UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Toward a unified framework of decision-making: the case of environmentally protective behaviour Axelrod, Lawrence J. 1994

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1994-953071.pdf [ 2.53MB ]
JSON: 831-1.0088024.json
JSON-LD: 831-1.0088024-ld.json
RDF/XML (Pretty): 831-1.0088024-rdf.xml
RDF/JSON: 831-1.0088024-rdf.json
Turtle: 831-1.0088024-turtle.txt
N-Triples: 831-1.0088024-rdf-ntriples.txt
Original Record: 831-1.0088024-source.json
Full Text

Full Text

TOWARD A UNIFIED FRAMEWORK OF DECISION-MAKING:THE CASE OF ENVIRONMENTALLY PROTECTIVE BEHAVIOURbyLAWRENCE J. AXELRODB.A., University of Illinois, 1980M.A, San Francisco State University, 1989A THESIS SUBMITI’ED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYINTHE FACULTY OF GRADUATE STUDIES(Department of Psychology)We accept this thesisto the requiredJuly 1994as conformingstandardTHE UNIVERSITY OF BRITISH COLUMBIALawrence J. Axeirod, 1994In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.Department of P5 yc4of VThe University of British ColumbiaVancouver, CanadaDate M “I;,DE-6 (2/88)ABSTRACTTheoretical determinants of decisions regarding one’s behaviour are integrated intoa comprehensive and inclusive framework. This framework is then employed to determinethe factors responsible for guiding decisions regarding the protection of the naturalenvironment. Included in the framework are constructs from four conceptual domains: (a)attitudinal phenomena, (b) efficacy beliefs, (c) functional motivators of behaviour (i.e.,outcome beliefs) and (d) personal values. In addition, the role of psychological motivatorsin the decision process is reviewed. Several advances in theory are suggested. First,attitudes towards objects (e.g., the natural environment) are differentiated from attitudestowards specific behaviours (e.g., recycling), and both concepts are incorporated into theframework. Second, it is recommended that different functional motives be defined andindependently assessed. In the present research, three domains of outcomes (i.e.,economic, social, and environmental) are specified and their motivational influenceexplored. Third, it is suggested that personal value orientations be included in acomprehensive study of behavioural decisions. A taxonomy of values specifying threedomains (i.e., economic, social, and universal) is proposed and the intluence of personalvalue orientations toward each domain on behavioural decisions is examined. Findingsfrom three experiments suggest that constructs from all four conceptual domains areinvolved in guiding decisions to perform environmentally protective behaviours.Specifically, beliefs regarding behavioural outcomes and efficacy were the most directlyassociated with these behavioural decisions. More positive outcome expectations andstronger beliefs of self-efficacy and behavioural accessibility were associated with decisionsto perform environmentally protective behaviours. Personal values also accounted for asignificant amount of variance in behavioural decisions. In general, economically-orientedsubjects were least likely to choose an environmentally protective course of action, whereasuniversally-oriented subjects were most likely to pursue environmental protection andpreservation. As expected, a persoiial value x outcome belief interaction was found thatiishowed that the decisions of economically-oriented subjects were consistently influenced byeconomic considerations, whereas the decisions of subjects in the other two value groupswere not. Finally, attitudinal phenomena appear to be least important in guidingbehavioural decisions. Theoretical considerations and implications regarding thepromotion of environmentally protective behaviour are discussed.iiiTABLE OF CONTENTSAbstract iiTable of Contents ivList of Tables viList of Figures viiAcknowledgement viiiINTRODUCTION 1PART I A Sociocognitive Framework ofGuided Behaviour: Overview 6Chapter One Attitudes 8Attitude-Behaviour Correspondence 11Attitudes as Predictors of EnvironmentallyProtective Behaviour 13Attitude Strength 15Threat Perception 19Chapter Two Self-efficacy and Behaviour Accessibility 21Efficacy Beliefs and Environmental Behaviour 23Chapter Three Behavioural Motives 26Functional Motives and Environmental Action 28Psychological Motives 32Chapter Four Personal Values 37Description of Prominent Value Theories 38A New Value Taxonomy 42Values and Environmentalism 47Part I Summary 51Part II Investigating the Factors that Guide Decisions RegardingEnvironmental Behaviour: Overview 53Chapter Five Measurement of Independent Factors 54Measuring Attitudinal Factors 54Measuring Value Orientations 55Chapter Six Study I - Determining the Factors that GuideEnvironmentally Protective Behaviour 60Hypotheses 61Method 62Results 66Conclusions 69ivChapter Seven Study 2- A Focus on the Role of Values and EconomicMotives in the Domain of Environmental Behaviour 71Hypotheses 71Method 72Results 73Conclusions 76Chapter Eight Study 3 - Exploring how Values Guide BehaviouralDecisions in Ecological Dilemmas 78Hypotheses 80Method 82Results 85Conclusions 93Chapter Nine Summary and Conclusions 98Theoretical Conclusions and Implications 98lmpl ications regarding promoting environmentallyresponsible behaviour 101References 103Appendix 1 Pre-test questions used to assess attitudinal ëonstructs 130Appendix 2 Revised items used to assess value orientations inpre-test for Studies 2 and 3 131VLIST OF TABLESTable 1 Univariate and partial correlation coefficientsfor the predictor variables regarding intentions toperform the criterion behaviour in Study 1 114Table 2 Mean importance ratings of the three behaviouraloutcomes for the different value orientations 115Table 3 Mean behavioural intention ratings for interactionbetween value orientation and economicoutcome condition in Study 2 116Table 4 Univariate and partial correlation coefficients for thepredictor variables regarding intentions to performthe criterion behaviour in Study 2 117Table 5 Value differences in the percentage of subjects selectingthe environmentally protective course of action inthree ecological dilemmas 118Table 6 Comparisons of the influence of context on behavioraldecisions for each value orientation 119Table 7 Mean importance ratings of select reasons found to beimplicated in guiding subjects’ decisions inthe Commons dilemma 120Table 8 Mean importance ratings of select reasons found to beimplicated in guiding subjects’ decisions inthe Harvest dilemma 121Table 9 Mean importance ratings of select reasons found to beimplicated in guiding subjects’ decisions inthe Waste dilemma 122Table 10 Percentage of subjects ranking different categoriesof motives as “most important” in influencingtheir decisions 123Table 11 Univariate and partial correlation coefficients forindependent factors as predictors of decisions inthe Harvest dilemma 124Table 12 Level of conflict experienced in the different outcomeconditions for each dilemma 125Table 13 Level of conflict experienced when subjects made valueconsistent decisions versus value inconsistent decisions 126viLIST OF FIGURESFigure 1 Sociocognitive framework of guided behaviour 127Figure 2 Description of Prominent Value Structures 128Figure 3 Proposed Value Taxonomy 129viiACKNOWLEDGEMENTSThis thesis could not have been written without the support and guidance of manypeople. I want to thank my advisor, Darrin Lehman, for providing me with theopportunity tO explore issues of personal interest and for his encouragement and inputthroughout the research process. I thank Peter Suedfeld for his counsel and contributions,and for consistently challenging me to think with clarity and purpose about the ideaspresented in this paper. I am grateful to Jennifer Campbell for her helpful adviseregarding various aspects of this research. I greatly appreciate the personal and intellectualsupport I received from my fellow graduate students in the Psychology Department. I amparticularly grateful to Barbara McGregor and Kyle Matsuba for their comments onprevious drafts of this thesis, and to David Mandel and Loraine Lavallee for their helpfulcomments and critiques on important aspects of this work. Special thanks are expressed toAndrew Starzomski who has stimulated, supported and constructively explored many ideaswith me, only some of which are discussed in this thesis. Finally, I want to acknowledgethe numerous researchers who have philosophically and empirically explored thepsychology of attitudes and behaviour regarding important issues of social concern.viii1IntroductionThe explanation and prediction of social behaviour has challenged philosophers andsocial scientists throughout the centuries. Wide variance in how individuals act in response toidentical stimuli has often been observed, even when the individuals originate fromhomogeneous populations. This incongruity can be noticed today in behaviours that haveimplications for pressing social concerns. Societal problems, such as crime, AIDS and otherhealth problems, discrimination, and environmental destruction have individuals concerned,but people’s responses to these dilemmas vary significantly. Even when clear actions areidentified that can prevent problems from escalating, many people are reluctant to alter theirbehaviour accordingly. For example, people continue to smoke cigarettes, engage in unsafesex, drink alcohol and drive, and abuse the natural environment. As our society grapples withthe myriad of social concerns, understanding the factors that guide people’s behaviouraldecisions becomes increasingly important. Although some variance in social issue behaviourcan be accounted for by differences in culture, demographics, and personality characteristics,a great deal is still left unclear.Throughout its history, social psychology has been the science most responsible fordeveloping theories that cogently explain the diversity in social behaviour from an individual,cognitive approach. This thesis continues that tradition. Its broad theoretical objective is toexpand our current understanding of the interplay among factors responsible for guidingdecisions regarding social behaviour. This theoretical goal will be pursued conjointly with aspecific applied concern- to understand more fully what beliefs, desires and values areassociated with behavioural decisions regarding an issue of both public and personal concern:the protection and preservation of the natural environment. Modern interaction with thenatural environment involves many difficult choices, many of which relate to new patterns ofindividual behaviour. The choices made and resulting consequences will likely have dramaticimplications for the functioning of society. These reasons highlight the need for, andsuitability of, psychological investigation of this issue.2In January 1989, Time magazine named ‘The Endangered Earth” as its person of theyear. On its cover, the Earth was graphically represented as a suffocating globe wrapped inplastic and twine. This pronouncement, twenty years after the first Earth Day, seeminglyplaced the eco]ogical crisis at the forefront of public awareness. Finally, the powerful andlasting effect that human behaviour has on our natural environment, and how this effect canjeopardize human life, was being recognized in mainstream society. Since then, discussionsof environmental concerns such as food contamination, air and water pollution, energyproduction and toxic waste, overexploitation of natural resources, and solid waste disposalhave become commonplace in scientific and political circles. The threats posed byenvironmental abuses have fostered the realization that virtually every action that we, asindividuals, perform has some, albeit small, environmental consequence. Taking a shower,watching T.V., driving our cars, heating our homes, throwing out garbage, how we eat, whatwe eat, all the basic activities that are part of our daily existence have an impact upon thenatural environment. Moreover, the collective impact of our way of life is creating manyenvironmental dilemmas with which we must contend.One might expect that this level of public exposure to ecological crises would havemobilized individuals to address these life-threatening predicaments- but has it? How haveindividuals in society responded to these challenges? Are we extinguishing behavioursresponsible for creating ecological problems and adopting behaviours that will amelioratethese problems, sustaining life well into the future? Although concern over environmentalproblems has apparently penetrated the mind of virtually every person (over 90% of thepublic claim to be environmentalists) and certain behavioural changes have been noted (e.g.,increased recycling), major behavioural change has not yet been observed (see, Angus ReidGroup, 1992; Gallup & Newport, 1990). This apparent discrepancy leads to certainquestions. For example, why do some people strongly pursue environmental preservation andprotection, and others do not? In addition, why do many people who report supporting3environmental protection and preservation not appear to act in a manner consistent with thoseattitudes?I began investigating these questions by attempting to determine the relevant socialpsychological theory which could provide the foundation from which the phenomena ofinterest could be studied and explained. This review identified numerous psychologicaltheories, primarily from the social psychological literature, that purported to explain variancein social behaviour. At the same time, I found no single theory, or general framework,adequate to account for the diversity and complexity of factors that influence behaviouraldecisions. Previous research examining determinants of social behaviour, as well as researchspecifically addressing environmental issues, has found limited success in explaining asubstantial portion of variance in social issue behaviour. Numerous theoretical positions havebeen investigated including attitude theories, concepts of behavioural efficacy and self-efficacy, expectancy/value models of motivation as well as theories of psychologicalmotivation, and to some extent human value systems. Although evidence supporting the rolethese psychological phenomena play in guiding behaviour has been noted, clarifying thecomplex interplay among the numerous behavioural determinants remains a significantchallenge.A number of prominent researchers have recently noted with dissatisfaction theabsence of “synthesis” in social psychological research (Aronson, 1992; Berkowitz & Devine,1989; Eagly, 1992). Aronson (1992) stated that the imbalance in favor of analysis oversynthesis has led to “a plethora of small theories with hardly anyone taking the trouble to tryto find the common ground among these theories (p. 309).” Berkowitz and Devine (1989)pointed out that the lack of synthesis is costly in that synthesis offers advantages in terms ofeconomy of thought and cohesion among different approaches. In the realm of attitudetheory, where debate continues regarding the operational definition and predictive utility ofthe attitude construct, Eagly (1992) noted that “a more comprehensive understanding ofattitude-behavior relations could emerge from joining ... traditions - that is, from formulating4causal sequences that take into account both attitudes toward behaviours and attitudes towardthe entities or targets toward which behaviors are directed” (p. 696). She also pointed outthat numerous moderator variables have for the most part not been integrated into explicittheories of attitude-behavior relations.Therefore, it was determined that synthesizing current theory into one unifiedframework would be a vital step in the research process. Such a framework could then beemployed to analyze the specific theoretical and applied issues of concern. This thesis isdivided into two parts corresponding with these research steps. Part I confronts the challengeof reviewing and synthesizing the vast psychological literature that attempts to explain andpredict social behaviour, integrating it into a single comprehensive framework. Specifically,in Part I, various factors found to be implicated in guiding behaviour are identified andobservations regarding the explanatory utility of each factor are presented. The relevantpsychological issues include: (a) attitude theory (Chapter 1), (b) theories of efficacy and self-efficacy (Chapter 2), (c) expectancy-value theories and other salient models of motivation(Chapter 3), and (d) theories of human value systems (Chapter 4). In these chapters Ielaborate upon the conceptualization of each theory, discuss the major research that hasexamined the relevant constructs embedded in them, and explore the utility of these constructsin explaining environmentally-relevant behaviour.Part II of this thesis describes the procedures and results of attempts to empiricallyinvestigate the phenomena under study. Three studies were conducted to explore both thetheoretical and applied questions. For each study, the rationale that guided its development isenumerated and specific hypotheses pertaining to the objectives of the study are presented.Prior to the description of the specific studies, methodologies and hypotheses common to allthree are presented (Chapter 5). Next, Study 1 is detailed, in which the influence that eachfactor noted in the framework has on a subject’s intentions to perform a specificenvironmentally protective action was investigated (Chapter 6). Study 2, which addressedunanswered questions from Study I is reported next, and specific conclusions based on these5findings are discussed (Chapter 7). These two studies revealed that decisions regardingenvironmentally-related behaviour are based on the complex array of factors presented in theframework. One notable result was the emergent role of human values in this decisionprocess. Consequently, Study 3 was designed to explore the specific influence of humanvalues on environmentally-related behavioural decisions, with particular emphasis on decision-making in circumstances of ecological-personal conflict (Chapter 8). The thesis concludeswith a discussion of how the proposed framework can contribute to the investigation of issue-oriented social behaviour (Chapter 9). In addition, implications regarding environmentally-related behaviour and the human decision process are discussed.It should be noted that in each of these studies, the criterion phenomena of focus is notactual behaviour, but its most closely related construct, behavioural intentions (Ajzen &Fishbein, 1980). Although understanding and explaining actual social behaviour in anaturalistic setting is the ultimate goal of behavioural research, it is often difficult, if notimpossible, to collect behavioural data in a context that allows for flexibility in the selectionof a criterion and control over the experimental setting. Therefore, researchers have oftenstudied the factors that influence behaviour choice rather than actual behaviour. It is assumedthat most social behaviours are under volitional control, and, therefore, behavioural intentionsare considered the direct antecedent of behaviour (Ajzen & Fishbein, 1980). Thus, explainingand predicting behavioural intentions and choice becomes a vital step in understanding actualbehaviour.6Part IA Sociocognitive Framework of Social Issue BehaviourOverviewThe framework presented in Figure 1 is the result of a comprehensive synthesis of thenumerous theoretical conceptualizations used to explain social behaviour and behaviouralchoice. Although there are distinct affective components inherent in the behavioural process,the framework is primarily concerned with delineating the beliefs, motives, and values thathave been identified by past social psychological research as implicated in the behaviouraldecision process. The development of the framework involved three steps. First, I reviewedrelevant social psychological theory and, where necessary, dissected each theory into itsindividual components. The most common operationalization of each component wasidentified in order to search for overlap among constructs.Once the essential components of each theory were identified, I found it necessary toreassess the terminology associated with certain concepts so that they would more accuratelyrepresent the exact nature of the construct. For example, the concept of “attitude” has beenused, as well as measured, in quite diverse ways. Some researchers assess attitudes in termsof affective evaluations of general objects or events, whereas others consider them useful onlyas they relate to specific, well-defined behaviours. To achieve a cohesive and understandablefinal model, each component of the framework needed to be distinguishable from all others.Substantial overlap among theories, and the use of common terminology does exist.Identifying the unique quality of each theoretical concept was a significant challenge. Thisprocess necessitated the selection of a single interpretation for each construct. In doing so, Irelied on the way that a construct had been consistently measured in previous research, ratherthan how it had been generally defined.The third step involved recognizing constructs that have been previously neglected inthis area of study but could reasonably be construed as playing an important role. The mostnoteworthy of these factors is personal values. Whereas the importance of value systems in7influencing human activity is well-noted (Rokeach, 1973), rarely has their role in this processbeen investigated in terms of how they relate to other constructs (see, Homer & Kahie, 1988for exception). I argue that the limited empirical attention given to values has beendeleterious to psychology’s attempt to develop a full understanding of the behavioural decisionprocess regarding social issues.The genesis of the framework originated in the specification and differentiation ofthree distinct domains of cognitive factors (2nd column in Figure 1). These are representedby the following statements: (a) I beJieve, therefore I act; (b) I can, therefore I act; (c) Idesire, therefore I act. The first statement refers to the concept of attitudes as guides tobehavioural choice. Although attitudes are presently construed as having an affective as wellas a cognitive component, I suggest that the phrase ‘I believe’ is a reasonable representationof the concept of attitude. This conclusion is based on the fact that attitudes are often inferredfrom an individual’s response to a series of belief statements (Ajzen & Fishbein, 1980). Thesecond phrase reflects the notion of self-efficacy or personal control as a determinant ofbehavioural intentions. The third statement identifies the motivational force provided by theneed or desire to attain certain outcomes from one’s actions. Within each domain, variouscomponents are posited in order to represent the diversity and complexity of each cognitiveguide.8Chapter 1AttitudesAttitudes are believed to play a vital role in individuals’ assessments of their world(Pratkanis & Greenwald, 1989). Although attitudes are recognized as a core construct insocial psychology, the exact nature of their function is still debated. Early theories ofattitudes specified numerous overlapping functions which attitudes serve (see Shavitt, 1989,for a review). These include a knowledge function (e.g., Katz, 1960; Katz & Stotland,1959), that is, attitudes help the individual organize and structure the environment andprovide consistency in one’s frame of reference. In addition, an attitude can guide behaviourby summarizing the outcomes associated with an object suggesting the path that maximizesrewards and minimizes punishments.Accordingly, Smith, Bruner, and White (1956) proposed an object-appraisal functionwhich presumes that attitudes serve as guidelines for classifying objects and structuring theenvironment to make responses available that maximize one’s own interests. They posited asocial adjustment function which suggests that attitudes help facilitate self-expression andsocial interaction. Attitudes also play a major role in maintaining self-esteem, a function thatSmith et al. called “externalization” and Katz termed “ego-defense.”Pratkanis and Greenwald (1989), in their recently developed sociocognitive model ofattitudes, integrated the early functional theories and concluded that attitudes serve threefunctions: (1) they provide a simple strategy for appraising an object, (2) they organize andguide complex behaviour, and (3) they define and maintain self-worth. A commonperspective among these theorists (e.g., Katz, 1960; Pratkanis and Greenwald, 1989; Smith etal, 1956) is that attitudes function as a guide to behavioural decisions.While Pratkanis and Greenwald’s model proposes that attitudes serve as causal agentsregarding behaviour, there is also support for viewing the causal path in the reverse direction.In fact, the behaviour-to-attitude link, derived from both Festinger’s (1957) cognitivedissonance theory and Bern’s (1972) self-perception theory, has been demonstrated to be quite9strong (e.g., Festinger & Carlsmith, 1959; Zanna, Olson, & Fazio, 1981). Additiona1support for this causal path is found in research on the effects of direct experience on attitudestrength and behaviour (Fazio & Zanna, 1981). There is little doubt that past behaviour caninfluence and/or change attitudes, but this conclusion does not conflict with the position thatattitudes can also function as a guide to behaviour. Thus, the causal path appears reciprocalin that attitudes function as a guide for future behaviour, and in turn may be influenced byany previous direct experience with the attitude object. This chapter is concerned primarilywith understanding the path going from attitudes to behaviour.In line with this conceptualization of attitude, I adhere to the definition of “attitude” asa general evaluation of an object, event, or issue- a combined affective and cognitiveappraisal that is best inferred from responses to general belief statements regarding thatobject. This definition, and approach to measurement, stands in contrast to that offered inFishbein and Ajzen’s prominent “Theory of Reasoned Action” (Fishbein & Ajzen, 1975) andits updated version, the “Theory of Planned Behaviour” (Ajzen, 1985). Their theory suggeststhat in order to understand and predict behaviour, the most salient attitude object is the actualtarget behaviour, and not the more general object or issue under study. Attitudes towardsmore global objects or issues which may be associated with the behaviour are several stepsremoved in the causal chain and are considered to have minimal utility in predictingbehaviour. For example, the association between attitudes toward capital punishment (pro orcon) and a corresponding behaviour (volunteering for a group lobbying for your position)would likely be minimal while significant correspondence would exist between attitudestoward the behaviour (i.e., pro or con regarding volunteering for the group) and thebehaviour itself. I argue that in order to fully understand complex patterns of socialbehaviour, the general attitude construct should be part of a comprehensive theoreticalframework. The relation between general attitudes and attitudes towards particular behaviours(this construct will be discussed in detail later) could then be examined.10Fishbein and Ajzen’s attitude concept is also considered a function of beliefs, but a setof beliefs quite different from the broader interpretation I have just noted. They postulate thata person who believes that performing a given behaviour will lead to a “positive” outcomewill hold a favorable attitude toward the behaviour, while a person who believes thatperforming the behaviour will lead to mostly negative outcomes will embrace an unfavorableevaluation of the behaviour. They specifically assess a person’s attitude toward a behaviourin terms of expectations regarding the accrual of various outcomes and the value placed onthose outcomes. One’s general attitude toward a behaviour is inferred from balancing thepositive and negative expected outcomes. To illustrate, a person who believes that recyclingwill help the environment, will save money, and will make the family happy (all positiveoutcomes) will likely hold a positive attitude toward recycling. A person who believes thatrecycling doesn’t really help the environment, has costs in terms of money and time, anddiverts attention away from real problems will likely embrace a negative attitude towardrecycling.Although a number of studies provide empirical support for Fishbein and Ajzen’sconceptual model (see Ajzen & Fishbein, 1973), one major concern with the “attitude-towardact” measure is its limited application (Schwartz & Tessler, 1972). Whereas the attitude-toward-act measure improves prediction of a given behaviour, it diminishes the strength of theoverall attitude concept. Within the present program of research, I am just as interested in therole that attitudes toward issues, objects, or events (i.e., the environment and environmentalpreservation) play in guiding behaviour, as in more specific beliefs regarding one particularaction.This is a case where clarification of terminology is required. In an attempt to resolvethis definitional quagmire, I thought it most reasonable to conceive of “attitudes” as thegeneral evaluation most current theorists construe it to be, and to deflne Fishbein and Ajzen’s“attitude” construct in terms of its two core components- outcome expectancy and outcome1]value. I will return to this issue when I discuss motivational and expectancy/value theories ofbehaviour.Attitude-behaviour correspondenceThe proposed framework identifies two general aspects of attitudes that prior work hassuggested are necessary for a complete understanding of how attitudes influence action. First,as previously noted, it is proposed that an individual’s general attitude toward a specific issue,object, or event predisposes that individual toward acting in a certain manner (Pratkanis &Greenwald, 1989). Thus, attitudes serve a heuristic purpose regarding behavioural decisions.Attitudes provide a cognitive base upon which people make decisions. A person who “likes”strawberry jam, is more likely to buy strawberry jam than someone who dislikes that flavor.The attitude “I support environmental protection,” in theory, would likely induce one toengage in specific environmentally-protective behaviours, such as recycling.Although this conceptualization of attitude is widely held, the connection betweenattitudes and behaviour has frequently been challenged. The study that kindled much of thiscritical view of the attitude construct was conducted by sociologist Richard LaPiere (1934).In his well-known study, LaPiere spenttwo years investigating the relation betweenprejudicial attitudes and acts of racial discrimination. He traveled the U.S. with a youngcouple from China visiting 251 eating and housing establishments. Only once did LaPiereobserve an act of discrimination, even though prejudice against Chinese people was prevalentat that time. In a follow-up questionnaire sent to the original establishments, over 90% of the128 replies reported that they would not accept members of the Chinese race in theirestablishment. These findings have been considered questionable (Dillehay, 1973) due to thepossibility that the people who provided the behavioural data in LaPiere’s field study may nothave been the same subjects who responded to the attitude questionnaire. Nevertheless, it hasbeen noted that these findings suggest that “people can hold abstract opinions which have littleor nothing to do with their actual behaviour” (PIous, 1993, p. 59).12In 1969, Wicker fueled an anti-attitude discourse with his review of 46 studies inwhich attitude-behaviour correspondence was explored. For his review, Wicker