were "very environmentally concerned," and participated in an average of six environmentally friendly behaviors such as recycling. Typical respondents also indicated that they had changed their buying behavior in response to environmental concern and that they were very affected by a company's environmental record. Once again
"typical" respondents represented only the norm with some respondents expressing more concern and involvement in environmental activities while others participated in fewer activities.
While U.S. Census data does not contain information on environmental concern, a number of national studies provide some insight into Americans in general. Concern for the environment is a relatively widespread phenomena. A 1988 Gallup Poll found that on average, over seventy-five percent of Americans were concerned about water and air pollution, ocean dumping, and solid waste disposal (Americans for the Environment et al, 1989). However, a
Roper Poll (1990) found that only sixteen percent of the nation on average avoids buying products from companies they consider environmentally irresponsible. In addition, environmentally concerned citizens appear to participate in far more environmentally responsible activities than does the public at large. According to the Roper Poll, only about one quarter of the population participates in activities such as: recycling newspapers, sorting trash, using biodegradable plastic bags, using low phosphate soaps and detergents, and avoiding aerosol products. This suggests then that while nationally environmental concern may be relatively high, the environmentally concerned citizen is substantially more active with respect to environmentally friendly behaviors than the population at large.
When asked about their level of interest and involvement in environmental and conservation organizations, "typical" respondents recalled the names of three organizations, most often representing some of the ten study organizations. While "typical" respondents would not have been members of environmental or conservation organizations, they had probably participated in some environmental activities such as letter writing or signing a petition. In 191 addition, "typical" respondents had probably given up to twenty-five dollars to an environmental/conservation organization in the past year, were members of other civic organizations, and had been interested in the environment for about fifteen years.
Data on environmental and conservation organization membership at a national level are very sketchy and often confused by the fact that multiple memberships are not uncommon
(Devall, 1970) specifically, within categorically similar groups (Babchuk and Booth, 1969).
Given this confusion, comparisons between this study's fifteen percent of respondents who were members in environmental/conservation organizations and the public at large is difficult. At best, it would appear that more environmentally concerned citizens are members of environmental/conservation organizations than in the general population at large (Manzo and Weinstein, 1987; Mitchell, 1980).
Environmentally concerned citizens in this study were more likely to be joiners than the public at large. While fifty-two percent of respondents indicated that they were members of at least one civic, fraternal, religious, or neighborhood organization, only forty-six percent of the public at large is a member (Wright and Hyman, 1958).
These findings ° Part of understanding more about interest and membership in environmental/conservation organizations can be obtained by looking at consumers preferences and perceptions of these 192 organizations in the decision making process. Choice sets are those sets of products that consumers compare at each stage in the decision making process. In this study, the development of choice sets for ten environmental and conservation organizations was studied at four levels: recall set, consideration set, preference set, and purchase set (see Table 66). Table 66 Rank Ordered Choice Sets for Environmental/Conservation Organizations Recall^ Consideration^ Preference^ Purchase^ Std. -1 Organization Set Set Set Set Dev. Ntl. Wildlife Federation 4 1 1 1 1.5 Ntl. Arbor Day Fnd. 7 6 6 2 2.2 Greenpeace 1 3 9 3 3.4 World Wildlife Fund 6 4 4 4 1 The Nature Conservancy 8 9 3 6 2.6 Ntl. Audubon Society 3 2 2. 5 1.4 Ducks Unlimited 5 14 10 8 3.8 Sierra Club 2 5 8 7 2.6 Wilderness Society 10 11 5 9 2.6 Cousteau Society 9 7 7 10 1.5 * | Organizations are listed in decreasing order of their membership size in population. ^ Organizations most frequently recalled ^ Consideration set included a list of twenty choices. ^ Organizations most preferred by respondents from 1= most prefer to 10=least prefer ^Organizations with the most membership in the sample (only 15% of the total for all other sets). 193 While no comparative literature was found on environmental/conservation organizations recalled, the consumer decision-making literature would suggest that construction of the consideration set is the first step in making a product purchase. While recognizing organizations from a fixed list would be one method of constructing the respondents consideration set, identifying the potential consumers unaided recall set can do this in addition to providing a measure of familiarity with the product category, especially a product category in which the names of the brands are not readily available to the consumer. Environmentally concerned citizens recalled in an unaided situation an average of 3.5 organizations of which the most frequently recalled were Greenpeace, Sierra Club, and National Audubon Society. This suggests then that these three organizations, while not necessarily the most preferred, have greater name recognition for potential consumers. Research on demonstrated recall measures found that recall of products within one day after the advertisement was aired on television was about twenty-four percent (Engel, et al, 1991). While the data are hardly directly comparable, the recall rates for the top three organizations in this study were between sixty-nine and forty percent. Those organizations whose recall rates were the lowest (e.g., Wilderness Society 5.0% and Cousteau Society 9.1%) will be those organizations who are less often included in consideration rates for membership or donations unless direct contact is made with the potential member at the time a decision to join is made. Moving the discussion from the recall set to the consideration set presents a different view of the environmental/conservation organization market. When presented with a list of twenty environmental/conservation organizations, those that environmentally concerned citizens stated most frequently that they would consider joining were: National Wildlife Federation, National Audubon Society, Greenpeace, and World Wildlife Fund. The differences in the relative order of the organizations from recall to consideration set demonstrates that the two are not perfectly correlated. Recall, however, is a necessity for consideration. 194 Discrepancies between the preference set and purchase set provide a view of what happens during the final stages of decision making. While the first three choice sets were constructed by the total sample, the actual purchase set was constructed by only the fifteen percent of the sample who were actual organizational members. Differences between the two sets may be explained therefore in two manners: 1) membership preferences for organizations may have been different than non-members but tended to be hidden in the larger set of responses, or 2) during the actual purchasing process, barriers or opportunities to purchase a specific organizational brand may have resulted in a brand substitution. For example, while the first choice organization (NWF) remained in the number one rank in the last two sets, the National Audubon Society (NAS) moved from preference rank two to purchase rank five. One possible explanation for this change was that although NAS may have been the preferred organization, a purchase barrier such as cost of membership may have intervened and consumers substituted a product (e.g., NADF) that was much cheaper. Additionally, post purchase evaluation may have caused members to re-evaluate their organizational preferences. The rank order of organizations purchased was very similar to national membership statistics. This suggests that the members of environmental/conservation organizations included in the sample are similar, at least in terms of organizational purchases, to the larger population of members. Research on choice sets provides some insight into what organizations can do to better assist their inclusion in the final purchase set (Spiggle and Sewall, 1987). Making that first cut in the choice sets model from the total set which includes all possible brands to the recognition or awareness set is crucial. In many traditional consumer goods situations, the brand choices need not be constructed solely from memory. In many shopping situations, consumers may scan shelves, or the yellow pages to construct their recognition set. The environmental/conservation organization, however, does not have this advantage. With few exceptions, there are no readily available published sources of lists of organizations that 195 consumers can use to construct their recall set. Thus, placement in an unaided recall set is necessary for eventual purchase. An organization can improve its chance of being placed in the awareness/recognition set simply through media exposure. Making the cut from consideration set to preference set is a reflection of the direction of the evaluation that the consumer places on the organization. If the evaluation is positive, the organization will be considered, otherwise it will become part of the inert or inept set. Attribution theory suggests that consumers are more likely to positively evaluate products toward which they take some action (Spiggle and Sewall, 1987) which suggests that the use of foot-in-the-door marketing campaigns or other involvement-oriented campaigns may assist the organizations likelihood of being purchased. An examination of the consistency of the rank order placement of the ten organizations across the four choice sets provides clues into what each organization could do to improve its placement in the final purchase set. A standard deviation measure across the scores shows that World Wildlife Fund (WWF), is the most consistent organization and Ducks Unlimited (DU) is the least consistent. National Wildlife Federation (NWF), also very consistent, ranked on the number one choice in the last three sets suggesting that if recall rates were higher, they may be included in more individuals considerations sets and purchases may increase. Greenpeace (GP), has a great deal of variance in its ranking. While it ranks highly in the recall, consideration, and preference sets, it received the second lowest purchase rating. A more detailed study may provide some insights into purchase barriers. Ducks Unlimited represents the organization in an unenviable position. It is consistently the lowest ranking organization with a great deal of choice sets variability emerging from its relatively high recall rate. Marketing research suggests that it is much easier to establish an image than to combat an existing negative one (Kotler, 1991), and this seems to be DU's situation. According to the rankings of this sample of environmentally concerned citizens, DU appears to be the misclassified organization. This is not necessarily surprising given that DU's mission and activities are much more conservative and recreation-oriented than any of the organizations studied (Snow, 1992). 196 Determining those things that a potential consumer prefers about a brand assists in the understanding of that brands placement in the market relative to other brands. While respondents generated a list of over sixty-one things that they liked about their number one preference choices, fourteen reasons were mentioned most often. An examination of these reasons provides insight into how consumers perceive organizations. Seven of those fourteen items, including four of the top five reasons, represent specific topical/species areas such as wildlife or trees. The second last reason represents tangible benefits offered by organizations, e.g., a magazine, while the other six reasons represent qualities such as "active stance" or "global in focus". Comparing organizations to the reasons why they were most preferred by respondents helps clarify the picture of who is identified with what attributes. Some organizations had clearly defined reasons of why they were most preferred. The National Arbor Day Foundation was associated with trees (86%); the National Wildlife Federation with wildlife (59%); and the National Audubon Society with birds (57%). It appears that given the relative consistency with which people associate a specific label with a particular organization that most of these organizations have a clearly identified perceptual key for respondents to remember. Those that are most commonly identified with a particular reason are those whose name contains, or relates directly to the reason. Thus, a key issue in marketing an organization may be the selection of an appropriate, distinct, and identifiable name that is readily associated with a desirable feature or attribute. Finally, choice sets, preferences and perceptions aside, the selection of self-labels by respondents lends support to the selection of the name environmentally concerned citizens for the respondents. An overwhelming majority of respondents (eighty-six percent), labeled themselves as "concerned citizens" while only a quarter of the respondents would ascribe the label "environmentalist" to themselves. This finding goes a long way in supporting the assertion that many feel that "environmentalist" may be a too value laden term and carries with it enough connotation such that the majority of people do not want to be so identified 197 (Snow, 1992). The following discussion of question three, incentives, will provide additional insights into why people join environmental/conservation organizations. Discussion of question two, market segments, is deferred until the end of the discussion section for the purpose of a clear ordering of the discussion. 198 Question Three. What incentives motivate people to join environmental/conservation organizations? Literature on interest group theory and voluntary associations (Lowi, 1971; Olson, 1977) suggested a number of different reasons why people join groups. Based on a theory of incentives (Olson, 1977), motivations for participation in environmental/conservation organizations were studied using a twelve item summated Likert-type scale. The instrument utilized to collect data for this question was a modification of the Dennis and Zube (1988) scale designed to measure incentives of participation in recreation and environmentally-oriented organizations by a sample of members of a large recreational equipment cooperative (REI). In its original form, two factors each consisting of six items, were identified by Dennis and Zube using factor analysis. These two factors were named by the authors, instrumental incentives, those incentives that obtain public goods, and expressive incentive, those that obtain personal benefits. The modified twelve item scale (see Table 18, p.89) was completed by environmentally concerned citizens and an exploratory factor analysis, and corresponding reliability tests, were computed. Using principal components factor analysis with orthogonal varimax rotation, three factors, not two as had been reported earlier by Dennis and Zube (1988), were produced in six iterations. Five of the same six items identified as expressive incentives by Dennis and Zube also loaded on a single factor in this study and thus, the factor was also named, "expressive incentives." Three of the six items that Dennis and Zube identified as the instrumental factor also loaded on the same factor. These three items had a much smaller focus or underlying common ground, however, than the concept of instrumental incentives implies and thus, this factor was named the "political efficacy" factor. The third factor identified using factor analysis contained only two items, one from each of Dennis and Zube's original factors. While these two factors are somewhat disparate in content, "normative compliance" was used as the general descriptive for this factor. 199 Two additional items, both originally from the instrumental factor in the Dennis and Zube analysis, did not load on any factor and were not used in any subsequent analysis. These items were: 1) Some important aspects of my life are affected by environmental/conservation problems. 2) Environmental/conservation organizations should actively involve themselves in environmental problems of special groups such as minorities and the economically disadvantaged. The difference in the factor loadings between the original application of this instrument by Dennis and Zube and those obtained in this study of environmentally concerned citizens suggests that there are some differences between these two populations and their incentives for participation in organizations. Part of the discrepancy between factor identification comes from the more liberal factor loadings used by Dennis and Zube. While the analysis for the current study used a conservative factor loading minimum of .50, Dennis and Zube allowed factor loadings as low as .35 to determine which items belonged to which factors. In fact, the two items that make up the third identified factor in the current study were items that loaded as .35 and .38 in the Dennis and Zube study. Consistencies between the identification of the expressive factor in both studies, and the correspondingly large Cronbach's alpha coefficients for this scale (r=.79 in the Dennis and Zube study; r=.89 in this study), suggest that expressive incentives are common constructs for both populations. The refinement of the second factor from instrumental incentives to the more narrowly defined political efficacy factor suggests either a difference in the study populations, or a difference in the meaning and/or strength of the items used to measure this factor. The emergence of the third factor in the current study is statistically not surprising given the factor loading differences mentioned above, however, this third factor, tentatively named "normative compliance" suggests that there are at least some different incentives or motivations for joining environmental/conservation organizations as compared to the environmental/outdoor recreation groups from the Dennis and Zube study. 