selected onlythose studies that measured attitudes and their expected corresponding behaviour at differenttimes, suggesting this was the only way to truly test attitude-behaviour relations. Heconcluded that attitudes are likely to be unrelated or only slightly related to actual behaviours.This conclusion was based on the finding that in relatively few cases was a correlation of over.30 observed, and that traditional attitude toward object measures rarely accounted for asmuch as 10% of variance in overt behaviour. Later, he went so far as to suggest that it mightbe prudent to abandon the attitude concept as a meaningful psychological construct.Fortunately, very few researchers heeded Wicker’s suggestion. Instead, attituderesearch began to focus on the reasons why attitude-behaviour discrepancies were frequentlyfound. One approach was to examine the methodology used in the assessment of bothattitudes and behaviours. When the studies cited in Wicker’s review were reanalyzed, hisconclusions were challenged due to numerous methodological concerns (e.g., Dillehay, 1973;Weigel & Newman, 1976). Weigel and Newman (1976) attempted to address several of theseconcerns in their study of attitude-behaviour relations. They found evidence that generalattitudes can accurately predict behaviour, but only when they are employed to predict ageneral pattern of behaviour and not a specific action. They administered a Likert styleattitude measure about the environment to 44 residents of a New England town. Follow-upbehaviour measures were collected after approximately twelve, eighteen, and twenty-sixweeks. The general attitude measure predicted the subjects’ general pattern of behaviourtoward the environment quite well (r = .62), but it was not a good predictor of any singleaction. Ajzen and Fishbein (1977), in their review of research examining attitude-behaviourcorrespondence, reported additional evidence that strong attitude-behaviour relations can beobtained when there is high correspondence between the attitude and behaviour entities (eithergeneral to general or specific to specific), especially between the target and the actionelements. Since that time, substantial research has focused on defining potential mediators13and/or moderators of attitude-behaviour correspondence. The results of this research will bediscussed later in this chapter and in Chapters 2 through 4.Although the ernpiricai relation between attitudes towards issues and correspondingbehaviour is at best uncertain, the theoretical connection between them is too strong todismiss. Thus, I support the position that attitudes predispose individuals to act in certainways. Often, though, other factors may be more directly involved in guiding one’sbehaviour, thereby limiting the influence that attitudes may have. The exact place for thisconstruct in relation to the other possible factors remains to be confirmed. One goal of thecurrent research was to examine these relations in the realm of environmental behaviour.Attitudes as predictors of environmentally-protective behaviourThere have been numerous studies investigating the relation between attitudes andenvironmental behaviour. Hines, Hungerford, and Tornera (1987) conducted a meta-analysisof environmental behaviour research to determine which variables appear to be mostinfluential in guiding individuals to engage in environmentally-protective behaviour. Theattitude-behaviour relation was a central feature of this review. For the purposes of thisanalysis, attitudes were defined as those factors that ‘dealt with the individual’s feelings, proor con, favorable or unfavorable, with regard to particular aspects of the environment orobjects related to the environment” (p. 4). This broad factor included assessments of generalattitude toward the environment as well as attitudes toward specific environmental actions.Results from the analysis of fifty-one measures found a corrected correlation between attitudesand behaviour of .35. Thus, it was shown that individuals with more positive attitudes weremore likely to report engaging in environmentally-protective behaviour.Hines et al. ‘s analysis attempted to address the distinction between general attitudesand attitudes regarding specific actions. They found that there were two types of attitudesbeing assessed by researchers: (a) attitudes toward the environment as an issue, and (b)attitudes towards taking environmental action (e.g., attitudes toward recycling, conservingenergy, etc.). Although they found a slightly stronger relation between attitude toward14actions and environmental behaviour (j = .38), the relation between general attitude towardthe environment and environmental behaviour was quite strong (r = .34). Thus, theyconcluded that “both of these types of attitudes were related to behaviour in an environmentalcontext” (p. 4). It is interesting to note that somewhat higher attitude-behaviour correlationswere found in situations where actual behaviours were measured (r .43) than were found instudies where behaviours were measured by self-reports (r = .33). One can conclude fromthis finding that both methods of measuring behaviour are valid, and that self-reports mayactually deflate, rather than inflate attitude-behaviour correspondence as some critics suggest(e.g., Wicker, 1969).Other studies have failed to find a significant relation between general attitudes andspecific environmental behaviour. As previously mentioned, Weigel and Newman (1976)found that subjects’ general attitude toward the environment did significantly predict whetheror not the subject engaged in any of three actions (i.e., signing a petition, roadside litterdisposal, recycling), but had little value in predicting the enactment of one specific action.Oskamp et al. (1991), in their investigation of household recycling behaviour, found similarresults: general proecology attitudes failed to discriminate between recyclers andnonrecyclers. In addition, various studies have found little or no relation between attitudestoward the environment and consumer-oriented environmental behaviour, such as installinghome insulation and purchasing environmentally-friendly products (Ba]derjahn, 1988).It is relatively clear that attitudes toward issues function as guides to behaviour, but itwould be most accurate to describe this relation as indirect and inconsistent. In fact, moststudies show that other behavioural determinants (e.g., efficacy beliefs, economic costs) eithermediate or moderate the attitude-behaviour relation. For example, people with a positiveattitude toward the environment appear to be willing to change their purchasing habits (buyenvironmentally-friendly brands) to protect the environment. However, this behaviour isblocked if the prices for environmentally-friendly brands are higher than other brands(Henion, 1972), leading to a reduced intluence of the attitude.15In an attempt to explore this process more comprehensively, Axeirod and Lehman(1993) measured general attitudes toward the environment as well as various other knownbehavioural determinants in both student and community samples. They also collected self-reports regarding the performance of 22 environmentally-protective behaviours. Thesebehaviours included political (e.g., donating money or time to environmental organizations)as well as household behaviours (e.g., recycling, buying environmentally-friendly products).They found that general attitudes were highly associated with an index of total behaviour (r =.45), but that this relation became nonsignificant (partial = 11) after partialling out thevariance accounted for by other factors (e.g., self-efficacy beliefs and outcome expectancies)in a multiple regression analysis.In sum, attitudes towards issues can be important guides to behavioural decisions. Asignificant association is likely to exist when a general attitude is related to a similarly generalcourse of behaviour. The relation is weakened when the behaviour of interest is restricted toone specific action. As well, when other guides to behaviour are included in a multivariateanalysis, the relation between attitudes and behaviour may be attenuated. Thus, although it ispresumed that peoples attitudes toward the environment provide a guide to decisionsregarding environmental behaviour, the directness of this relation is questionable. Althoughsome clarity is emerging regarding the role that general attitudes play in guiding social issuebehaviour, further exploration is needed to isolate other salient guides to behaviour andincorporate them into a comprehensive behavioural framework. The rest of this chapter andChapters 2 and 3 describe the factors posited to be more direct determinants of socialbehaviour, and therefore believed to mediate and/or moderate the attitude-behaviour link.Attitude StrengthOne of these factors is the second component described in the attitude domain of theproposed framework. Based upon research by Fazio (e.g. Fazio & Zanna, 1978), Krosnick(1989), and others (see Raden, 1985, for a review), it is suggested that the strength of one’sattitude may have a direct effect on behaviour, but more likely it either mediates or moderates16the attitude-behaviour relation. For example, the stronger one’s favorable attitude towardenvironmental protection, the more likely he or she will act in an environmentally protectivemanner. A number of dimensions related to attitude strength have been reported to influencetheir predictive utility. Attitude strength has been conceptualized in terms of several cognitivedimensions (see Krosnick, Boninger, Chuang, Berent, & Carnot, 1993 for a review). Theseinclude: (a) the extremity of the evaluation, (b) the certainty or confidence with which theattitude is held, (c) the accessibility of the attitude, (d) the amount of direct experience anindividual has had with the attitude object, (e) the extent to which the individual has a vestedinterest in the attitude, and (f the level of importance placed on the attitude object by theindividual.Extremity refers to the extent to which an individual’s evaluation of an attitude objectdeviates from the midpoint on an attitude scale (e.g., Judd & Johnson, 1981). Extremity ofattitudes is rated along scales such as favorable-unfavorable and strongly agree-stronglydisagree.Certainty refers to the confidence held by an individual that his or her attitude towardan object is correct (Krosnick et al., 1993). Fazio and Zanna (1978) provided evidence thatcertainty can moderate the attitude-behaviour relation. For persons expressing high certaintyregarding a specific attitude, significant attitude-behaviour correlations were found, whereasfor individuals expressing low certainty no relation was found.Several theorists have proposed that the accessibility of an attitude is a measure ofattitude strength (Fazio & Williams, 1986; Fazio, 1989; Krosnick, 1989; Sherman, 1987).Attitude accessibility refers to the strength of an object-evaluation link in memory. It hasbeen measured by the length of time it takes subjects to report their attitudes. Fazio (1989)conceptualized attitude strength as a continuum ranging from nonattitudes to highly accessibleattitudes. Strength of attitude is defined by the object-evaluation link in memory and ismeasured by the speed and ease with which the attitude can be accessed from memory(Krosnick et al., 1993). As with certainty, strong attitude-behaviour correspondence is found17for highly accessible attitudes, while little or no relation is observed for attitudes that aredifficult to access (Fazio, Chen, McDonel, & Sherman, 1982; Fazio & Williams, 1986).Kaligren and Wood (1986) found that attitude accessibility influenced attitude-behaviourconsistency in the domain of environmental activism. Specifically, subjects with relativelyhigh levels of access to their attitudes regarding environmental preservation (mostlysupportive) were likely to act in a manner consistent with their attitudes. Subjects with littleaccess demonstrated low attitude-behaviour consistency.Regan and Fazio (1977) have argued that attitudes can be distinguished by the mannerin which they were formed. Some attitudes are formed through direct behavioural experience(e.g., having a crime committed against oneself may influence attitudes about crimeprevention) while others are based on accumulating information through indirect, ornonbehavioural means (e.g., news reports of crime). Their research, along with that of Fazioand Zanna (1978), shows that attitude-behaviour consistency is higher for attitudes based ondirect experience than attitudes formed in other ways. Attitudes formed through directexperience have been found to be more accessible (Fazio et al., 1982) and held with greaterconfidence (certainty) than attitudes formed through other means. Since these dimensionshave been found to be interrelated (see Raclen, 1985), it is not known whether greater attitude-behaviour correspondence is a direct effect of one of these specific strength dimensions ormediated through a central dimension.Sivacek and Crano (1982) have argued that when an individual has no vested interestwith respect to the attitude object or the behaviours suggested by it, the attitude-behaviour linkshould be weak. Their research revealed that high vested-interest subjects had higher attitudebehaviour correlations than subjects in low or moderate vested-interest groups.A number of studies (e.g. Granberg, 1985; Snyder, 1982) have shown the significantinfluence of the importance (salience) of the attitude object on the ability of the attitude topredict relevant behaviour. Attitude importance has been defined as the extent to anindividual cares about the attitude object. It is usually operationalized by self-reports of the18importance ascribed to the attitude object (Krosnick, Boninger, Chuang, Berent, & Carnot,1993). Research has shown that people are more likely to act on issues that are personallyimportant to then then they are to act on issues with which they are not personally connected(e.g., Krosnick, 1989, 1990; Sherif, 1980; Sivacek & Crano, 1982).In their multivariate study of the factors that guide environmental behaviour, Axeirodand Lehman (1993) included a measure of issue importance. They found that subjects’assessment of the importance of the environment as a social issue strongly predicted theirperformance of environmentally-protective behaviours (r = .55). In addition, the predictiveutility of the issue importance construct remained significant when analyzed simultaneouslywith eight other behavioural determinants (partial r = .26). As previously noted, this was notthe case for genera] attitude, whose strong univariate correlation with self-reported behaviourwas reduced to non-significance in the multivariate model.Thus, the exteiit to which an attitude object is perceived as more or less important, aswell as the broader notion of attitude strength, has been shown to influence the attitude-behaviour relation. It should be noted that although the described aspects of attitude strengthare typically thought to reflect underlying dimensions of one overall concept, it has recentlybeen found that these dimensions reflect distinct, but related constructs (Krosnick et aL,1993). Krosnick et al. (1993) conclude that “it seems reasonable to use the term attitudestrength as a shorthand way of saying that some attitudes are stable and consequential andothers are not, or as a term for describing the group of constructs that differentiate strongattitudes from weak ones (p. 1143). They go on to suggest that each construct should beindependently assessed and that “researchers should stick to terminology and conceptualapproaches that are closer to the operationalizations they use in a particular study” (p. 1143).The present framework suggests that the construct of attitude strength may be an importantfactor in driving a person’s decision to act, and should be investigated in some manner.19Threat PerceptionAnother factor that may be implicated in the behavioural decision process is beliefstoward other objects or issues that may bear on the decision. In fact, variance in behaviourwithin certain domains may be more directly accounted for by alternative beliefs about anissue rather than ones general evaluation regarding that issue. For example, to predict jampurchase behaviour, it would be prudent to assess attitudes towards a series of objects, notsimply the flavor of the jam. For instance, the brand name and the cost of the product couldbe important evaluations to consider. Thus, a positive attitude toward strawberry jam couldbe overshadowed by a stronger incongruent attitude toward the brand or the package style anddesign of the product.One specitc evaluation that has been found to be salient when investigating issuesrelated to health protection and/or lifestyle preservation is beliefs pertaining to the level ofrisk posed by a potential threat. This construct has been found to provide added explanatoryutility when predicting protective behaviour (e.g., Axelrod & Newton, 1991; Becker, 1974;Paterson & Neufeld, 1987; Rogers, 1975; Wolfe, Gregory, & Stephan, 1986). Threatperception is currently viewed as a composite of beliefs regarding the: (a) likelihood of theevent in question; (b) the appraised severity of the event; and (c) the immediate nature of thethreat (Paterson & Neufeld, 1987). A number of behavioural models emphasize threatperception factors as important predictors of protective behaviour (e.g. Health Belief Model,Becker, 1974; Protection Motivation Theory (PMT), Maddiix & Rogers, 1983). It isassumed that high perceptions of risk should lead individuals to perform actions intended toreduce or eliminate the risk.Wolf et al., (1986) and Axeirod and Newton (1991) applied PMT to predict intentionsto engage in efforts to prevent nuclear war. Both studies found a greater readiness to engagein protective behaviours (e.g., donations to peace organizations) when nuclear war wasconsidered likely and its effects judged as severe. Axelrod and Lehman (1993) found similarresults in the domain of environmental behaviour. Combining perceptions of the likelihood,20severity, and immediacy of risk posed by environmental destruction into one scale termed“threat perception,” these researchers found that this construct explained a significant portionof variance in the reporting of environmentally-protective behaviour within a 9 factormultivariate model.SummaryThe attitude literature is fraught with inconsistencies regarding the function of attitudesand, more specifically, the influence attitudes have on behaviour. Much of this confusionstems from how one defines the concept of attitude. Some researchers believe that attitudesare only useful in guiding behaviour when they portray evaluative beliefs regarding a specificaction. Others believe that this interpretation is too restrictive, and that much can be learnedabout social behaviour through the assessment of attitudes toward the issues/objects ofinterest. In the proposed framework, I address this conundrum by separating attitudes towardissues from attitudes towards actions (to be discussed in terms of outcome expectancies andvalues), as other researchers have recommended (e.g., Hines et a!. 1987). In addition, theframework extends this attitudinal domain to reflect the importance of other salient constructs,including attitude strength and evaluations regarding other relevant issue/objects - in this casethe potential risk posed by environmental harm. Both theoretical and applied research suggestthat general attitudes towards issues do serve as guides to behaviour, but that this relation isusually indirect and is observable only when relating them to an equally general assessment ofbehaviour (e.g., Weigel & Newman, 1976). Other attitudinal variables, such as attitudestrength and threat perception, may actually influence behavioural decisions more directly, butthe nature of this interactive process is not yet clear.21Chapter 2Self-efficacy and behaviour accessibilityWhereas individuals may hold strong attitudes on an issue of personal importance, theymay not possess (or believe they possess) the abilities necessary to act consistent with theirattitudes. In fact, an individual may not believe that any corresponding actions are available.People’s beliefs about what they can do, as well as their perceptions regarding the ease withwhich an action can be performed, are viewed as important determinants of what they will do.Most social-behavioural theories now either center on or include the construct of self-efficacyor personal control in their model (e.g., Self-efficacy Theory, Theory of Planned Behaviour,Protection Motivation Theory). Work by Ajzen (Ajzen & Madden, 1986; Ajzen & Timko,1986), Bandura (see Bandura, 1986, for a review), and Maddux (Maddux, Norton, &Stoltenberg, 1986), and their associates has clearly established that changes in self-efficacybeliefs are linked with behavioural change, and that self-efficacy is a strong predictor ofbehaviour. In addition, experimental research has supported the importance of self-efficacy asa determinant of behaviour (Bandura, Reese, & Adams, 1982; Davis & Yates, 1982; Maddux& Rogers, 1983).For example, Bandura et al. (1982) examined the influence of self-efficacy on copingbehaviour and fear arousal. In two experiments, they initially manipulated self-efficacy intodifferent levels in phobic subjects. In a second phase, they successfully raised beliefs of self-efficacy to designated levels within the same subjects. After each manipulation, measures ofcoping behaviour were collected. They found that coping behaviour correspondedsignificantly to self-perceptions of efficacy, such that higher levels of perceived self-efficacywere associated with increased performance. This relation was found across various modes ofmanipulation, different types of behaviour, and in both intragroup and intrasubjectcomparisons.Although the concept of efficacy is well established, different components ofefficacious beliefs have yet to be clearly defined. The basis of the efficacy concept stems22from the notion that self-referent cognitions mediate the relation between knowledge andaction (Bandura, 1986). It is suggested that how people judge their own capabilities willaffect their motivation and behaviour. As Bandura states, ‘people tend to avoid tasks andsituations they believe to exceed their capabilities, but they undertake and perform assuredlyactivities they judge themselves capable of handling” (p. 393).The present framework divides efficacious beliefs into two constructs- self-efficacyand channel efficacy. The definition of self-efficacy suggested here was initially set forth byBandura (1986). He defined perceived self-efficacy as “a judgment of one’s capability toaccomplish a certain level of performance” (p. 391). In other words, it refers to theexpectation that one is able to engage competently in the specific action of concern, ratherthan one’s ability to achieve a certain outcome. The latter concept, often referred to as eitherresponse efficacy or response-outcome expectations, refers to a judgment regarding the likelyconsequence of a behaviour. For example, the belief that one conserve energy is an self-efficacy judgment, whereas the anticipated economic savings, social rewards, orenvironmental benefit derived from such an action would be considered outcome expectations.The former conceptualization of self-efficacy identifies it as an independent construct,distinguishable from outcome expectancies. I suggest that Bandura’s outcome expectancyconstruct parallels Fishbein and Ajzen’s behavioural belief concept, both representing theinfluence that expected behavioural consequences have on behavioural choice. I have placedoutcome expectancy under the domain of motives because it most parsimoniously representsthe influence of outcome goals on behavioural choice.A second efficacy construct refers to perceptions of external barriers to action, ratherthan a focus on internal capabilities as with self-efficacy. This notion, which I have termed“channel efficacy,” posits that one’s belief about the ease with which behaviours areaccessible to the individual is an independent influence on behavioural choice. Kurt Lewinwas the first to acknowledge the importance of seemingly minor but actually important aspectsof a situation that may affect behaviour, which he called “channel factors” (see Ross &23Nisbett, 1991 for a review). Lewin recognized that behaviour is often produced by theopening up of some channel and sometimes closed by the blocking of a channel. I suggestthat the influence on behaviour exerted by channel accessibility manifests itself through one’sbeliefs regarding the accessibility of a behaviour. In other words, it is not simply howaccessible a behaviour is, it is how accessible a behaviour is perceived to be. For example,individuals with equal access to a behavioural channel (such as a recycling depot) are likely tovary greatly in their beliefs regarding how accessible the depot is. It is believed that thisappraisal will directly influence the individual’s intentions to act, and represents aconceptually independent aspect of control.A recent study by Boyd and Wandersman (1991) supports the unique importance ofchannel efficacy in guiding behaviour. They assessed the utility of various factors aspredictors of subsequent condom use, and found that beliefs regarding the convenience ofusing a condom accounted for a substantial portion of the variance in actual condom use overa three month period (18%), more than any other single belief. They also found that beliefsin self-efficacy (i.e., condoms are easy to use) significantly explained a portion of variance(5.8%). Thus, there is reason to believe that self-efficacy and channel efficacy representindependent constructs.Efficacy beliefs and environmental behaviourStrategies used in the public sector to promote environmentally-responsible behaviouroften focus on making behaviours seemingly more efficacious. Many interventions attempt toimprove the infrastructure with respect to a certain behaviour by making the actions moreaccessible and by building a sense of self-efficacy. For example, communities across NorthAmerica have noted that curbside recycling programs (collecting recyclables from individualresidences) have significantly increased recycling activity. It can reasonably be asserted thatcurbside recycling programs increase recycling behaviour in part because they make the actionmore accessible (i.e., channel efficacy). Another tactic often used is to provide people withthe information necessary to feel competent to perform a desired behaviour. For example,24brochures are created to provide consumers with specific instructions on how to conserveenergy, thereby building a sense of self-efficacy or personal competence necessary to engagein the desired behaviour.In addition to this non-scientific evidence, there is empirical support for theimportance of efficacy in this domain. In their meta-analysis, Hines et al. (1987) reported on15 studies that investigated the relation between individual efficacy perceptions (which theytermed “locus of control“) and environmental behaviour. The corrected correlation acrossthese studies was = .36, indicating that those individuals who had a strong sense of personalcontrol or efficacy were more likely to engage in environmentally protective behaviours thanwere individuals exhibiting a lower sense of personal efficacy.Axeirod and Lehman (1993) found that both self-efficacy and channel efficacyregarding environmental actions accounted for significant amounts of explained variance insubjects’ reports of past environmental behaviour. Numerous other studies have shown thatefficacious beliefs can have a significant influence on the performance of environmentally-responsible actions (Huebner & Lipsey, 1981; Kantola, Syme, & Nesdale, 1983; Thompson& Stoutemyer, 1991).SummaryVBeliefs regarding the efficacy of performing a certain behaviour clearly influencebehavioural decisions. Two efficacy constructs are proposed. The first construct, selfefficacy, has long been construed as an important determinant of behaviour. Correlationalresearch has found a strong association between self-efficacy and corresponding behaviours.In addition, evidence supporting the contention that self-efficacy is a causal agent in thebehaviour decision process has been found through experimental study. The second construct,“channel efficacy,” although previously neglected in much social issue research, is nowreceiving its due as a construct that contributes uniquely to the explanation of socialbehaviour. Both self-efficacy and channel efficacy have been found to be significantlyassociated with the performance of environmentally-protective behaviours. Multivariateresearch is needed to explore the exact nature of the relation among efficacy beliefs,attitudinal beliefs, and other salient behavioural guides, such as outcome expectations.2526Chapter 3Behavioural MotivesThe third domain of behavioural determinants identified in the present frameworkrefers to the influence of individual needs, desires, and expectations on behaviour. Clearlypeople’s actions are guided by a strong yearning to secure outcomes that satisfy basic needsand fulfill individual desires. Sustenance, pleasure, wealth, friendship, knowledge, respect,self-respect and peace of mind are just some of the vast number of goals found to motivatesocial behaviour. The notion that people’s behaviour is directly motivated by the pursuit ofneeded and desired outcomes seems almost too obvious to research. Yet, the link betweenfunctional motives and behaviour can be quite complex, and, at times, unpredictable. Thechallenge, as I see it, regarding the psychological investigation of motivation and socialbehaviour, is to determine what contextual and personal characteristics influence the intensitywith which differing motives are experienced.Past research has shown that the amount of influence motives have on behaviour isdependent on the extent to which a desired outcome is expected as a consequence of one’saction (outcome expectancy) and how much the individual values that outcome (outcomevalue). Most models that purport to explain social behaviour rely to some extent onperceptions regarding outcome expectancies and values (e.g., Ajzen & Fishbein’s Theory ofreasoned action, 1980; Bandura’s Self-efficacy Theory, 1986; Triandis’s Theory of SocialBehaviOur, 1980). For example, Maddux, Norton, and Stoltenberg (1986) investigated thedirect influence that outcome expectancies, outcome values, and self-efficacy had on subjects’decision to perform a specified behaviour. Much of their reasoning was based on Bandura’swork. Written communications were used to manipulate each factor independently into highand low levels. The essays described an interpersonal communication technique that couldincrease one’s success in dealing with problematic social encounters. Subjects in the highoutcome expectancy condition were led to believe that the technique was extremely effective,while subjects receiving the high outcome value essay learned that the use of this technique in27interpersonal situations would increase success in these interactions (this manipulationassumed that “success” was of value to subjects). It was argued to subjects in the high self-efficacy condition that the technique is very easy to learn and perform. Subjects in the lowconditions of each factor received essays arguing the opposite position (i.e., not effective, oflittle value, difficult to learn and use). Although experimental results supported thecontention that outcome expectancies significantly influence behavioural intentions, they failedto do so for outcome value and self-efficacy. In contrast, correlational analyses found that allthree variables were significant, and roughly equivalent predictors of intentions.This study showed that expectations regarding the accrual of positive and valuedoutcomes significantly motivate behavioural decisions. However, it is not surprising thatpeople who believe they will accrue positive outcomes from an action intend to perform thataction. Therefore, investigating the mu] ti faceted nature of expectancy/value beliefs is thoughtto be a worthwhile endeavor. This approach was prompted by the fact that the complexnature of outcome expectancies and values as behavioural rnotivators has not beensystematically studied. Outcome expectancies have generally been analyzed in simplisticterms of valence (e.g., positive/beneficial versus negative/harmful), as was done in theMaddux et a]. study. I suggest that the next step in this line of research is to delve deeperinto people’s motives for performing an action by classifying outcome expectancies in termsof substantively different, and possibly conflicting, outcome domains. I contend that differentclasses of outcomes must be recognized and systematically studied within the same researchprogram in order to establish a more comprehensive understanding of how different motivesmay influence behaviour through expectancy/value beliefs.The field is currently experiencing a renewed interest in the study of behaviouraloutcomes in terms of their functional utility in motivating social issue behaviour. Forexample, Omoto and Snyder (1990) have applied a functional approach, more commonlyidentified with attitudinal theory, to the study of AIDS volunteerism. Functional theoristshave argued that the same attitude may serve different psychological functions for different28people (e.g., Katz, 1960; Smith, Bruner, & White, 1956; Snyder & DeBono, 1987, 1989).Similarly, Ornoto and Snyder (1992) assert that a functional approach may hold great promisefor unraveling the complex web of personal and social motivations that serve as thefoundations of volunteer activity. They go on to state that one of the key elements in afunctional approach, and one of the primary reasons they feel it has utility beyond the field ofattitudes, is its explicit concern with motivation. A functional approach creates theopportunity to understand how thoughts and actions that share the same surface features mayactually reflect different motivational processes for different individuals. This approach canbe used insightfully in all areas of social issue behaviour, and, as we shall see, is particularlyrelevant in understanding the array of motivations underlying environmental action.People certainly can engage in the same behaviours for very different reasons and withdifferent outcomes in mind. By examining the root motives that may underlie people’soverall evaluation of outcomes to be gained from a certain action, we can identify preciselythe most salient and/or powerful motives, as well as explore potential individual differences inhow those motives are manifested. The initial step in this process was to group possiblemotivations in general categories, which could serve as a foundation for analyzing specificareas of behaviour. It should be noted that both outcome expectancy and outcome value referto beliefs regarding outcomes derived from a specific situation. It is anticipated that the traitnotion of personal values, more so than the state belief of outcome value, may be useful inexplaining behavioural decisions. This expectation is based on the notion that althoughoutcome expectancies are clearly reliant on aspects of a specific situation, the extent to whichan outcome is valued is set by an internal, stable individual belief system. Personal valueswill be discussed in Chapter 4.Functional Motives and Environmental ActionDrawing upon previous research and theory, I have identified three broad domains ofhuman needs and desires that will serve as the foundation from which to understand themotives involved in the realm of environmental action. Different behavioural domains will29necessarily involve different possible outcomes. Therefore, I focused on those outcomesreasonably associated with environmental action. First, in accordance with a central tenet ofequity theory (Walster, Berscheid, & Walster, 1973) and economic utility theory (Simon,1957), people are believed to be motivated to maximize their own economic or material gainfrom personal action. This class of outcomes can be thought of as tangible in nature,primarily referring to outcomes such as economic or material rewards and/or avoidance ofeconomic, material, or time costs. For example, people would be prone to actenvironmentally protective if there were some economic benefit linked with their actions(e.g., energy conservation leading to reduced energy bills).Various studies have found that the use of lotteries, prizes, and other economicrewards or messages produces increased levels of environmentally protective behaviour (e.g.,Jacobs & Bailey, 1982; Luyben & Bailey, 1979; Margolin & Misch, 1978). Recently,Needleman and Geller (1992) compared the influence of eight different intervention strategiesto increase people’s recycling activity in a workplace setting. The interventions includedenvironmental appeals, inducements such as a small, non-economic gift, goal setting, andraffles for prizes. In fact, they found that the economic intervention (i.e., raffles) was theonly effective one. In their study of recycling behaviour, Vining and Ebreo (1990) found thatwhereas both recyclers and nonrecyclers were motivated by a concern for the environment,nonrecyclers were more concerned with economic aspects of the situation. Thus, economicmotives seem clearly implicated in motivating, and possibly blocking, environmentallyprotective behaviour.The second class of outcome expectancies originates from theories of social influence(e.g., see Aronson, 1991 for a review) and is consistent with Ajzen and Fishbein’s (1973)notion of subjective norms. This class of outcomes holds that people are motivated to act by‘social” factors, such as seeking belongingness and acceptance in a social group and avoidingsocial embarrassment. In contrast to other theories, where the influencing path is not directlyrelated to perception of outcomes, the present framework defines social influences in terms of30their correspondence with related outcome expectancies and values. I believe it is, in fact,advisable to investigate social intluence in this manner in that social outcome expectancies,such as avoiding embarrassment, enhancing one’s image, or gaining respect from others, canplay a significant role in guiding behaviour.Social influence strategies have often been employed to promote environmentallyprotective behaviour. Near]y all of us, at some time, feel pressure from others to act in the“socially prescribed” way. One reason for the success of curbside recycling programs acrossNorth America, in addition to the obvious reason that recycling is made more convenient, islikely to be the social aspects of the behaviour. People tend to observe whether or not theirneighbors are recycling, and are likely to be motivated by the example set by them. Socialpressure can foster intensified environmentally protective behaviour. For example, Aronsonand O’Leary (1983) found that students would conserve water more (by modifying theirmanner of taking a shower) when their actions were observabl and more conservation-oriented behaviour had recently been modelled by a fellow student. Vining and Ebreo (1988)found that social pressure presented by one’s peers was reported to be an important reason forrecycling.Although few scholars would dispute the strong influence social pressures can exert onbehaviour, it is often difficult to ascertain the extent of their effect. For example, Vining andEbreo (1990) failed to find a difference between recyclers and nonrecyclers in theirrecognition of social motives. This finding contrasts with the results from their earlier studyin which social pressures were reported as influential (Vining & Ebreo, 1988). Oneexplanation for this inconsistency is that people may, at times, underestimate the influence ofsocial pressures because admitting that social pressure affected one’s behaviour can beperceived to reflect a weakness in ones character. I suggest that the desire for socialacceptance and belongingness is basic to all people (at least to some extent), and that thismotive can be a positive force. The difficulty in observing this phenomenon, particularly incontrived situations or when relying on self-reports, presents the researcher with a formidable31challenge. Nevertheless, the impact of social goals should be included in a comprehensivediscussion of functional motives and social behaviour.The third class of outcomes identifies consequences that are intended to benefit someexternal entity. The motivation underlying this class of outcomes sterns from the personalsatisfaction people accrue from contributing to a cause they value. Individuals are motivatedby a desire to pursue objectives regarding the specific issue of concern, and are notnecessarily concerned with accruing any tangible or social benefit. Krosnick (1990) andOmoto and Snyder (1990) have noted a similar motive in their research in the areas of politicsand volunteerism, respectively. (An extensive review of the theoretical literature on personalvalues and value-behaviour relations is presented in Chapter 4.) The influence of this class ofoutcomes can be observed in the expectation the individual has regarding the extent to whichtheir behaviour benefits the source they support (e.g., the environment). For example,buying environmentally protective products may be personally more costly and have noperceived social benefits, but people may do it because of their desire to improveenvironmental conditions- an outcome likely to be consistent with a high value placed on theenvironment.In fact, a desire to help the environment is the motivation most often presumed tounderlie environmentally-protective beli avi our. This contention has received substantialempirical support (e.g., De Young, 1986a, 1986b; Dunlap, Grienecks, & Rokeach, 1983).Thus, one’s expectancy that this outcome would be achieved is presumed to be an importantfactor guiding one’s behavioural decisions. Although this expectancy is strongly implicated,there is clearly more involved in these behavioural choices. We have already seen howeconomic and social factors can exert a direct influence on the performance of environmentalbehaviour. What is important to determine is the manner in which these different functionalmotives may independently and interactively influence individual behaviour.Whereas these three classes of outcomes do not exhaust all possible behaviouralmotives (e.g., pleasure, intellectual growth), past theorizing and research suggest that this32typology represents a valid grouping of outcomes. One, two, or all three classes of outcomesmay be perceived as antecedents of a single behaviour. Moreover, people are likely to bemotivated to some extent by each of the outcome classes. Thus, for example, recycling hasbeen promoted by the use of each class of outcome: economic (deposits on returnablebottles); social (setting of community standards); environmental (linking recycling to helpingto protect the environment).The relation between each class of outcome and behaviour can be, and should be,independently assessed. This is especially important in situations where positive outcomesmay be perceived in one class (e.g., economic benefits), and negative outcomes (e.g., socialembarrassment) in a second class. It should be noted that no evaluative connotation isintended in this classification system. Incorporating the potential for conflict among differentoutcomes is considered a vital step in understanding the role of outcome expectancies onbehaviour.Psychological MotivesAlthough the proposed framework focuses on functional motives, it would beneglectful not to include some discussion about the role of psychological motivations as guidesto social behaviour. In this section, I will briefly describe two prominent theoreticalapproaches to understanding psychological motivation, dissonance theory (Festinger, 1957)and self-affirmation theory (Steele, 1988), and explore their explanatory value in the realm ofenvironmental behaviour. I will introduce an alternative perspective regarding these theories-one which reveals a potential commonalty that has not as yet been pursued.Dissonance theory states that if a person holds two cognition s that are psychologicallyinconsistent, he or she will be motivated to reduce that negative state just as a person needs toreduce hunger or other drives (Festinger, 1957). Dissonance theory arose as a way ofexplaining how behavioural changes can produce attitudinal change, rather than the reversepath. The claim of dissonance theory is that most people strive for three things: (1) topreserve a consistent and predictable sense of self, (2) to preserve a competent sense of self,33and (3) to preserve a morally good sense of self. As Aronson (1992) has explained it “whatleads me to perform dissonance-reducing behaviour is having done something that (a)astonishes me, (b) makes me feel stupid, or (c) makes me feel guilty” (p. 305). Dissonance isstrongest and clearest when it involves not just two cognitions but, rather, a cognition aboutthe self and a behaviour that violates that self-concept.Dissonance theory, as applied to environmental behaviour, would suggest that the bestway to promote environmentally-protective behaviour would be to create a sense ofdissonance in people that would motivate them to reduce dissonance by performing desiredactions. Considering that while most people today report being concerned about theenvironment but few are actually highly active (Angus Reid Group, 1992; Gallup, 1990), theopportunity to promote environmentally-concerned behaviour through dissonance mechanismsseems substantial. This possibility was tested in a study that attempted to establish whetherhigh consumers of energy would adapt their behaviour to conserve energy when placed in acognitively dissonant situation (Kantola, Syme, & Campbell, 1982). The energy use ofsubjects, divided into four experimental groups, was compared. The four groups were: (a) adissonance group, where people were informed that an inconsistency existed betweenpreviously measured attitudes toward conservation and their high use of electricity, (b) afeedback group, where people were notified that they were high consumers of electricity, (c) atips only group, where people were sent information on ways to conserve electricity (also sentto the dissonance and feedback groups), and (d) a control group. While the dissonance groupconserved more electricity than all other groups in an initial two-week measurement period, itdiffered from only the control group in a second two-week measurement period. It isreasonable to conclude that dissonance reduction was a motivator only in the immediate timeframe following the intervention. Once a consonant state was reestablished, subjects’conservation behaviour regressed, although still higher than baseline. At this point, subjectsseemed to be motivated by other factors- the same factors that motivated the other two34experimental groups (i.e., awareness of high energy use in general and concrete knowledge ofways to conserve energy).Although this study appears to confirm the motivational influence of cognitivedissonance, the source of dissonance may not have been simply attitude-behaviourinconsistency. It may have been that the ability to reduce energy expenses was more clearlyrecognized (i.e., enhanced efficacy) in the dissonance group than in the other experimentalgroups. This recognition may have stemmed from a higher level of scrutiny of the subjects’behaviour in this group prompted by the attitude-behaviour inconsistency. This enhancedrealization (i.e., the opportunity to save money) could have been the factor directlyresponsible for prompting them to reduce energy usage. In addition, they may have feltcertain social pressures to conserve considering that their high usage of electricity wasobviously public knowledge. The possibility that these outcomes motivated the behaviouralchange is consistent with a theory developed by Cooper and Fazio (1984). They noted that inall previous dissonance experiments, not only was inconsistency present, but so were aversiveconsequences. They asserted that it was these aversive consequences that aroused feelings ofdissonance.In a recent study examining the influence of dissonance on attitude-behaviourrelations, Scher and Cooper (1989) found that experiencing dissonance can motivate attitude-behaviour consistency, but that dissonance arises from the perceived consequences of one’sactions (aversive vs. nonaversive), and not attitude-behaviour inconsistency. Their studyshowed that subjects who freely chose to perform acts that could produce aversiveconsequences experienced feelings of dissonance and changed their initial attitudes regardlessof whether these actions were consistent or inconsistent with their original attitudes. Incontrast, subjects who were led to believe that their actions were unlikely to have aversiveconsequences experienced little or no dissonance and did not alter their opinions in eitherconsistent or inconsistent situations.35Steele and Liu (1983) have also proposed a modification of dissonance theory. Theyagree that a need for consistency motivates behaviour, but they suggest that the need forpsychological consistency is not part of the dissonance motivation. Their research suggeststhat self-affirmation objectives are the key to a consistency motivation, in that dissonanceappears strongest when highly valued aspects of the self are most relevant. Recently, Steele(1988) has shown self-affirmation to be a distinct source of motivation. He found that ifpeople are put in dissonance-arousing situations, not all of them will attempt to reducedissonant feelings by changing their attitudes. Specifically, subjects who were given anopportunity to affirm some valued aspect of their self-concept as an alternative to attitudechange were able to maintain their original attitudes even though the attitudes wereincongruent with their behaviour. Self-aftrmation theory suggests that a state of consistencyis not essential (and inconsistency is tolerable) if an important aspect of the self in notinvolved. Steele and Spencer (1992) argue that maintaining one’s integrity, and not self-consistency per Se, is the primary motive influencing subjects in the majority of dissonance-related studies.All of the proponents of these dissonance-related theories would agree that a dissonantstate is psychologically troublesome, motivating a person experiencing this state to somehowreestablish cognitive balance, but they differ about what occurrences arouse this negativestate. Whichever view is most accurate, their commonalty is that dissonance is most likely tooccur when something that is highly valued by an individual is threatened (i.e., an aspect ofthe self-concept or negative consequences). An aspect of this phenomenon that has not yetbeen fully considered is the notion of “value systems” as they may influence dissonancearousal and resulting behaviour. What are the aspects of the self that are highly valued andhow might these core values influence social behaviour? This issue seems important becausepeople would likely look favourably upon behavioural outcomes that are consistent with, orenhance, their conceptions of themselves, and perceive outcomes that are inconsistent withtheir self-concepts as either negative or irrelevant.36To fully understand how dissonance functions as a motivation, it seems important toknow more about the individual than his or her level of self-esteem. For example, if self-esteem is strongly linked to economic status, and not at all related to academic abilities, thendissonance would more likely occur when this individual’s belief in their economic prowess isthreatened. In addition, this individual would likely pursue economic outcomes in order to beconsistent with, and possibly enhance, the perception of self. In fact, in this case, whetherbehaviour is seen as motivated by self-consistency, self-affirmation, self-enhancement, orbehavioural consequences, the same predictions regarding the person’s behaviour could bemade: that is, dissonance will occur if the person’s economic self (in terms of consistency orcompetence) is in some way threatened. The influence of personal values in the dissonanceprocess, as well as in the broader behavioural picture, seems important and in need of furtherinvestigation.37Chapter 4Personal ValuesPersonal values, as suggested in the proposed framework, function as a foundation onwhich more specific attitudes, beliefs, and ultimately behaviour, are rooted. As stated bySchwartz (1992), most theorists view values as “the criteria people use to select and justifyactions and to evaluate people (including the selt) and events” (p. 1). Values, as defined byRokeach, (1980) are standards or criteria which guide action as well as other psychologicalphenomena such as attitudes, judgements, and attributions. Values are considered deeper andmore stable than attitudes representing standards of “oughts and shoulds” (p. 272), and areviewed as determinants of attitudes. Rokeach refers to values as “core conceptions of thedesirable” (p. 2), and posits that the number of human values is relatively small and capableof being embraced differently by individuals within a society. This conceptualization ofvalues is consistent with Kluckhohn (1951), who defines values as “conceptions of thedesirable means and ends of actions.” Schwartz (1992) has added that “the primary contentaspect of a value is the type of goal or motivational concern that it expresses” (p. 4). Thesedefinitions suggest that the values people embrace are responsible for guiding their pursuits inlife. More specifically, values are construed as determinants of both attitudes and outcomebeliefs. Accordingly, Rokeach (1980) has stated that a meaningful discussion of attitudes,behaviour, and attitude/behaviour relations is not complete unless it considers the influence ofvalues.Although the psychological importance of personal values has been noted, researchexamining the influence of values on social behaviour has not yet been fully explored. Ibelieve this is due to problems that arise in operationalizing value systems and incorporatingvalue orientations as factors in behavioural research. In other words, how does a researcherassess a subject’s value priorities and then relate them to a behaviour of interest (e.g.,volunteerism, recycling, voting)? Are personal value systems a social construct, or part of anindividual’s personality? Notwithstanding these difficulties, the potential importance of38personal values in guiding behavioural decisions should not be overlooked. A review of theprominent value and motivational structures is presented below (see Figure 2 for moredetails).Description of Prominent Value TheoriesOne of the first value taxonomies dates back to 1931. Developed primarily by GordonAilport, the Study of Values (Ailport, Vernon, & Lindzey, 1960) distinguished six basicinterests or personality motives: economic, social, theoretical, aesthetic, political, andreligious. Briefly summarized, the economic person values that which is useful, judgingthings by their tangible utility; the social person most values love and connection with others;the theoretical person values the discovery of truth; the aesthetic person most values beautyand harmony; the political person values power and influence as an end; and the religiousperson most values unity, seeking communion with the cosmos. The Study of Valuesmeasured the relative strength of each value by having respondents complete a series ofstatements in which they rank ordered options (either 2 or 4 options) that represented differentvalue orientations. For example, one item read “To what extent do the following famouspersons interest you: (a) Florence Nightingale; (b) Napoleon; (c) Henry Ford; or (d)Galileo.” Each response option represented a different value (i.e., social, political, economic,and theoretical, respectively). Although the reliability and validity of this value measure hasreceived substantial empirical support (e.g., Cantril & Allport, 1933; Duffy, 1940), thecontent of the measurement instrument is quite dated in terms of some of its response options(today, most people probably could not identify all of the noted historical figures). This, inpart, accounts for its limited use in recent years.A second dominant theory of values was proposed by Maslow (1970, 1971). Maslowconceived of human behaviour as motivated by certain needs or values (in fact, he used thetwo words interchangeably), which were hierarchically ordered. At the first level of thehierarchy were physiological and safety needs - the things that people need to live physicallyhealthy and secure lives. Once people satisfy their basic needs, they then seek to satisfy39desires for belongingness and love. The two higher-order needs involved developing self-esteem and ultimately seeking to become fully themselves - to “actualize” all that theypotentially can be. Maslow’s theory defines concisely what can be considered coremotivational domains, although certain concerns with its structure can be identified. First,there is no allowance for individual differences in the extent to which a need is valued.Second, it is unclear how one would assess whether or not a certain need has been satisfied.For example, the amount of security or love needed to satisfy an individual’s need is moreaccurately understood as a psychological construction, rather than some definitive, universallevel. Thus, substantial individual variance in the perception of this level is likely to exist.Maslow’s theory’s most relevant contribution is its recognition of those needs presumed tomotivate all people in some manner.Milton Rokeach pioneered much of the recent study and delineation of value systemsby developing the Rokeach Value Survey (1967). Rokeach (1973) described the nature ofhuman values as guided by five assumptions: “(1) the total number of values that a personpossesses is relatively small; (2) all people everywhere possess the same values to differentdegrees; (3) values are organized into value systems; (4) the antecedents of human values canbe traced to culture, society and its institutions, and personality, (5) the consequences ofhuman values will be manifested in virtually all phenomena that social scientists mightconsider worth investigating and understanding’ (p. 3). Rokeach argued that the valueconcept, more than any other, should occupy a central position across all the social sciences.Rokeach divided values into two domains: desirable modes of conduct (instrumental)and desirable end-states of existence (terminal). He proceeded to identify 18 values in eachdomain, which he claimed made up the totality of human values. Because the notion ofvalues as portrayed in the present framework focuses on their potential motivational role,Rokeach’s terminal values are highlighted. Examples of the values denoted by Rokeachinclude: a comfortable life, a sense of accomplishment, equality, family security, maturelove, self-respect, and true friendship (see Figure 2 for a complete list). To assess an40individual’s value structure, respondents simply rank-order all 18 values from most importantto least important.Rokeach (1973) has reported a substantial body of data, which shows that one or moreof these values significantly predicts corresponding behaviours of interest. These behavioursinclude participating in civil rights demonstrations, church attendance, and political activism,among many others. While the ranking of one value (equality) seems to be important in agreat many behaviours, Rokeach has found that any one behaviour is likely to be predicted bymany values. In total, over 33% (252 out of a possible 684) of reviewed value-behaviourrelations were found to be significant. In addition, Rokeach’s values have been found topredict significantly many theoretically related attitudes. In fact, Rokeach (1980) argues that“it is difficult to identify socially important behaviours or attitudes that are nor predicted byone or more values” (p. 274).Although Rokeach’s value survey has been widely used, some researchers haveexpressed various concerns with his paradigm (e.g., Clawson & Vinson, 1978; Kahle, 1983).These criticisms include a loss of information because of rank orderings, difficulty of thelengthy ranking task, and ambiguous relevance of many of the values to daily life. In aneffort to resolve some of these concerns, Kahle (1983) developed an alternative, simplifiedstructure which he called the List of Values (LOV). The LOV defines nine values thought tobe relevant to daily life and construed as motivational in nature. Kahle also suggested thatvalues be rated independently, rather than rank-ordered.In a recent study of consumer behaviour, Homer and Kahle (1988) examined theinfluence that values have on consumer attitudes and behaviours regarding the purchase ofnatural food products. Using factor analyses, they were able to determine three underlyingfactors that accounted for 64.4% of the variance. Factor 1 comprised four values: (a) selffulfillment, (b) excitement, (c) sense of accomplishment, and (d) self-respect. Factor 2included three values: (a) a sense of belonging, (b) being well-respected, and (c) security.Factor 3 included: (a) fun and enjoyment and (b) warm relationships. They identified Factor412 as representing an external dimension of values, whereas Factors 1 and 3 were consideredmore internally oriented. Using this factor structure in a causal modelling analysis, theyfound that certain values were important in predicting attitudes towards the product ofinterest, and that those attitudes, in turn, led to corresponding behaviour. Thus, attitudesplayed a mediating role between values and behaviour. Several precautions regarding thespecific nature of their study were presented, thereby limiting the generalizability of theirconclusions. These included recognizing that these relations were studied in only one specificcontext, and that social behaviour should not be seen as merely an expression of values and noother determinants. They suggest that “now the challenge is to verify these findings in othersituations” (p. 645).An impressive attempt to construct a universal value structure has recently beenconducted by Shalom Schwartz and his associates (e.g., Schwartz, 1992; Schwartz & Bilsky,1987, 1990). Schwartz began his search for a typology of universal value domains bytheorizing that values could be derived from “the universal requirements reflected in needs(organi Sm), social motives (interaction), and social institutional demands.” Schwartz’sapproach to defining values relied on three underlying ideas: (1) that values must representthe interests of some person or group, (2) that a distinction between individualistic (e.g.,ambition) and collectivistic (e.g., helpfulness) interests is useful, and (3) that the differentmotivational domains in which universal human requirements are expressed provide a basisfor distinguishing among value contents. Using the values suggested by Rokeach (1973) asmarkers, Schwartz and Bilsky (1987) initially derived seven universal and distinctivemotivational domains of values (enjoyment, security, social power, achievement, selfdirection, prosocial, restrictive conformity), plus an eighth domain they termed “maturity”which represented goals that people reach only through experiencing life (cf., Maslow’s selfactualizing need).Schwartz and Bilsky (1990) have since revised their value theory numerous timesbased on a series of cross-cultural studies that included respondents from 20 countries,42representing varied cultures. In the majority of these studies, respondents were asked to rateeach of 30 terminal values and 26 instrumental values in terms of their influence as a guidingprinciple in their lives on a nine point scale ranging from “of supreme importance” to“opposed to my values.” From these studies, an updated factor structure has been determinedthat includes 10 value domains (seen in Figure 2), two of which probably need someclarification. First, Tradition refers to respect, commitment, and acceptance of customs inone’s culture. This content domain stemmed from the notion that groups in every culturedevelop practices that represent their shared experience. Complying with tradition becomesan important motive in many individuals. Second, Universalism incorporates the idea ofmaturity and part of the prosocial domain described in an earlier structure (Schwartz &Bilsky, 1988, 1990). The motivational content of this value is perceived as theunderstanding, appreciation, tolerance, and protection of all people and of nature. Incontrast, Benevolence refers specifically to concern over the welfare of close others ineveryday interaction. After much research, Schwartz (1992) has concluded that the content,measurement, and structure of values represented in this value typology seems “sufficientlywell established to justify their adoption as the basis for future research...