200 Market segment profiles based on membership, donations, and extreme environmental concern had mean scores for all three derived factors that were greater than their segments counterparts. This suggests that each of these three segments would be more greatly motivated by these incentives than nonmembers, non-donors, and the very environmentally concerned. The relatively higher rankings by members, donors, and extremely concerned individuals on all factors, however, tells us little about these segments other than their increased tendency to rate any motivation higher than their segment counterparts. Paired t- tests to determine the relative importance for each segment helped further differentiate between the segments. Each segment pair; members and nonmembers, donors and non-donors, and extremely versus very environmentally concerned citizens were compared to see if there were any differences between the relative importance, or rank order, of each incentive. In each case the results were the same. Members, donors, and extremely concerned citizens ordered the incentives such that political efficacy was the most important incentive, followed by the normative compliance incentive, and lastly, the expressive incentive. The incentives to participate in an organization because of the potential impact a member of a group can have on influencing the political and social order (political efficacy incentive) was then the dominant motivator for group membership, while personal gain benefits (expressive incentive) were least important. Each segments counterpart; nonmembers, non-donors, and very environmentally concerned citizens ordered the incentives with expressive incentives in last place, but the relative order of the political efficacy incentive and normative compliance incentives were undetermined. Once again, personal gain benefits were not deemed as as important by these segments. This suggests, that those individuals more prone to be active in organizations, including actual members, donors, and those who were extremely concerned about the environment are more highly motivated by political efficacy incentives than by personal gain incentives. This finding is consistent with those of Dennis and Zube (1988) who found that instrumental incentives were the strongest incentives for those that held membership in conservation- 201 oriented voluntary organizations. However, these results go beyond the simple instrumental incentives explanation and are more in line with proponents of interest group liberalism who suggest that collective goods that may occur as a result of interest group membership are insignificant, rather most members join to gain a sense of political efficacy (Moe, 1980). This is consistent with Milbrath (1981) and Manzo and Weinstein (1987), who found that environmental/conservation organization members and activists had greater feelings of political efficacy. While expressive incentives such as the personal benefits received from organizational publications and the social aspects of groups are important, they are less important to members and potential members than are the feelings of political efficacy one gets or is able to express, by joining a group (Dennis and Zube, 1988). Organizations seeking to capitalize on this finding while still continuing to offer expressive incentives, should profile more highly the political efficacy benefits one receives from being a part of a group. The relationship between the normative compliance incentives and the other two incentives is less clear. This factor was named normative compliance based on the literature on reference group behavior. Normative compliance refers to the influence of significant others on an individuals behavior (Engel et al, 1991). This theory suggests that reference groups influence belongingness because of: 1) value-expressive influences such that the image of the consumer is enhanced in the eyes of the reference group if the behavior is adopted or the product purchased, or 2) informational influences such that consumers accept the opinion of others as credible (Caldor and Burnkrant, 1977). This second justification has often been found to be especially important when a product is difficult to assess. Bruner and Pomazal (1987) support the importance of information seeking and normative compliance as it relates to product purchases. Specifically, they suggest that reference groups can often initiate a change in the desired state (e.g., motivation to purchase) in the problem recognition stage of decision making. The fact that the political efficacy incentive was more important than the normative compliance incentive for members, donors, and extremely concerned citizens suggests 202 that these groups have higher feelings of or needs for political efficacy. Perhaps experience with the product (e.g., as a member or through giving) is what determines this. In an attempt to improve the discriminant functions derived for the segmenting variables in the previous question, the three incentive factors, political efficacy, normative compliance, and expressive incentives were added to the discriminant functions. Overall, the addition of the incentive factors improved the classification of segments only marginally. Given the added difficulties of collecting incentives data compared to most of the other discriminating variables and the marginal results, the incentive factors do not appear useful for classification. 203 Question Four. What are the characteristics of an ideal environmental/conservation organization? Understanding the ideal environmental/conservation organization was accomplished through analysis of three areas that examined preferences for products, services, and opportunities, as well as ideal characteristics and actions of organizations. There are many similarities between the factors derived for these three areas as well as with the previous research question on incentives suggesting that there are some general patterns of needs, and wants across all segments. Product. Service, and Opportunity Benefits Using factor analysis, four distinct product, service, and opportunity needs (or benefit packages) were identified that respondents were looking for from organizations. These benefit packages were comprised of: 1) items that described an environmental/conservation organization with a local or chapter structure, 2) items that provided respondents an opportunity for social involvement, 3) items that provided tangible benefits such as magazines and gift or catalog services, and 4) larger order items that provided hedonic benefits such as the chance to help solve problems, to educate others and to financially support a cause. What these factors show is that respondents classify benefits into like types, or bundles of benefits (Kotler, 1991). These derived benefit factors are similar to those reported in the literature. Consumer behavior literature in general recognizes two broad types of needs or benefits sought: utilitarian needs, such as the functional product attributes like a magazine or catalog service, and hedonic or experiential benefits obtained from contributing to a cause, and helping to solve problems (Engel et al, 1990; Hirschman and Holbrook, 1982). The factors based on social aspects and tangible services are similar to a study by Manzo and Weinstein (1987), who found that less active members found satisfaction in receiving a magazine while 204 more active members reported satisfaction from social aspects of members. Differences for preferences of the benefit factors varied by segment. However, summated scores across all segments suggest that the most important needs, in descending order, were the hedonic, tangible, social, and local structure needs. If these needs are related to incentives theory the results parallel Moe (1980) who found that purposive incentives, that would motivate an individual to look for products that offer hedonic benefits, predominated in noneconomic groups. The presence of tangible physical product offerings as the number two benefit sought is also supported by a number of researchers who found that in addition to the affective (hedonic) and normative (social) benefits that are part of participation in an interest group or organization, organizations must also offer utilitarian incentives (hedonic benefits) that are visible, valuable, and utilized (Burgess, and Conway, 1961; Knoke and Wright-Isak, 1981; Clark and Wilson, 1961). Defining products in terms of the needs and benefits sought from a consumer perspective is important because people seek the benefits products provide not the products themselves (Day, Shocker, and Srivistava, 1979). Ultimately then, while there are some differences between market segments based on these benefit bundles, in general the most desirable attributes of an organization, that is those rated most highly by respondents, are those that offer hedonic benefits followed by those that offer tangible benefits. In order to develop product offerings or to evaluate existing product offerings, an organization should determine how well it rates on the specific benefits sought by consumers. Characteristics of Organizations Factor analysis was also used to analyze respondents scores for a list of fifteen characteristics of the ideal environmental/conservation organization. Four common groupings or descriptors of the ideal organization were uncovered. The first factor contained six items that included a range of characteristics including: global, education, nature appreciation, preservation, wise- 205 use, and environmental problem-solving. This factor was identified as the traditional factor since it included common descriptors of organizations. Interestingly enough, two items that were grouped together in this factor included preservation and wise-use, words that are often thought of as opposites in the environmental/conservation literature (Hays, 1987; NWF, 1990). Conservation, a descriptor more in keeping with the wise-use label was not extracted in the analysis. This suggests that respondents must define these terms differently, and/or they do not see the characteristics of preservation and wise-use as incompatible. The second factor included two items, liberal and pacifist, and was correspondingly named the pacifist factor. The substantially lower reliability score for this item (alpha =.48) suggests that this factor is relatively weak and little reliance should be placed in this construct. The items in the third factor, radical and militant produced a more reliable factor (alpha =.64) even though it contained only two items. This suggests that respondents see a great deal of similarity, or at least association, between these two characteristics and at least some portion of the population, albeit a small one, identify them as characteristics of an ideal organization. This factor parallels previous research (Milbrath, 1984; Morrison, 1980; and Mitchell, 1980). While the traditional factor items are undoubtedly the most popular, combinations both either the liberal or radical factors with it are also possible and not necessarily incompatible. In an attempt to determine the degree of focus the ideal environmental/conservation organization ought to possess, the characteristics "single-issue" and multi-issue" were included in the choice list along with "global" and "local." While global loaded on the traditional factor, and single-issue was extracted as its own factor, the characteristics local and multi-issue were not extracted. There is little related literature to compare these derived factors to, with the most common organizational attributes that were identified in previous research being those that have something to do with a single-issue mission statement or topical orientation (NWF, 1990). Also relatively common are organizations that are identified as either global or local (Anderson, 1989). 206 Within this study, the traditional factors were for most people the ideal set of organizational characteristics. Analysis by segment suggests some opportunities for organizations with other combinations of characteristics but further research, in a forced choice format, would be needed to determine the optimal characteristics for other segments. Appropriate Actions Three distinct factors were extracted in the analysis of appropriate actions for solving environmental problems that were identified as low-risk activities, demonstration activities, and eco-terrorist activities. These clustered activities were surprisingly similar to hypothesized results with the first set of activities being the more traditional types of responses to environmental concerns. These activities involved little risk to the participants involved and required limited involvement. The second category of activities, the demonstration activities, were public protest-oriented and represented a much higher level of activism and risk. Lobbying as an activity did not load on any of the factors although it was most closely associated with the demonstration activities. Finally, the third set of activities, those classified as eco-terrorist activities, involved a much greater degree of personal involvement and personal risk. Very few respondents supported these activities, but those that did also preferred the radical factor in the previous question. The bulk of the writing on environmental actions has been anecdotal, however, the most often identified clusters of activities are the radical or militant factors (Milbrath, 1984; Anderson, 1989). 207 Question Five. How are environmental/conservation organizations perceived by consumers? Determining respondent preferences and perceptions for environmental/conservation organizations was accomplished by the analysis of three separate items: organizations respondents would consider joining, ranked preferences for the list of ten study organizations, and similarities data for all possible pairs of these ten organizations. Discussion, recommendations, and implications for each of these questions is included below. O rganizations Respondents Would Consider Toining Although rank order data on organizations respondents would consider joining were discussed in the choice sets analysis for question one, factor analysis was used to help uncover any common groupings. The original list of ten organizations was expanded to twenty o rganizations for this purpose. These additional organizations were selected based not only on percent of membership but on hypothesized similarities between organizations. Inclusion of this broader list of organizations in a manner that still allowed respondents to process the data, had the potential to show: 1) that certain organizations were commonly selected by respondents based on some topical similarity or approach similarity, and/or 2) that respondents may select an array of organizations representing a range of topical areas and approaches. Factor analysis of respondents ratings of these twenty organizations revealed six factors that grouped together common organizations. While the first two factors had relatively strong reliability scores (alpha = .77 and .66 respectively), the remaining factors had substantially lower reliability alpha, thus, the strength of these factors should be interpreted with caution. At best, the final four factors may be useful in identifying common clusters of organizations that could be further studied by the inclusion of other related organizations or by the use of a more discriminating scale (e.g. ordinal ranking of factors). 208 Factor one identified five organizations that were grouped under the common name "wildlife organizations." With the exception of The Wilderness Society (WS), all of these organizations could clearly be identified as having wildlife as a focal interest and as a portion of their namp. While it is not incompatible, based on organizational missions, for WS to be included in this list, the focus of this organization is much more at the habitat/ecosystem level than at the species level. In addition, similarities rankings on the perception question also suggest that it would have been expected that The Wilderness Society would have more closely clustered with a "similar" organization (e.g., The Nature Conservancy). Recall of WS was the lowest of all ten organizations, thus, it may be possible that respondents chose this organization on the basis of only a portion of its name or with a lack of understanding of its mission, and consequently were most likely to cluster it as a wildlife organization. Further research is recommended to probe consumers awareness and understanding of this organization. What this strong clustering of wildlife organizations suggests, is that environmentally concerned citizens have a strong recognition, and undoubtedly, an affinity, for wildlife. The second group of organizations identified included five organizations from the National Toxics Campaign to Earth! First!. These organizations were identified as "action" organizations because all have very action-oriented agendas. Again, within this grouping there were some anomalies. It was hypothesized that Environmental Defense Fund (EDF) although very much an action-oriented organization, would be more likely to be clustered with the third factor, those organizations that work within the political and legal framework. The League of Conservation Voters (LCV) and the Natural Resource Defense Council (NRDC), the two organizations in this third cluster, work within the same political and legal framework as EDF. Similarly, Sea Shepherd Conservation Society, although a very small and little known organization, has a topical orientation and style of activism that is much more similar to Greenpeace and Earth! First! yet it was loaded as a factor in and of itself. The only hypothesized reason for this is that the low awareness and frequency of choice of this organization by respondents resulted in it loading as its own factor. Once again 209 perceptions of consumers should be probed to clarify their understanding of these organizations in order to assist in understanding these groupings. Naming the remaining three factors, especially given the few organizations included in each, and the low reliability scores, is much more difficult. Factor five, included the National Arbor Day Foundation (NADF), the National Audubon Society (NAS), and Ducks Unlimited (DU), and could be identified as the single-issue factor since all focus primarily on a type of organism. The fourth factor contained two longstanding organizations, the Cousteau Society and the Sierra Club. One possible name for these organizations may be, "establishment" organizations, although it would be expected that the Audubon Society would have also been clustered on this group. "Founders influence" may also be a good name for this dimension. The importance of each of these organizations founders (Jack Cousteau and John Muir respectively), and the recall of these founders names by respondents was evident when respondents were asked why they most preferred their first choice organization. One of the two organizations that were not extracted in the analysis is also puzzling. While Conservation International, is not particularly well known and, thus, it was not surprising that this organization was not extracted, The Nature Conservancy (TNC) had a relatively high recall rate. It was suggested earlier in this discussion that based on similarities scores of organizations a grouping of TNC and The Wilderness Society would not have been surprising. Given this was not the case, TNC, which was clearly identified with the key words "buy" or "preserve" land when asked why the organization was preferred, may have been left by itself as the only real estate-oriented organization. Once again, however, WS and Ducks Unlimited ought to have clustered with The Nature Conservancy if this were the common link between these organizations because all three have a strong real-estate orientation. This may suggest that TNC has a much more clear association with the "preservation" element while the other two organization's work in this area is not as well known. 