“(p. 60).A New Value TaxonomyAlthough Schwartz’s value structure appears to be quite reliable and comprehensive,its utility as a tool to explain and predict social behaviour is questionable. No procedure foridentifying the value orientation or preferences for an individual is identified, and no linkbetween the value domains and social behaviour has been offered. These limitations, incombination with the concerns noted regarding the other value theories, suggest a need for analternative value taxonomy, one that would be more in line with the requirements ofmultivariate behavioural research. The results of an endeavour to create such a taxonomy canbe seen in Figure 3.Three primary objectives were considered in determining this new structure and itsassociated assessment process. First, the value domains identified in the structure should43involve concrete motivational content in terms of daily behavioural decisions. Many of thevalues noted in the Rokeach, Kahle, and Schwartz models have motivational implications thatare either unclear (e.g., equality) or can vary greatly in what an individual perceives asmeeting that goal (e.g., what stimulates or excites one person may not stimulate or excitesomeone else). Thus, each domain should be consistently construed as representing the samegoals, even though the extent to which each goal is valued will vary across individuals.Second, the list of domains should be simplified to include only those core domainsthat seem to be universal goals that people find important at least to some extent. Schwartzand Bilsky (1987) identified three core domains which they referred to as “universal humanrequirements.” These included (a) biologically based needs, (b) social requirements, and (c)group welfare and survival. These three requirements are consistent with motivational guidesidentified in the social psychological literature (reviewed in Chapter 3), and therefore theyserved, in part, as a foundation on which the proposed value taxonomy was constructed.Third, it was deemed desirable to be able to ascribe a “value orientation” (i.e.,identifying one value domain as most important when compared to all others) to individuals,in the tradition of the Study of Values (Allport et al., 1960). Representing human values asan individual difference construct would create the opportunity to analyze behavioural patternsacross individuals who embrace different, and possibly conflicting, value orientations. Aspreviously noted, values are considered to be deeply rooted and stable predispositions. Theyare considered an integral part of a how a person defines oneself. By assigning a dominantvalue orientation to individual subjects, the influence that values have on behaviouraldecisions can be explored in a different, possibly more revealing, manner than typically donein value research. For example, differences in the behaviour of independent groups asdefined by their value orientations (e.g., economically-oriented vs. socially-oriented) can beexplored using experimental procedures. This design enables the researcher to examine thelink between values, behavioural decisions, and other salient factors. A complete review of44the procedures used to assess individual value orientations for the proposed value structure ispresented in Chapter 5.The proposed value structure identifies a triarchal classification of motivationaldomains: (a) economic, (b) social, and (c) universal. Although derived from a differentconceptual base, it is interesting that this structure closely parallels the classification schemedescribed earlier for outcome desires. One important difference is that the earlier valuetaxonomy was developed to describe generic motivational domains. In contrast, the latterclassification was developed to describe the motivations that specifically guide environmentalbehaviour.The first domain refers primarily to goals such as economic security or achievement,material rewards and/or avoidance of economic, material, or time costs. It parallels thesustenance needs identified by Schwartz and Bilsky (1987) and Maslow (1970, 1971),although it more accurately reflects the value placed on economic and material desiresregardless of one’s actual need situation. This is an important distinction because the pursuitof economic gain appears to motivate behaviour well beyond a time when physical needs aresatisfied. The desire for economic outcomes can be seen as a psychological need, andachievement in this area is highly valued by people who place it at the top of their valuehierarchy. This motivational domain is considered most consistent with Rokeach’s “acomfortable life” and Ailport et al.’s “economic person.”Economic goals can influence many, if not all, issue-oriented social behaviouraldecisions. In terms of the present study, economic goals, both personal and societal, are oftenimplicated in environmental decisions. For instance, on a personal level, purchasingenvironmentally-protective products may necessitate spending more money. On a societallevel, reducing the amount of logging may mean the loss of income for numerous people.Although this motivational domain is usually construed as external in nature, it is suggestedthat people also seek and receive internal satisfaction from attaining these outcomes.Economically-oriented people judge themselves based on their economic status and success.45The second motivational domain represents the social aspects of life. It specifiesdesires regarding social consequences from one’s actions and includes both belongingness andconformity drives (see Aronson, 1991 for a review) as well as aspects of social altruism (e.g.,Schwartz, 1977) and benevolence (Schwartz, 1992) motives. It postulates that the motivationto seek belongingness and acceptance from others is a central guiding force in decisions toact. Each of the value theories reviewed earlier includes at least one domain that involvessocial needs. It is postulated that conformity and belongingness values, as well as a part ofbenevolent values, are rooted in relationship needs and desires. Thus, actions in accordancewith these desires would, theoretically, lead people to pursue goals such as the welfare ofclose others as one means of maintaining and/or enhancing one’s feelings of connection withothers.Social values and desires are frequently noted in discussions of environmentalbehaviour (see discussion in Chapter 3 on social motives). For instance, conformity can beobserved in the influence of modelling on environmentally protective behaviour, the desire forbelongingness can induce people to act in a manner conSistent with valued others, and thevalue placed on benevolence may prompt socially-oriented people (people who place socialvalues at the top of their hierarchy) to act environmentally-protective when they believe theiractions can help minimize the plight of other people. Although the proposed social domainincludes several components, a single factor representing this value is consistent with thesocial requirement identified by Schwartz and Bilsky (1987), the belonging and love needproposed by Maslow (1970), and the social person described by Aliport et al. (1960). Ibelieve that the benevolence aspect of the social domain is rooted in the desire for personalacceptance and beiongingness.The third motivational domain, “universal,” is most consistent with Schwartz’s (1992)“universalism” domain. The motivational content of this value type involves the pursuit ofself-respect garnered from making a contribution to the betterment of the world, particularlyas it pertains to pursuing and attaining outcomes that correspond with universalistic-type goals46(e.g., equality, environmental preservation). Pursuing these goals may, in fact, involvecertain social or economic costs, which the universally-oriented individual is willing to incur.For example, protesting the harvesting of a section of forest may mean a loss of a job andhave no perceived social benefits, but people may do it in response to their desire to improveenvironmental conditions - an outcome consistent with a desire to act in a universalisticmanner. This motivational domain is most strongly reflected in those people willing to violatelaws and court injunctions and risk substantial fines in order to achieve a certain goal - in thiscase preventing the continued destruction of the natural environment.Concisely stated, humans are believed to be motivated by three fundamental terminalgoals: economic, social, and universal. People who highly value one domain over the othertwo will likely be oriented to act in ways consistent with that goal. Thus, behaviouraldecisions will be most influenced by the motivational domain each individual values mosthighly. The importance an individual assigns to each goal will be the foundation upon whichher or his behavioural choices are rooted.One cautionary consideration regarding the construction of this value taxonomy shouldbe addressed. First, although each domain is presumed to be an independent source ofmotivation, overlapping values among the domains may occur. To illustrate, universal goalsmay involve principles such as social justice - a goal that blurs the line between the social anduniversal value orientations. This confusion can be addressed by recalling the basicmotivations associated with each domain. People with a social orientation are most concernedwith maintaining and enhancing connection with others. When applied to social issueconcerns, this goal is considered consistent with pursuing outcomes that are believed tobenefit a majority of people. People adhering to a universal value orientation embrace acontributory ethic - one which emphasizes the pursuit of a personal conception of universalgoals. Perhaps the most obvious place this distinction can be observed is when a trade-offbetween universal goals and either social or economic outcomes is involved.47More specifically, value distinctions can be understood by examining particular issuesin which different value orientations would necessarily lead to different behavioural decisions.One example is the situation where social good is in conflict with environmental good. In thiscase, it would be predicted that someone with a social orientation would pursue outcomesconsistent with perceived social good (e.g., immediate benefit for a majority of people),whereas someone with a universal orientation, who also strongly values the environment, willlikely pursue environmental good, even if it means social or personal sacrifice. Thishypothesis, as well as the role that value orientation plays in guiding issue-oriented socialbehaviour, is explored later in this thesis.One final distinction that can be made regarding these value orientations is therelativity of each domain. The universal domain, and to a lesser extent the social domain, aresubstantially context free. They can be attached to any issue and may imply differentcorresponding behaviours depending on the normative ideals of the group being investigated.Universally-oriented people are particularly likely to hold strong beliefs that may or may notbe laudable. This would depend on whether their beliefs and goals are consistent with one’sown ideals. In contrast, the economic domain is mostly context dependent where the pursuitof economic goals would be less dependent on the ideological norms of a society. Normativeinfluences may come into play, but only in the extent to which economic goals are valued,and would not vary much from issue to issue.Values and EnvironmentalismThe proposed taxonomy of values parallels two recent value/ethical classificationsdeveloped specifically with respect to the environment (Merchant, 1992; Stern, Dietz, &Kalof, 1993). Stern et al. (1993) proposed a social-psychological model that presumes thatthe pursuit of environmental quality may stem from any of three value orientations: Egoistic,Social-altruistic, and Biospheric. They posit that egoistic values predispose people to perceivedecisions regarding environmental protection in terms of how they, personally, would beaffected. In contrast, people who adhere to social-altruistic values judge decisions in the48environmental context based on an evaluation of the costs and benefits for a wide socialgroup. These orientations can be differentiated by the outcome of most immediate and directconcern. Egoistic people emphasize personal gain, whereas Social-altruistic people are mostdirectly concerned with social good. People guided by biospheric values judge environmentaldecisions based on perceived costs or benefits to ecosystems of the biosphere.In a similar structure, Merchant (1992) outlines three ‘Grounds for EnvironmentalEthics” (p. 64) as: Egocentric, 1-lomocentric, and Ecocentric. An Egocentric ethic refers tothe maximization of self-interest under which people adhere to a philosophy that what is goodfor each individual will benefit society as a whole. People who embrace a Homocentric ethicbelieve in “the greatest good for the greatest number of people,” adopting social justice astheir guiding doctrine. An Ecocentric ethic holds that rational, scientific belief systems wouldbe based on laws of ecology and in the “unity, stability, diversity, and harmony” of theecosystem. Merchant suggests that mdi vidual/society environmental practices will beconsistent with the ethical grounding to which individuals/societies adhere.Whereas the similarities between Stern’s and Merchant’s models and the proposedvalue taxonomy are significant, an important difference is that the proposed value taxonomywas developed generically - not based on any single issue. Therefore, this value structureoffers a more basic motivational model with which to explore individual differences inpursuing environmental preservation, the issue of interest in this thesis.Clearly, if only beneficial economic, social, and universal outcomes could be derivedfrom the same course of action, little conflict would exist among people who embracedifferent value orientations, as well as within an individual. Unfortunately, economic wellbeing is often at odds with environmental preservation, as in the case of the “commonsdilemma.” Although people are certainly influenced by all three motivational domains tosome extent, they are likely to develop hierarchies in their values which guide their decisionsregarding specific concerns such as ecological dilemmas.49Whereas specific expectations regarding the relation among values, other behaviouraldeterminants, and decisions regarding environmental behaviour will be presented for eachstudy, several general hypotheses can be identified. First, economically-oriented individualsare likely to engage in environmentally protective behaviour when it is linked with sometangible benefit. If no positive environmental-economic link is apparent, economically-oriented people would not be expected to act in environmentally protective ways. When thereis no apparent link between economics and environment, economically-oriented people may ormay not act in support of environmental protection. In this case, their actions would beguided by other behavioural cues. This pattern would be true for the other value orientationsas well. Socially-oriented people would be predisposed to actively seek environmentalpreservation primarily when there is some perceived social benefit associated with the decision- either for other persons or for themselves. In contrast, universally-oriented people would bepredisposed to favor environmental protection under any circumstances, as it is presumed tobe a goal that they are likely to value strongly.Second, it is hypothesized that people who embrace a universal value orientation willbe most likely to actively pursue environmental preservation, whereas economically-orientedpeople would be least likely, due to the conflict that often (but not necessarily) exists betweeneconomic and ecological concerns. This conjecture is based on the noted similarities betweenthe proposed value taxonomy and the environmental value structures identified by Stern et al.(1993) and Merchant (1992), as well as other research that has explored the connectionbetween values, motivation and the environment (e.g., De Young 1986a, 1986b; Dunlap etal. 1983, Neuman, 1986). For example, Dunlap et al. (1980) found that people whoperformed an environmentally protective behaviour (i.e., recycling) emphasized principledvalues such as aesthetics and self-actualization (values consistent with a universal orientation)and deemphasized personal values such as safety and security. This research suggests thatpreserving and protecting the natural environment is a widely embraced normative ideal thathas primacy with universally-oriented Peolle in our culture. In addition, it is presumed that50when conflict between environmental concerns and personal needs exist, people will be guidedby the motivational domain that they most highly value. For example, economically-orientedpeople will likely emphasize personal economic interest over potential environmental harm.SumniaryPersonal values are an important. factor to consider when investigating issue-orientedsocial behaviour. They are believed to be deeply rooted, stable beliefs with strongmotivational content. Thus, attitudes, outcome expectancy beliefs, and social behaviour arelikely to be strongly influenced by one’s value orientation. Three value domains aresuggested to represent global motivational domains that guide issue-oriented behaviouraldecisions: economic, social, and universal. People who embrace one value domain as mostimportant will be biased toward pursuing behavioural outcomes that correspond with thatvalue. The relation among values (as defined by the proposed taxonomy), other behaviouraldeterminants, and environmental behaviour is a central focus of the present course ofresearch.51Summary - Part IFive specific features thought to refine our current understanding of various cognitiveconstructs and how they may influence behavioural decisions have been proposed. First, it issuggested that a general attitude measure (toward issues, objects) be incorporated into anycomprehensive attempt to explain and predict social behaviour. The assessment of thisgeneral attitude should not be confused with measuring attitudes toward specific actions.Many recent behavioural, models have focused on expectancy-value beliefs or attitudestowards specific actions as determinants of behaviour neglecting the more global attitudeconstruct. This may be due to the fact that its relation with social behaviour has been foundto be inconsistent, even though widely accepted attitude theory contends that attitudesfunction, in part, as guides to behaviour. It is suggested that the term attitude be used torepresent general evaluations of objects or issues, and that Ajzen and Fishbein’s (1980)attitudes-toward-action construct be defined in terms of its operational components: beliefsregarding outcome expectancies and outcome values.Second, the framework suggests that two corollary attitudinal constructs be assessed inconjunction with a general measure of attitude. These are the strength of one’s attitude, andattitudes toward related objects (e.g., the threat stemming from environmental destruction).Both theoretical and empirical work have demonstrated that these constructs can be importantpredictors of social behaviour. One manner in which they may influence behaviouraldecisions is by mediating the relation between one’s general attitudes and the behaviour inquestion.Third, the designation of an underexplored efficacy concept termed “channel efficacy”is suggested. This construct represents an individual’s belief regarding the accessibility of abehaviour. Although there is a long history of research on the impact that accessibility of abehaviour has on its enactment, beliefs regarding accessibility have rarely been studied as anindependent factor.52Fourth, a systematic exploration of different types of outcomes (i.e., economic, social,and environmental) in terms of their motivational influence is recommended. This structurerecognizes the complex and dynamic interaction that may exist between the motivational forceexerted by different types of outcomes. It also proposes that the relation between socialbehaviour and expectancy/value beliefs can be more productively studied and understoodwhen desires regarding different outcome domains are investigated independently within aninclusive multivariate design.Fifth is the recognition that personal values may play an important role in thebehavioural decision process. It is suggested that much social behaviour is rooted in thesedeep-seated, stable predispositions. Environmental behaviour, in particular, appears to berelated directly to individual values, and a great deal of explanatory information can be lost ifpersonal values are not incorporated into a multivariate analysis of behavioural decisions. Atriarchic taxonomy of values is proposed that identifies three core motivational domains:economic, social, and universal.In general, the proposed framework should be viewed as a research tool rather than asan explanatory model. It outlines a multivariate structure in which the salient constructsnecessary for conducting a comprehensive investigation of decision-making regarding socialbehaviour are presented.53Part IIInvestigating the factors that guide decisions regarding environmental behaviour: A testof the sociocognit.ive frameworkOverviewThe procedures and results of three studies that explored the association that eachfactor outlined in the framework has with behavioural decisions are presented. The first twostudies focus on identifying the model that most parsimoniously explains decisions to performone specific environmental action. The third study broadens the investigation to explain andpredict decisions of more personal and societal consequence. Specifically, this studyexamines the influence that personal values, attitudes, and outcome conditions have on thebehavioural decisions of subjects placed in hypothetical Commons Dilemma type situations.Theoretical and applied implications of these studies are then discussed.54Chapter 5Measurement of independent factorsIn this chapter, the assessment techniques employed to measure the independent factorscommon to all three studies are described. These factors include the attitudinal variables(i.e., general attitude, attitude importance, and threat perception) and subjects’ valueorientations. The pre-test questionnaires used to assess these variables were administeredapproximately one month prior to the implementation of each study. Specifically, there weretwo pre-test administrations that employed the same procedures: one provided data for Study1 (n = 645) and the second provided data that were used in both Study 2 and Study 3 (n =622). Pre-test questionnaires were distributed to undergraduate psychology students during aclass period. Students were asked to complete the questionnaire and return it at the next classperiod. Respondents received course credit for their participation.In an effort to maintain consistency regarding the testing of these constructs, theassessment methods varied only slightly across the two pre-tests. Certain measurementconcerns observed in the pre-test for Study 1 prompted slight modifications in the content ofthe assessment instruments used in Studies 2 and 3. These variations dealt primarily withimproving reliability. Variables specific to each study (i.e., efficacy beliefs, motivationalconstructs) are detailed in the corresponding chapter.Measuring Attitudinal FactorsThree attitudinal variables were assessed in the pre-tests: (a) general attitude towardthe natural environment and environmental protection, (b) the importance of environmentalissues, (i.e., measure of attitude strength), and (c) threat perception (see Appendix 1 forlisting of exact items). In both pre-tests, general attitude was assessed using a scale consistingof six items, which measured respondents’ general beliefs regarding the environment andbeliefs regarding how human beings should interact with the natural world. Developing anattitude scale from responses to belief statements has been found to be a reliable and validmeans of assessing attitudes (e.g., Ajzen & Fishbein, 1980). Moreover, because recent55polling evidence suggests that most people would support a simplified attitudinal statementsuch as “I support environmental protection,” an attempt was made to promote variance inattitudes by having subjects respond to more controversial belief statements. Thus,respondents were asked to indicate how they felt about six belief statements regarding theenvironment and its protection on a scale ranging from “strongly disagree” to “stronglyagree.” Adequate internal consistency was found in the first pre-test (Cronbach’s Alpha of.69). Slight wording modifications resulted in increased reliability in the second pre-test(Cronbach’s Alpha of .75).Issue importance was assessed by two items that measured the absolute importance ofthe environment to the individual as well as its relative importance compared to numerousother social concerns, such as AIDS and poverty. These two items were correlated, = .49in Pre-test 1 and r = .38 in Pre-test 2.The construct of threat perception was formed by combining responses pertaining toperceptions of the likelihood, severity, and immediacy of environmental destruction, (3 totalitems). Acceptable scale reliability was found in both pre-tests (Cronbach’s Aiphas of .73and .72, respectively).Measuring Value OrientationsPersonal value orientations were assessed using a new ipsative measure developed bythe author. Measuring values by having subjects rank order the importance of value domainsin relation to one another has been suggested as the most appropriate manner with which toassess value systems (Rokeach, 1973). This new device, modeled after the “Study of Values”by Aliport, Vernon, and Lindzey (1960), provided the means by which individuals could beassigned a value orientation (See Appendix 2 for item wording). The instrument calls forsubjects to assign a numeric value to each of three response categories (representing each ofthe three value domains) so that the total numeric value applied across all three responses isequal to 4. There are eight items in total. A sample item from the instrument is “Which ofthe following goals do you pursue more strongly: (a) achieving a sense of belonging and56acceptance from others; (b) attaining a comfortable and financially secure life; or (c)maintaining a sense of self-respect derived from acting in accord with deeply held values?”For example, a highly economically-oriented subject might assign a 4 to response (b), and avalue of 0 to the other two responses. Someone who is less economic in orientation, but stillvalues economic goals more highly than the other two value domains, might respond byassigning a 2 to (b), and a 1 to each of the other two options. Scale scores for each valuedomain were computed and four value groupings were created: economic, social, universal,and a mixed group in which no value domain dominated.Subjects’ value orientations for Study 1 were assigned initially based on whether or nottheir scores were greater or equal to one-half SD above the mean on one of the three valueorientation scales. After recruitment of subjects for this study was completed, a review ofsubjects’ value orientations suggested that dividing subjects into four groups based upon amore stringent criterion (greater or equal to one SD above the mean) would provide a moreuseful classification.In an attempt to classify subjects with more confidence, two criteria were employed todetermine whether a subject was to be assigned a dominant value orientation in Study 2.First, a score on a particular scale had to be at least one SD above the mean. Second, thescore needed to be higher than the scores on the other two scales. Approximately 52% of thepre-test subjects who provided usable data (n = 531) could be classified according to thisprocedure as either economically-oriented ( = 89), socially-oriented (n = 66), oruniversally-oriented (n = 120).Reliability and validity tests were conducted for the three value orientation scales. Inthe first pre-test, both the economic and universal scales had adequate internal consistency(Cronbach’s Alphas of .75 and .62, respectively). The reliability of the social orientationscale fell below usually desirable levels (Cronbach’s Alpha = .33). Implementation of slightmodifications in wording improved reliabilities in the second pre-test (Cronbach’s Alphas of.80 for economic, .74 for universal, and .48 for social). The lower reliability observed for57the social scale may suggest that this motivational domain is more complex in nature than theother two. Notwithstanding the relatively low reliabilities for the social scale, taken as awhole, the instrument appeared to exhibit acceptable internal consistency in classifyingsubjects’ value orientations.The validity of the instrument was explored in two ways. First in Pre-test 1,concurrent validity was examined by comparing the scales with subjects’ responses on theRokeach Value Survey (1967). Oneway analyses of variance were used to explore possiblegroup differences in the ranking of three terminal values selected a priori as the beliefs mostparallel to the three value orientations. Lower scores on the Rokeach values indicate a higherranking. Significant differences were found in the rankings of “a comfortable life”, F(80) =25.67, p < .001, “true friendship”, F(80) = 5.93, p K .01, and “equality”, F(80) = 2.93, p< .06, where the latter two Rokeach values were considered the most analogous to the social•and universal orientations, respectively. Tukey’s HSD tests revealed that economically-oriented subjects ranked “a comfortable life” higher (M = 4.2) than did both the social (M =10.9) and universal (M = 11 .5) value groups. Socially-oriented subjects ranked “truefriendship’ (M 4.4) significantly higher than did the economically-oriented subjects (M =7.7), but only marginally different than did the universally-oriented subjects (M = 5.3).Finally, universally-oriented subjects ranked “equality” (M = 8.7) significantly higher thandid economically-oriented subjects (M = 11.5), but only marginally higher than did subjectsin the social value group (M = 9.9).Second, in Pretest 2, subjects were asked to rate the importance of various life goalsusing a 9-point Likert scale. Three goals presumed to parallel the value orientations wereincluded: (1) “economic prosperity,” (2) “quality of friendships,” and (3) “acting in accordwith your values.” Analysis of variance procedures revealed, as would be expected, thatthere was a high degree of consistency between subjects’ value orientations and their ratingsof importance for the parallel goals. Thus, economically-oriented subjects rated “economicprosperity” significantly higher (M = 8.34) than socially-oriented (M = 7.03) or universal-58oriented (M 6.16) subjects, (283) = 43.99, p < .001; socially-oriented subjects rated“quality of friendships’ higher (M 8.63) than economically-oriented (M = 7.82) oruniversal-oriented (M = 8.35) subjects (though the difference was not significant in the lattercase), F (283) = 10.88, p < .001; and universally-oriented subjects rated “acting in accordwith your values” higher (M = 8.5) than economically-oriented (M = 7.05) and socially-oriented (M = 7.5) subjects, E(283) = 37.99, p < .001.Although the potential for socially desirable responding regarding one’s valueorientation must be considered (i.e., pursuing universal goals can be considered morecommendable than pursuing social acceptance or financial security), it can be concluded thatsocially desirable responding did not prevent subjects from endorsing “less” desirable optionsto any problematic degree. This conclusion is based on (a) the ease with which subjects couldbe assigned a value orientation, (b) the author’s attempt to present all the response options ina positive frame, and (c) the significant differences among the value groupings and theirrankings or ratings of parallel categories in the validity checks. To illustrate, even thoughmeans for the economic scale were, in general, lower than means for the social and universalscales (8.6 versus 11.7 and 11 .6, respectively in Pre-test 2), there was a substantial portion ofthe sample who readily met the criteria for being identified as economically-oriented. Thus,although there may have been some bias toward the universal and social response options, thisbias did not preclude a significant minority of respondents from reporting a high value for theeconomic goals. This pattern of responses may also stem from the use of a student sample inthat economic issues may be less salient to university undergraduates than to the population ingeneral.The relation between attitudes and values was examined prior to conducting theexperimental tests. Both pretests revealed similar patterns. Therefore only the results fromPretest 2 will be reported. As anticipated, favourable attitudes toward the environment wereassociated positively with a universal value [r(6l4) .35, p < .001] and negatively with aneconomic value [r(614) = -.34, p < .001]. General environmental attitudes were59uncorrelated with the social value. When comparisons were made among subjects whoembraced a particular dominant value orientation, similar findings were obtained.Specifically, analysis of variance procedures revealed a significant difference among the threevalue orientation groups [E(281) = 17.27, p < .001]. Thus, universally-oriented subjectsreported the most favourable attitude (M = 5. 