210 What these factors suggest about consumers is that they tend to cluster organizations together based on a variety of attributes including: topical organization and action strategies. These derived factors provide a way of defining the market structure, and the competitive structure from the consumers point of view. There are two broad marketing implications to the resulting competitive structure: 1) that organizations with whom any particular organization is clustered represent a source of membership given the tendency for environmental/conservation organizations to retain multiple memberships (Devall, 1970 and Faich and Gale, 1971), and 2) that the organizations within the same factor are direct competitors (Day, Shocker, and Srivistava, 1979). In order to successfully attract members, an organization identified as one of many similar organizations must clearly differentiate itself from its competitors. Perhaps the most enviable position then for an organization is as an organization that does not load with many other organizations or is identified, by a substantial number of people (this would therefore preclude Sea Shepherd Conservation Society), as loading on a factor by itself. Preferences for Organizations In the second part of analysis for this research question, respondents were asked to rank the ten study organizations from most preferred (one) to least preferred (ten). An examination of the consistency of rank order of these organizations across all segments provides us with some clues as to which organizations are generally most preferred. Examination of the mean ranks of the organizations for the total sample show that National Wildlife Federation (NWF), National Audubon Society (NAS), and The Nature Conservancy (TNC) were the three most preferred organizations. The two least preferred organizations were Ducks Unlimited (DU) and Sierra Club (SC). Both relative placement and consistency across all segments can provide insight into how an organization is positioned, what opportunities there are within specific segments, and who its competitors are. Rank orders of the most preferred (NWF) organization and the least preferred organization (DU) were consistent across all segments. While this suggests that NWF is in a very stable, and ideal 211 position across all market segments, it also suggests that there are no opportunities for Ducks Unlimited to improve its ranking given the way the segments are currently defined. Organizational Similarities Analysis of perceived similarities of these organizations was measured by computing perceived distances between all possible pairs of the ten organizations. The resulting ordinal comparisons were converted into squared Euclidian distances and plotted on the minimum number of dimensions needed to mathematically solve these perceptual distances. Perfect fits of these distances were sacrificed in order to aid in the interpretation of the results. The resulting multidimensional scaling maps are maps that show organizational similarities on 'n' number of dimensions. These dimensions represent an attribute or combination of attributes varying together that differentiate two organizations. Multidimensional maps (MDS) were plotted for the four dichotomous, categorical dependent variables: members and nonmembers, donors and non-donors, joiners and non-joiners, and extremely versus very concerned citizens. A number of general conclusions can be made from examination of the eight resulting maps. Each of the eight maps was plotted on only two dimensions given the criteria for stress levels and r-square values established a priori. While the addition of another dimension may have resulted in a more accurate plot, the tradeoff between minimized stress levels and the added dimension was readability. Within even a two dimensional limitation, however, r-square values, the primary goodness of fit statistic used to assess the dimensionality were very strong associations (r=.90 or higher). This indicates that there was very little difference in the plot between the actual numerical distances in the data and the data compressions required to plot the solution on two dimensions. The implications of this finding are that consumers in all segments use a relatively simplistic, two attribute (or dimension) method of judging the similarity of the organizations. This finding is in keeping with Rosch (1978) who 212 stated that based on the principle of cognitive economy, the fewer the number of attribute categories that relate to the functions of the products that can be used will be. Other researchers in the leisure and recreation fields have found solutions ranging from two to four dim ensions (Ritchie, 1975; Williams, 1988; and Russel and Hultsman, 1987). The use of only two dimensions suggests three possible conclusions: 1) that consumers clearly identify two attributes upon which to compare organizations, 2) that consumers have limited dimensionality because they are unfamiliar with the attributes that could be used to make similarities judgments, or 3) that consumers use an overall judgment of the product categories and are thus unable to recall specific attributes to compare them by (Engel et al, 1991). The results chapter suggests tentative names for these two dimensions that appear to be relatively consistent across all of the maps. The names are, however, only tentative since one of the great weakness of MDS is that the naming of the resulting attributes is very subjective (Fox, 1988; and Green, 1975). The vertical dimension was named the "organism" dimension since the organizations ranged from zoological species to botanical species. Determining a name for the horizontal dimension was much less clear with two names distinct possibilities. "Habitat" dimension was proposed since it appeared that what differentiated the organizations on this dimension was topical coverage as it ranged from avian habitats, through terrestrial habitats, to aquatic habitats. Alternately, an action or political dimension from conservative to liberal, or conservation to preservation was suggested since the organizations appeared to range from the most conservative organization, Ducks Unlimited, to the most liberal organization, Greenpeace. The lack of spread between any of these organizations on the horizontal dimension made this dimension much harder to explain. Previous work on organizational dimensions, although non-research based in nature has identified organizational dimensions that are more sophisticated. Hoesterey and Brown (1976) labeled the array of organizations from outdoor recreation-type to environmental advocacy-type while others have used dimensions such as: wilderness orientation (Hendee, 213 Catton, Marlow, and Brockman, 1968), political radicalism (Morrison, 1980; Mitchell, 1980; Outside, 1991), or alternative technology orientation (Mitchell, 1980). None of these previous studies, however, was based on membership, or a similar respondent groups perceptions of environmental/conservation organizations. Additionally, each of these other "maps" were uni-dimensional in nature. So while these alternative dimension names are intriguing, they appear to be more complex than consumer defined dimensions in terms of attributes but less complex in terms of the number of dimensions. It should be noted that those organizations that are located exactly on a dimension are those that would appear to possess no value for that dimension. For example, World Wildlife Fund is located on the midpoint of the horizontal dimension on the vertical dimension line and this suggests that it could not be identified, or differentiated from any other organization based on the habitat dimension. In addition, on several of the maps, an organization or organizations does not appear on the plot. This usually occurs when the organization is too far out on one of the ranges and thus can not be plotted. Additionally, it can mean that the placement of the missing organization on the plot would violate the acceptable stress levels for that solution, that is, it may be located on its own dimension. While determining what the dimensional names are requires a great deal of creativity, understanding what these similarities rankings mean for organizations is much more clear. Those organizations that cluster together are perceived as most similar while those that are farthest apart are most dissimilar. Throughout all of the segments maps the relative placement of the organizations is very consistent. The exception to this is the dimensional map for the "very environmentally concerned" segment although the organizations on the two dimensions appears to be simply reversed perhaps the result of the rotation of the mathematical solutions. In most maps, The Wilderness Society (WS) and The Nature Conservancy (TNC) cluster very close together suggesting that respondents across all segments, see these two organizations as very similar. Greenpeace (GP), National Arbor Day 214 Foundation (NADF), and Ducks Unlimited (DU), are the three most dissimilar organizations from respondents point of view, each locating in a different quadrant of the perceptual maps. At times, (e.g. on the member/nonmember map), Greenpeace is so dissimilar to the other organizations that it is not even plotted on the solution. This visual measure of organizational similarity once again provides a method of viewing the organizational market from the consumers perspective (Day, Shocker, and Srivastava, 1979). Potential competitors or cooperators can be identified from these maps and a preliminary understanding of how consumers compare organizations is revealed. By understanding how consumers rate each organization on a given set of dimensions, organizations can make strategic marketing decisions such as: modifying the product, altering beliefs about the product, or calling attention to the neglected attributes (Petty and Cacioppo, 1981; and Green, 1975). For example, if respondents join multiple similar organizations, then the WS and TNC could expect a great deal of overlap, but not necessarily competition, in membership. If these two organizations are seen as too similar, they will be direct competitors to each other and in order to create a secure market niche at least one of the organizations will have to perceptually differentiate itself from the other, perhaps by making consumers aware of a third, salient dimension. Organizations that are perceptually dissimilar, for example DU and NADF, are less likely to be direct competitors. If they both meet different consumer needs, while still possessing some common attribute (unlike DU and Greenpeace who possess strong hunter/anti-hunter values), there may be opportunities for the two organizations to engage in some creative network marketing. 215 Question Two. How do individual and environmental differences vary by market segment, and what is the best combination of variables to identify these market segments? While a discussion of "typical" respondents characteristics' is useful from the perspective of getting a general overview of who environmentally concerned citizens are, such a homogeneous view of the results hides the market opportunities that exist within the population. To maximize potential, segmentation must occur. There are two important considerations regarding these market segments: 1) What is the optimal combination of independent variables to identify these segments? and 2) What are the profiles of the resulting segments? Discriminating Variables Stepwise discriminant analysis for the categorical segmenting variables and stepwise multiple regression analysis for the continuous segmenting variables were used to help identify the characteristics that aided in classification and prediction of the dependent variables used to define the market segments. Variables used in both the classification functions and the regression equations were selected because they represented all available continuous variables or categorical variables that could be coded as dummy variables. The independent variables included: total environmental activities, total number of environmentally friendly behaviors, social class, market size, family size, length of interest in the environment, donation behavior, joining behavior, type of residence, race, level of concern, and likelihood of changing behavior. Initially geographic region, and social group type were also i. eluded in these equations, however, when they were recoded as dummy variables the hbsi us were somewhat muddied with only marginal increases, thus, they were excluded from the final analysis. Each discriminant function or regression equation produced somewhat different results (see Table 67). A holdout sample was used with discriminant analysis to determine how well the 216 final discriminant function could classify respondents based on the dependent variable. Classification rates varied for each segment from the highest for members to the lowest for joiners. That is, on the basis of the identified variables, the chance of predicting group membership for members (71.4%) and nonmembers(79.8%) was much better than for joiners (54.5%) and non-joiners (66.0%). Given that chance alone would yield a fifty percent correct classification rate if n's were equal, the discriminant functions for these segments are not all equally useful. The segments constructed around joining behavior specifically have the poorest classification rate, a result that is not particularly surprising given that fewer statistically significant differences were found between joiners and non-joiners on other analysis. Similar results were found for the continuous level dependent variables social class and length of interest. Those variables included in the final regression equations accounted for twenty-four percent of the total variance in social class and thirteen percent of the total variance in length of interest in the environment. Even when the motivation incentives were added to the equations, the discriminant and regression functions failed to yield significant improvements. Other studies of populations of environmentally concerned citizens have accounted for between 34% and 16% (Devall, 1970; Dennis and Zube, 1988; Wright, 1988; and Gomon, 1991) of the variance using a combination of sociodemographic, lifestyle, and incentive variables. While poor classification and predication rates are to be expected when using sociodemographic variables, it was thought that the addition of a number of psychographic and lifestyle variables would have improved the classification rates more. This suggests then that in order to improve classification and prediction rates, more discriminating variables will have to be found. 217 Table 67 Classification and Prediction of Market Segment Variables Discriminant Analysis Segment/Group Correct Classification M embership Member 71.4% Non-Member 79.8% Environmental Concern Extremely Concerned 69.2% Very Concerned 59.3% Joining Behavior Joiners 54.5% Non-Joiners 66.0% Donating Behavior Donors 60.3% Non-Donors 75.3% Multiple Regression R-Square Social Class .24 Length of Interest in the Environment .13 Variables used in the final discriminant functions and the regression equations also varied (see Table 68). Of the eleven variables included in any of the functions, five were included in eighty percent of the functions and one was included in all of the functions (organizational activities). Determining which variables to use in identifying respondents who are likely to be part of a particular market segment should be made on the basis of the variables included in that particular segments discriminant function or regression equation. While organizational activities, donation behavior, and social class would be most useful in identifying members, those who were extremely environmentally concerned could be best identified by willingness to change behavior, environmentally friendly behaviors, and social class. Organizational activities was most frequently the first variable included in the 218 equations, followed by social class. Other variables changed place relative frequently. Membership was not included in all of the equations because it had a high degree of multicolinearity with all other variables. Based on these results we can conclude that those variables most useful in identifying the market segments were organizational activities, social class, environmentally friendly behaviors, and length of interest in the environment. Although there is little comparable research on many of these variables, Devall (1970) found that social class was still a useful and persistent variable in predicting environmental/conservation organization membership specifically, with respect to larger and older nationwide organizations. Table 68 Variables Included in Discriminant Functions and Regression Equations V ariable Segments* Membership Donations Joining Environ. Social Length Concern Class of Interest Organizational Activities 1 1 1 5 5 Environmental Behaviors — 3 — 2 7 4 Donation Behavior 2 * 5 — 1 1 Social Class 3 2 2 3 * --- Household Size 4 ——— 2 2 Market Size 5 7 3 — 5 — Length of Interest 6 4 4 4 — — Race 7 —— 6 3 — Environmental Concern 8 —— * 4 — Joining Behavior — 5 * — 6 3 Change Behavior ------1 --- — i Variables are ranked in order of inclusion in the final equation. * Variable not included in own equation. Variable not utilized in final equation. 219 Market Segment Profiles Six dependent variables; membership, joining behavior, donating behavior, environmental concern, social class, and length of interest in the environment were selected as the basis upon which market segments were constructed. Each segment view, for example members and nonmembers, overlaps the other segment views, thus, the reader should not interpret the resulting segments as twelve different unique segments but rather six different ways to view the larger market. The selection of an appropriate market segment view will depend ultimately on the purpose for segmentation. If the purpose is to increase membership, understanding the differences between members and nonmembers will be of assistance. If the purpose is to build donations, understanding who gives versus who does not will help in the design of campaigns. Some of the other segment views provide an opportunity to explore more theoretically-oriented market segments. Does social class or length of interest in the environment help differentiate between environmentally concerned citizens? Do organizational joiners in general have some unique characteristics that may be useful in helping to understand not only who joins organizations but why? A summary of the significant findings and the implications for each of the six market segment views is presented below. For the four dichotomous market segments, these summaries will profile a segment's differences. For the two multi-category market segments, the summaries will profile the general trends across the segments. Membership Fifteen percent of the sample were members in at least one environmental/conservation organization. Members tended to have a higher family income, more often hold white collar jobs, and have more education (specifically graduate education) than nonmembers. They were white, from non-family households, and grew up in an urban/suburban environment. 220 Members were most likely to give money to an environmental/conservation organization apart from regular membership dues and they were likely to be members of other civic, fraternal, or religious organizations. They expressed a greater degree of concern for the environment and participated in more environmentally friendly behaviors and more environmental organizational activities. They recalled more organizations and had been interested in the environment for a longer period of time. Members ranked all three motivation incentives higher than nonmembers and were more motivated, in descending order, by the political efficacy incentive, the normative compliance incentive, and the expressive incentive. Members rated the tangible and hedonic needs as more important than nonmembers and were most likely to describe their ideal environmental/conservation organization with a combination of traditional and liberal characteristics. They were more likely to support low risk and demonstration activities than were nonmembers and expressed interest in a whole range of other organizations. Members most often preferred Greenpeace, The Nature Conservancy, and National Wildlife Federation while nonmembers most preferred the Audubon Society, National Wildlife Federation and National Arbor Day Foundation. Members and non-members appeared to use relatively consistent dimensions when comparing organizational similarities and members perceived The Nature Conservancy and the Wilderness Society as most similar and Greenpeace and Ducks Unlimited as least similar. Based on the above profile, a number of strategic marketing implications are suggested for organizations wishing to increase membership. Given the persistence of the conservation elite phenomena, organizations wishing to continue attracting this profile of individuals should offer products and services targeted to higher class occupations and more highly educated individuals. This may include offering not only environmental/conservation organizational products that are targeted to these needs but also engaging in cooperative marketing strategies with companies whose products are also sought by these individuals (e.g., The Nature Conservancy and the Nature Company). If organizations wish to target a broader base 221 of society, or specifically target unrepresented segments then they must develop products and meet needs of the lower class segments. The segmentation profile by social class may help in clarifying some of these needs. Since current members are those individuals who are more active, both in organizational activities and environmentally responsible behaviors, environmental/conservation organizations could target individuals who exhibit these behaviors. Organizations with local chapter structures could promote, operate, or support local recycling activities in communities, or local car pooling efforts. In addition, since members are more more prone to participate in environmentally friendly behaviors, organizations should promote how joining an environmental/conservation organization is an environmentally friendly behavior. Differences in incentives and needs sought, also present opportunities for target marketing. Since political efficacy incentives are most important to current members, promoting the benefits of feelings of political efficacy one receives from joining an organization is one way of attracting those individuals who are motivated by this need. In addition, product development and corresponding promotional strategies should ensure that not only tangible benefits are offered but more importantly, the hedonic benefits that these individuals value in environmental/conservation organizations. Joining Behavior Slightly over half (52.2%) of all respondents indicated that they were a member of at least one civic, fraternal, religious, or neighborhood organization. Joiners scored significantly higher on the social class index, a finding consistent with much of the voluntary association literature (Babchuk and Booth, 1969). Similar to results found by Florin, Jones, and Wandersman (1986), this analysis found that joiners were also more likely to be white. Additionally, joiners were more likely to have donated money to an organization, and to have been interested in the environment for more years. 222 Joiners recalled more organizations, had more current memberships and participated in significantly more organizational activities. There were no differences between joiners and non-joiners on the basis of incentives for participation in organizations and product, service, and opportunity needs, a finding that was inconsistent with previous research (Knoke, 1981; Olson, 1972; Smith and Reddy, 1973) that found that joiners had higher levels of political efficacy than non-joiners. Joiners also rated traditional characteristics of organizations higher than did non-joiners and also rated low-risk activities significantly higher. Unlike the other market segment views, the relatively few significant differences between joiners and non-joiners provides little opportunity for achieving any benefits by segmenting in this manner. Thus, while the literature suggested that segmentation by joining behavior may have some utility (Knoke, 1981; Smith and Reddy, 1973; Babchuk and Booth, 1969), there appeared to be too few significant and relevant differences in this analysis for a segmentation strategy based on joining behavior to have utility. Donating Behavior Approximately thirty percent (28.9%) of the sample indicated that they had given money to an environmental/conservation organization in the past year. Donors tended to score higher on the social class index and on all of its subcomponents and have been interested in the environment longer than those who were non-donors. While donors had fewer current organizational memberships than non-donors, they recalled more organizations and participated in significantly more environmentally friendly and organizational behaviors than did non-donors. Donors also rated all three incentives for participation higher than did non-donors and ranked them in order of importance as: political efficacy, normative compliance, and expressive. They rated all four need categories significantly higher than did non-donors and they also rated all three characteristic factors higher than did non-donors. Donors much 223 preferred the low risk and demonstration activities and were significantly more likely to indicate that they would consider joining almost any organization compared to non donors. It would appear, based on these findings, that environmental/conservation organizations have two separate client bases: members and donors. This finding is consistent with a number of other studies (Manzo and Weinstein, 1987; Mitchell, 1980) which found distinct categories of involvement in environmental/conservation organizations that included: subscribers, members, supporters, and donors. This differentiation between members and donors suggests that donating is not synonymous with membership and perhaps those who donate have needs that are different from those who are members. While both represent important market segments for organizations to pursue, there do not seem to be many differences between members and donors within this data set that would suggest how organizations could target donors separately from members. The only obvious fundamental difference is that while members and donors could be targeted the same way, donors should be encouraged to give financially in exchange for the benefits that political efficacy and the satisfaction of hedonic needs provides. This suggests that future research could examine other needs and benefits sought by donors and members in order to more effectively target them. Environmental Concern Those respondents who stated that they were extremely environmentally concerned represented only about twenty percent of the sample. Those who were extremely concerned were more likely to have a higher income than those who were less concerned, however, they scored lower on the social class index. This finding is interesting, although unexplainable with current data, because it suggests that those who are extremely concerned about the environment come from a relatively wealthy but overall lower class (e.g., more blue collar) households. While there were no other significant sociodemographic differences between these segments, those who were extremely concerned were more likely to have indicated that they had changed their buying behaviors and were more likely to have participated in both 224 organizational and environmentally friendly behaviors. In addition, they were interested in the environment for a significantly longer length of time. Those who were extremely concerned rated the political efficacy motive higher as well as the tangible needs. They rated both low risk and demonstration actions significantly higher as well as lobbying. There were no significant differences between organizational preferences. While there were no significant differences between environmental concern and donating behavior, in spite of the higher reported incomes, those who were extremely concerned about the environment were more likely to be members. This suggests that those who are environmentally concerned represent an excellent market for potential memberships in environmental/conservation organizations. Social Class A social class index constructed on the basis of income, education, and occupation proved useful in discriminating between market segments. Upper class respondents were more likely to be married and live in larger market areas. In addition, upper class respondents were more likely to be joiners, donators, and have changed their behavior in response to environmental concerns. In contrast, lower class respondents were more likely to be minorities, and rank expressive incentives higher than upper class respondents and prefer fewer of the low risk or demonstration activities. While segments constructed on the basis of social class may have little practical utility for organizations, it was hoped that segmenting on the basis of social class may provide some insight into the "conservation elitism" argument. Demographically, the findings related to social class, market size, joining, donating, membership behavior, and minority status are similar to the results of other studies (Mitchell, 1980; Perry et al, 1975; Harry, Gale, and Hendee, 1969). No studies were found, however, that examined social class with respect to 225 organizational preferences. The fact that lower class respondents ranked expressive incentives or motives for participation in environmental/conservation organizations higher than the upper classes suggests that those organizations seeking to attract membership from a broader cross-section of society ought to emphasize the expressive benefits that organizational membership provides. Length of Interest in the Environment Length of interest in the environment was also used to provide a market segment-based view of environmentally concerned citizens. Those respondents who indicated that they were interested in environmental issues for between ten and twenty-nine years were more likely to be donors, joiners, to have changed their behaviors in response to environmental concerns, to be members of organizations, to have participated in more environmentally friendly behaviors, and to score higher on the social class index. Those who indicated zero years of interest in the environment ranked hedonic needs lower as well as traditional organizational characteristics and low-risk activities. In contrast those who were interested in the environment for between ten and thirty-nine years rated the political efficacy incentives higher as well as the liberal characteristics and the demonstration characteristics. These findings suggest that length of interest in the environment closely parallels environmental behaviors and involvement and interest in environmental/conservation organizations. Most significant differences were found between those who indicated zero years of interest in the environment and those who had been interested for between ten and thirty-nine years. Environmental behaviors and activism appeared to peak between the ten to thirty-nine year length of interest time frame and decline significantly after forty years of interest in the environment. This may have occurred because those who indicated over forty years of interest in the environment had significantly different interests or needs than those who had been interested for between ten and thirty-nine years or it may be the result of apathetic behavior or changing concerns after such a length of interest in the environment. 226 Those environmental/conservation organizations seeking to rejuvenate this market will need to further examine the needs and wants of the "older" market. Initially, it was suspected that those respondents who indicated zero years of interest in the environment were simply those who chose not to answer the question. This seemed the logical explanation since all respondents had been previously screened on the basis of environmental concern. While the purpose and wording on this question was different (e.g., length of interest in environmental/conservation issues) than the screening question (e.g., how environmentally concerned are you), it was felt that the length of interest question would be very appropriate for a sample pre-screened for environmental concern. Given the number of significant differences between those who indicated zero years of environmental concern and those who listed a specific length of interest this doesn't seem to be the case. Possible explanations for this discrepancy that: 1) at least some respondents felt that environmental concern and length of interest in environmental issues were not necessarily parallel thoughts, 2) respondents did not answer truthfully to the environmental concern question on the screening questionnaire and were in fact not interested in environmental issues, 3) that different household members completed the screening questionnaire versus this study's questionnaire. Further analysis of this segment, while small, would be advisable. It would appear from these findings that those who had been interested in environmental/conservation issues for between ten and twenty-nine years represent the best target for attracting members and donators to environmental/conservation organizations, or at least the current form of these organizations. Developing products and services for respondents who have just recently developed an interest in the environment or for those who have been interested for a longer length of time may also represent good markets. 227 CONCLUSIONS Based on a research design that utilized a nationally representative sample, the results of this study can be generalized to the population of self-expressed, environmentally concerned citizens in the U.S. A high response rate and subsequent non-response checks helped ensure high external validity for the data. The reader should be cautioned that the use of a market research panel necessitated a non-random sample, however, the representative nature of these panels and the checks for external validity led to the use of statistics for analytical purposes and to generalizations of the results to the larger population of environmentally concerned citizens. While the results of this study may be particularly useful for strategic marketing decisions for environmental organizations, they are equally as interesting for those involved in the study of the organized portion of the environmental movement. • Environmentally concerned citizens have more years of education and have higher status occupations than the population in general lending support to conservation elitism arguments. Income, does not appear, however, to be correspondingly higher for this group. • Environmentally concerned citizens are those individuals in the later stages of the family lifecycle who have no children at home, i.e., empty-nestersi. • While expressed environmental concern is high in the general population, this sample of environmentally concerned citizens participated in substantially more environmentally friendly behaviors. • While few respondents were currently members of environmental/conservation organizations, many had participated in an activity such as writing a letter or signing a petition. 228 • Environmentally concerned citizens were more likely to be members in civic, fraternal, religious, or neighborhood organizations than the public at large. • Environmentally concerned citizens recalled on average 3.5 organizations, with some organizations having recall rates above sixty percent. • Those organizations that were most frequently recalled were not necessarily those organizations that were most frequently preferred. • Reports of environmental/conservation organization membership levels in this study were relatively representative of membership levels in these organizations in the population at large. • There were marked differences between the organizations in respondents' consideration sets versus the actual purchase sets suggesting that there were either barrers to the actual purchase or that that subset of respondents who purchased memberships had different preferences. • Those organizations that were most commonly identified with a particular reason for preference were those whose name represented the organizations unique selling proposition (USP) such as the National W ildlife Federation or the National Arbor Day Foundation. • The label "environmentalist" is less preferred as a descriptor of this population compared to "concerned citizens" or "conservationists." • There are three basic incentives that motivate people to join environmental/conservation organizations: political efficacy incentives, normative compliance incentives, and expressive incentives. Overall, the expressive incentives are the least important to all respondents and political efficacy is most important. 