12), economically-oriented subjects the leastfavourable attitude (M = 4.29), and the attitude of socially-oriented subjects fell between theother two groups (M = 4.8). Planned comparisons revealed that differences between pairs ofgroups were significant in all cases, p < .05. In the next three chapters, the role of thesefactors as guides to behavioural intentions is investigated in a multivariate format that includesother salient cognitive factors.60Chapter 6Study 1: Determining the factors that guide environmentally protective behaviourOverviewThis study was designed to assess the relative contributions of each factor presented inthe proposed framework as predictors of intentions to engage in an environmentally protectivebehaviour. Two analytic approaches were employed. First, expectancy/value beliefsregarding three types of possible behavioural outcomes, and self-efficacy beliefs wereexperimentally manipulated in order to examine their causal role in influencing behaviouralintentions. These constructs were selected because they were presumed to be beliefs that weremore susceptible to the planned manipulations than the other constructs. The effects of theseconstructs were also analyzed in terms of possible interactions with subjects’ valueorientations. Second, measures of outcome expectancies, outcome values, and efficacy beliefswere collected in order to assess their unique predictive utility in a multivariate procedure thatalso included the constructs measured by the pre-test.The behaviour of interest was subjects’ intentions to attend a workshop that wouldteach people how to recycle common household waste into handmade paper. This behaviourwas selected for three primary reasons. First, it represented an activity that would beperceived as beneficial to the environment, while also involving possible economic and socialbenefits. Second, although a workshop that teaches the paper-making process actually exists,very few students know about it. Therefore, it was presumed that the subjects recruited forthe experiment would enter the study with equally low information regarding thisenvironmental activity. This assumption was confirmed by the subjects during debriefing.Third, attending the workshop and using the paper-making process are activities that could beconvincingly presented as accessible to all subjects, regardless of possible demographicdifferences.Self-efficacy beliefs and outcome expectancies were manipulated factorially usingwritten descriptions of the workshop. This methodology was based on Maddux et. al (1986).61Self-efficacy was manipulated into high and low beliefs regarding subjects’ perceptions oftheir capability of learning and using the paper-making process. In addition, four outcomeconditions were created, three of which made salient one of three possible behaviouraloutcomes participants of the workshop would accrue if they attended the workshop: (a)economic, operationalized as saving money, (b) social, operationalized as impressing andearning respect from friends, or (c) environmental, operationalized in terms of helping theenvironment. A fourth description, which did not refer to any outcome expectancy, wasincluded as a control. Measures of self-efficacy, channel efficacy, outcome expectancies andoutcome values were collected, using questionnaires completed after the workshopdescriptions were presented.HypothesesBased on the framework, it was anticipated that correlational analyses would revealthat both efficacy beliefs and all three outcome expectancy variables would have significantunivariate correlations with intentions to perform the targeted behaviour. Thus, higher beliefsof self-efficacy and channel efficacy, and higher expectancies regarding the possibility ofsaving money, receiving social benefits, and helping the environment would all be associatedwith a greater intent to attend the workshop. Of the outcome value measures, only the valueplaced on helping the environment was expected to correlate with behavioural intentions. Theextent to which subjects valued economic or social outcomes as measured during theexperiment was not perceived as providing a direct motivation related to environmentallyprotective behaviour. Finally, only minimal, but possibly significant, univariate relationsbetween the general attitudinal constructs (i.e., general attitude, issue importance, and threatperception) and the target behaviour were expected. This hypothesis is based primarily on thedifferences in measurement (i.e, attitude objects were general and behaviour was veryspecific). Multivariate regression analyses were expected to show that both efficacy beliefsand outcome expectancies regarding all three outcome classes would explain significant levels62of variance in intentions to act, while any variance explained by the attitudinal variables andoutcome va]iie constructs would be negated.With regard to the experimental manipulations, two main effects were expected. First,a main effect for self-efficacy was anticipated, with subjects in the high efficacy conditionreporting stronger behavioural intentions than subjects in the low efficacy group. Second,considering that positive outcome expectancies should promote the performance of abehaviour, a main effect was predicted for the outcome manipulation (i.e., all threeexperimental groups would report higher behavioural intentions than the control group). Dueto the lack of previous research bearing on this question, no predictions were made regardingwhich outcome classes might inspire stronger intentions to act. An interaction was anticipatedbetween outcome condition and personal value orientation. Specifically, subjects exposed to aconsistent outcome condition (e.g., economically-oriented subjects presented with theeconomic outcome description) were expected to report stronger behavioural intentions thansubjects who received an inconsistent outcome description (e.g., economically-orientedsubjects receiving the social outcome description).MethodDesign and SubjectsThis experiment employed a 2 X 4 between-subjects factorial design, with high (n =60) and low (11 = 57) levels of self-efficacy, and three outcome conditions that made salientthe rewards to be derived from the specific behaviour representing each class of outcome:economic (11 = 31), social (ii = 29), environmental (n =30), and a control condition (n =27). These subjects were recruited from the population of respondents (n = 531) whocompleted the pre-test questionnaire. A total of 117 students participated in the experiment(over 80% of those contacted). The subjects ranged in age from 17 to 70, with an averageage of 20. Seventy-one percent of the subjects were female and 29% were male. Theexperimental. paradigm is typical of those used in verbal persuasion research (e.g. Maddux etal, 1986).63Stimulus MaterialsWritten communications were designed to manipulate the independent variables ofself-efficacy and outcome expectancy. These communications were presented as part of astudy sponsored by the fictitious UBC Department of Environmental Education to obtainopinions from students regarding several new workshops related to the environment that weresupposedly being considered by the university. It was thought that this cover story would beaccepted by the subjects and would control for possible demand characteristics that might biassubjectst responses. The material presented to each subject described, in detail, one newworkshop, though subjects were given the impression that other workshops were also beingevaluated. The workshop, entitled ‘Recycled Handmade Paper,” described a 1/2 day coursethat would instruct participants on how to use low-tech, low budget methods of turning muchof their household refuse into handmade paper products such as greeting cards, note pads, andwriting paper. In addition, to alleviate any artistic biases in evaluations, the basic descriptionindicated that no artistic ability or inclination would be needed to successfully use thistechnique.More detailed information was then provided regarding the workshop to help thesubjects evaluate their interest in it. The high self-efficacy description stated that the processtaught in the course is easy to learn, requires no special tools, and is almost always successful(“over 90% of all past workshop participants report that they learned and regularly use theprocess with their waste material”). The low self-efficacy description stated that the processcan be difficult to learn, requires some special tools, and is only sometimes successful (“onlyabout 10% of all past workshop participants report that they learned and regularly use theprocess with their waste material”).It was presumed that subjects would perceive implicit in the title of the workshop arelation to the environment. Therefore, the objective of the outcome expectancy manipulationwas to make salient a different outcome for each orientation. In the “economic” outcomeexpectancy condition, it was noted that using the technique is an efficient way to save money64(by making one’s own paper products), avoid the hassles of stores, and also possibly earnmoney (“handmade paper is a highly valued product and can be sold for extra income”). Inthe “social” outcome condition, the social consequences of attending the workshop wereemphasized. These included the ability to impress one’s friends and family, and join others inhelping the environment. In addition, it was noted that community leaders strongly endorseefforts to reduce waste, and encourage everyone to do their part. The “environmental”outcome condition noted the significant environmental benefits that can be gained by wastereduction and how this technique offered workshop participants the ability to help theenvironment. Although it was expected that all subjects would perceive the environmentalbenefits, it was thought that the environmental outcome description would increase subjects’expectancies regarding helping the environment. The lengths of the descriptions were allapproximately equal.Dependent MeasuresThe dependent measure and manipulation checks were assessed using a WorkshopEvaluation Questionnaire. A check on the validity of the self-efficacy manipulation consistedof two items concerned with one’s perceived ability to “learn” and “use” the handmade paper-making process. Subjects were asked to rate their agreement on a nine-point scale rangingfrom “strongly disagree” to “strongly agree” with the following statements: (a) “The processof making paper from waste materials would probably be easy for me to learn,” and (b) “Theprocess would probably be difficult for me to use” (reverse scored). These two items werecombined into one efficacy scale (r(l 17) = .57). Channel efficacy was assessed using oneitem, which asked subjects about their accessibility to the workshop in terms of timeconsiderations (i.e., whether or not they agreed or disagreed with the following statement,“Finding the time to attend this kind of workshop would be difficult for me.”) Themanipulation check for outcome expectancy consisted of three items that assessed the abilityof the course participants to benefit themselves or the environment as specified by the threeoutcome conditions (i.e., save money, impress friends, and help the environment). For65example, the environmental outcome expectancy item asked subjects to indicate how stronglythey agreed or disagreed with the statement “1 can contribute to reducing environmentalproblems by taking this workshop and learning the paper-making technique.” It was expectedthat subjects would identify the salient outcome (the one presented in their description) asmore likely to be gained from taking the workshop than those of the other two types ofoutcomes. Outcome values specific to the workshop were also assessed with three items thatasked subjects to rate the personal importance of each type of outcome to them (e.g., “Howimportant is it to learn ways to better your economic situation?”). Each item was assessed ona nine-point Likert scale. In addition to serving as manipulation checks, these itemsrepresented continuous measures of these constructs, which were used as independentvariables in correlational analyses.The major dependent measure, intention to attend the workshop, was assessed with oneitem asking subjects to indicate their interest in taking the workshop along a 9-pointcontinuum from “not at all interested” to “very interested.”ProcedureSubjects were recruited from the pre-test population based on their outcome valueorientations so that each outcome expectancy condition contained approximately equalnumbers of those with economic, social, universal, and mixed orientations. Under theprocedure as previously outlined, four vahie groups were created: economic orientation ( =31), social orientation (n = 23), universal orientation (n = 30), and mixed orientation (thosesubjects who did not score above one SD on any scale, n = 33). The different manipulationswere randomly assigned within these groups.Subjects were asked to report to a laboratory in the psychology building. When theyarrived, they were informed that the psychology department had been asked by the UBCDepartment of Environmental Education to study students’ opinions regarding some newworkshops being considered. Each subject was handed an envelope which contained thedescription of one of the new workshops, and asked to read the description and complete the66accompanying questionnaire. This task was completed in a private room with no observers.After subjects completed the questionnaire, which took approximately 15 minutes, allmaterials were collected.ResultsManipulation ChecksSelf-efficacy. A main effect was found for the self-efficacy manipulation.Specifically, subjects in the high self-efficacy condition reported higher levels of self-efficacythan did the subjects in the low self-efficacy condition, F(l,109) = 8.10, p < .01 (Ms =6.93 and 6.15, respectively). It is important to note that although the efficacy manipulationdid influence subjects’ beliefs of their capabilities of learning and using the paper-makingprocess, no group differences were found for the measure of channel efficacy.Outcome expectancy. A comparison among the three outcome conditions and thecontrol group on each outcome manipulation check item revealed that only the economicmanipulation seemed to work near expectations. A main effect was found for the economicmanipulation check: subjects who read the economic outcome description indicated that theyexpected to accrue more economic outcomes if they attended the workshop than did subjectsin the control condition, F(3,109) = 2.67, p < .06. Although subjects in the economiccondition did indicate higher levels of economic outcome expectancy than subjects in thesocial and environmental outcome conditions, these differences were not significant. Nodifferences were found for the social and environmental outcome expectancy manipulationchecks. This pattern of results suggested that subjects were not swayed in their expectanciesas a function of the manipulations. On average, subjects recognized that all three outcomescould result from the behaviour (means on al] three scales were above the scale midpoint).Environmental benefits were seen as most likely (M = 7.26), economic benefits were second(M 6.38), and social outcomes were the least expected (M = 5.56).Outcome value and value orientation. A comparison of subjects’ value orientationsand their responses to the outcome value items revealed expected patterns of convergent67validity. Universally-oriented subjects reported a higher value placed on beneficialenvironmental outcomes (M = 8.10) than did the economic group [M = 7.29, (3,l01) =2.92, p < .05], although they did not significantly differ from the other two groups[M(social) = 7.91; M(rnixed) = 7.61)]. Economically-oriented subjects reported a highervalue placed on “saving money’ (M = 8. 16) than the other groups [M(social) = 7.57,M(universal) 6.70, M(mixed) = 7.21; F(3,101) = 5.52, p < .011. Finally, socially-oriented subjects reported a significantly higher value on social outcomes (M = 6.74) thanuniversally-oriented subjects (M 5.13), F(3, 101) = 3.52, p < .02. The ratings ofsocially-oriented subjects did not significantly differ from either the economic group M =6.52) or the mixed group (M = 6.39). These results provide additional support for thedistinctions made among subjects regarding individual value orientations.Experimental analysesA 2 (efficacy) X 4 (outcome condition) X 4 (value orientation) Anova was performedon the behavioural intention item, yielding a main effect for self-efficacy, (1,85) = 3.65, p< .06, and no main effect for the outcome manipulations or for value orientation. Aspredicted, the high self-efficacy group reported stronger behavioural intentions than did thelow self-efficacy group (Ms = 6.7 and 6.0, respectively). Although no significant maineffects were found for outcome condition and value orientation, certain trends were observed.Subjects in the economic outcome condition reported somewhat higher behavioural intentionsthan all other groups including subjects in the environmental outcome condition - Ms = 6.74(economic), 6.38 (social), 6.13 (environmental), and 6.15 (control). In addition, universallyoriented subjects reported the strongest intentions to perform the behaviour: M = 6.97 ascompared to the other value orientation groups, M = 6.39 (economically-oriented), M =6.30 (socially-oriented), and M = 5.91 (mixed orientation). No significant interaction effectswere found.68Correlational analysesAs predicted, the behavioural intentions measure was correlated positively andsignificantly with self-efficacy, all three outcome expectancies, and outcome value pertainingto environmental outcomes, but not with the social or economic outcome value measures (seeTable 1). In addition, channel efficacy and the pretest measures of general attitude and issueimportance were positively correlated with behavioural intentions. Threat perception did notcorrelate with intentions to attend the workshop. A review of the intercorrelations among thepredictor variables revealed some interesting results. Most intriguing was the low (marginallysignificant) correlation between self-efficacy and channel efficacy [r(l 15) = .22, p < .05],supporting the notion that these variables represent different constructs. In addition, nosignificant correlations were found between general attitudes and the efficacy and outcomeexpectancy constructs. General attitude was correlated with environmental outcome value[(ll5) = .34, p < .01], but not with the other two outcome value domains.In order to examine the unique variance explained by each factor, a simultaneousmultiple regression analysis was performed. Results indicated that five factors accounted forthe significant portion of the explained variance in behavioural intentions (see Table 1). Aspredicted, both efficacy beliefs explained unique variance, whereas only two of the threeoutcome expectancies explained significant levels of variance. These were economic outcomeexpectancies and environmental outcome expectancies. In addition, in an enigmatic result,threat perception accounted for a significant portion of explained variance. Contrary topredictions, expectancies regarding social outcomes did not significantly explain variance inbehavioural intentions in the regression model. Consistent with expectations, no outcomevalue beliefs or attitudinal beliefs explained a significance amount of variance in theregression model. Overall, 54% of variance in behavioural intentions was explained [R(109)= .74J by the regression model.69ConclusionsThis study assessed the multivariate association among the factors presented in theproposed framework as cognitive guides to behavioural intentions. In addition, it examinedthe effects of self-efficacy and different types of expected outcomes, by independentlymanipulating each construct. Self-efficacy was successfully manipulated and had a maineffect on behavioural intention, adding to the evidence that supports the notion that self-efficacy has an important influence on behavioural intentions.Only one outcome manipulation (i.e., economic) appeared to be effective inmanipulating outcome expectations. This finding calls into question the strategy used in thisstudy to manipulate outcome expectancies. No main effect was found for outcomeexpectancy; however, an interesting trend emerged. Specifically, subjects who received theeconomic outcome manipulation reported the strongest intentions to act. This findinghighlights the importance of economic considerations when contemplating performing abehaviour directly involving environ mental protection.I predicted that subjects who received an outcome description consistent with theirvalue orientation would report higher intentions to act. In fact, the interaction betweenoutcome condition and value orientation was not significant in this study. This may be due toa lack of power as well as a result of ineffective manipulations. Further investigation isclearly needed to explore this hypothesis. One manner of effectively examining thisphenomenon would be to isolate an outcome class and manipulate expectancies regarding thatclass into different levels across the different value orientations. This design would be morepowerful in detecting the nature of the influence of a specific outcome class within each valuegrouping.The correlational analyses in this study confirm a number of past findings, and offersome new possibilities regarding the factors that guide behavioural decisions. First, furthersupport was found for the importance of self-efficacy and outcome expectancy as independentpredictors of behavioural intentions. Second, support was found for the new construct of70channel efficacy as an independent source with which variance in intentions to act may beexplained. Third, although the attitudinal factors did not account for any significant amountof explained variance in the regression procedures, their marginal but significant univariatecorrelations with the dependent variable provide evidence for their inclusion in theframework. Whereas the efficacy constructs and outcome expectancies combined to form themost parsimonious predictive model in this study, attitudinal factors are likely to beimplicated in the decision process. Their influence, however, is likely to be minor whenpredicting intentions to perform any one specific behaviour.In addition to these theoretical implications, the results found in this study haveimplications for the promotion of environmentally protective behaviour. They suggest thatthose interested in creating behavioural change should try to understand how individuals thinkregarding each behavioural guide, and subsequently focus their efforts on addressing thosebeliefs that may be blocking the desired action. In the environmental domain, where attitudesseem to be overwhelmingly pro-environmental (Gallup, 1990), current programs shouldattempt to increase levels of self-efficacy and channel efficacy regarding environmentalaction, as well as using economic incentives to motivate behaviour.The results of Study 1 also support the importance of multivariate investigations ofsocial behaviour. In particular, they begin to validate the utility of the proposed frameworkas a foundation from which researchers can investigate the influence of various factors asguides to social behaviours of interest. Whereas certain results were reasonably strong, othersmerely represented trends that require further study. Exploring some of these trends,particularly the influence of value orientations, will be a primary focus of Studies 2 and 3.71Chapter 7Study 2: A focus on the role of values and economic motives in the domain ofenvironmental behaviourOverviewStudy 2 was designed to address some of the ambiguous results found in Study 1. Itemployed the same procedures used in Study I with three important changes. First, in lightof the inability to influence either the social or the environmental outcome expectancy inStudy 1, the outcome expectancy manipulation in Study 2 focused solely on economicconsiderations. Specifically, economic outcome expectancies were manipulated into high andlow conditions in order to more directly assess the influence that economic considerationshave on an environmental behaviour. This methodological focus also allowed for a specificexploration of the interaction between value orientation and expectancies regarding thisimportant motive. Expectancies regarding all three outcomes were still measured so thatcorrelational analyses, which paralleled those in Study 1, could be performed.Second, in response to the high correlation between channel efficacy and behaviouralintentions found in Study 1, an attempt was made to examine the causal nature of thisconstruct. Therefore, beliefs regarding channel efficacy were manipulated by altering theaccessibility of the workshop into convenient and inconvenient conditions. The self-efficacymanipulation was retained to provide an opportunity to replicate the findings of Study 1.The last change involved increasing the power of the experimental design. Twomethods were used. First, the number of subjects per condition was increased. Second, thenumber of experimental groups was reduced. Thus, the outcome manipulation involved twogroups rather than four, and only subjects who embraced a dominant value orientation wereincluded in the study, thereby reducing the number of value groups from four to three.HypothesesMain effects for all independently manipulated constructs were predicted.Specifically, higher levels on each variable, self-efficacy, channel efficacy, and economic72outcome expectancy, would be associated with stronger behavioural intentions. I expectedthat universally-oriented subjects would report higher intentions to take the workshop thaneconomically-oriented subjects, and that the difference in behavioural intentions between highand low economic outcome expectancies would be greater for economically-oriented subjectsthan for universally-oriented subjects. In other words, it was anticipated that economically-oriented subjects would be significantly influenced by the economics of the situation, whereasuniversally-oriented subjects would not be influenced by the economic manipulation. Areplication of the correlational results found in Study 1 was expected.MethodDesign and SubjectsThis experiment employed a 2 X 2 X 2 between subjects factorial design, with high (= 72) and low (II 72) levels of self-efficacy, high (n = 73) and low (n 71) levels ofchannel efficacy and high (n = 71) and low ( = 73) levels of economic outcome expectancy.Subjects were recruited from the population of Pre-test 2 respondents ( = 531) who correctlycompleted the questionnaire (a different subject population than used in Study 1). As in Study1, the average age of the subjects was 20 with approximately 70% of the sample beingfemale. Using the criteria as specified in Chapter 5, the sample was approximately equallydivided among the three dominant value orientations: economic, 48; social, n = 47;universal, = 49.Stimulus MaterialsWritten communications were again used to manipulate the independent variables ofself-efficacy, channel efficacy, and outcome expectancy, and were presented under the sameguise as in Study 1. Self-efficacy was manipulated exactly as in Study 1. Beliefs regardingchannel accessibility were manipulated through changing the location where the workshopwould be held. The high channel efficacy group was informed that the workshop would beheld on the university campus, whereas subjects in the low channel efficacy condition were73informed that the workshop would be located at a relatively distant off-campus location -though one that was still accessible to students by mass transportation.With respect to the outcome manipulation, the high economic outcome expectancycondition group received the same description as presented in the economic outcome conditionin Study 1. The communication for the low economic outcome condition also referred to thepossibility of saving money, but in addition, stated “that using the (paper-making) process hasits setup costs in terms of purchasing material and tools that you may not currently own.”Emphasizing the potential costs was predicted to lower subjects’ overall expectanciesregarding the economic consequences from attending the workshop and using the paper-making process, while still appearing to be within reason. If the negative condition wasperceived as too strong, skepticism regarding the intent of the study may have been raised inthe subjects’ minds, and this could have contaminated the results.Dependent MeasuresThe dependent measures and manipulation checks remained the same as in Study 1,with the addition of a new measure of behavioural intention. This new measure askedSubjects more directly, “If this workshop was offered later this year, how likely is it that youwould attend?” This revision was intended to enhance the face validity of the dependentmeasure. Responses to this item were correlated r( 140) = .92 with responses to the sameitem as used in Study 1. Clearly, no significant differences existed between these measures.Thus, all results described in this report are based on responses to the new behaviouralmeasure.ProceduresRecruitment and experimental procedures were identical to those used in Study 1.Resu ItsManipulation ChecksSelf-efficacy. The self-efficacy manipulation was successful in influencing selfefficacy beliefs. Subjects in the high condition (M 13.06) reported higher efficacy beliefs74than subjects in the low condition (M = 11.36), F(120) = 10.4, p < .01. Channel efficacybeliefs and outcome expectancies were unaffected by this manipulation. Furthermore, novalue differences were observed regarding self-efficacy beliefs.Channel efficacy. The channel efficacy manipulation was not successful in alteringperceptions of accessibility to the workshop. This may be due to the fact that many studentslive off campus at UBC and would actually have had equal access to the two workshoplocations. Unfortunately, this possibility was not recognized at the time of the research.Economic outcome expectancy. No main effect was found for economic outcomeexpectancies, but a very anomalous significant interaction was observed between the outcomemanipulation and value orientations [F(120) = 3.3, p < .05J. As expected, economically-oriented and universally-oriented subjects in the high economic outcome condition (Ms forboth = 6.25) reported higher expectancies than their counterparts in the low outcomecondition (Ms = 5.21 and 5.6, respectively). Quite unexpectedly, the reverse pattern wasobserved for socially-oriented subjects. These subjects appeared to have greater economicoutcome expectancies in the low condition (M = 6.5) than in the high condition (M 5.64).None of these differences was significant at the p < .05 level, nevertheless the trend is aparadoxical finding.Outcome value and value orientation. The consistency between subjects general valueorientations and the importance they ascribe to economic, social, and environmental outcomesas reported during the study was explored. As in Study 1, significant group differences werefound for all three outcome value measures in expected directions [economic, F(120) = 7.02,p < .01; social, F(120) = 8.74, p < .01, and environmental, F(120) = 4.13, p < .02, seeTable 2 for group comparisons). Post-hoc analyses showed that economically-orientedsubjects reported higher importance for economic outcomes than did universally-oriented orsocially-oriented subjects. In contrast, universally-oriented subjects rated environmentaloutcomes as more important than did the other two value groups. Finally, socially-oriented75subjects rated social outcomes as more important than did universally-oriented subjects, butthey did not differ from subjects with an economic value orientation.Experimental AnalysesAs predicted, a main effect was found for self-efficacy. Thus, subjects in the highself-efficacy group (M 5.82) reported significantly higher intentions to attend the workshopthan did subjects in the low efficacy group (M = 4.94), F(l20) = 4.83, p < .05. No maineffect was found for channel efficacy, probably due to the ineffective manipulation. Inaddition, no main effect was found for economic outcome expectancy, although there was amarginal outcome expectancy by value orientation interaction that will be described later.There were significant differences in behavioural intentions across the three value orientations[(120) 4.37, p < .02). As anticipated, universally-oriented subjects reported the highestbehavioural intentions (M = 6. 