229 • While members, donors, and extremely concerned citizens are motivated most by the political efficacy incentives, the incentives studied do not discriminate particularly well between segments. • Respondents identified four major bundles of benefits that organizations offer: hedonic benefits, local structure benefits, tangible benefits and social benefits. • Hedonic benefits are most important to respondents followed by tangible benefits. • Respondents identified four common clusters of organizational descriptors: traditional, pacifist, radical and single-issue. • Respondents grouped together two characteristics: preservation and wise-use, that the literature would suggest were incompatible unless respondents view preservation as wise-use of the resource. • Respondents identified three common groupings of actions that could be used to solve environmental problems that included: low risk/low involvement activities, public demonstration-oriented activities, and eco-terrorist activities. Very few respondents supported the eco-terrorist activities. • Respondents identified two strong clusters of organizations that they would consider joining: wildlife organizations, and action-oriented organizations. Other clusters of organizations had reliability scores that were weak to moderate in strength. • The most preferred organizations were the National Wildlife Federation, National Audubon Society, and The Nature Conservancy. The least preferred organizations were Ducks Unlimited and Sierra Club. 230 • Some organizations (e.g., National Wildlife Federation) had a relatively stable placement in all choice sets, while others (e.g., Ducks Unlimited) were very unstable. • Respondents had relatively simplistic, two attribute methods of comparing organizations on the basis of similarity. • The dimensions and placement of the organizations on these dimensions was relatively similar across all market segments. • The Nature Conservancy and the Wilderness Society and National Wildlife Federation and World Wildlife Fund were perceived as the most similar pairs of organizations. Ducks Unlimited, National Arbor Day Foundation and Greenpeace were the least similar organizations. • Those variables most useful for discriminating between market segments were social class, length of interest in the environment, environmentally friendly behaviors, and organizational activities. • Discriminant functions were at best able |o achieve an eighty percent correct classification rate. Review of these results within the context of the five steps of the consumer decision-making model (Engel et al, 1991) suggests the following things about purchasing a membership or giving money to an environmental/conservation organization. In the need recognition stage of decision-making, environmentally concerned citizens are motivated by political efficacy incentives, normative compliance incentives, and lastly, expressive or personal gain incentives. An examination of the specific organizational needs respondents desired, which also reveals data about the alternative evaluation portion of decision-making, suggests that hedonic needs are more important to environmentally concerned citizens than tangible needs. 231 While only one portion of the search process, source of information, was examined in this study, the primary sources of information were external in nature: receiving a solicitation in the mail, examining a magazine, reading a news report or magazine, and word of mouth. Data from the product, service, and opportunity needs as well as the ideal characteristics of an organizations and the perceptual maps provides insight into the third stage of decision making, alternative evaluation. As was indicated above, hedonic needs dominated the expressed needs as did the traditional organizational characteristics such as global in focus, education-oriented, nature appreciation-oriented, preservation-oriented, and a wise-use orientation. Analysis of the perceptual maps revealed, however, that respondents utilized relatively few, and relatively simplistic dimensions to compare organizations. Two dimensions were identified that appeared to be relatively topical in nature: a habitat orientation, and an organism level focus. Those organizations that were more general in nature, in both dimensions appeared to be those that were most preferred. In the purchase and post-decision evaluation stages of the model, few respondents (15%) actually purchased memberships in organizations and only thirty percent indicated that they were organizational donors. Memberships purchased, however, were representative of national organizational membership at large. An examination of differences between purchase and preference sets suggests that some post-purchase evaluation may have occurred that resulted in re-evaluation of the actual purchased product and subsequent revision of preference sets. One standard decision-making model does not address the differences that exist within a heterogeneous population. It appears that those individual and environmental differences that affect this decision-making process in particular are: social class specifically education and occupation, environmental activity, race, market size, environmental concern, and length of interest in the environment. 232 RECOMMENDATIONS AND IMPLICATIONS Recommendations based on the results of this study are grouped into two broad areas: recommendations for strategic marketing decisions, recommendations for future research. Recommendations for Strategic Marketing Decisions The following recommendations provide guidance to environmental/conservation organizations to assist in making strategic marketing decisions. Organizations should: • Promote and advertise the incentives that they provide to members, especially political efficacy incentives. Specifically, organizations should engender in individuals in all social classes feelings of political efficacy and make available the resources required to influence the socio-political system • Offer, and promote both hedonic and utilitarian benefits to consumers. Most organizations promote excessively the tangible benefits that they provide at the expense of what appear to be more desired: the hedonic benefits and incentives • Examine perceptual mapping and choice sets placement to determine a consumer defined competitive structure as well as the market opportunities that exist within that structure • Use a combination of product differentiation and market segmentation strategies to develop a unique selling proposition (USP) in order to differentiate their organization from their competition. Those organizations with a strong topically-oriented USP were most frequently identified, however, those organizations that were more general in focus were most preferred • Try and improve relative placement in consumers unaided recall sets and overall consistency across all choice sets 233 • Develop environmental/conservation organization products delivered through conventional channels or through a local chapter structure that cater to families with children • Concentrate on turning the large empty-nester population into donors and volunteers since these individuals are the ones with available time and disposable income • Promote environmental/conservation organization membership as a responsible environmental behavior to capitalize on high levels of expressed environmental concern and activism within this population • Examine barriers to purchase of organizational memberships specifically the timing on when people join and who makes the decision to purchase the membership Recommendations for Future Research The analysis of this data set has generated questions for future research that can be grouped into several broad areas which call for research on: environmentally concerned citizens; research on market segments; research on choice sets; research on incentives; research on perceptual differences; and research on product, service, and opportunity bundles. Such research would build a strong theoretical understanding of why people join as well as provide an enhanced marketing approach for organizations. Environmentally Concerned Citizens Analysis of this data set on environmentally concerned citizens confirmed previous research on elitism within the conservation movement but also found that these respondents were predominantly older individuals with few children at home. These findings plus others suggest the following opportunities for future research: 234 • examine the relationship between income and environmental concern to further examine the conservation elite argument. Specific attention should be be paid to the work of Mohai (1985) and Van Liere and Dunlap (1980) with respect to the differentiation between concern and activism and the roll of political efficacy and resource availability on this connection • examine how stage in family life cycle affects environmental concern and activism within organizations and specifically what factors affect decreased activism by families with children at home and what methods can be utilized to increase activism • develop an index of actual environmental behavior (e.g., Green Behavior Scale) in order to discriminate more accurately between concern and activism • examine the perception of the labels "environmentalist, " conservationist," and "concerned citizen" to respondents to determine what implications these identifying labels have in marketing environmental and conservation products Market Segments Analysis of market segments for this study utilized an a priori segmentation design based on psychographic and behavioral variables. This analysis found that most of the market segments provided the same view of respondents and thus the results were not particularly useful for practical market segmentation. Recommendations for further analysis or future study include: • further analysis of sub-segments of the market for example: active members versus non active members, and empty-nesters • further analysis of the data using a cluster-based design to see if natural market segments that may occur in the data have more utility than those specified a priori 235 • future studies of other psychographic or behavioral variables or combinations of variables that provide data more useful to environmental/conservation organizations • future studies of other variables that could be used to classify or predict membership in any particular segments and thus used to more easily identify members Choice Sets An examination of the choice sets respondents constructed of environmental/conservation organizations found that: organizations varied substantially in recall rate; that a high recall rate did not necessarily imply high placement in consideration or preference sets; and that preference sets and purchase sets were different. Further research is recommended to: • study how recall sets of environmental/conservation organizations are constructed and what marketing practices can be used to improve recall rates • examine why different market segments rank organizations differently • determine what influences the differences between preference and purchase sets, specifically, are there any barriers to purchase that prevent consumers from purchasing memberships in their preferred organizations Incentives for Participation In this study, a twelve-item scale based on an incentives theory that supported expressive and instrumental incentives was used. Analysis found that environmentally concerned citizens did not identify two types of incentives, but rather three. In addition, while segments rank ordered these three incentives, the incentives found did not discriminate between segments particularly well. Further research is recommended to: 236 • clarify whether respondents seek political efficacy incentives by joining an organization or whether respondents who join organizations have high feelings of political efficacy • clarify the role of normative compliance as an incentive for participation in organizations, specifically, do reference groups encourage compliance through value-expressive needs or through informational transfer. • study a more complete range of incentives for participation, perhaps based on other theories of incentives (e.g., Olson, 1977; Moe, 1980) to determine if different incentives motivate market segments differently • study a broader range of incentives in a forced choice fashion to determine which incentives are more important to different respondents • examine differences in incentives between those who are members or activists in small, local organizations versus those involved in larger, national organizations Product, Service, and Opportunity Needs Analysis of respondents preference for specific product, service, and opportunity needs discovered fc ur broad bundles of benefits. Future studies are recommended to: • develop a more comprehensive list of needs sought • study these needs in a forced choice fashion to help determine which attributes are more important than others • examination of how these need preferences vary under specific usage situations (e.g., when giving a membership as a gift, for a family, or for an adult individual) 237 • examine differences in needs between those who are members or activists in small, local organizations versus those involved in larger, national organizations Perceptual Differences Between Organizations An examination of perceptual differences between organizations using multidimensional scaling revealed that: consumers use relatively simplistic dimensions and attributes in making decisions; that the attributes or dimensions used are topical in nature; that there is very little difference between perceptual maps across all segments. Based on these findings and the other study findings related to incentives and products, the following research areas are suggested: • use the derived factors based on salient attributes, needs, and services in this study to examine respondents perceptions of organizations on an attribute by attribute basis (e.g„ utilizing conjoint analysis to determine optimal combinations of the attributes for specific segments) • future research to examine how respondents cluster organizations perhaps using a broader list of organizations with a more discriminating (e.g., ordinal level or forced choice) measure • conduct more detailed studies of perceptions and preferences by members of specific organizations • in-depth, perhaps qualitative studies to help determine how perceptions of organizations are formed • studies of the perceptions of environmental/conservation organizations from an organizational standpoint 238 • research to compare the perception of organizations and how they are rated by members and organizational staff on specific attributes compared to the reality of provision of those attributes APPENDIX A GLOSSARY 239 240 The following terms are defined as a reference tool for this dissertation. Many of the definitions come from the glossary in Engel, Blackwell, and Miniard, 1990. Alternative Evaluation: The third stage of the decision-making process in which a choice alternative is evaluated and selected to meet consumer needs (Engel, Blackwell, and M iniard, 1990). Agglomerative Method: A hierarchical clustering procedure which starts with each object or observation in separate clusters. In subsequent steps, object clusters which are closest together are combined to build a new aggregate cluster (Hair, Anderson, Tatham, 1987). Aggregate Analysis: An approach to multi-dimensional scaling, in which the analyst generates perceptual maps of the respondent's evaluations of stimuli. In an attempt to create fewer maps, the analyst seeks to cluster subjects' responses to find a few "average" or representative subjects (Hair, Anderson, Tatham, 1987). Algorithm: A set of rules or procedures; similar to an equation (Hair, Anderson, Tatham, 1987). Attribute: A characteristic or property of a product; generally refers to a characteristic that serves as an evaluative criterion during decision making (Engel, Blackwell, and Miniard, 1990). Attribute Evaluation Measures: Measures used to assess the goodness or badness of an attribute (Engel, Blackwell, and Miniard, 1990). Attribute Importance Measures: Measures used to assess the concept of salience or potential influence of product attributes (Engel, Blackwell, and Miniard, 1990). Benefit Segmentation: A marketing strategy oriented toward meeting a benefit or felt need in a target market segment (Engel, Blackwell, and Miniard, 1990). Compensatory Strategy: A strategy in alternative evaluation in which a perceived weakness on one attribute may be compensated for or offset by strength on others (Engel, Blackwell, and Miniard, 1990). Compositional: This is one of the most popular approaches to modeling consumer attitudes towards attributes. The compositional approach uses subjective attribute perceptions and values to predict attitudes or preferences (Malhotra, 1988). Conjunctive Decision Rule: A noncompensatory decision rule involving processing by brand in which cutoffs are established for each salient attribute and each brand is compared to this set of cutoffs (Engel, Blackwell, and Miniard, 1990). Consideration Set: The set of alternatives from which choice is made (Engel, Blackwell, and M iniard, 1990). Consumer Behavior: Those actions directly involved in obtaining, consuming, and disposing of products and services, including the decision processes that precede and follow these actions (Engel, Blackwell, and Miniard, 1990). 241 Cutoff: A restriction or requirement for acceptable attribute values, used in alternative evaluation (Engel, Blackwell, and Miniard, 1990). Decision Rule: A strategy that a consumer uses to make a selection from the choice alternatives (Engel, Blackwell, and Miniard, 1990). Decompositional: Decompositional methods used evaluative judgments to infer the values of experimentally manipulated attribute levels. These techniques are most often used in new product development (Malhotra, 1988). Determinant Attribute: A salient attribute on which choice alternatives differ in their performance (Engel, Blackwell, and Miniard, 1990). Dimensions: Features of a phenomenon (product, service, image, etc.). May be both perceived (subjective) dimensions and objective dimensions (Hair, Anderson, Tatham, 1987). Disaggregate Analysis: An approach to MDS in which the researcher generates perceptual maps on a subject-by-subject basis(Hair, Anderson, Tatham, 1987). Divisive: A clustering procedure which begins with all objects in a single cluster. This is opposite of agglomerative procedures. The procedure starts with a single large cluster that is divided into separate clusters based on objects that are most dissimilar (Hair, Anderson, Tatham, 1987). Environmental Organization: A non-governmental social group whose members work to support some aspect of the environmental movement. Euclidean Distance: The most commonly used measure of the similarity between two objects. Essentially, it is a measure of the length of a straight line drawn between two objects (H air, Anderson, Tatham, 1987). Evaluative Criteria: The standards and specifications used by consumers to compare different products and brands (Engel, Blackwell, and Miniard, 1990). Evoked Set: see consideration set. Habitual Decision Making: Decision making based on habits of repeat purchasing, often formed to simplify decision process activity (Engel, Blackwell, and Miniard, 1990). Hedonic Benefits: The symbolic value of a consumption object in terms of emotional response, sensory pleasure, daydreams, or aesthetic considerations (Engel, Blackwell, and Miniard, 1990). Ideal Points: The point on a perceptual map which represents the most preferred combination of perceived attributes (according to respondents) (Hair, Anderson, Tatham, 1987). Image Analysis: The examination of consumers' knowledge or beliefs about a product's properties (Engel, Blackwell, and Miniard, 1990). Lexicographic Decision Rule: A noncompensatory rule involving processing by attribute in which brands are compared on the most important attribute; if more than one brand 242 qualifies, the next most important attribute is selected until the tie is broken (Engel, Blackwell, and Miniard, 1990). Mahalanobis Distance: A standardized form of Euclidean distance. Data are standardized by scaling responses in terms of standard deviations and adjustments are made for inter correlations between the variables (Hair, Anderson, Tatham, 1987). Marketing: The process of planning and executing the conception, pricing, promotion, and distribution of ideas, goods, and services to create exchanges that satisfy individual and organizational objectives (Engel, Blackwell, and Miniard, 1990). Marketing Mix: This term refers to a unified strategy integrating production, price, promotion, and placement (Engel, Blackwell, and Miniard, 1990). Market Segmentation: A marketing strategy involving viewing each segment as a distinct target with its own requirements for product, price, promotion, and placement (Engel, Blackwell, and Miniard, 1990). Multi-attribute Attitude Models: Models that propose that overall attitude depends on beliefs about the attitude object's attributes weighted by the salience of these attributes (Engel, Blackwell, and M iniard, 1990). Network Marketing: The development of a firm's marketing mix in close relationship to the marketing program of other firms (Engel, Blackwell, and Miniard, 1990). Noncompensatory Strategy: A strategy in alternative evaluation in which a brand's weakness on one attribute cannot be offset by a strength on another attribute (Engel, Blackwell, and Miniard, 199G). Objective Dimensions: Physical characteristics of a phenomenon (Hair, Anderson, Tatham, 1987). Perception: The act of apprehending by means of the senses or o the mind; cognition; understanding. Immediate or intuitive recognition. Perceptual Map: The visual representation of a respondent's perceptions of two or more dimensions or features (Hair, Anderson, Tatham, 1987). Perceived Dimensions: Respondents may "subjectively" attach features to a phenomenon (H air, Anderson, Tatham, 1987). Preference: A practical advantage given to one over others. The act of preferring. Preference Data: Preference implies that stimuli are judged by the respondent in terms of dominance relationships: that is, the stimuli are ordered in preference with respect to some property (Hair, Anderson, Tatham, 1987). Rational Decision Making: Problem solving based on the careful weighing and evaluation of utilitarian or functional product attributes (Engel, Blackwell, and Miniard, 1990). 243 Retrieval Set: A consideration set obtained by recall of alternatives from memory (Engel, Blackwell, and Miniard, 1990). Salience: The potential influence that a criterion exerts during the alternative evaluation process, often measured in terms of importance (Engel, Blackwell, and Miniard, 1990). Similarity Data: When collecting this type of data, the researcher is trying to determine which of the items are most similar to each other and which are the most dissimilar (H air, Anderson, Tatham, 1987). Simple Additive Decision Rule: A compensatory decision rule under which the consumer counts the number of times each alternative is judged favorably in terms of the set of salient evaluative criteria (Engel, Blackwell, and Miniard, 1990). Spatial Maps: MDS techniques that enable the researcher to represent respondents' perceptions spatially (Hair, Anderson, Tatham, 1987). Stress Measure: A measure of the proportion of the variance of the disparities which is not accounted for by the MDS model. The stress measure helps to determine the appropriate number of dimensions to include in the model (Hair, Anderson, Tatham, 1987). Utilitarian Benefits: Benefits resulting from purchase or other consumer decisions that are objective, functional product attributes (Engel, Blackwell, and Miniard, 1990). Weighted Additive Decision Rule: A compensatory decision rule in which judgments about an alternative's performance on evaluative criteria are weighted by the relative salience of the evaluative criteria (Engel, Blackwell, and Miniard, 1990). APPENDIX B SCREENING QUESTIONNAIRE 244 ANSWER OTHER SIDE FIRST 7. Which, if any, of the following activities do you or your household engage in on a regular basis? (✓ All That Apply) 01 □ Recycle aluminium cbQ Use public transportation m 0 Recycle glass * 0 Recycle newspaper « □ Use water conservative devices in home » □ Combine errands to save gas 070 Buy energy efficient appliances « 0 Consider environmental Impact when buying products oo0 Recycle used motor oil/paints » 0 Recycle plastic bottles/oorrtainers n 0 Reuse plast&paper bags from supermarket « 0 Buy only organicatly.gnown fruit s/vegetables • 0 None of the above 8. How much does a company's environmental record affect yourlfcelihood of buying that company's products? (✓ One Box) i 0 Extremely affected * O Very affected * 0 Somewhat affected 4 0 Not at all affected 9. How environmentally conscious/ concerned do you consider yourself or your household? ( / One Box) i 0 Extremely concerned t 0 Very concerned a O Somewhat concerned 4 0 Not at all concerned 10.a.Have you or anyone in your household changed their buying habits due to environmental concerns? i 0 Yes - (Continue) * 0 No - (Stop - Return Card) lO.b.What items or products, If any, have you or your household stopped buying or buy less of because of environmental considerations? Please list the first three products that come to mind. 1.______APPENDIX C NFO PANEL BACKGROUND QUESTIONNAIRE 246 247 NFO TELEPHONE MEMBERSHIP NATIONAL FAFAMILY NUMBER NljMBc R OPINION, INC 1 t TOLEDO. OHIO 43654►-0999 A. is your name, address and phone number printed correctly? O Yes ONo 4 It No. please make corrections. ------► B. Is your delivery address the same as the mailing address? O Yes O N o -> If No. please write in your delivery address ------1 C. If I need to contact you by phone, when would be the best time to reach you? Monday - Friday Saturday • Sunday CD Morning CD Morning Dear NFO Member, It's time again for me to ask you to continue as one of my valued members by completing this Household Information sheet As you know, occasionally I need to update my files by asking you to answer these questions so that we can reflect vour opinions accurately to manufacturers and service providers I hope you will take a few minutes to complete and return this questionnaire in the enclosed postage-paid envelope so I can continue to send you questionnaires to answer As always, your answers are held in strict confidence and used tor market research purposes only Please read the special instructions below before starting to fill out this document Be sure to answer the questions about your nam e and address at the top of this page before continuing. I d like to thank you for your continued help as a National Family Opinion member I look forward to hearing from you so o n Sincerely. P S Do you have any questions about answering your Household Information My staff sheet> will be happy to assist you Call 1 -800-537-4097 toll-free In Ohio. calM -800-472-4000 Monday through Friday, 8 00 A M to4 30PM Eastern time Special Instructions: PLEASE USE ONLY A NO. 2 PENCIL WHEN COMPLETING THIS FORM. • Do not use ink, ballpoint or felt tip pen. • C om pletely fill in th e oval w ith a dark m ark. • Erase unwanted marks cleanly and carefully. • Make no stray marks on this form. Correct Mark • Incorrect Marks (ETCCD® ( HWM'l ...... PLEASE OO NOT WfifTE IN THIS AREA B oooqqoooqoqqooqqqqqoqoo 1 5 7 1 7 - 0 2 248 Correct Mark A Incorrect Marks (B ^ O ® 1. Please list all m em bers of your household or in the arm ed forces do NOT iist uersui c _ = = = = ^ ■ m n r N iik birth (DO NOT include day of birth) indicz CODE LIST A MANAGE RIAL/PROFESSIONAL/EXECUTIVE SERVICE: PLE PRIN EXECUTIVE/ADMINISTRATIVE/MANAGERIAL. 3 5 CLEANING/8UIIDING/H0USEHQL0 SERVICE FIR^ .JAME Ql ACCOUNTANTS,'AUDITORS/FINANCIAL MANAGERS 3 6 FOOD SERVICE 0 2 administrators 3 7 WEALTH SERV1CE/ASSISTANTS/AIOES MONTH 03 MANAGERS 3 8 PROTECTIVE SERVICE MARKING EXAMPLE AND 04 SUPERVISORS 39 BARBERS/BEAUTCIANS/C0SMET0L0GISTS 0 5 ALL OTHER EXECUTtVE/AOMINlSTRATiVE/ 4 0 ALL OTHER SERVKE OCCUPATIONS YEAR MANAGERIAL OCCUPATIONS FARMING / FORESTRY / FISHING. Jane OF 41 FARM OPERATORS/MANAGERS PROFESSIONAL SPECIALTY YEAR BIRTH 0 6 ARTISTS 4 2 ALL OTHER FARMING/FORESTRY/FISHW G 07 BANKERS/BROXERS OCCUPATIONS CD Jan W rite th e 0 8 CLERGY CRAFTSMAN/REPAIRMAN: n u m b e r s 09 0ENTISTS CD Feb 4 3 CONSTRUCTION TRA0E ( h e b o x e u» tjir 10 OX TITIAN 44 MECHANICS/REPAIRERS GDGD 11 ENGINEERS 12 LAWYERS 4 5 PRECISION OCCUPATIONS (E Apr 13 MATHEMATICS/COMPUTER SCIENTISTS 4 6 ALL OTHER CRAFTSMAN REPAIRMAN OCCUPATIONS 14 PERFORMERS/ENTERTAINERS OPERATOR/LABORER: CS> May 15 PHARMACISTS 4 7 GENERAL LABORERS CD Jun ofiesooiici 16 PHYSCIANS 4 8 MACHINE OPERATORS 1 : PROFESSORS/INSTRUCTORS 4 9 TRANSPORTATION/MATERIAL MOVING CD Jul 18 PSYCHOLOGISTS, COUNSELORS 5 0 ALL OTHER OPERATOR/LABORER OCCUPATIONS CD Aug CD CD 19 REGISTERE0 NURSES A LL O T H E R : 2 0 TEACHERS PRIMARY CD Sep CD rp 21 TEACHERS SECONDARY 5 1 AflMEO FORCES 2 ? VETERINARIAN 52 HOMEMAKERS Oct 23 WRITERS/JOURNALISTS 53 STUOENTS CD Nov 2 4 OTHER HEALTH PROFESSIONAL 5 4 ALL OTHERS 2 5 A ll OTHER PROFESSIONAL SPECIALTY © Dec OCCUPATDNS l RELATIONSHIP TECHNICAL/SALES/ADMINISTRATIVE SUPPORT TO 2 6 ADMINISTRATIVE SUPPORT/CIERCAL/QFFICE HELP 2 7 COMPUTER OPERATORS YOU 28 COMPUTER PROGRAMMERS 2 9 HEALTH TECHNICIANS 2. Is this person employed' 3Q INSURANCE A O JU ST Q R S.'W ll COLLECTORS 31 RETAIL SALES/CASHIERS 32 SALES-FINANC1AL/BUSWESS YES. full-time 3 3 TRAVEL AGENTS YES, part-tim e 3 4 ALi OTHER TECHNICAL/SALES/ADMINISTRATIVE SUPPORT OCCUPATIONS NO. retired • (Skip to Queshi. _ NO. not em ployed - (Skip t o Question 3. CODE LIST 8 a. Which one of the codes from Code 01 AGRICULTURE / FORESTRY/FISHING R E T A l TRADE List A best describes this person s 0 2 MMMG 38 APPLIANCE RETAIL STORES occupation' 0 3 CONSTRUCTION 3 9 RETAR BAKERS 4 0 RfTAX OfiUG STORES/PHARMAOES Write in the MANUFACTURING 41 MASS MERCHANDISE/06C0UNT/0EPARTMEWT 0 4 AGAl-CHEMCAl /PESTCO E/FER TIllZER STORES appropriate number 0 5 APPLIANCE MANUFACTURE 4 2 RESTAURANTS/0RWXING ESTABLISHMENTS 0 6 BOTTLING FRM S 4 3 GROCERY RETA1/FOOO RE TAX from Code List A 0 7 CANOY/fflOZEN CONFECTIONERIES MANUFACTURE 4 4 ALL OTHER RETAIL TRADE and fill in the 0 8 CEREAL FINANCE/INSURANCE/REAL ESTATE 0 9 COFFEE/TEA MANUFACTURE corresponding ovals >0 COOUE/CRACXER/BiSCUIT MANUFACTURE -'BAKERIES 4 5 BANKS/SAVM GS & LOANS 11 COSN'ETC MANUFACTURE 4 6 INSUFIANCE COMPANIES completely 12 Q A M S/CHEESE MANUFACTURE 4 7 ALL OTHER FW ANOAL/INSURANCE/ REAL ESTATE 13 OETEriGEWT/SQAP/ClEAMNG AJOS MANUFACTURE INOUSTOES 14 FEMININE PROTECT ON PRQOUCTS BUSINESS AND REPAR SERVICES 15 FRACAANCE /PERFUME DESCN/MANUfACTURE 48 ADVERTISM3 16 HEALTH I BEAUTY AJOS MANUFACTURE 4 9 APPLIANCE REPAIR 17 GFIAM COMPARES/GRAIN MILLS 5 0 EXECUTIVE SEARCH FIRMS b. Which one of the codes from Code liist B 18 ORANGE JUCE MANUFACTURE 51 MARKET RESEARCH best describes the company this persbn 19 PAPER PROOUCTS MANUFACTURE/PAPER MILLS 5 2 ALL OTHER BUSNESS OR REPAIR SERV O 20 PRWTWG/PUBLISHING a n t S T W S works for' 21 PHARMACE UTCAL / VTTAMtN MAN UF ACT URE 22 SALTY WACK/POPCORN/POTATO O t f MANUFACTURE PERSONAL SERVICES 2 3 SlAfOCO/MEAT PR0CESSN3 5 3 E0UCATXM 24 SOFT ORNK/OTHER BEVERAGE MANUFACTURING 5 4 HOTELS/MOTELS/W4S Write in the 25 SUGAR PROCESSMG 5 5 A U OTHER PERSONAL SERVCES appropriate number 26 ANY OTHER FOOO/GROCtRY MANUFACTURING/ 56 ENTEJTTAJMMIffT AND RECREATION from Code List B PROCESSMG SERVICES 27 ALL OT>€R MANUFACTURMG PROFESSIONAL AND RELATED SERVICES and fill in the TRANSf-ORTATION/COMMUtaCATION/PUBLC UTUJTIES 5 7 HEALTH SERVCES 28 8RQAiXAST M£OU RAOt)/TV corresponding ovals 5 8 A U OTHER PRQFESSCNAL AMD 2 9 . TRUCK WG/WARE HOUSING RELATEO SERVCES com pletely 3 0 ANY 4JTXR TRANSPORTATION/C0MMUNJCATBN/PU8LC U T im tS INOUSTRXS 5 9 PlIBUC ADMUMSTRATTON/GOVERNNREIfT 31 WHOLESALE TRADE-EXJAABLE GOODS 6 0 NOT CLASSIFIED ABOVE WHOLESALE TRADE-NONDURABLE GOODS c. Write in name of company/organization 3 2 COSMETIC PACK AGING/SUPPLY employed by (If self-employed, 33 FOQOPACXAGfC 34 FOOD ASSOCLATOiS/COOPS write in self-employed) 3 5 GROCERY W H O LESA LE/C tSTW nQ N 36 O T K R WHOLESALE PACXAONG 3 7 (JT>€8 MPOURA8LE GOODS WHOLESALE 249 living at hom e List yourself in the first section. (Do NOT list persons w ho are aw ay at college . What is your marital status7 ' ns at home only on vacation ) Print each person's first name. Indicate their month and year of ffi Now married ate their sex and their relationship to you. © Never married OTHER HOUSE HOLD MEMBERS ffi Divorced | YOURSELF SPOUSE 1 ffi Widowed j ffi Separated j I M O N T H Y E A R MONTH MONTH 4. Are you i <33 Jan ® Jan 1 9 - ® Jan <33 Jan 1 9 - CD Jan 1 9 - ffi Ja n 19- ffi Ja n 1 9 - ffi White i 'ffl Feb I ® Feb © Feb ffi Feb ffi Feb ffi Feb ffi Feb © Black/African - American ! Spouse Other Member: 250 8. Indicate the h ig h est grade or year of school (or equivalent) completed by you/your spouse (FILL IN ONE OVAL) Yourself Spouse Elementary Less than 8 years ...... CD CD Elementary: 8 years (Graduate) . .. . CD CD High school 1 - 3 years . CD CD High school 4 years (Graduate) . CD CD College 1 - 3 years (Attended college or associate degree) ...... CD CD College: (Graduate) . . CD CD College: (Post-graduate studies) ...... CD CD 9. Please fill in the oval below to show your Household's YEARLY INCOME in dollars Be sure to combine the total income for all household members living at home-such as wages or salaries, income from self-employment, rents, dividends, etc -BEFORE tax deductions. (FILL IN ONE) © Under $7.500 CD $ 2 0 ,0 0 0 ■ - $22,499 © $35,000 ■• $37,499 ffi $ 7 5 ,0 0 0 - $ 8 4 ,9 9 9 ffi $7,500 -$ 9 .9 9 9 ffi $22,500 ■• $24,999 ® $37,500 - $39,999 ffi $85,000 - $99,999 ffi $ 1 0 ,0 0 0 - $ 1 2 ,4 9 9 ffi $25,000 ■$27,499 © $40,000 -$ 4 4 ,9 9 9 ffi $ 1 0 0 ,0 0 0 - $ 1 2 4 ,9 9 9 ffi $ 1 2 ,5 0 0 - $1 4 ,9 9 9 ® $27,500 -■ $29,999 ffi $45,000 ■■ $49,999 ffi $ 1 2 6 ,0 0 0 - S 1 4 9 .9 9 9 CD$15,000 - $1 7 .4 9 9 © $ 3 0 ,0 0 0 ■ ■$ 3 2 ,4 9 9 © $50,000 ■■ $59,999 © $ 1 5 0 ,0 0 0 - $ 1 7 4 9 9 9 CD $ 1 7 ,5 0 0 - $ 1 9 .9 9 9 © $ 3 2 ,5 0 0 - $ 3 4 ,9 9 9 ffi $60,000 ■ ■$ 7 4 ,9 9 9 ffi $175,000 and over 1 0 . Please fill in the ovals below to show the major credit cards you hold CD MasterCard - Regular CD American Express - Green (2) AT & T Universal C i O th e r CD MasterCard - Gold/Premium CD American Express - Gold ffi J C Penney C i N o n e CD MasterCard - Business Card CD American Express - Platinum ® S ears CD VISA - Regular ffi American Express - Optima ffi Diner s Club CD VISA - Gold/Premium C® American Express - Corporate ® Discover CD VISA - Business Card ® Oil C om pany Card 11. During the past 12 months, how many trips by commercial airline did the m ost frequent traveler in your household mrv i -iL ake -o for Domestic (U.S.) Business O no trips 0 1-3 0 4 - 9 O 10 or more trips D om estic (US) Pleasure O no trips 0 1-3 0 4 - 9 O 10 or more trips International Business O no trips 0 1-3 0 4 - 9 O 10 or more trips International Pleasure O no trips 0 1-3 0 4 - 9 O 10 or more trips 1 2 . Mark all of the following which apply to you or any member of your household OWN: OWN. WEAR: CD Dog CD Automatic Dishwasher © Hard Contact Lenses CD C at CD Mobile Home © Soft Contact Lenses CD Microwave CD Motor Home/Travel Trailer/RV ® Gas Permeable Contact Lenses CD Automatic Clothes Dryer ® VHS Video Cassette Recorder © Extend Wear Contact Lenses CD Agitator (Top Loading) ® Cellular Phone (Car Phone) ® Dentures Automatic Washing Machine © IBM Compatible Personal Computer SUBSCRIBE TO: CD Front Loading ® Non-IBM Compatible Personal Computer © Cable TV - Basic Service Automatic Washing Machine ® Automatic Drip Filter Coffee Machine © Cable TV - Premium Channels © Business Operated from Home SUFFER FROM: © Arthritis 13. Please mark the name of the long distance company you use from your © A sthm a h o m e, for personal calling (FILL IN ONE ONLY) ® Diabetes © AT&T ® GTE ® High Blood Pressure CD MCI CD ITT © Migraines ffi US Sprint CD O ther © Ulcers CD ALLNET 1 4 . Do you. or does anyone in your household, smoke cigarettes? CD Yes CD No Pfinud in U S A Mark H k lU .- by NCS MP86984:321 ______PIE A SE DO NOT WAITE IN TWIS AREA jojooooooooooooooooooooooo APPENDIX D SURVEY QUESTIONNAIRE 251 252 "MARKET RESEARCH THROUOH I I M IIIHTtTI V I HOUIIHOIDI' NATIONAL P. O. Box 474 Toledo, OH 43654 lA f lf lB V TOLL-FREE NUMBERS OUTSIDE OHIO: 1-600-637-4097 OPINION. INC INSIDE OHIO: 1-600-472-4000 MONDAY THROUGH FRIDAY 600 A.M TO 430 P.M. EASTERN TIME 17411 Dear NFO Member, Today I am interested In your opinions regarding environmental and conservation organizations. As a person concerned about the environment, your opinions are very important to me. I think you will find the questions Interesting to answer. Most ot the questions can be completod simply by checking a box. Whore I have provided write-in lines, please be as specific as possible. Since this questionnaire is being mailed to only a select group ot people, your answers are especially important to me. Some of my questions refer to specific environmental or conservation groups. Your answ ers are important to me EVEN IF YOU ARE UNFAMILIAR with the organization. For those organizations that are unfamiliar to you, please respond based upon what you think the group is like. After you have answered all of my questions, please return the questionnaire in the postage-paid envelope I have enclosed. Thank you so much for your cooperationl Sincerely, $ M 4 l ' Carol Adams V MEMBER OT AMERICAN MARKETING ASSOCIATION TOLEDO CHAMBER OK COMMERCE 1 . To start with, please list below ifl ol the environmental/conservation organizations that coma to mind. The following questions ask you about membership In any environmental/conservation organization. For the purposes ot this survey, membership means that you pay annual dues to an organization. 2. Are you a member ol any environmental/conservation organization? i OYcs • (Continue) i □ H o - (Skip To Qu. 5) 3. List all ol the environmental/conservation organizations ol which you are currently a member and indicate the years during which you havo been a member. Hame Ql Organization Years Ol Membership 4. How did you decide to become a member ot these environmental/oonservationorganizatbns? (Check ALL ThatA pply) ‘ Qnooommcndod by a friend • QRoceived a solicitation In the mall * □ Received as a gift t QSaw a T.V. program * □Impressed by theirmagazine or other • QSaw/read news reports/articles publication • QDo not recall « QSaw an advertisement OOther - (Specify): ______i QWent to a meeting 5. Relative to environmental conservation organizations, have you ever . . . (Check ALL That Apply) i QAttended a local meeting • □Held a position or office i □ Participated In a sponsored trip » QAttended a national meeting > □ Written letters on environmental Issues $ QSigned a petition « QSubmitted an artide or photo to a QOlher - (Specify): ______pubf'cation . QNone of the above * QVoiunteored any time 6. Have you donated any money to support environmental/conservation organizations or their projects or goals In the last year (this Is In addition to any membership dues you may pay)? i QYes -* Approximately how much In the last year? (Wrtte In ONE Dollar Amount) $ ______•□ N o 7. For approximately how long have you been Interested in environmental/conservation Issues? # Of Years: ______ 8. Do you belong to any civic, fraternal, re'-gbus. or neghborhood organizations? i CJYes -♦ Approximately how many? #______i QNo 254 Listed below are pairs of environmental/conservation organizations. We want your llrst Im pression ol now mucn one org an izatio n Is lik e a n o th e r . T ak e a look a t th e p airs ot o rg an izatio n s below . S o m e o l th e s e n a m e s m ay b e m ore lam ihar than others — that is okay, we simply want your overall Im pression as to how similar each pair is There are many dillerent ways to make these comparisons. Please use the criteria or characteristic^) you are most comfortable with. Please rate the following pairs of environmental/conservation organizations on a scale of 1 to 5 where • f m eans not at all similar and "S' m eans extremely similar. You may choose any number between 1 and 5 (Check ONE Box For EACH Pair Ol Organizations) Not At All Extremely -Similar. _Slmllar. Compare Creonpeacet TO -lAudubon Sodety: & i C o m p are Audubon Society TO 'Sierra C l u b i. C o m p are Nature Conservancy TO iWodd Wildlife Fund '. C o m p are Greenpeace;.1 « TO .Cousteau Society v .. C om pare : 'Jatfpnat Wifdflfa F e d e ra tio n ® TO Ducks Unlimited. ,v, " f-K W\'' vM , jf-y .