10), followed by the socially-oriented group (M = 5.34) andthe economically-oriented group (M = 4.69). Post-hoc comparisons revealed that only theuniversal value group and the economic value group differed significantly, p < .05.A marginally significant trend was found for the interaction between value orientationand economic outcome condition [F(120) = 2.16, = .12]. Planned post-hoc comparisonsrevealed, as predicted, that only economically-oriented subjects appeared to be significantlyinfluenced by the manipulation (see Table 3). Thus, economically-oriented subjects in thehigh economic condition reported stronger intentions to attend the workshop than dideconomically-oriented subjects in the low economic outcome condition. In contrast,universally-oriented subjects were not influenced by the economic outcome manipulation. Nosignificant differences were found for socially-oriented subjects.Correlational AnalysesThe correlational analyses in this study replicated the approach employed in Study 1.Consistent with hypotheses, expectancies regarding all three classes of outcomes and bothself-efficacy and channel efficacy had significant univariate correlations with behaviouralintentions (see Table 4). Once again, only a moderate correlation was found between self-76efficacy and channel efficacy [r(140) = .29, p < .05], confirming the relative independenceof these constructs, Of the attitudinal variables, issue importance and threat perceptionshowed small but significant correlations, whereas general attitude was not correlated withintentions to act. This pattern was slightly different than that observed in Study 1.To isolate the unique contribution of each of these factors a simultaneous entrymultiple regression analysis was performed (results shown in Table 4). Following the patternfound in Study 1, both efficacy variables and two of the three outcome expectancies accountedfor significant variance in behavioural intentions. In contrast to the findings in Study 1,econornic outcome expectancies did not remain significant in the regression model, whereassocial outcome expectancies did account for a substantial portion of explained variance. Thisresult shows the susceptibility of the regression procedure to slight changes in the correlationsamong the independent variables and the criterion. In all, 54% of the variance in behavioralintentions was explained, primarily by differences in efficacy beliefs and outcomeexpectancies.ConclusionsResults confirm many of the findings from Study 1, and suggest some additionalconclusions. Evidence supporting the importance of self-efficacy as a determinant ofbehavioural intentions was once again found. Although experimental results werecompromised due to the ineffective manipulation, correlational analyses supported theindependent role that beliefs regarding channel accessibility play in guiding behaviouraldecisions.The influence of outcome expectancies and value orientations on behavioural intentionswas much stronger in Study 2 than found in Study 1, both in terms of group differences andinteractions between the two constructs. Universally-oriented subjects were clearly the mostinterested in performing the environmentally protective action, and were unaffected bycontextual variations iii economic outcome beliefs. In contrast, whereas economicallyoriented subjects were least likely to report intentions to engage in the behaviour of interest,77their likelihood of performing the behaviour increased significantly when more beneficialeconomic outcomes were perceived. At present, the responses of socially-oriented subjectsare difficult to understand. Nevertheless, based on the correlational evidence, expectanciesregarding social outcomes do appear to be salient guides to subjects’ behavioural decisions.Perhaps the most interesting finding in this study is the involvement of personal valuesin these decisions. Personal value orientations seem to be directly involved in guidingbehavioural decisions regarding the environment, as well as interacting with outcomeexpectancies to influence behavioural intentions. Whereas the role of attitudinal factors inthese decisions is best described as indirect and inconsequential, deeply rooted convictionsregarding personal desires are strongly implicated. These results suggest certain appliedimplications. It appears that people who embrace a universal value orientation will bepredisposed to act in environmentally protective ways, probably because doing so satisfiestheir need to contribute to the ecological betterment of their world. In contrast, economicreinforcement is an important factor in prompting economically-oriented individuals to engagein environmentally protective behaviours. Social considerations appear to play a role in thesedecisions, but how they manifest themselves is uncertain.In sum, results of Study 2 lend further support to the utility of the proposedframework as a valid research tool, particularly the influence of self-efficacy, outcomeexpectancies, and personal values on behavioural decisions. Attitudinal factors were found tobe weakly and indirectly related to behavioural intentions. These findings are consistent withother research that has attempted to predict a specific behaviour from a general attitude. Ofthe three attitudinal factors, the attitude importance construct seemed most implicated.78Chapter 8Study 3: Exploring how values guide behavioural decisions in ecological dilemmasOverviewIn the first two studies, the framework was shown to be useful in explaining andpredicting intentions to perform one specific environmentally protective behaviour. Twogroupings of predictors, efficacy and outcome expectancies, accounted for the most variancein behavioural intentions, whereas personal values were also implicated in these decisions.Study 3 investigated more thoroughly the influence of such values in different behaviouraldecisions regarding the environment. Exploring the interaction between personal values andthe specified domains of outcome expectancies was of particular interest.Specifically, the goal of this study was to ascertain how individuals react when placedin hypothetical situations of personal ecological conflict, and to determine which behaviouralguides account for variance in their responses. Motivated by the often noted suggestion thatcurrent ecological problems are a result of a “crisis” in human values (e.g., Armstrong, 1972,Milbrath, 1986), a secondary focus of this research was to determine how personal values areimplicated in important and controversial ecological behavioural decisions. Several issueswere addressed with respect to this objective: (1) how might individual differences in theweight placed on different values affect decisions regarding environmental preservation -particularly when preservation is in conflict with other personal desires; (2) how mightdifferent contexts, in terms of personal economic and social circumstances and/or outcomeexpectancies, influence individual choice, and (3) how might attitudinal factors be involved inthese decisions?To examine these issues, three hypothetical dilemmas involving ecological issues weredeveloped. The use of hypothetical scenarios in research examining psychological processeshas a long and productive history, particularly in the area of moral development andbehaviour (e.g., Forsyth, 1985; Kirtines, 1986, Kohlberg, 1958) and decision-making (e.g.,Tversky & Kahneman, 1981). They can present diverse settings in which subjects can be79called upon to make decisions in a format conducive to empirical examination. It wastherefore reasoned that this methodology would be appropriate for the purposes of the presentstudy.The scenarios were developed based on two guidelines. First, the situations werepresented in very simple terms, providing only a limited amount of detail. Describing thedilemmas in this manner required subjects to rely on their own beliefs and values as theprimary bases for their responses. Second, the scenarios represented dilemmas that to someextent model actual situations that people and/or communities may actually encounter.The first scenario described a variant of the “Commons’ dilemma (Hardin, 1968). Arancher has to decide how to act when a short-term economic gain can be obtained, but at thecost of neglecting commitment to the other ranchers who share common grazing land andpossibly contributing to the irreversible depletion of the shared resource. A second situation(termed the “Harvest” scenario) models the conflict faced by many timber communities. Itpresents the choice between economic rewards (i Se., employment) and environmentalpreservation (preventing the harvesting of old-growth trees). The third scenario (referred toas the “Waste” scenario) represents the real-life problem of waste management. It asks peopleto decide between two methods of waste management, one that is considered moreenvironmentally protective (intensive recycling) versus one that ignores the mounting wasteproblem, but would be less intrusive into people’s lives.Each scenario pits economic motivations against those of environmental preservationand protection, while also examining the role that social outcome factors play in decisionmaking. In order to test the causal relation between these motives and individual choice, twoversions of each scenario were examined. In one version, the scenario described a situationwhere the choice in favor of economic gain was in direct conflict with the environmentallyprotective choice. In addition, in two of the three dilemmas social pressures were describedthat supported the “economic” decision. This manipulation was included in order to exploresubjects’ responses to social influences. No social influence manipulation was included in the80Waste dilemma, because the scenario was thought to be too personally relevant for ahypothetical social influence intervention to have an effect. In contrast, the other versionpresented a situation where economics should only factor minimally into one’s decision andno social barriers were placed in conflict with the environmentally-responsible choice.The goal of the manipulation was to create a high level of conflict between ecologicalgoals and other motivations in one condition, and low conflict in the other condition. For thisreason, economic and social influence pressures were combined in order to maximize thepotential for conflict. For example, in the high conflict version of the Commons dilemmascenario, the rancher owns a small ranch, is in significant economic need, and has no socialties to the other ranchers who share the common grazing land. In the low conflict version,the rancher owns one of the larger ranches, is doing quite well economically, and is friendswith the other ranchers. It should be noted that the social manipulation addressed theinfluence that important others may have on people’s decisions, and not the perceived socialbenevolence of the situation. A social benevolence interpretation of each scenario would leadto specific predictions independent of the social influence manipulation. As it turned out, thesocial influence manipulation seemed to be greatly overshadowed by the responses to theeconomic manipulation. Thus, I refer to the experimental manipulation in terms of theconflict between the economic and envi ron mental interests.HypothesesFour primary hypotheses were tested in this study:1. Subjects value orientations will significantly influence their decisions in eachscenario. Specifically, I predict that economically-oriented subjects will be more likely toselect the economical I y-respon sib] e option while universal] y-oriented subjects will tend tochoose the environmentally-protective option. This pattern will be observed in all threedilemmas. The decisions of socially-oriented subjects will vary. In the Commons scenario,socially-oriented subjects will select the environmentally protective option primarily becauseof the implied social contract between the ranchers. In the Harvest scenario, they will81respond in the economically responsible manner because this option will be perceived asoffering the greatest good for the greatest amount of people. No predictions were maderegarding this group and the Waste scenario.2. Whereas the experimental manipulations (particularly the economic situation) mayhave a main effect on subjects’ behavioural intentions, it will be the interaction between thecontext and individual value orientations that is of interest. It is expected that the differentcontexts will interact with subjects’ value orientations to influence their decisions. Thus,universally-oriented subjects will not be influenced by changes in economic need or by anysocial pressures (the experi mental manipulations), whereas economical] y-oriented subjects willbe influenced by the changes in economic situation. Although it is presumed that socially—oriented subjects will be intluenced by personal social pressures inherent in a situation, thismanipulation was considered exploratory and not much confidence was held in manipulatingthese pressures. Therefore, specific hypotheses regarding its influence were not offered.3. It is expected that motives that represent all three value orientations will bereported and rated as important. However, strong value differences in the extent to whichthese motives are reported as influential should be found. Thus, the economically-orientedsubjects will emphasize economic need and their right to economic gain as central to theirdecisions, the socially-oriented will report social benevolence and social influence factors asguiding their decisions, and the universally-oriented subjects will rate ecological concerns, aswell as adhering to one’s core values, as most important.4. A moderate relation between attitudinal factors (i.e., general attitude and issueimportance) and behavioural decisions is expected, especially in the Harvest and Wastedecisions. This expectation is partly due to the high degree of relevance of attitudes towardthe environment and the issues presented in these scenarios, issues perceived as more relevantto attitudes than the criterion in the previous two studies. No attitude-behavioural intentionrelation is expected in the Commons dilemma decision. It is believed that the economic and82social aspects implicated in this particular dilemma will overshadow any potential influencethat attitudes may have in guiding responses.The methods employed in this study also allowed for an examination of certainhypotheses regarding the experience of cognitive dissonance. It was presumed that thedilemmas would create a certain amount of internal value conflict within each decision-maker.This conflict would be experienced differently based on the value system embraced by thesubject and the scenario in which she or he was placed. In general, it was anticipated thatsubjects would experience more conflict when positive environmental outcomes were pairedwith negative economic outcomes than when they were not. In addition, it was hypothesizedthat subjects who make a value-consistent decision will report less inner conflict than willsubjects who select a value-inconsistent option. So, economically-oriented subjects will reportmore conflict when they chose the environmentally-protective option, thereby passing up aneconomic opportunity. In contrast, both universally-oriented and socially-oriented subjectswould report more conflict when they chose the option that was not consistent withenvironmental protection.MethodExperimental Design and SubjectsThis experiment employed a 2 X 3 mixed factorial design, in which high (n = 72) andlow (n = 72) levels of economic-ecological conflict were presented to subjects representingthe three value orientations- economic (n = 48), social (11 = 47), and universal (n = 49) foreach of three hypothetical situations. The subjects recruited for Study 2 were employed asparticipants for this study as well. Subjects were informed that the two studies were part ofdifferent and unrelated research programs.Stimulus MaterialsSubjects were administered questionnaires containing descriptions of all threehypothetical ecological di lern mas. The “Corn mons” di leni ma described a situationencountered by a rancher who shares a common grazing land with nine other ranchers.83Specifically, the dilemma was described as follows: (Note: bracketed sentences represent themanipulations where the high conflict condition is described first and the low conflict isdescribed second.)“Bill shares a common area of grazing land with nine other ranchers. [Though all theranchers are business associates, they rarely socialize and are not really friends with eachotherj/[As well as being business associates, all the ranchers are friends]. All the ranchersknow that the common grazing land is the perfect size for the total amount of cattle thatthey collectively own. Any more cattle would overuse the grazing land, and if all theranchers bought extra cattle the land would be ruined. The ranchers have an “unwritten”commitment not to overuse the common grazing land. Bill comes across a special dealwhere he can purchase a number of additional cattle for a very low price. [Bill owns oneof the smaller ranches and has been hurt by the bad economic times.]/[Bill owns one of thelarger ranches and has actually done well financially even during the bad economic times.]He knows he would be violating the ranchers’ agreement, but he plans to keep the newcattle for only a short time. Then everything would return to normal.In the Harvest Dilemma, a woman who lives in a small town in British Columbia isfaced with a decision of whether or not to try to prevent the harvesting of a large area of oldgrowth forest near her home. This dilemma read as follows:“Jane has lived in a small town in British Columbia all her life. The main source of jobs inthe town is the local lumber company. Jane has found out that the company is planning toharvest a large area of old growth forest land which the company owns, that up untilrecently has been set aside as parklaiid. It’s been. a slow period for the town and this newproject means jobs and income for a number of years. [Most of Jane’s friends andprevious co-workers are very excited about the new harvest, and want to see the projecthappen. Both Jane and her husband were recently laid off by the company and will berehired when this new harvest begins.]/[Most of Jane’s friends are very concerned aboutthe new project, and are troubled by the harvesting of old growth forest land. Althoughthe harvest does mean jobs for many people in the town, both she and her husband alreadyhave well paying jobs and would be unaffected by the new project.] She has heard that oldgrowth forests can’t be regenerated and the current ecosystem would be severelydamaged.”The third dilemma dealt with the current problem many communities face regardingwaste disposal. This dilemma was described as follows:Tom and Mary live on the outskirts of a medium size city. The regional area has agrowing problem with what to do with its garbage. The landfill site is almost at capacity.The local governmental department of waste management has developed two proposals todeal with this situation. The first is to build a second landfill site. The best location forthe new site is an area currently designated to be a new park. [The land already belongs tothe regional district and no new fees wotld be required under this proposal.]/[This84proposal would result in the residents of the greater regional area being charged with amonthly fee for garbage collection specifically to pay for the new site.] The secondproposal would involve creating a new collection system that would force all people toseparate their garbage into seven categories - six for recycling and one for composting.Containers would be provided by the waste department and substantial fines would becharged to anyone who doesn’t abide by the new regulations. [Additionally, the residentsof the greater regional area would be charged with a monthly fee in order to pay for theadded costs necessary to run this waste collection system.]/[No new fees would be requiredunder this system.] The government is planning to hold a public vote on this issue nextweek.Dependent MeasuresAfter reading each scenario, subjects responded to four sets of issues. First, theyindicated what behavioural decision they would make if they were in that situation. For the“Commons” dilemma, this meant either purchasing or not purchasing the additional cattle.For the Harvest dilemma, subjects had to choose between “actively” supporting the harvestand “actively” opposing the harvest. For the Waste dilemma, the choice was between votingfor building the new landfill or voting for creating the new collection system. Second,immediately following their decisions, subjects were asked to rate how much conflict theywould feel if they were in that situation on a 7-point scale ranging from “no conflict” to “a lotof conflict.”Third, subjects were asked to “describe,” in an open-response format, the factors thatinfluenced their decision . Subjects were prompted to list all the factors that they thoughtwere important. This procedure was employed! in order to let the subjects indicate the reasonsthat influenced their decisions without prompting from the experimenter. Responses from 20subjects were coded by two independent raters, and general categories were identified.Reliability for the three dilemmas was: (a) Commons: 91.4%, (b) Harvest: 91.2%, and (c)Waste: 94%. Once reliability was established and disagreements were resolved, one ratercoded the remaining data.Fourth, for each of the three dilemmas subjects were asked to rate how important eachof twelve reasons (found through pilot testing to be implicated in guiding people’s decisions)were in making their decisions. These ratings were made on 5-point scales, which rangedfrom “no importance” to “great importance.” Included were items that represented aspects of85each of the three value domains. Examples of items used for the Commons dilemma are:“Your short term financial needs,” “Whether the rights of the other ranchers would beviolated,” and “Whether purchasing the cattle violates one’s moral or ethical code.”Examples from the Harvest dilemma are: “How the decision would affect your family’sfinancial situation,” “One’s relationship with the townspeople,” and “What the costs to theenvironment would be.” Examples from the Waste dilemma are: “Your ability to pay thenew fee,” “Which proposal is supported by your friends and family,” and “How eachproposal would affect the environment.” Finally, from this list of twelve reasons subjectswere asked to rank one as ‘most important” in making their decision.ProcedureParticipating subjects were instructed to read each scenario, and then complete thequestionnaire that followed. By random assignment, the Commons scenario was alwayspresented first, followed by the Harvest scenario, and then the Waste scenario. Theexperimental conditions were counterbalanced, resulting in eight possible groupings (i.e., highor low conflict in the Commons scenario, high or low conflict in the Harvest scenario, andhigh or low conflict in the Waste scenario). No order effects were found. When subjectscompleted all three scenarios, they returned their completed questionnaires, and were fullydebriefed as to the intent of the study.Resu ItsAcross conditions and subjects, there was a strong tendency to select theenvironmentally protective options. In the Commons dilemma, 82% of the sample reportedthat they would purchase the cattle, while 18% indicated they would accept the “specialdeal.” The responses to the Harvest dilemma were more balanced. Sixty-four percent of thesample indicated that they would actively oppose the harvest, while 36% supported theharvest. Overwhelming support for the recycling option was found in the Waste dilemma(91 %), whereas only 9% of the sample decided to vote for the new landfill.86Influence of Value OrientationsTable 5 presents the percentage of subjects that selected the environmentally protectiveoption for each value orientation. Significant value differences were found for both theCommons ( = 6.05, p < .05) and Harvest dilemmas (XZ = 15.1, p < .001). Plannedcomparisons among the three groups revealed patterns consistent with the first hypothesis.Thus, universally-oriented subjects were more likely to report that they would engage in theenvironmentally protective option than were subjects with an economic orientation. Thisoccurred in both the Commons dilemma and in the Harvest dilemma.Also as predicted, the responses of sociaLly-oriented subjects varied as a function ofthe social benevolence aspects of the dilemmas. In the Commons scenario, they selected theenvironmentally-protective option (as did the universally-oriented subjects) at a significantlyhigher rate than did the economically-oriented subjects. In contrast, their responses paralleledthe economically-oriented subjects in the Harvest dilemma, selecting the environmentally-protective option less often than did the universally-oriented subjects. Although notsignificant, the results for the Waste dilemma did show a trend similar to the findings in theCommons dilem ma, where socially-oriented and universally-oriented subjects favored theenvironmentally protective action more than did economically-oriented subjects.Effects of ManipulationsThe only significant manipulation effect was found in the Waste dilemma, where theenvironmentally-protective option (recycling) was supported less (85% compared to 97%)when it was associated with the additional economic cost (X = 5.29, p < .02). Subjectsalso supported the economically-advantageous option somewhat more often in the highconflict condition than in the low conflict condition of the other two dilemmas (22% to 14%in the Commons dilemma; 42% to 30% in the Harvest dilemma), but the differences were notstatistically significant.Of greater interest was the interaction between the experimental manipulations and thevalue groupings. Table 6 presents the results of these analyses. Although only the Harvest87dilemma yielded a significant interaction (X = 6.75, p < .0 1), a similar pattern could beobserved in all three dilemmas. Thus, as predicted, it was found that economically-orientedsubjects were influenced by the contextual manipulations, whereas universally-oriented andsocially-oriented subjects were not influenced by these manipulations. This interaction wasparticularly strong in the Harvest dilemma, where 29.2% of economically-oriented subjectssupported preserving the forest in the high conflict condition, whereas 70.8% supported thepreservation option in the low conflict condition. Clearly, the economically-oriented subjectswere substantially influenced by the personal economic need factor, whereas the subjectsembracing a different value orientation were not. The underlying justifications for subjects’decisions are discussed in the next section.Underlying MotivesIn general, the motives that subjects described as important and as influencing theirdecisions were consistent with their value orientations. The relation between underlyingmotives and value orientations was assessed in three ways. Responses to the open-endedquestion were explored first, using a content analysis approach to determine global categoriesof reasons for each dilemma. From these data, I examined individual value differences interms of the frequency with which certain motives were reported.Next, comparisons among value orientations were conducted on the 12 importanceratings presented to the subjects. These analyses were performed for each dilemmaindependently, using 3-way Anovas with condition, value orientation, and decision as theindependent factors, and the rating of each reason as the dependent factor. Because the focusof this study is on the role that value orientation plays in these decisions, only differencesregarding value orientation are reported. When overall significant differences were observed,Tukey’s HSD tests were used to determine specific differences between individual groups.Last, value differences in the ranking of the “most important” factor influencingsubjects’ decisions were explored. For these analyses, individual factors were grouped intothree categories that directly paralleled the three motivational domains (i.e., economic, social,88universal). Each category was comprised of two, three, or four individual reasons. Oncecategorized, the percentages of subjects that ranked the motives in a category as mostimportant were compared across the three value orientations.Open-ended responses. For the Commons dilemma, seven categories of motives werefound. The two most often cited reasons were ‘economic considerations” (e.g., personalfinancial need) and “social considerations” (e.g., effect on other ranchers) with 74.5% of thesample in each case reporting the factor as important in influencing their decisions. The nextmost reported factor (50% of the subjects) was adhering to “the agreement” not to graze anymore cattle. The other four reasons given were: (a) the “effect on the environment” (47%),(b) “moral considerations” (i.e., subject must have specifically referred to “moral,” “ethical,”or personal “principles” as influencing their decision- 20%), (c) “general futureconsiderations” (i.e., with respect to no specitc outcome of concern- 10%), and (d) “notgetting caught” (2.8%). Counter to expectations, the pattern of responding in this dilemmadid not significantly differ across value orientations. One marginal difference was found for“economic consideration SI, where economically-oriented subjects (85.4%) reported thisconcern more often than did socially-oriented (63.8%) or universally-oriented subjects[(73.5%), = 5.82, p < .06)].In the Harvest dilemma, eight categories of motives were identified. Here the mostcited factor was “damage to the ecosystem” (e.g., harvesting the old-growth forest woulddestroy the ecosystem) with 63.4% of the subjects reporting this influence. The othercategories in order of reporting were: “economic needs of the community” (43.4%),“personal economic need” (35.9%), “general harvest considerations” (e.g., can the harvest bealtered in some way- 29.7%), “personal social considerations” (e.g., what friends think-13.1%), “possibility of finding jobs elsewhere” (16.6%), “forests are a renewable resource”(6.2%), and “moral considerations” (4. 