■ ■> t - C o m p are Audubon Society - * TO C ousteau Society C o m p are Wilderness Society •, } TO G reenpeace ...... C om pare National Arbor Day Foundation TO World Wildlife Fund . C om pare Cousteau Society ro N ature Conservancy C om pare Ducks Unlimited • TO SiOfU Club 1 , , C om pare Wilderness Society TO Audubon Society. . C om pare Ducks Unlimited TO National Arbor Day Foundation C om pare Nature Conservancy ■ - . TO Groonpeaco ...... C om pare Cousteau Society TO Sierra club; ...... C om pare Sierra Club • TO Of jo n p >c i C o m p are C ousteau Society -w- TO National WiioMo F ed eratio n . C o m p are Audubon Society TO Nature Conservancy . .,... C o m p are World Wttdlito Fund TO Aoc.e onr - i, C o m p are National Wllolifo Federation :■ 10 Groi rirn r C om pare Wilderness S ociety • •; ; - TO Sloriitlii : C om pare Nature Conservancy. TO National Arbor Day Foundation C om pare National Arbor Day Foundation TO national w.iiouia reoeratiqn . C o m p are World Wlldllle Fund TO Sierra Club :v ,.. C om pare Wilderness Society ‘ TO National Wildlife Federation , ; C om pare I'tallonal Wildlife Federation TO Audubon Society C om pare Nature Conservancy TO Sierra Club , C om pare Greenpeace . w .; TO National Arbor Day Foundation C om pare World Wildlife Fund A TO Wilderness Society ...... C om pare Ducks Unlimited ■ • > TO Audubon Society ,1 C om pare National Arbor Day Foundation TO Wilderness Society, r '.■ •• ; . -1 -.1 - C o m p are Wilderness Society -- ■; TO Cousteau Society... C o m p are National Wildlife Federation TO Nature Conservancy ’ C o m p are World Wildlife Fund TO Greenpeace C o m p are National Anxir Day Foundation TO Cousteau Society. National Arbor Day Foundation TO C o m p are j4» ...... ; Audubon Society C om pare Sierra Club TO National Wildlife F ed eratio n .’. C o m p are Nature Conservancy . / TO Dusks Unlimited .....___ :. C o m p are Ducks Unlimited y TO World Wildlife Fund >...,. ... C om pare National Wildlila FoderalionVip TO World Wildlife Fund C o m p are Nature Conservancy TO yekunV ' Wilderness Society--. C om pare Greenpeace '■ ' TO Ducks Unlimited : •. C o m p are Ducks Unlimited - ’ ■ 1 - .. ‘ I TO Wilderness Socioty. ,,.,;. •., C om pare Cousteau Society , fk TO World Wildlife Fund / ...... C om pare Sierra Club • • • • TO National Arbor Day Foundation C o m p are Cousteau Society 1 ‘ > TO Ducks Unlimited f 10. In this question, we are Interested In knowing which organizations from the list provided below you would most prefer to join. (Please Rank From 1 = “Most Preferred" to 10 = “Least Preferred". Please make sure you rank each organization and use ALL num bers between 1 and 10.) 1 = Most Preferred 10 = L e a s t P r e fe r re d Ducks Unlimited ...... National Wildlife Federation . National Audubon Society ...... Greenpeace ...... Nature Conservancy ...... W ilderness Society ...... World Wildlife F u n d ...... Cousteau Society ...... National Arbor Day Foundation ____ Sierra C lub ...... 11. What do you like best about your num ber one choice in Question 10? (Please Be As Specific As Possible) 17411 255 12. Whal products and services would you m ost want from an environmental/conservation organization? Please rate me following Items from 1 to 5 where * 1* m eans not at ail Important and ‘5* m eans extremely important. You may choose any number between t and 5. (Check ONE Box For EACH Product/Service) N o t A t All E x tre m e ly unp^am lmo|nani a. M a g a z i n e ...... -□ >n >n *n b. Local chapter ...... •n •n *n >□ c. Local outings ...... - □ »n >n •n • □ d . Local chapter m eetings ...... • □ • □ *□ >□ • □ e. Conferences or trainings ...... - □ »n >n •n • □ f. Adventure travel ...... •□ •n >n ■n a Gift/catalog services ...... »n >n 13. Other than those products and services listed above in Question 12. what are the m ost Important opportunities an environmental/conservation organization can offer you? Please rate the following items from 1 to 5 where T m eans not at all Important and "5" m eans extremely Important. You may choose any number between 1 and 5. (Check ONE Box For EACH Opportunity) Not At All Extremely Important Important 4------a. Opportunity to m eet people with similar Interests ...... >□ >n >n ♦n •n b . Opportunity to volunteer ...... •□ >n »n «n •n c. Chance to become more aware and knowledgeable ...... -□ >n .□ •n •n d . E n te rta in m e n t, en jo y m en t, f u n ...... ,□ > □ > □ «□ »□ e. Opportunity to financially support a c a u se ...... □ >n >n 14. if you were to describe your Ideal environmental/conservation organization, what characteristics would it have? (Check ALL That Apply) oiQGtobal »□ Radical ■ □ Liberal io Q Nature appreciation ■□Single-issue ■□Preservation ■□Pacifist uQWise-use ■□ Educational ■□Militant cOLocal ■□ Legislative influence ■□Conservative ■□ Environmental problem-solving ■□Multi-issue OOther - (Specif):______ 15. Please review the following list and place a check In the box of each environmental/conservation organization that you would consider joining. (Check ALL That Apply) ■□Sierra Club ■ □ Greenpeace ■ □ S ea Shepherd Conservation Society ■□National Arbor Day Foundation ■□D ucks Unlimited ••□Rainforest Action Network ■□Friends of Animals nQEarlhl Firstl •□ W orld Wildlife Fund >•□ W ilderness Society ■□National Audubon Sodety ■□League of Conservation Voters ■□National Toxics Campaign ■□Nature Conservancy ■□N atural Resource Defense Council ■□National Wildlife Federation ■□D efenders of Wildlife ■□Conservation international io Q Environmental Defense Fund □O ther - (Specify):. nQCousleau Society . □None of the above 16. For each of the organizations listed below, please Indicate to what degree you agree or disagree the organization possesses each ol the 3 characteristics - education oriented, politically oriented, and protest oriented. (For Each Organization, Check ONE Box For Each Characteristic) E d u c a tio n Politically P r o te s t - O rie n te d O rie n te d O r ie n te d « £«> pS ? i r2S So S o II fj II 5.£ic&rB J4 Sierra C lu b ...... • - n o C l C l C l > n C l 4 1 4 1 4 1 C l C l 4 1 4 1 4 1 Cousteau S odety ...... • > n i 1 C l C l C l C l C l C l C l C l ■ n 4 1 C l 4 1 4 1 W ilderness Society ...... c i C l C l C l C l > n C l 4 1 C l C l C l 4 1 4 1 4 1 4 1 Greenpeace ...... • < n C l C ) 4 1 4 1 ■ n C l 4 1 C l 4 1 ■ n 4 1 C l 4 1 4 1 National Wildlife F ederation ____ . - n o a o 4 3 >□ < 3 4 3 4 3 4 3 ' □ 4 3 4 3 4 3 4 3 National Arbor D ay Foundation . • > n C l C l C l 4 1 C l C l C l C l 4 1 >n 4 1 C l C l 4 1 W orld Wildlila Fund ...... • C l C ) C ) C l C l C l C l C l C l C l > n C l • C l 4 1 C l Nature C onservancy ...... ,r I >11 C l 4 1 C l C l C l C l 4 1 4 1 - n 4 1 C l 4 1 4 1 National Audubon Sodety ...... • Cl Cl Cl Cl' 41 C l 4 1 C l C l C l • n C l 4 1 4 1 4 1 Ducks Unlimited ...... • C l C J o CD 4 3 >□ 4 3 4 3 4 3 4 3 >□ 4 3 4 3 4 3 4 3 © 17411 256 17. The following statem ents ask you how you feel atxiut envitonmentaVconservalton organizations. Pleas* tat* the follow ing s ta te m e n ts o n a s c a le o l 1 to 5 w h e re "1 'm e a n s d isa g re e co m p letely a n d *5* m e a n t a g re e c o m p letely . Y ou m ay choose any number between 1 and 5. (Check ONE Bo* For EACH Statem ent) Disagree Agree Completely completely — a. Membership contributions help environmental/conservation organi zations Influence government action on conservation problem * ...... >□ '□ 0. Organized groups are very effective In Influencing nallonal environmental tssuot ...... ’□ »□ •□ •a c. People become members of envlronmentat/oonservatlon organizations because thoir Iriends encourage them to belong ...... *□ »□ «□ d. II people like me contribute to an environmental or conservation organization, the group will be better able to work kx improved environmental quality ...... >□ *□ •□ e. Some Important aspects of my tile at* alfected by environmental/ »□ *□ <□ »□ I. Environmental/conservation organizations should actively Involve themselves in environmental problems ol special groups such as minorities and the economically disadvantaged ...... -□ •□ g. Environmental/conservation magazines and other publications are one ol the most important reasons why I would )oin these organizations ...... -n *□ >□ «□ •□ h. M embers gain much Irom the information they receive Irom these organizations ...... *□ »□ •a i. I think membership in an environmental/conservation organization Is important because many knowledgeable and influential people support these groups ...... «□ >□ •□ j. Belonging to these groups m akes members leel better about what .. ■□ »□ k. If environmental/conservation organizations achieve their goals. my lile and the lives ol the next generation will ben efit ...... >□ *□ •□ I. Belonging to these groups is one ol the most important things >□ >□ «□ •□ 18. Which ol the following actions do you consider are appropriate for solving environm ental problem s? (Check ALL That Apply) oinL efar writing campaigns « □ Publicity raQPetitions ijQ Boycotting viQM arching. picketing, ot sitting in ••□Fundraising o-QLawsuits u Q S tr ik e s « □ Property damage ijQV iolent actions against people (■□Testifying at hearings ••□Education ••□Supporting research □O ther - (Specify):______(■□Lobbying . □ N o n e 19. How would you describe the area where you currently live? (Check ONE Bos) • □ R u ra l i □Suburban i □ U rb a n 20. How would you describe the area where you grew up? (Check ONE Box) i □ R u r a l > □Suburban » Q U r b a n 21. Which ol the following do you consider yoursell? (Check ALL That Apply) i □Environmentalist »□Concerned citizen i □Eco-activist •□Naturalist > □Conservationist nOther _-(Specify):__ • □Educator . □Nona cl tha above Thank youl Please return this questionnaire to me In the enclosed postage-paid envelope. 17411 REFERENCES Ajzen, I., and Fishbein, M. 1980. Understanding attitudes and predicting social behavior. Englewood Cliffs, N. J.: Prentice-Hall, Inc. Alpert, Mark I. 1971. Identification of determinant attributes: A comparison of methods. Journal of Marketing Research. Vol. 8:184-191. Americans for the Environment, Louis Harris, V. Lance Tarrance, and Celinda C. Lake. 1989. The Rising Tide: Public Opinion, Policy and Politics.Unpublished paper. Amis, William D., and Samuel E. Stern. 1974. A critical examination of theory and functions of voluntary associations. Journal of Voluntary Action Research. Vol. 3(4):91-99. Anderson, Walter Truett. 1989. Environmentalist rift: As everyone joins, 'movement' splits into factions. Rocky Mountain News. December 23. p. 53. Andrews, Richard N. L. 1980. Class Politics or Democratic Reform: Environmentalism and American Political Institutions. Natural Resources Journal. Vol. 20:221-241. Babchuk, Nicholas and Alan Booth. 1969. Voluntary Axxociation Membership: A Longitudinal Analysis. American Sociological Review, pp. 31-45. Bartell, T., and A. St. George. 1974. A Trend Analysis of Environmentalists' Organizational Commitment, Tactic Advocacy, and Perceptions of Government. Journal of Voluntary Action Research. Vol. 3:41-46. Becker, Boris W. 1976. Perceived similarities among recreational activities. Journal of Leisure Research. Vol. 8 (2): 112-122. BereJson, Bernard and Gary A. Steiner. 1963. Human behavior: an inventory of scientific findings. New York: Harcourt, Brace and World, Inc. Bergier, Michel J. 1981. A Conceptual Model of Leisure-Time Choice Behavior. Journal of Leisure Research. Vol. 13 (2): 139-158. Berry, J. M. 1977. Lobbying for the people. Princeton: Princeton University Press. Bettman, James R. 1979. An Information Processing Theory of Consumer Choice. Reading, MA: Addison-Wesley Publishing Co. 257 Bettman, James R., and C. Whan Park. 1980. Effects of Prior Knowledge and Experience and phase of the Choice Process on Consumer Decision Processes: A protocol analysis. Journal of Consumer Research. 7(Dec.): 234-248. Biehal, Gabriel, and Dipankar Chakravarti. 1983. Information Accessibility as a Moderator of Consumer Choice. Journal of Consumer Research. Vol. 10: 1-14. Booth, A., and N. Babchuk. 1969. Personal influence networks and voluntary association affiliation. Sociological Inquiry. Vol. 39 (Spring):179-188. Bonomo, Thomas V., and B. P. Shapiro. 1983. Segmenting the Industrial Market. Lexington Books. Brown, Alan and Angus Deaton. 1972. Models of consumer behavior: a survey. T h e Economic Journal. Vol. 82 (Dec.): 1145-1236. Bruner, Gordon C , and Richard J. Pomazal. 1988. Problem recognization: The crucial first stage of the consumer decision process. The Journal of Consumer Marketing. Vol. 5 (1): 53-63. Bugelski, B. 1971. The Definition of the Image in Imagery: Current Cognitive Approaches. N. Y.: Academic Press. Burgess, Philip M., and Richard Conway. 1973. Public Goods and Voluntary Associations. Beverly Hills: Sage. Buttel, Frederick, and William L. Flinn. 1974. The Structure of Support for the Environmental Movement, 1968-1970. Rural Sociology. Vol. 39, No. 1, pp. 56-69. Buttel, Frederick H., William L. Flinn. 1976. Environmental Politics: the Structuring of Partisan and Ideological Cleavages in Mass Environmental Attitudes. T h e Sociological Quarterly. Vol. 17, pp. 477-490. Buttel, Frederick H., and William L. Flinn. 1978. The Politics of Environmental Concern: The Impacts of Party Identification and Political Ideology on Environmental Attitudes. Environment and Behavior. Vol. 10(l):17-36. Calder, Bobby and Robert Bumkrant. 1977. Interpersonal Influence on Consumer Behavior: An Attribution Theory Approach. Journal of Consumer Research. Vol. 4 (July): 29-38. Carson, Rachel. 1962. Silent Spring. Boston: The Riverside Press. Clark, Peter B. and James Q. Wilson. 1961. Incentive systems: A theory of organizations. Administrative Science Quarterly. Vol. 17 (March): 1-25. Coleman, Richard P. 1983. The continuing significance of social class to marketing. Journal of Consumer Research. Vol. 10(Dec.):265-280. Consumer /Industrial Research Service. 1987. Find out how your product is viewed. Medical Marketing & Media. April, pp. 23-29. Cook, V. J., and J. D. Nerniter. 1972. NON MAD, or how consumers behavior. Sloan Management Review. Spring: 77-97. Coombs, D. 1972. The Club Looks at Itself. Sierra Club Bulletin. Vol. 57(7):35-39. Cowley, J., and R. Schreyer. 1986. Visitor preferences for interpretive media at archeological sites. Paper presented at the First National Symposium on Social Science in Resource Management, Corvallis, Oregon, May. Day, George S., Allan D. Shocker, and Rajendra K. Srivastava. 1979. Customer- oriented approaches to identifying product-markets. Journal of Marketing. Vol. 43 (Fall): 8-19. Dennis, Steve and Ervin H. Zube. 1988. Voluntary association membership of outdoor recreationists: An exploratory study. Leisure Sciences. Vol. 10:229-245. Devall, W. 1970. Conservation: An Upper-Middle Class Social Movement: A Replication. Journal of Leisure Research.ol. V 2(2):123-126. D evall, W. 1970. The Governing of a Voluntary Organization: Oligarchy and Democracy in the Sierra Club. Doctoral Dissertation, University of Oregon. Devall, W. 1980. The Deep Ecology Movement. Natural Resources Journal. Vol. 20:299-322. Dickson, Peter R. 1982. Person-situation: Segmentation's missing link. Journal of M arketing. Vol. 46(Fall):56-64. Dillman, D. A. 1978. Mail and Telephone Surveys: The Total Design Method. New York: John Wiley and Sons. Driver, B. L., P. J. Brown. 1978. The opportunity spectrum concept and behavioral information in outdoor recreation supply inventories: A rationale. In, Integrated Inventories of Renewable Natural Resources. USDA Forest Service General Technical Report RM-55. Ft. Collins, Colo.: Rocky Mountain Forest and Range Experiment Station, pp. 24-31. Dunlap, Riley E., and Kent D. Van Liere. 1978. The "New Environmental Paradigm." Journal of Environmental Education, pp. 10-19 Dutsch, M., and R. M. Kraus. 1965. Theories of social psychology. Basic Books. Engel, James F., David T. Kollat, and Roger D. Blackwell. 1970. Research in Consumer Behavior. Holt, Rinehart and Winston, Inc. Engel, James F., D. T. Kollat, R. D. Blackwell. 1973. Consumer Behavior. Holt, Rinehart, and Winston, Inc. Engel, James F., R. D. Blackwell, and P. Minaird. 1991. Consumer Behavior (6th Ed.). Orlando, FL: Dryden Press. 260 Faich, R., and R. Gale. 1971. The Environmental Movement: From Recreation to Politics. Pacific Sociological Review. Vol. 14(3): 270-287. Fishbein, Martin. 1967. Attitude and the prediction of behavior. In, Readings in Attitude Theory. M. Fishbein (ed.). New York: Wiley. Fishbein, M. 1972. Toward an understanding of family planning behaviors. Journal of Applied Psychology. Vol. 2(3):214-227. Fishbein, M. and I. Ajzen. 1975. Belief, Attitude, Intention, and Behavior. Reading, Mass.: Addison-Wesley. Florin, Paul. Eric Jones and Abraham Wandersman. 1986. Black participation in voluntary associations. Journal of Voluntary Action Research. Vol. 15(l):65-86. Fox, Richard J. 1988. Perceptual Mapping Using the Basic Structure Matrix Decomposition. Journal of the Academy of Marketing Science. Vol. 16(l):47-59. Frank , Ronald E., William F. Massy and Yoram Wind. 1972. Market Segmentation. Englewood Cliffs, New Jersey: Prentice-Hall, Inc. Friesema. 1975. Environmental Group Fragmentation and Administrative Decision- M aking. Paper presented at the annual meeting of the American Society for Public Administration, Chicago. Garson, G. D. 1978. Group Theories of Politics. Beverly Hills: Sage. Gensch, Dennis H., and Rajshekhar G. Javalgi. 1987. The influence of involvement on disaggregate attribute choice models. Journal of Consumer Research. Vol. 14 (June): 71-82. Godwin, K. and R. C. Mitchell. 1982. Rational models, collective goods and nonelectoral political behavior. Western Political Quarterly. Vol. 35(2):161- 181. Goman, Barbara. 1991. Characteristics and motives of volunteer facilitators of Project Wild in Ohio. Unpublished Master's Thesis at The Ohio State University. Gordon, C. W. and N. Babchuk. 1959. A typology of voluntary associations. American Sociological Review. Vol. 24(l):22-29. Green, Paul E. and T. W. Morris. 1969. Individual Models in Multidimensional Scaling. The Wharton School, University of Per.nyslvania. Green, Paul E. and Frank J. Carmone. 1970. Multidimensional Scaling and Related Techniques in Marketing Analysis. Allyn and Bacon, Inc.: Boston, Massachusetts. Green, Paul E. 1975. Marketing applications of MDS: assessment and outlook. Journal of Marketing. Vol. 39 (January), pp. 24-31. Green, Paul E.; and V. R. Rao. 1972. Applied multidimensional scaling. New York: Holt, Rinehart, and Winston. Green, Paul E., and Yoram Wind. 1973. Multiattribute decisions in marketing: A measurement approach. Hinsdale, Illinois: Dryden Press. Green, Paul E., Donald S. Tull, and Gerald Albaum. 1988. Research for Marketing Decisions. Englewood Cliffs: New Jersey. Prentice-Hall, Inc. Groliers. 1991. Groliers Electronic Encylopeadia. Grunig, James. 1989. Sierra Club study shows who become activists. Public Relations R eview . Vol. 15(3):3-24. Grunig, James. 1989. A situational theory of environmental issues, publics, and activists. Environmental Activism Revisited. NAEE Monograph V. Grunig, Larissa A. 1989. Activism in the Northeast: Surveying the effects of public relations on conflict resolution. Environmental Activism Revisited. NAEE M onograph V. Hair, Joseph E, Ralph E. Anderson, and Ronald L. Tatham. 1987. Multivariate Data Analysis. New York: NY. Macmillan Publishing Company. Haley, Russel I. 1968. Benefit segmentation: A decision-oriented research tool. Journal of Marketing. Vol. 32 (July): 30-35. Harry, Joseph, Richard Gale, and John Hendee. 1969. Conservation: An Upper- middle class social movement. Journal of Leisure Research, pp. 246-254. Harry, J., R. Gale, and J. Hendee. 1971. Reply to McEvoy: Organized Conservationists are Upper-Middle Class. Journal of Leisure Research. Vol. 3(2):129-131. Hauser, John and Frank S. Koppelman. 1979. Alternative perceptual mapping techniques: Relative accuracy and usefulness. Journal of Marketing Research. Vol. 16(Nov.):495-506. H aw es, Douglass K. 1974. An Exploratory nationwide mail survey of married adults leisure time behavior patterns and the satisfactions derived from leisure-time pursuits. Unpublished Doctoral Dissertation, Ohio State University. Hays, Samuel. 1987. Beauty, Health, and Permanence. Cambridge University Press. Hendee, J,. W. Catton, L. Marlow, and C. Brockman. 1969. Wilderness Users in the Pacific Northwest - Their Characteristics, Values, and Management Preferences. (USDA Forest Service Research Paper PNW-61). Portland, Oregon.: Forest Range Experimental Station, U.S. Forest Service. Hirschman, Elizabeth C., and Morris B. Holbrook. 1982. Hedonic Consumption: Emerging Concepts, Methods, and Propositions. Journal of Marketing.V61. 45:92- 101. Hoesterey, J., and J. Bowman. 1976. The Environmental Message of "Audubon" and the "Sierra Club Bulletin." The Journal of Environmental Education. Vol. 7(3):61-65. Homans, George. 1961. Social Behavior: Its Elementary Forms. New York: Harcourt. Howard, John A. 1963. Marketing Management Analysis and Planning. Homewood, Illinois: Richard D. Irwin, Inc. Howard, John A., and Jagdish N. Sheth. 1969. The Theory of Buyer Behavior. New York: John Wiley and Sons. Humphreys (1982) in Malhotra, N. 1988. Some observations on the state of the art in marketing research. Journal of the Academy of Marketing Science. Vol. 16 (1): 4- 24. Huber, Joel, M. B. Holbrook, and B. Kahn. 1986. Effects of competitive context and of additional information on price sensitivity. Journal of Market Research. Vol. 23(August):250-260. Hyman, H. and C. Wright. 1971. Trends in voluntary association memberships of American adults: Replication based on secondary analysis of national sample surveys. Americau Sociological Review. Vol. 36:191-206. Jaccard, James, David Brinberg, and Lee J. Ackerman. 1986. Assessing Attribute Importance: A Comparison of Six Methods. Journal of Consumer Research. Vol. 12. 463-468. Jacoby, Jacob, R. W. Chestnut, and W. A. Fisher. 1978. A behavioral process approach to information acquisition in nondurable purchasing. Journal of Marketing Research. Vol. 15(Nov.):523-544. Jacoby, A. P. and N. Babchuk. 1963. Instrum ental and expressive voluntary associations. Sociology and Social Research. Vol. 47(4):461-471. Jacoby, Jacob, Donald E. Speller, and Carol A. Kohn. 1974. Brand Choice Behavior as a Function of Information Load. Journal of Marketing Research. Vol. 11:63-69. Jain, Arun K. and Christian Pinson. 1976. The effect of order of presentation of similarity judgements on multidimensional scaling results: An empirical examination. Journal of Marketing Research. Vol. 13 (November): 435-439. Kassarjian, Harold H., and Thomas S. Robertson. 1973. Perspectives in Consumer Behavior. Scott, Foresman and Company: Glenview, Illinois. Klanderman, P.G. 1983. Rotter's I.E. Scale and Socio-Political Action-Taking: The Balance of 20 Years Research. European Journal of Social Psychology. Vol. 15:399-415. Knoke, David. 1981. Commitment and Detachment in Voluntary Associations. American Sociological Review. Vol. 46, pp. 141-158. Knoke, David and David Prensky. 1984. What Relevance Do Organization Theories Have For Voluntary Associations? Social Science Quarterly. Vo* 65. No.l:3-20. Knoke, David and Christine Wright-Isak. 1981. Individual motives and organizational incentive systems, in S. B. Bacharach, ed., Perspectives in Organizational Sociology: Theory and Research. Vol. 1:209-254. Koenig, Daniel J. 1975. Additional Research on Environmental Activism. Environment and Behavior. Vol. 7(4):472-485. Kohl, Daniel H. 1975. The Environmental Movement: What Might it be. N a tu ra l Resources Journal. Vol. 15, pp.327-351. Kollat, David T., Roger D. Blackwell, and James F. Engel. 1970. Research in Consumer Behavior. Holt, Rinehart, and Winston, Inc. Koontz, H., and C. O'Donnell. 1968. Principles of Management: An Analysis of Managerial Functions. New York: McGraw-Hill. Kotler, Philip. 1975. Marketing for Nonprofit Organizations. Englewood Cliffs, New Jersey: Prentice-Hall, Inc. Kotler, Philip. 1991. Marketing Management: Analysis, Planning, Implementation, and Control (7th ed.). Englewood Cliffs, New Jersey: Prentice-Hall, Inc. Lancaster, Kenneth. 1971. Consumer Demand: A New Approach, New York: Columbia University Press. Larson, Mark A., Mary Forrest, and Lloyd Bostian. 1978. Participation in Pro- Environmental Behavior. Journal of Environmental Education, pp. 21-24. Larson, Mark A., Don Zimmerman, and Cliff Scherer. 1982. Communication Behavior by Environmental Activists Compared to Non-Active Persons. Journal of Environmental Education, pp. 11-20. Leonard-Barton, D. 1981. Voluntary Simplicity Lifestyles and Energy Conservation. Journal of Consumer Research. Vol. 8(3): 243-252. Lowi, T. J. 1979. The End of Liberalism: The Second Republic of the United States (2nd ed.). New York: Norton. Malhotra, Naresh K. 1987. Validity and structural reliability of multidimensional scaling. Journal of Marketing Research. Vol. 24 (May):164-173. Malhotra, Naresh K. 1988. Some observations on the state of the art in marketing research. Journal of the Academy of Marketing Science. Vol. 16(l):4-22. Maloney, M., and M. Ward. 1973. Ecology: Let's Hear From the People. Am erican Psychologist. Vol. 28(7):583-586. M anfredo, M. J , 1979. Wilderness experience opportunities and management preferences of three Wyoming wilderness areas. Ft. Collins, Colo.: Colorado St. Univ. Unpublished doctoral dissertation. Manzo, Lynne C. and Neil D. Weinstein. 1987. Behavioral Commitment to Environmental Protection. Environment and Behavior. November, pp. 673-694. Markham, E. M. Horton, and A. McLanahan. 1980. Classes and collections: Principles of organization in the learning of hierarchical relations. Cognition. Vol. 8:227-241. Maslow, A. H. 1943. A Theory of Human Motivations. Psychological Review. July: 370-396. Maslow, A. H. 1954. Motivation and Personality. New York: Harper and Row. Massy, William F., D. B. Montgomery and D. G. Morrison. 1970. Stochastic models of buying behavior. Cambridge: MIT Press. May, Frederick E., and Richard E. Homans. 1977. Evoked Set Size and the Level of Information Processing in Product Comprehension and Choice Criteria. In, Advances in Consumer Research, Vol. 4, ed., W. D. Perrault, Jr., Atlanta: Association for Consumer Research, pp. 172-175. McDonough, M. 1982. The influence of place on recreation behavior: An ecological perspective. In D. W. Lime (ed.), Forest and River Recreation: Research Update (pp. 120-123). St. Paul, Minn.: Minn. Agric. Expt. Stn., Univ. of Minn. Misc. Pubn. McEvoy, J. 1971. A Comm ent: Conservation: An U pper-m iddle Class Social Movement. Journal of Leisure Research. Vol. 3(2):127-128. McFarland, A. S. 1976. Public interest lobbies: Decision making on energy. Washington, D.C.: American Enterprise Institute of Public Policy Research. Milbrath, L. 1975. Environmental Beliefs, Perceptions, and Actions. Journal of Social Relations. Vol. 10(1 ):139-149. M ilbrath, Lester W. 1981. General Report: U.S. Component of a Comparative Study of Environmental Beliefs and Values. Mimeo, Occasional Paper Series, Environmental Studies Center, State University of New York at Buffalo. M ilbrath, Lester W. 1984. Environmentalists: Vanguard for a New Society. Albany, NY: State University of New York Press. Mitchell, Robert C. and J.C. Davies. 1978. The United States Environmental Movement and Its Political Context: An Overview. Mimeo, The Conservation Foundation discussion paper D-32, Washington, D.C. Mitchell, Robert C. 1979. Silent Springs/Solid Majorities. Public Opinion. Vol. 2 (August/September): 16-55. Mitchell, Robert C. 1980. How "Soft," "Deep," or "Left?" Present Constituencies in the Environmental Movement for Certain World Views. Natural Resources Journal. Vol. 20: 345-358. Moe, T. 1980. The organization of interests. Chicago: University of Chicago Press. Mohai, Paul. 1985. Public concern and elite involvement in environmental- conservation issues. Social Science Quarterly. Vol. 66(4): 820-838. Morrison, Denton E. 1980. The Soft, Cutting Edge of Environmentalism: Why and How the Appropriate Technology Notion is Changing the Movement. N a tu ra l Resources Journal. Vol. 20: 275-298. Myers, James H. and m. L. Alpert. 1976. Semantic confusion in attitude research: Salience vs. Importance vs. Determinance, In, Advances in Consumer Research. Proceedings of the Seventh Annual Conference of the Association of Consumer Research. October. 1976. pp. 106-110. Narayana, Chem L., and Rom J. Markin. 1975. Consumer behavior and product performance: An alternative conceptualization. Journal of Marketing. Vol. 39 (October): 1-6. N ational Wildlife Federation. 1990. Conservation Directory. Washington, D.C. Neidell, Lester A. 1972. Procedures for obtaining similarities data. Journal of Marketing Research. Vol. 9 (August): 335-337. Nicosia, Francesco M. 1966. Consumer Decision Process. Englewood Cliffs, N.J.: Prentice-Hall. Norussis, Marija J. 1990. SPSS Base System User's Guide. SPSS, Chicago, 111. Olson, Mancur, Jr. 1965. The Logic of Collective Action. Cambridge: Harvard University Press. Olson, M. 1970. Social and political participation of blacks. American Sociological R eview . Vol. 35:682-697. Olsen, Marvin. 1972. Social participation and voting turnout: a multivariate analysis. American Sociological Review. Vol. 37:317-333. Payne, John W. 1976. Heuristic search processes in decision making. In, Advances in Consumer Research, Vol. 3, ed., Beverley B. Anderson, Ann Arbor: Association for Consumer Research, pp. 321-327. Perry, R., C. Cleveland, D. Gillespie, and R. Lotz. 1975. The Organizational Consequences of Competing Ideologies: Conservationists and Weekenders in the Sierra Club. Annals of Regional Science: A Journal on Urban, Regional and Environmental Research. Vol. 9:14-25. Peterson, G. L. 1974. Evaluating the quality of the wilderness environment: Congruence between perception and aspiration. Environment and Behavior. Vol. 6:169-193. Peterson, N. 1982. Developmental Variables Affecting Environmental Sensitivity in Professional Environmental Educators. Unpublished Masters Thesis, Southern Illinois University at Carbondale. Petty, Richard E., and J. T. Cacioppo. 1981. Attitudes and Persuasion: Classic and Contemporary Approaches. Dubuque, Iowa: W. C. Brown Co. Piaget, J. 1951. Play, Dreams, and Imitation in Childhood. London: Routledge and Kegan Paul Ltd. Reddy, Richard D., and David Horton-Smith. 1973. Who participates in voluntary action? Journal of Extension. Fall:17-23. Ries, Al, and Jack Trout. 1982. Positioning: The Battle for Your Mind. NY: Warner Books. Ritchie, J. R. Brent. 1974. An exploratory analysis of the nature and extent of individual differences in perception. Journal of Marketing Research. Vol. 11 (Feb.): 41-49. Ritchie, J. R. Brent. 1975. On the derivation of leisure activity types — A perceptual mapping approach. Journal of Leisure Research. Vol. 7(2): 128-140. Rogers, E., and F. Shoemaker. 1971. Communication of innovations. New York: The Free Press. Rosch, E. 1978. Principles of categorization. In E. Rosch and B. Lloyd (Ed.), Cognition and Categorization. Hillsdale, N.J.: L. Erlbaum Associates, Inc. Rosenberg, M. J. 1956. Cognitive structure and attitudinal affect. Journal of Abnormal and Social Psychology. Vol. 53.: 367-372. Rothschild, Michael L. 1974. The effects of political advertising on the voting behavior of a low involvement electorate. Unpublished Ph.D. dissertation, Graduate School of Business, Stanford University, Stanford, CA. Russell, Ruth V., and John T. Hultsman. 1988. An empirical basis for determining the multidimensional structure of leisure. Leisure Sciences. Vol. 10: 69-76. Salisbury, R. H. 1975. Research on political participation. American Journal of Political Science. Vol. 19(2):323-341. Scholl, M. 1983. A Survey of Significant Childhood Learning Experiences of Suburban/Urban Environmentalists. Paper presented at the 12th Annual Conference of the North American Association for Environmental Education, Ypsilanti, Michigan. Seventh Generation, (no date) Field Guide to More Than 100 Environmental Groups. Seventh Generation. South Burlington, VT. Shepard, Roger N., A. Kimball Romney, and Sara Beth Nerlove. 1972. Multidimensional scaling: theory and applications in the behaviors sciences. New York: Seminar Press. Shaiko, Ronald G. 1987. Religion, politics, and environmental concern: A powerful mix of passions. Social Science Quarterly. 68(2):244-262. Sia, Archibald P., Harold R. Hungerford, and Audrey N. Tomera. 1984. Selected Predictors of Responsible Environmental Behavior: An Analysis. Journal of Environmental Education, pp. 31-40. Silk, Alvin J. 1973. Preference and Perception Measures in New Product Development, pp. 42-64. In, Kassarjian, Harold H., and Thomas S. Robertson. 1973. Perspectives in Consumer Behavior. Scott, Foresman and Company: Glenview, Illinois. Sills, David L. 1968. "Voluntary associations: sociological aspects," in David Sills (ed.). International Encyclopedia of the Social Sciences. Vol. (16). N ew York: Macmillan and Free Press. Smith, David Horton. 1966. The importance of formal voluntary organizations for society. Sociology and Social Research. Vol. 50(July):483-494. Smith, Constance and Anne Freedman. 1972. Voluntary associations: perspectives on the literature. Cambridge, Massachusetts: Harvard University Press. Snow, Donald. 1992. Inside the Environmental Movement: Meeting the Leadership C hallenge. Washington, D.C.: Island Press. Spiggle, Susan and Murphy A. Sewall. 1987. A choice sets model of retail selection. Journal of Marketing. Vol. 51 (April): 97-111. Srivastava, Rajendra K. 1981. Usage-situational influences on perceptions of product markets: Theoretical and empirical issues. In, Advances in Consumer Research. Vol. 8. Assoc, for Consumer Research, pp. 106-111. Stafford, James. E. 1966. Effects of group influences on consumer brand preferences. Journal of Marketing Research. Vol. 3 (Feb.): 68-75. Stallings, Robert A. 1973. Patterns of Belief in Social Movements: Clarifications from an Analysis of Environmental Groups. The Sociological Quarterly. Vol. 14:465-480. Sudman, Seymour, and Norman M. Bradburn. 1985. Asking Questions. San Francisco: California. Jossey-Bass Publishers. Tanner, R. 1979. Formative Influences in the Lives of Citizen Conservationists. In A. Sacks and C. Davis (Ed.), Current Issues V: The Yearbook of Environmental Education and Environmental Studies (pp. 189-204). Columbus, Ohio: ERIC/SMEAC. Tesh, S. 1984. In support of single-issue politics. Political Science Quarterly. Spring, pp. 27-44. Tillman, A. A. 1973. The Program Book for Recreation Professionals. Los Angeles: National Press Books. Tucker, L. 1978. The Environm entally C oncerned Citizen: Some C orrelates. Environment and Behavior. Vol. 10(3):389-418. Turner, Ralph H. 1970. Determ inants of social m ovement strategies, in Tamotsu Shibutani (ed.) Human Nature and Collective Behavior: Papers in Honor of Herbert Blumer. Englewood Cliffs, N. J.: Prentice-Hall. Turner, Ralph H. and Lewis M. Killian. 1972. Collective Behavior. (2nd ed.). Engelwood Cliffs, N. J.: Prentice-Hall. Truman, D. 1951. The Government Process. New York: Knopf. Tversky, Amos. 1972. Elimination-by-aspects: A theory of choice. Psychological R eview . Vol. 79(July):281-299. Van Liere, Kent D., and Riley E. Dunlap. 1980. The social bases of environmental concern: A review of hypotheses, explanations, and empirical evidence. Public Opinion Quarterly, pp. 181-197. Walsh, E. J. and R. H. Warland. 1983. Social movement involvement in the wake of a nuclear accident: Activists and free riders in the TMI area. Am erican Sociological Review. Vol. 48(6): 764-780. Weigel, Russell, and Joan Weigel. 1978. Environmental Concern: The Development of a Measure. Environment and Behavior. Vol. 10(1):3-15. Williams, Daniel R. 1988. Measuring perceive similarity among outdoor recreation activities: A comparison of visual and verbal stimulus presentations. Leisure Sciences. Vol. 10:153-166. Wilson, J. Q. 1973. Political Organizations. New York: Basic Books. Wind, Yoram. 1976. Preference of relevant others and individual choice models. Journal of Consumer Research. Vol. 3 (1): 50-57. Wind, Yoram. 1978. Issues and advances in segmentation research. Journal of Marketing Research. Vol. 11 (Aug.):317-337. Wind, Yoram and P. J. Robinson. 1972. Product positioning: An application of multidimensional scaling. Attitude Research in Transition, pp. 155-175. Woodside, Arch G. and William H. Motes. 1981. Sensitivities of market segments to separate advertising strategies. Journal of Marketing. Vol. 45 (Winter), pp. 63- 75. Wright, Charles R., and Herbert H. Hyman. 1958 Voluntary Association Memberships of American Adults: Evidence from National Sample Surveys. American Sociological Review, pp. 284-295. Wright, Pamela. 1988. Market segmentation of interpretive participants at National Park Service sites. An unpublished master's thesis from The Ohio State University. Wright, Peter. 1975. Consumer Choice Strategies: Simplifying vs. Optimizing. Journal of Marketing Research. Vol. 12:60-76. Young, Shirley, Leland Ott, and Barbara Feigin. 1978. Some practical considerations in market segmentation. Journal of Marketing Research. Vol. 15 (Aug.): 405-412. Young, Forrest W. and Rostyslaw Lewyckyj. 1979. ALSCAL User's Guide. Data Analysis and Theory Associates. Zimmer, Basil G., and Amos H. Hawley. 1959. The significance of membership in associations. The American Journal of Sociology, pp. 196-201.