1 %).Value differences, in expected patterns, were found regarding two factors underlyingsubjects’ decisions in this dilemma. Universally-oriented subjects listed “damage to the89ecosystem” (75.5%) more often than did socially-oriented (66%) and economically-oriented(47.9%) subjects (X = 8. 17, p < .02). As would be expected, this pattern was reversed for“personal economic considerations.” Whereas 47.9% of economically-oriented subjects listedthis as an important factor, only 36.2% of socially-oriented subjects and 24.5% ofuniversally-oriented subjects reported this factor as influencing their decision (X 5.77, p< .06).Nine categories of reasons were described as influencing subjects’ decisions in theWaste dilemma. The most often cited reason was “economic considerations” with 62.8% ofthe subjects listing this concern. This motive was closely followed by “effect on theenvironment” where 53.8% of the sample reported this outcome as important in guiding theirdecision. The other seven reasons were as follows: (a) “importance of the community park”(33.8%), (b) “landfill oniy a temporary solution” (31 %), (c) “promote recycling” (28.3%),(d) “recycling not difficult to do” (e.g., does not take a lot of time - 25.5%), (e) “socialconsiderations” (9.0%), (f) “recycling too difficult to do” (e.g., takes too much time- 6.9%),and (g) moral considerations (6.9%). Two areas of value differences were found. First,universally-oriented (59.2%) and socially-oriented (66%) subjects listed “effect on theenvironment,” whereas only 37.5% of the economically-oriented subjects did so = 8.5, p< .02). Similarly, universally-oriented (38.8%) and socially-oriented (42.6%) subjects weremore likely to refer to the importance of the park than economically-oriented subjects (18.8%;7.05, p < .05).Importance Ratings. Substantial support for the third hypothesis was found whendifferences in importance ratings were examined for each of three dilemmas. In theCommons Dilemma (see Table 7), economically-oriented subjects rated “short-term financialneeds” as more important than did socially-oriented and universally-oriented subjects, (129)= 4.71, p < .001. In contrast, socially-oriented subjects rated ‘the relationship one has withthe other ranchers,” F(l31) 6.71, p < .002, and “what one’s family thinks” E(130) =4.02, p < .05, as more important than both economically-oriented and universally-oriented90subjects. In addition, socially-oriented subjects rated “Whether the rights of the otherranchers would be violated” and “How much purchasing the cattle could hurt the otherranchers financially” as more important than did economically-oriented subjects, but thesocially-oriented subjects did not differ from universally-oriented subjects on these reasons,F(130) = 8.14, p < .001; F(129) = 4.07, p < .05, respectively. Universally-orientedsubjects rated “adhering to one’s core principles” as substantially more important than dideconomically-oriented subjects and marginally more important than socially-oriented subjects,F(131) = 8.01, p < .01. Furthermore, universally-oriented subjects rated the importance ofmoral and ethical considerations as more important than did economically-oriented subjects,F(131) = 9.21, p < .001. Various other significant and non-significant comparisonsfollowed this pattern.Consistency between motives and value orientations was found when the importanceratings were examined for the Harvest decision (see Table 8). In this case, economically-oriented subjects rated “How the decision would affect your family’s financial situation” asmore important than did socially-oriented and universally-oriented subjects, F(l30) = 8.02, p< .01. In contrast, universally-oriented subjects rated the importance of “What the costs tothe environment would be” as significantly higher than economically-oriented subjects, andmarginally higher than socially-oriented subjects, F(130) = 3.74, p < .05. In addition,socially-oriented subjects rated “One’s relationship with the townspeople” as significantlymore important than did economically-oriented and universally-oriented subjects, F(129) =2.54, p = .08. No other significant differences were observed.A similar pattern of responses was found in the Waste Dilemma (see Table 9).Economically-oriented and socially-oriented subjects rated “how much each proposal will costthe residents of the community” as more important than did universally-oriented subjects,F(131) = 3.55, p < .05. In contrast, universally-oriented subjects rated “adhering to coreprinciples related to the decision,” F(13l) 5.67, p < .01, and “how each proposal wouldaffect the environment,” F(132) = 4.5, p < .02, as significantly more important than did91economically-oriented subjects and marginally, but not significantly, more than did socially-oriented subjects.“Most important” ratings. Further support for these value differences was found in thesubjects’ rankings of the “most” important reason (see Table 10). In all three dilemmas,economically-oriented subjects ranked economic factors as “most important” in guiding theirdecisions significantly more often than did socially-oriented and universally-oriented subjectsin all three dilemmas. Socially-oriented subjects ranked social factors as “most important”more often than did economically-oriented and universally-oriented subjects in both theCommons and Harvest dilemmas, but not in the Waste dilemma. Finally, as was anticipated,universalistic (e.g., protecting the environment, adhering to core values) factors were rankedas most important by a great majority of universally-oriented subjects in all three dilemmas.These percentages were significantly higher than the percentages of economically-oriented andsocially-oriented subjects who did so.Attitudes, values, and behavioural decisionsThe attitude-behavioural decision relation was analyzed independently for eachdilemma. As predicted, both general attitude (r = .35, p < .05) and issue importance (r =.37, p < .05) were significantly predictive of decisions in the Harvest dilemma, howeverneither was significantly correlated with decisions in the Commons dilemma. Contrary topredictions, no relation existed between either attitudinal factor and behavioural decisions inthe Waste dilemma.To explore more fully the different contributions made by attitudinal factors and valuesin accounting for variance in behavioural decisions, a simultaneous entry multiple regressionprocedure was employed to analyze the Harvest dilemma decision, the only scenario whereattitudes were implicated. In addition to including both attitudinal variables in the regressionmodel, five other independent constructs were included. These were: (a) a variable thatrepresented the different outcome conditions from which the decision was made; (b) twovariables that reflected the differences found between value groups; and (c) two variables that92represented the interaction between values and the outcome manipulation. Coding methodsoutlined by Cohen (1968) were used to include these variables in the regression model.Results from this analysis showed that each independent factor accounted for asignificant portion of explained variance (see Table 11). Overall, 29% of variance in thebehavioural decision was explained [F(13l) = 9.06, p < .001]. Personal values accountedfor 16.4% of variance in the Harvest decision. Regression analyses performed on thedecisions made in the other two dilemmas found no single factor or group of factors able toexplain a significant portion of variance.Conflict analysesOverall, the highest degree of conflict was reported in response to the Harvestdilemma (M = 4.73), the Commons dilemma was second (M = 4.41), and the least amountof conflict was reported in response to the Waste dilemma (M = 3.11). Tn two of the threedilemmas, more conflict was reported in the high conflict condition than in the low conflictcondition (see Table 12). Specifically, more contlict was reported by subjects when economicneed was high in the Commons dilemma and by subjects in the Waste dilemma wheneconomic outcomes were in conflict with the environmentally protective option (i.e.,recycling would cost more money). No difference was found in response to the Harvestdilemma.When the interaction between subjects value orientations and their behaviouraldecisions was compared in terms of the level of conflict experienced, trends were found thatfollowed predictions in two of the three dilemmas (see Table 13). No differences wereobserved in the Commons dilemma and, therefore, specific results are not reported. In theHarvest dilemma, a marginal overall interaction was found {F(131) = 2.91, p < .06]. Posthoc analyses showed a trend consistent with expectations. Specifically, it was found thateconomically-oriented subjects reported higher levels of conflict when they chose to preservethe forest, the value inconsistent option, than when they chose the harvest option, 1(131) =1.64, p < .10. In contrast, socially-oriented subjects reported more conflict when they chose93to harvest than when they decided to preserve, although this difference did not approachsignificance. No difference was found for universally-oriented subjects.In the Waste dilemma, a similar trend was found, F(131) = 3.38, p < .05.Economically-oriented subjects felt more conflict when they chose the recycling option thanwhen they endorsed building the new landfill. In contrast, both universally-oriented andsocially-oriented subjects reported more conflict when they chose the landfill option.Although these results are intriguing, for the most part they represent only trends, andare based on very small cell sizes in certain groups. Therefore, care should be taken indrawing conclusions from these findings. Nevertheless, the results suggest that the experienceof dissonance and its motivational influence on behaviour is associated with one’s valueorientation. Research that focuses on this interaction, as well as on the relation betweencognitive dissonance and environ mental decision —making, is warranted.ConclusionClearly, value orientations play a key role in guiding individuals’ decisions regardingecological issues. Although people recognize, and are motivated by, needs and desiresconsistent with all three motivational domains, when a dominant value orientation isembraced, it is likely to have the greatest influence on an individual’s thinking andbehavioural decisions. In addition, personal values may influence the way in whichpsychological motivations, such a cognitive dissonance, manifest them selves.Whereas strong relations between personal values and behavioural decisions werefound, the relation between attitudes and behavioural decisions was less reliable. Asignificant relation between attitudes and behavioural decisions may exist, but this appears tobe the exception rather than the rule. Thus, in the Harvest dilemma, the constructs of generalattitude and issue importance appeared to play significant roles in guiding subjects’ harvestdecisions. In contrast, these factors were not found to be associated with decisions in theother two dilemmas. It may be the case that certain issues are more attitudinally salient thanothers. In this study, the issue of forest preservation may have activated subjects’ attitudes94toward environmental preservation more so than the other two issues. These findings mayalso be explained by the skewed results in the Commons and Waste dilemmas, where adisproportionate percentage of subjects favoured the environmentally protective option (thepercentages were more balanced in the Harvest dilemma). I suspect that in the real-world ahigher percentage of people would endorse the economically advantageous option. If thatwere the case, attitudinal factors may be useful in accounting for variance in behaviouraldecisions. Therefore, it is argued that general attitudes should be included in a multivariatemodel of social behaviour and decision-making.In addition to these theoretical implications, the results of Study 3 have implicationsregarding confronting current ecological dilemmas. Although a “crisis” in values may not beresponsible for all our environmental troubles, personal value systems do influencebehavioural choice regarding these issues. For example, when environmental preservationand economic outcomes are in conflict, economically-oriented people will be more biased infavour of pursuing the economically-advantageous course of action than will people who havedifferent value oTientations. As Merchant (1992) suggests, this may stem from a belief that ifeach individual attempts to maximize personal economic outcomes, the society as a whole willbenefit. Adherents to this value orientation probably consider it their right to pursueeconomic goals, a right central to the effective functioning of society, and believe thatenlightened self-interest will guide us to a sustainable future.In contrast, socially-oriented individuals appear to place the needs of the many beforethe needs of any one individual or the natural environment. They seek maximum benefit forthe maximum number of people- a goal consistent with a highly valued desire for personalconnection with others. These pursuits are directed towards satisfying the needs of otherpeople, and are not necessarily consistent with preventing ecological problems. Peopleadhering to this value orientation will be most likely to support ecological preservation whenit is socially optimal to do so (in terms of personal and/or societal rewards), as seen insubjects’ responses to the Commons dilemma. They may also support an economically-95advantageous course of action when it is perceived to present more beneficial outcomes formore people, as was observed in responses to the Harvest dilemma. For socially-orientedpeople, their sense of social need, as well as their desire for personal social connection, serveas the underlying motives on which they base their decisions in ecological dilemmas.The group most likely to ardently pursue environmental preservation, even whennegative economic and/or social outcomes may result, are those who have developed auniversal value orientation. These people seek internal satisfaction from knowing theycontributed to resolving important societal/global concerns. In terms of ecological concern,they appear to be predisposed to protecting the natural world more consistently thanindividuals who embrace the other two value orientations, even when it means personalsacrifice.Our society has strong ties to an economic system that promotes the maximization ofpersonal economic outcomes as a desired (and respected) end. Many people’s judgementsregarding the appropriateness of their actions are rooted in this system, leading them to adopta value systeiii dominated by economic motives- a value system, as we observed in thisstudy, that will often contlict with ecological goals. Although this system may not necessarilyimpede a harmonious relation with the natural environment, a basic conflict does appear toexist.The question arises as to how society can most effectively resolve this conflict. Oneapproach contends that although a paradigm shift in personal values (i.e., toward a universalvalue orientation) may be most effectual in achieving preservation of the natural world, thisshift is not likely to occur in the near future. Rather than approaching ecological problemsfrom the position that human values are in “crisis,” it may be more efficacious to develop abetter understanding of the psychology that motivates decisions regarding these matters.Therefore, an initial step towards resolving ecological dilemmas would be a genuinerecognition of the various interests (motives) involved in such dilemmas and acceptance ofthose interests as legitimate by all involved parties. From this understanding, resolutions can96be sought through open and respectful communication between the various interests. Anapproach to resolving ecological dilemmas which is sensitive to, and respectful of, thecompeting interests (and personal values) involved is necessary to create a livable balanceamong the various needs at stake. This approach necessarily places a high level of confidencein the notion of ‘enlightened self-interest” (i.e., that once environmental hazards arerecognized, people would be motivated to ameliorate them, for it would be in their interest todo so). There are notable examples where enlightened self-interest has alerted people topotentially destructive practices (e.g., recognition of the effects of chloroflorocarbons on theozone shield has led to a reduced use of products containing this gas).However, in light of psychology’s research on behaviour in Commons type dilemmas(see Pious, 1993 for review), this enlightened self-interest approach may not be enough. Aprimary feature of a Commons dilemma is the inability of those involved to recognize thebroader, collective negative implications of their actions. As PIous (1993) has recently noted,“certain collective traps, such as those involving ... environmental degradation.. .have adisturbing air of permanence, and it remains to be seen whether humanity will be able to solvethem” (p. 251). Even if collective concerns are noted by a majority of people, the results ofthe present study imply that only individuals who are guided by a universal value orientationand who adhere to society’s present normative ideal regarding environmental protection willbe consistently willing to incur personal sacrifice in order to protect the natural world.Two limitations of this research deserve comment. First, the restricted nature of thesample (i.e., students) impedes generalizing the findings to the community at large,particularly with respect to the descriptive results. Reactions to these scenarios could varysystematically depending on the population that is being sampled. Second, hypotheticalsituations were used to elicit responses. Certainly, additional pressures would be felt if thesedecisions needed to be made in real life, pressures that may alter they way an individualwould respond. The students in the sample most likely have not had to contend with thesespecific situations. Therefore, their responses may lack insight into the true dynamics of the97described situations. Nevertheless, the importance of values in guiding ecological choice hasbeen demonstrated. Future research could aim at clarifying trends found in this study byexploring these important issues with a broader and more diverse sample, and in actual fieldsettings.98Chapter 9Summary and ConclusionsA primary goal of this thesis was to integrate, clarify, and expand upon socialpsychological theory that attempts to describe the beliefs, desires, and personal values thatinfluence social behavioural decisions. Pursuing this theoretical objective is important in lightof the many problems that modern society faces, problems that directly stem from individualhuman beliefs and activity. These predicaments include, but are not limited to, healthconcerns such as the spread of AIDS, economic difficulties that create higher levels ofpoverty, hunger, homelessness, prejudice and discrimination, and the primary focus of thepresent course of research, environmental destruction. Resolution of most, if not all, of theseproblems involves understanding the ways individuals think, feel, and act regarding issues ofsocial concern.The current research highlights the multifaceted and complex nature of decision-making regarding social behaviour. Much human behaviour involves planned reactions tospecific circumstances. Thus, behavioural decisions are usually motivated by a complex arrayof factors including beliefs about one’s world, beliefs about oneself, as well as individualdesires and needs. Although one factor may dominate the behavioural decision in any onesituation, it is understanding the diversity and complexity among these many factors that willenable social scientists to account most effectively for the variety of behaviour exhibited inimportant social situations. The framework offered here outlines some of the beliefs andpredispositions that are likely to influence individual action, and suggests the manner in whichthese factors are implicated in the decision-making process. Although it is often necessary toconcentrate one’s research on specific processes, it may be through multivariate, contextualinvestigation that the antecedents to human action can be most accurately understood.Theoretical conclusions and implicationsAll of the general domains of cognition identified in the framework (i.e, attitudes,efficacy beliefs, outcome beliefs, and personal values), were found, in some respect, to guide99behavioural decisions. The factors that were most consistently implicated were efficacybeliefs, outcome expectancies, and personal values. Specifically, both efficacy constructs(self-efficacy and channel efficacy) were found to be strong and direct predictors ofbehavioural intentions. Experimental results found self-efficacy, in particular, to be apowerful determinant of behavioural choice. These results build on, and add to, thesubstantial body of evidence that has shown self-efficacy and personal control factors to bestrongly implicated in guiding social behaviour.In general, beliefs regarding behavioural outcomes (i.e., outcome expectancies) werealso found to be consistent predictors of behavioural decisions. In addition, the recommendedprocess of systematically exploring the influence of different classes of outcomes that maymotivate a behaviour was found to be worthwhile. The three classes of outcomes (economic,social, and environmental) investigated in the current program of research were all found, atsome point, to be implicated in guiding behavioural choice. This association was particularlyevident for economic outcomes, even though the issue of concern was environmentalprotection. Although the results of the present research do not provide suitable evidence fordetermining which outcome domain is the most (or least) powerful predictor, they do confirmthe multifaceted nature of outcome expectancies. It is suggested that future researchexamining social behavioural decisions should attempt to explore the impact of variousindependent functional motives, rather than assessing only one global outcome expectancy.In contrast to efficacy beliefs and outcome expectancies, the linkage betweenattitudinal factors and behavioural intentions was found to be indirect and inconsistent.Although the attitudinal factors (general attitude toward the environment, importance of theenvironmental issue, and threat perception) enjoyed strong univariate associations withbehavioural decisions, their influence was dramatically reduced when they were integratedinto a more complex multivariate model. A notable exception in Study 3 was the influencethat general attitudes and issue importance had on decisions in the Harvest dilemma scenario.In this case, both constructs were found to account for significant portions of explainedS 100variance. Although attitudinal factors may not always, or even often, directly predictbehavioural decisions, their potential role in the decision process should not be dismissed.They may, in fact, function like a heuristic in that they exert more influence on behaviouraldecisions when beliefs regarding other, more direct, factors are uncertain or not formed.Future multivariate research should continue to explore these relations.Of all the independent constructs outlined in the framework, the findings regardingpersonal values are the most intriguing. Significant differences in the responses of the threevalue groupings were found in Study 2, and in two out of the three situations in Study 3.Non-significant trends consistent with these results were observed in Study 1 and in the thirdscenario in Study 3. Thus, personal values, in terms of the value placed on three genericmotivational domains (i.e., economic, social, universal), were found to guide decisionsregarding one specific environmental behaviour, as well as environmental decisions of morepersonal and societal consequence. Results indicate that personal value systems provideindividuals with a motivational foundation in which behavioural decisions are rooted.Although the value structure employed in the current research was found to bereasonably reliable and relevant, its explanatory utility regarding other issues and behaviouraldomains remains to be determined. This is particularly true for the social value group. Twoconcerns regarding this value orientation were observed. First, the scale used to assess thisvalue orientation had an internal consistency that fell below usually acceptable standards.Second, the responses of subjects assigned to this group were, at times, quite enigmatic (e.g.,reaction to the economic outcome manipulation in Study 2). Nevertheless, personal valuesseem to be an integral part of the decision process and should be included in research effortsattempting to explain behaviour, particularly in a social issue context.It should be pointed out that these conclusions extend only to behavioural decisiQnsand intentions, and not to actual behaviour. None of the studies described in this thesis usedactual behaviour as a criterion. Future research should attempt to explore the constructspresented in the framework as they account for differences in actual behaviour.10].Implications regarding promoting env iroiimentally protective behaviourAn original objective of this program of research was to explain the apparentdiscrepancy between strong public support for environmental protection and the moderateamount of action being undertaken in this regard. The research also attempted to address thequestion, why do some people strongly pursue environmental preservation and protection, andothers do not? I believe this research enables me to offer some insight into these issues.First, pro-environment attitudes are only one factor, and a marginal one at that, in theprocess that leads people to engage in environmentally protective behaviour. Each individualresponds to a myriad of concerns that ultimately motivate her or his actions. These include:how important environmental protection is to the person, whether or not the individualexpects that their actions would contribute to creating better environmental conditions, andwhat other possible outcomes may also result from the behaviour. In fact, people may engagein environmentally protective actions because they can accrue positive economic and/or socialoutcomes, and not necessarily because they wish to help the environment. The results of thisresearch suggest that an effective method of promoting environmentally protective behaviourwould be to inform people about the short-term and long-term economic costs that can resultfrom environmental abuse and destruction. This information could persuade people thatpursuing economic interests does not have to conflict with protection of the environment. Onthe other hand, unacceptable economic or social costs (as defined by each individual) canblock an individual from acting in an environmentally protective manner. Furthermore,people must believe they are capable of performing the desired behaviours, and that theactions are feasible (i.e., can reasonably fit into their already over-burdened daily schedule).In my view, unless the majority of people are convinced that economic needs are bestsatisfied by environmental protection, environmental protection will only have priority withthose people who have a value orientation that places universal concerns and a contributoryethic over other values, in particular the value placed on economic need. Until that time, thebattle over the exploitation of natural resources and the need to regulate behaviour willcontinue to grow.102103ReferencesAjzen, I. (1985). From intentions to action: A theory of planned behavior. In J. Kuhi & J.Beckman (Eds.), Action-control: From cognition to behavior (pp. 11-39). Heidelberg:Springer.Ajzen, I., & Fishbein, M. (1973). Attitudinal and normative variables as predictors ofspecific behaviors. Journal of Personality and Social Psychology, 7, 4 1-57.Ajzen, I., & Fishbein, M. (1977). Attitude-behavior relations: A theoretical analysis andreview of empirical research. Psychological Bulletin, 4, 888-918.Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social behavior.Englewood Cliffs, NJ: Prentice-Hall.Ajzen, I., & Madden, T. J. (1986). Prediction of goal-directed behavior: The role ofintention, perceived control, and prior behavior. Journal of Experimental SocialPsychology, 22, 453-474.Ajzen, I., & Timko, C. (1986). Correspondence between health attitudes and behavior.Basic and Applied Social Psychology, 7, 259-276.Ailport, G., Vernon, P., & Lindzey, G. (1960). Study of values. Test booklet and manual.Boston: Houghton Mifflin Co.Angus Reid Group, (1992). Canadians and the Environment. Vancouver, B.C.: Angus ReidGroup.Armstrong, T. R. (1972). Why do we still have an ecological crisis? Englewood Cliffs, NJ:Spectrum Books.Aronson, E. (1991). The Social Animal. New York: W.H. Freeman and Company.Aronson, E. (1992). The return of the repressed: Dissonance theory makes a comeback.Psychological Inquiry, 3, 303-311.Aronson, E., & O’Leary, M. (1983). The relative effects of models and prompts on energyconservation. Journal of Environmental Systems, 12, 219-224.104Axeirod, L. J., & Lehman, D. R. (1993). Responding to environmental concerns: Whatfactors guide individual action? Journal of Environmental Psychology, 13, 149-159.Axeirod, L. J., & Newton, J. W. (1991). Preventing nuclear war: Beliefs and attitudes aspredictors of disarmist and deterrentist behavior. Journal of Applied Social Psychology,21, 29-40.Balderjahn, 1. (1988). Personality variables and environmental attitudes and predictors ofecologically responsible consumption patterns. Journal of Business Research, 17, 51-56.Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change.Psychological Review, 84, 191-215.Bandura, A. (1982). Self-efficacy mechanism in human agency. American Psychologist, 7,122- 147.-JBandura, A. (1986). Social foundations of thought and action: A social cognitive theory.New York: Prentice-Hall.Bandura, A., Reese, L., & Adams, N. E. (1982). Microanalysis of action and fear arousal asa function of differential levels of perceived self-efficacy. Journal of Personality andSocial Psychology, 4, 5-21.Becker, M. H. (1974). The health belief model and personal health behavior. HealthEducation Monographs, , 326-473.Bern, D. J. (1972). Self-perception theory. In L. Berkowitz (Ed.), Advances inExperimental Social Psychology (Vol. 6, pp. 1-62). New York: Academic Press.Berkowitz, L. (1992). Even more synthesis. Psychological Inquiry, , 312-313.Berkowitz, L., & Devine, P. G. (1989). Research tradition, analysis, and synthesis in socialpsychological theories: The case ol dissonance theory. Personality and Social PsychologyBulletin, 493-507.Boyd, B., & Wanclersman, A. (1991). Predicting undergraduate condom use with theFishbein and Ajzen and the Triandis attitude-behavior models: Implications for publichealth interventions. Journal of Applied Social Psychology, 21, 18 10-1830.105Cantril, H. & Allport, G. (1933). Recent applications of the study of values. Journal ofAbnormal and Social Psychology, , 259-273.Clawson, C. J., & Vinson, D. E. (1978). Human values: An historical and interdisciplinaryanalysis. In H. K. Hunt (Ed.), Advances in consumer research (Vol. 5, pp. 396-402).Ann Arbor, MI: Association for Consumer Research.Cohen, J. (1968). Multiple regression as a general data-analytic system. PsychologicalBulletin, 70, 426-443.Cooper, J., & Fazio, R. H. (1984). A new look at dissonance theory. In L. Berkowitz(Ed.), Advances in Experimental Social Psychology (Vol. 17, pp. 229-266). New York:Academic Press.Davis, F. W., & Yates, B. T. (1982). Self-eft1cacy expectancies versus outcomeexpectancies as determinants of performance deficits and depressive affect. CognitiveTherapy and Research, 6, 23-25.De Young, R. (1986a). Encouraging environmentally appropriate behavior: The role ofintrinsic motivation. Journal of Environmental Systems, j5, 28 1—292.Dc Young, R. (1986b). Some psychological aspects of recycling: The structure ofconservation satisfactions. Environment and Behavior, j, 435-449.Dillehay, R. C. (1973). On the irrelevance of the classical negative evidence concerning theeffect of attitudes on behavior. American Psychologist, 28, 887-893.Duffy, E. (1940). A critical review of investigations employing the Allport-Vernon study ofvalues and other tests of evaluative attitudes. Psychological Bulletin, 3j, 597-6 12.Dunlap, R. E., Grienecks, J. K., & Rokeach, M. (1983). Human values and proenvironmental behavior. In W.D. Conn (Ed.), Energy and material resources: Attitudes.values, and public policy (pp. 145-168). Boulder, CO: Westview Press.Eagly, A. H. (1992). Uneven progress: Social psychology and the study of attitudes.Journal of Personality and Social Psychology, 63, 693-7 10.106Fazio, R. H. (1989). On the power and functionality of attitudes: The role of attitudeaccessibility. In A. R. Pratkanis, S. J. Breckler, & A. G. Greenwald (Eds.), Attitudestructure and function (pp. 153-179). Hillsdale, NJ: Eribaum.Fazio, R. H., Chen, J., McDonel, E. C., & Sherman, S. J. (1982). Attitude accessibility,attitude-behavior consistency, and the strength of object-evaluation association. Journal ofExperimental Social Psychology, 18, 339-357.Fazio, R. H., & Williams, C. J. (1986). Attitude accessibility as a moderator of the attitude-perception and attitude-behavior relations: An investigation of the 1984 presidentialelection. Journal of Personality and Social Psychology, j, 505-5 14.Fazio, R. H., & Zanna, M. P. (1978). Attitudinal qualities relating to strength of theattitude-behavior relationship. Journal of Experi mental Social Psychology, j.4, 398-408.Fazio, R. H., & Zanna, M. P. (1981). Direct experience and attitude-behavior consistency.In L. Berkowitz (Ed.), Advances in Experimental Social Psychology, (Vol. 14, pp. 164-181). New York: Academic Press.Festinger, L. (1957). A theory of cognitive dissonance. Stanford: Stanford UniversityPress.Festinger, L., & Carlsmith, J. (1959). Cognitive consequences of forced compliance.Journal of Abnormal and Social Psychology, 58, 203-2 10.Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention and behavior: An introduction totheory and research. Reading, MA: Addison Wesley.Forsyth, D. R. (1985). Individual differences in information integration during moraljudgment. Journal of Personality and Social Psychology, 49, 264-272.Gallup, G. Jr., & Newport, F. (1990). Americans strongly in tune with the purpose of EarthDay 1990. Gallup Poll, 4, 1-5.Granberg, D. (1985). An anomaly in political perception. Public Opinion Quarterly, 49,504-516.Hardin, G. J. (1968). The tragedy of the commons. Science, i, 1243-1248.107Henion, K. E. (1972). The effect of ecologically-relevant information on detergent sales.Journal of Marketing Research, 9, 10-14.Hines, J. M., Hungerforcl, H. R., & Tomera, A. N. (1987). Analysis and synthesis ofresearch on responsible environmental behavior: A meta-analysis. Journal ofEnviron mental Education, 18, 1-8.Homer, P. M., & Kahie, L. R. (1988). A structural equation test of the value-attitude-behavior hierarchy. Journal of Personality and Social Psychology, 4, 638-646.Huebner, R. B., & Lipsey, M. W., (1981). The relationship of three measures of locus ofcontrol to environmental activism. Basic and Applied Social Psychology, , 45-58.Jacobs, H. E., & Bailey, J. S. (1982). Evaluating participation in a residential recyclingprogram. Journal of Environmental Systems, j, 14 1-152.Judd, C. M., & Johnson, J. T. (1981). Attitudes, polarization, and diagnosticity: Exploringthe effects of affect. Journal of Personality and Social Psychology, 41, 25-36.Kahle, L. R. (Ed.). (1983). Social values and social change: Adaptation to life in America.New York: Praeger.Kallgren, C. A., & Wood, W. (1986). Access to attitude-relevant information in memory asa determinant of attitude-behavior consistency. Journal of Experimental Social.Psychology, 22, 328-338.Kantola, S. J., Syme, G. J., & Campbell, N. A. (1982). The role of individual differencesand external variables in a test of the sufficiency of Fishbein’s model to explain behavioralintentions to conserve water. Journal of Applied Social Psychology, 12, 70-83.Kantola, S. J., Syrne, G. J., & Nesdale, A. R. (1983). The effects of appraised severity andefficacy in promoting water conservation: An informational analysis. Journal of AppliedSocial Psychology, 13, 164-182.Katz, D. (1960). The functional approach to the study of attitudes. Public OpinionQuarterly, 24, 163-204.108Katz, D. & Stotland, E. (1959). A preliminary statement to a theory of attitude structure andchange. In S. Koch (Ed.), Psychology: A study of a science (Vol. 3, pp. 423-475).New York: McGraw-Hill.Kirtines, W. M. (1986). Moral behavior as rule governed behavior: Person and situationeffects on moral decision making. Journal of Personality and Social Psychology, 50, 784-791.Kluckhohn, C. (1951). Values and value orientations in the theory of action. In T. Parsons& E. A. Shils (Eds.), Toward a general theory of action. Cambridge, MA: HarvardUniversity Press.Kohlberg, L. (1958). The development of modes of moral thinking and choice in the yearsten to sixteen. Unpublished doctoral dissertation, University of Chicago, Chicago.Krosnick, J. A. (1989). Attitude importance and attitude accessibility. Personality and SocialPsychology Bulletin, 15, 297-308.Krosnick, J. A. (1990). Government policy and citizen passion: A study of issue publics incontemporary America. Political Behavior, 12, 59-92.Krosnick, J. A., Boninger, D. S., Chuang, Y. C., Berent, M. K., & Carnot, C. G. (1993).Attitude strength: One construct or many related constructs? Journal of Personality andSocial Psychology, , 1132-1 151.LaPiere, R. T. (1934). Attitudes versus actions. Social Forces, j3, 230-237.Luyben, P. D., & Bailey, J. S. (1979). Newspaper recycling: The effects of rewards andproximity of containers. Environment and Behavior, 11, 539-557.Maddux, J. E., Norton, L. W., & Stoltenberg, C. D. (1986). Self-efficacy expectancy,outcome expectancy, and outcome value: Relative effects on behavioral intentions.Journal of Personality and Social Psychology, 51, 783-789.Maddux, J. E., & Rogers, R. W. (1983). Protection motivation and self-efficacy: A revisedtheory of fear appeals and attitude change. Journal of Experimental Social Psychology,, 469-479.109Margolin, J. B., & Misch, M. R. (1978). Incentives and disincentives for ridesharing: Abehavioral study. Washington, D.C.: U.S. Government Printing Office.Maslow, A. H. (1970). Motivation and Personality. New York: Harper & Row.Maslow, A. H. (1971). The Farther Reaches of Human Nature. New York: Viking Press.Merchant, C. (1992). Radical ecology: The search for a livable world. New York:Routledge, Chapman, & Nail, Inc.Milbrath, L. (1986). Environmental beliefs and values. In M. Herman (Ed.), PoliticalPsychology (pp. 97-138). San Francisco: Jossey-Bass.Needlernan, L. D., & Geller, E. S. (1992). Comparing interventions to motivate work-sitecollection of home-generated recyclables. American Journal of Community Psychology,, 775-785.Neuman, K. (1986). Personal values and commitment to energy conservation. Environmentand Behavior, 18, 53-74.Omoto, A. M., & Snyder, M. (1990). Basic research in action: Volunteerism and society’sresponse to AIDS. Personality and Social Psychology Bulletin, 16, 152-165.Oskamp, S., Harrington, M. J., Edwards, T. C., Sherwood, D. L., Okuda, S. M., &Swanson, D. C. (1991). Factors influencing household recycling behavior. Environmentand Behavior, 23, 494-519.Paterson, R. J., & Neufeld, R. W. J. (1987). Clear danger: Situational determinants of theappraisal of threat. Psychological Bulletin, 101, 404-4 16.Pious, S. (1993). The psychology of judgment and decision-making. New York: McGraw-Hill.Pratkanis, A. R., & Greenwald, A. G. (1989). A sociocognitive model of attitude structureand function. In L. Berkowitz (Ed.), Advances in Experimental Social Psychology (Vol.22, pp. 245-285). New York: Academic Press.Raden, D. (1985). Strength-related attitude dimensions. Social Psychology Quarterly, 4,3 12-330.110Regan, D. T., & Fazio, R. F. (1977). On the consistency between attitudes and behavior:Look to the method of attitude formation. Journal of Experimental Social Psychology,i, 28-45.Rokeach, M. (1967). Value Survey. Sunnyvale, CA: Haigren Tests.Rokeach, M. (1973). The nature of human values. New York: The Free Press.Rokeach, M. (1980). Some unresolved issues in theories of beliefs, attitudes, and values. InM. M. Page (Ed.), 1979 Nebraska Symposium on Motivation. Lincoln: University ofNebraska Press.Rogers, R. W. (1975). A protection motivation theory of fear appeals and attitude change.The Journal of Psychology, 91, 93-114.Ross, L., & Nisbett, R. E. (1991). The person and the situation: Perspectives of socialpsychology. New York: McGraw-Hill, Inc.Scher, S. J., & Cooper, J. (1989). Motivational basis of dissonance: The singular role ofbehavioral consequences. Journal of Personality and Social Psychology, , 899-906.Schwartz, S. H. (1977). Normative influences on altruism. In L. Berkowitz (Ed.), Advancesin Experimental Social Psychology (Vol. 10, pp. 22 1-279). New York: Academic Press.Schwartz, S. H. (1992). Universals in the content and structure of values: Theoreticaladvances and empirical tests in 20 countries. In M. P. Zanna (Ed.), Advances inExperimental Social Psychology (Vol. 25, pp. 1-65). San Diego: Academic Press.Schwartz, S. H., & Bilsky, W. (1987). Toward a universal psychological structure of humanvalues. Journal of Personality and Social Psychology, 53, 55-562.Schwartz, S. H., & Bilsky, W. (1990). Toward a theory of the universal content andstructure of values: Extensions and cross-cultural replications. Journal of Personality andSocial Psychology, , 878-891.Schwartz, S. H., & Tessler, R. C. (1972). A test of a model of reducing measured attitudebehavior discrepancies. Journal of Social and Personality Psychology, 4, 225-236.111Shavitt, S. (1989). Operationalizing functional theories of attitude. In A. R. Pratkanis, S. J.Breckler, & A. G. Greenwald (Eds.), Attitude structure and function (pp. 311-337).Hilisdale, NJ: Eribaum.Sherif, C. W. (1980). Social values, attitudes, and involvement of the self. 1979 NebraskaSymposium on Motivation (pp. 1-64). Lincoln: University of Nebraska Press.Sherman, S. J. (1987). Cognitive processes in the formation, change, and expression ofattitudes. In M. P. Zanna, J. M. Olson, & C. P. Herman (Eds.), Social Influence: TheOntario symposium, Vol. 5 (pp. 75-106). Hillsdale, NJ: Erlbaurn.Simon, H. A. (1957). Models of man: Social and rational. New York: Wiley.Sivacek, J., & Crano, W. D. (1982). Vested interest as a moderator of attitude-behaviorconsistency. Journal of Personality and Social Psychology, 43, 2 10-221.Smith, M. B., Bruner, J. S., & White, R. W. (1956). Opinions and personality. New York:Wiley.Snyder, M. (1982). When believing means doing: Creating links between attitudes andbehavior. In M. P. Zanna, E. T. Higgins, & C. P. Herman (Eds.), Consistency in socialbehavior (P1). 105-130). Hillsdale, NJ: Eribaum.Snyder, M., & Debono, K. G. (1987). A functional approach to attitudes and persuasion. InM. P. Zanna, J. M. Olson, & C. P. Herman (Eds.), Social Influence: The OntarioSymposium (Vol. 5, pp. 107-125). Hillsclale, NJ: Erlbaum.Snyder, M., & Debono, K. G. (1989). Understanding the functions of attitudes: Lessonsfrom personality and social behavior. In A. R. Pratkanis, S. J. Breckler, & A. G.Greenwald (Eds.), Attitude structure and function (pp. 339-359). Hillsdale, NJ:Erlbau iii.Snyder, M., & Omoto, A. M. (1992). Who helps and why? The psychology of AIDSvolunterism. In S. Oskamp & S. Spacapan (Eds.), Helping and being helped in the realworld: Naturalistic studies. Newbury Park, CA: Sage.112Steele, C. (1988). The psychology of self-affirmations: Sustaining the integrity of the self.In L. Berkowitz (Ed.), Advances in Experimental Social Psychology (Vol. 21, pp. 261-302). San Diego: Academic Press.Steele, C., & Liu, T. (1983). Dissonance processes as self-affirmation. Journal ofPersonality and Social Psychology, 45, 5-19.Steele, C., & Spencer, S. J. (1992). The primacy of self-integrity. Psychological Inquiry, ,345-346.Stern, P. C., Dietz, T., & Kalof, L. (1993). Value orientations, gender, and environmentalconcern. Environment and Behavior, 25, 322-348.Thompson, S. C., & Stoutemyer, K. (1991). Water use as a commons dilemma: The effectsof education that focuses on long-term consequences and individual action. Environmentand Behavior, 23, 3 14-333.Triandis, H. C. (1980). Values, attitudes, and interpersonal behavior. In M. M. Page (Ed.),1979 Nebraska Symposium on Motivation (pp. 195-259). Lincoln: University ofNebraska Press.Tversky, A. & Kahneman, D. (1981). The framing of decisions and the psychology ofchoice. Science, 21, 452-458.Vining, J., & Ebreo, A. (1988). A survey of residents’ responses to a citywide curbsiderecycling program before and after expansion. Report to the City of Champaign, Ill.Vining, J., & Ebrea, A. (1990). What makes a recycler?: A comparison of recyclers andnonrecyclers. Environment and Behavior, 22, 55-73.Waister, E., Berscheid, E., & Walster, G. W. (1973). New directions in equity research.Journal of Personality and Social Psychology, 25, 15 1-176.Weigel, R. H., & Newman, L. S. (1976). Increasing attitude-behavior correspondence bybroadening the scope of the behavioral measure. Journal of Personality and SocialPsychology, 33, 793-802.113Wicker, A. W. (1969). Attitudes versus action: The relationship of verbal and overtbehavioral responses to attitude objects. Journal of Social Issues, 25, 4 1-48.Wicker, A. W. (1971). An examination of the “other variables” explanation of attitude-behavior inconsistency. Journal of Personality and Social Psychology, j, 18-30.Wolf, S., Gregory, W. L., & Stephan, W. G. (1986). Protection motivation theory:Predictions of intentions to engage in anti-nuclear war behaviors. Journal of AppliedSocial Psychology, j, 310-321.Zanna, M. P., Olson, J. M., & Fazio, R. H. (1981). Self-perception and attitude behaviorconsistency. Personality and Social Psychology Bulletin, 7, 252-256.114Table 1Univariate and partial correlation coefficients for the predictor variables regarding intentions toperform the criterion behaviour in Study 1(R = .74)U nivariate PartialPredictor r rEnvironmental outcome expectancy .57 * .36*Economic outcome expectancy •49* . 18!Social outcome expectancy •37* .05Self-efficacy .29* .18!Channel efficacy 49* •33*Environmental outcome value •44* .09Economic outcome value 15 .12Social outcome value .05 -.01General attitude . 19* .07Issue importance .27*. 15Threat perception .04*p < .05; ! < .10Note: Partial correlation coefficients derived from simultaneous entry regression procedure.115Table 2Mean importance ratings of the three behavioural outcomes for the different value orientationsValue orientationOutcome value Economic Social UniversalEconomic 777a 677bSocial 673a 649a 520bEnvironmental 7°6a “ ‘5a 782bNote. Significant differences within rows are indicated by different subscripts, < .05.116Table 3Mean behavioural intention ratings for interaction between value orientation and economicoutcome condition in Study 2Value orientationEconomic outcome Economic Social UniversalConditionLow 4’3a 5.73 6.20High 5•25b 5.00 6.00Note. Significant differences within columns are indicated by different subscripts, < .05.117Table 4Univariate and partial correlation coefficients for the predictor variables regarding intentions toperform the criterion behaviour in Study 2(R = .74)U nivariate PartialPredictor r rEnviron mental outcome expectancy .5 1 * .25 *Economic outcome expectancy .48* .02Social outcome expectancy .58* .40*Self-efficacy 45* .27*Channel efficacy .50* .29*Environmental outcome value •44*. 16Economic outcome value .07 -.03Social outcome value -.04 -.06General attitude. 11 -.01Issue importance .20*-.00Threat perception .21 * .04*p < .05Note: Partial correlation coefficients derived from simultaneous entry regression procedure.118Table 5Value differences in the percentage of subjects selecting the environmentally protective courseof action in 3 ecological dilemmasValue orientationDilemma Economic Social UniversalCommons 71a 89b 86bHarvest 50a 559a 85bWaste 85°a 94°aNote. Signitcant differences within a row are indicated by different subscripts, p < .05.119Table 6(omnrisons of the influence of“t on hehviorl decicions for each value--r- SJJI.flL-—--- -Value OrientationDilemma Economic! Economic Social UniversalEnvironmentConflictHigh 62.5% 87.5% 83.3%CommonsLow 79.2% 90.9% 88.0%High 29•a 56.5% 87.0%Harvest-Low 708b 54.2% 84.0%96.O%b*Note. Tabled values are percentage of subjects supporting the environmentally-protectivecourse of action. Significant differences within a column are indicated by different subscripts,p < .01, p < .10.WasteHigh 75.O%a*Low92.0%95.5%87.5%100%120Table 7Mean importance ratings of select reasons found to be implicated in guiding subjects’ decisionsin the Commons dilemmaValue OrientationReason Economic Social UniversalShort-term financial needs 293bRelationship with other ranchers 3•85a 450bWhat one’s family thinks 2•98a 348bRights of other ranchers violated 370a 433bHurt other ranchers financially 3•63a 425bAdhere to one’s principles 390a 433bPurchasing cattle violates moral code 362a 446b 455bIs agreement legally binding 288a 264aExpectations regarding other ranchers 360a 350aRight to make a better life 3•85a 355aRight to pursue economic interests 3. 13a 270aOne’s long-term goals 432a 389aNote. Significant differences within rows are indicated by different subscripts, < .05.121Table 8Mean importance ratings of select reasons found to be implicated in guiding subjects’ decisionsin the Harvest dilemmaValue OrientationReason Economic Social UniversalHow affect personal financial situation 4.‘5a 374b 323cOne’s relationship with townspeople 281a 321bThe costs to the environment 388a 430bWhether any laws would be violated 3•27a 283aEffect on friends and neighbors 3. 10a 338aLong-term needs of society 390a 4.02aEffect on animal species 367a 4°2a 430aWhat other townspeople want 275a 3°0aA person’s right to have ajob 344a 352a 283aEconomic responsibility of town 335a 324aShort-term economic interests of town 321a 3‘7aAvailability of alternative industries 385a 387a 376aNote. Significant differences within rows are indicated by different subscripts, < .05.122Table 9Mean importance ratings of select reasons found to be implicated in guiding subjects’ decisionsin the Waste dilemmaValue OrientationReason Economic Social UniversalCost to community residents 370a 3 ‘9bAdhering to core principles 346a 391b 4How proposal would affect environment 4. 13a 448b 467bWhat the majority want to do 271a 261a 2°8bWhat the government recommends 233aWhat family and friends support 273a 280aAbility to pay new fees 358a 343aLong-term needs of city 421a 450aRight of government to regulate 292a 291a 296aConvenience of recycling O1)tiOfl 335a 350a 252bSpace limitations in one’s home 298a 266abAesthetics of the neighborhood 3•26a 347a 294aNote. Significant differences withit rows are indicated by different subscripts, p < .05.123Table 10Percentage of subjects ranking different categories of motives as “most important” ininfluencing their decisionsValue OrientationDilemma Motive Economic Social UniversalEconomic 58•3a 128°bCommons Social 83°b 298°aEnvironmental 20.8a 447 °b 67°°cEconomic 39•6a 14•9bHarvest Social 21b 17°°a 10°abEnvironmental 33°°a 447 °ab 63.3 bEconomic 208°a ‘°6°abWaste Social 27°91a 2559’aEnvironmental 37.5 a 575 0b 81•6cNote. Significant differences within rows are indicated by different subscripts, < .05.124Table 11Univariate and partial correlation coefficients for independent factors as predictors of decisionsin the Harvest dilemmaUnivariate Partial VarianceCorrelation Correlation ExplainedPredictor Variables (R = .57)General attitude .41 .24 4.0%*Issue Importance .39 .24 4.0%*Manipulation effect .09 .30 6.8%*Value difference #1 .33 .26 47%*(universal vs. others)Value difference #2 -.13 .22 34%*(social vs. others)Value difference #1 -.14 -.26 4%*by manipulationValue difference #2 .28 -.22 35%*by manipulation* < .02125Table 12Level of conflict experienced in the different outcome conditions for each dilemmaLevel of economic-environ mental conflictDilemma Low HighCommons 397aHarvest 470aWaste 270aNote. Significant differences within rows are indicated by different subscripts, < .05.126Table 13Level of conflict experienced when subjects made value consistent decisions versus valueinconsistent decisionsValue OrientationDilemma Decision Economic Social UniversalHarvest 3.71(24)a 5.33(21) 5.00(7)HarvestPreserve 4.81(26) 5.02(41)Landfill 2.50(6) 4.67(3)WasteRecycle 3.51(41) 3.12(43)Note. Significant differences within columns are indicated by different subscripts, p < .10.Figures in parentheses represent number of subjects selecting that option.127Figure1SociocognitiveframeworkofguidedbehaviourStatementaboutTheoreticalRelationGeneralMotivationSpecificConstructsDefinitionof constructstoBehaviouralDecisionGeneralAttitudeGlobalevaluationofissue/objectIndirectandinconsistent.huttheoretically-relevantIbelieveAttitudeStrengthAccessibility,importance,orDirectintermsofmediatingpersonalrelevanceofattitudeattitude-behaviourrelationOtherrelevantattitudesEvaluativeheliefsaboutrelatedPossiblydirectdependingonissuesorobjects(e.g..threattypeofbehaviourposedbyattitudeobject)Self-efficacyCapabilityofperfiwmingDirectandcausalspecifiedbehaviourIcanChannelefficacyPerceptionof accessibilityDirectandcausaltospecifiedbehaviourOutcomeexpectancyExpectationsthatcertainoutcomesDirectandcausalwillhederivedfromactions.I desire/expect(Beliefsregardingspecificoutcomedomainsshouldbesystematicallyanalyzed)OutcomevalueValueof expectedoutcomesAssociated,but relationcurrentlyunclearValueindividualplacesonIndirectthroughattitudes,IvalueIndividualvaluecore,genericterminalgoalsoutcomebeliefs;possiblyorientations(i.e.,economic,social,universal)directaswell128Figure2DescriptionofProminentValueStructuresAllport.Vernon.LindzcyMaslowsHierarchyRokeachKahle(1983)Schwartz’sUniversal(1992)StudyofValues(1960)ofNeeds(1970.71)ValueSurvey(1968)ListofValuesStructureofHumanValuesDescriptionofValueStructureDescribesthesixDescribesaneedIdentifiesall’theSpecifiesthevaluesDescribesuniversallistbasicinterestsorhierarchythatmotivatesterminalvaluesimportantmistrelevanttoofhumanvaluesprsiinaltvmotiveshumanbehaviourinhumsia.1dalvIifeormotivationaltypesContentDomainsinValueTheoriesTheoreticalPhysiologicalNeedsAComfortableLifeSelf—liiltilmentSelf—direction(searchfortruth)AnExcitingLiIExcitementStimulationEconomicSaftvNeedsASenseofContributionASenseofAccomplishmentHedonism(valuesthingstangible)AWorldatPeaceSelf-respectAchievementAestheticBelonging&LoveNeedsAWorldofBeautyASenseofBelongingPower(valtiesbeauty)EqtialitvBeingWell—respectedSecuritySocialSelf-esteemNeedsFamilySecuritySecurityConformity(altrtiisticlove)FreedomFun&EnovmeniTraditionPoliticalSeI f—acttoili/alionHappinessWarmRelationshipsBenevolence(power&influence)InnerHarmonUniversalismReligiousMattireLove(spirituality&unity)NationalSecurityPleasureSalvationSdf-respectSocialRecognitionTrueFriendshipWisdomAssessmentTechniquesPersonalvaluetypeNomeasurementValuehierarchiesValuehierarchiesFactorstructuremeasuredbyrelativescaleestablishedbyidentifiedbyindependentderivedfromratingspreferencerankingsofrankorderingallratingsof eachvalueof 56individualvaluesalternativesrepresenting18valueafrommosteachinterestover45itemstoleastimportant129Figure 3Proposed Value TaxonomyDescription of Value StructureSpecifies three core value domains which have clear motivational attributesContent Domains in Value Theories1. Economic/Self-enhancing- values things tangible- values economic security2. Social- benevolence to community- a sense of belonging3. Universal- a sense of contribution- self-respectAssessment TechniqueIndividual orientations assessed by preference rankings of three alternatives representing eachof the three motivational domains.130Appendix 1Pre-test questions usedto assess attitudinal constructs.General Attitude itemsNote: All items were responded to on a six-point scale ranging from “strongly disagree” to“strongly agree.”(1) “It is more important for human life to progress than it is to protect animal or plant life.”(2) “The resources of the earth exist for the use of humankind.”(3) “Environmental protection must not stand in the way of providing economic opportunityfor everyone.”(4) “The earth will always be able to provide the resources necessary for the human race tosurvive.”(5) “The behaviour of humans needs to become more sensitive to the environment if thehuman race is to survive.”(6) “The extinction of animal ahd plant species is acceptable as long as alternative resourcesexist.”Issue Importance Items(1) “How important is environmental protection to you?” Rated on six-point scale rangingfrom “not very important” to “very important.”(2) “Please rank the following issues from ‘1’ (the most important to you) to ‘10’ (the leastimportant to you).” The issues were: AIDS, Civil Rights, Crime, Drugs, Economics(Jobs), the Environment, the Homeless, Hunger, Nuclear Weapons, and Poverty.Threat Perception Items(1) “How likely is it that major environmental destruction will happen?” Rated on six-pointscale ranging from “very unlikely” to “very likely.”(2) “How severe will the consequences of major environmental destruction be to nature andlife?” Rated on six-point scale ranging from “not very severe” to “very severe.”(3) “How soon, if at all, will major environmental destruction occur?” Rated on six-pointscale ranging from “never” to “very soon.”131Appendix 2Revised items used to assess value orientations in pre-test for Studies 2 and 3.InstructionsA number of questions regarding your views of yourself are given below, each with threepossible responses. Please indicate how you feel about each alternative by assigning a scorenext to each response (0, 1, 2, 3, or 4) so that the total amount across all three responsesequals 4. You should give higher numbers to the answers that you feel better representyour thoughts.For example, you could assign answers in the following ways:Example 1 Example 2 Example 3 Example 4a=3 a=2 a=l a=0b=l b=l b=O b=4c=0 c=l c=3 c=OExample (1) represents the situation where you most strongly support answer (a), slightlysupport answer (b), and do not at all support answer (c). The responses in example (2) wouldindicate that you support each alternative to some extent, but that answer (a) is moreconsistent with your views than either (b) or (c).Please assign values as you see fit keeping in mind that you must distribute exactly 4 pointsacross the three responses.1. Which of the following do you think defines success in life?(a) contributing something back to society(b) having close and satisfying relationships with others(c) achieving a high level of prosperitya= b= c=2. If you should see the following headlines in your newspaper, which would you read moreattentively?(a) GREAT IMPROVEMENT IN MARKET CONDITIONS(b) HUNGER STRIKE CALLS INTO QUESTION GOVERNMENT POLICY(c) GOOD RELATIONSHIPS: THE KEY TO HAPPINESSa= b= c3. When you think about being rewarded for your actions, which do you more highly value?(a) financial rewards(b) maintaining respect from family and friends(c) a sense of self-respectb= c=1324. Which of the following goals do you pursue more strongly:(a) achieving a sense of belonging and acceptance from others(b) attaining a comfortable and financially secure life(c) maintaining a sense of self-respect derived from acting in accord with deeply heldvaluesa= b= c=5. When deciding how to act with regard to global issues, you would definitely:(a) keep in mind the social responsibility of your actions(b) attempt to protect your economic well-being(c) look to the behavior of others as a guide to the appropriateness of your actionsa= b= c=6. When unsure about how to act in an unfamiliar situation, you would consider:(a) how your actions would affect your financial situation(b) how others may perceive what you do as either good or bad(c) how consistent your actions would be with what you valuea= b c=7. If you were given an award hoiiouring a contribution you made to society, you would bemost satisfied by:(a) sharing your honour with close friends and family(b) receiving the large cash award(c) knowing you’ve worked to benefit your societya=8. When considering a position in your chosen career, you would be most interested in:(a) working in a friendly, accepting environment(b) working for a socially-responsible organization(c) earning a good salarya= b= c


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items