How materialism affects environmental beliefs, concern, and environmentally responsible behavior

This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

http://www.elsevier.com/copyright

Author's personal copy

How materialism affects environmental beliefs, concern, andenvironmentally responsible behavior

William Kilbourne ⁎, Gregory Pickett ⁎

Clemson University, United States

Received 1 August 2006; received in revised form 1 March 2007; accepted 1 September 2007

Abstract

This article examines the relationship between materialism, environmental beliefs, environmental concern, and environmental behaviors. Thestudy used a random telephone survey of 337 US adults. Using a causal modeling approach, the study demonstrates that materialism has anegative effect on environmental beliefs, and these beliefs positively affect environmental concern and environmentally responsible behaviors.The article then provides implications of the results for consumer and environmental policy.© 2007 Elsevier Inc. All rights reserved.

Keywords: Materialism; Environment; Environmentally responsible behavior; Environmental concern

1. Introduction

The role of the environment in market behavior has takenmany turns over the last thirty years. Prior foci of researchinclude identifying environmentally concerned consumers(Alwitt and Pitts, 1996; Kinnear et al., 1974), green marketingstrategies (Menon and Menon, 1997), socially responsibleconsumption (Fisk, 1973), energy conservation (Leonard-Barton, 1981), and sustainable consumption (Kilbourne et al.,1997) among many others. Many argue that environmentalawareness and concern have increased since the early 1970's,but an attitude–behavior gap still exists (Alwitt and Pitts, 1996).This gap refers to the fact that “environmentally concerned”consumers do not seem to show any consistent preference forenvironmentally friendly products in their purchase behavior.Smith (1999) and Dowie (1995) suggest that while, on thesurface, environmentalism appears to be increasing in the US,the environmental movement fails to deliver substantial changesin behavior.

This raises two vexing questions. Why does the attitude–behavior gap persist? What public policy will encourageconsumers to be more environmentally benign in their purchasebehavior? Certainly, many studies examine the relationshipbetween consumer action and the environment. However,consensus is lacking as to why negligible progress occurs intransitioning to more sustainable consumption behavior or whyefficacious policy alternatives have not been forthcoming. Theone consistent premise in much of the research on theenvironmental consequences of market behavior is that boththe quality and quantity of consumption in Western industrialsocieties are complicit in the environmental problem (Capra,1982; Jones, 1987; Porritt, 1984).

Carson (1962) characterizes environmental decline as aneconomic phenomenon, and Fisk (1973) brings the environmentinto the marketing literature as a consumption problem.However, most of the research that followed was not systemicand tended to focus on symptoms of environmental degradationsuch as pollution, resource decline, and waste disposal ratherthan root causes. Kilbourne and Beckmann (1998) provide asummary of environmental studies within the marketing lite-rature that clearly demonstrates this. Porritt (1984) and Jones(1987) make the same argument suggesting that a thoroughexamination of the institutions of Western society is necessaryfor enduring changes toward environmental stability.

Available online at www.sciencedirect.com

Journal of Business Research 61 (2008) 885–893

⁎ Corresponding authors. Kilbourne is to be contacted at Clemson University,343-B Sirrine Hall, Clemson, SC 29634-1325. Pickett, Clemson University, 245Sirrine Hall, Clemson, SC 29634-1325, United States.

E-mail addresses: [email protected] (W. Kilbourne),[email protected] (G. Pickett).

0148-2963/$ - see front matter © 2007 Elsevier Inc. All rights reserved.doi:10.1016/j.jbusres.2007.09.016


Arguably, the institution most in need of examination andcritique is consumption itself. The literature only recently relatesconsumption to the environment per se. The present articlefocuses on the problem of materialism, or more specifically, thecentrality of consumption in theWestern industrial lifestyle and itsrole in individuals' willingness to adopt more environmentallybenign consumption types.

Numerous authors address the question of consumption in theenvironmental context (e.g., see Capra, 1982; Daly, 1991; Porritt,1984; Trainer, 1985). All argue from a conceptual frameworkrather than an empirical one however, and all address theconsequences of excess consumption from the perspectives ofpollution, waste, resource depletion, or some other side effect ofconsumption behavior. None addresses the impact of consump-tion practices from the perspective of the values and beliefs thatguide individuals' consumption behaviors.

The present article redresses this deficiency by examining therole of certain consumption patterns and the values that drivethem in the formation of environmental beliefs, expressions ofenvironmental concern, and environmentally responsible con-sumption behaviors (ERBs). In developing the model, theparticular form of consumption referred to as materialism is pre-sented first. Certainly, not all levels or types of consumption areequally complicit in environmental degradation. The marketingliterature, for example, substantiates green marketing efforts well,although Alwitt and Pitts (1996) question their efficacy inreducing environmentally damaging consumption practices.

2. Materialism

A diverse set of materialism constructs evolved over the lastfew decades. Bredemeier and Toby (1960) argue that consumersin the US believe the acquisition of material goods leads to thefulfillment of life. They also argue that materialism is the causeof many social problems. More recently, Belk (1985)characterizes materialism as the importance attached to worldlypossessions. Others suggest that possessions affect perceptionsof well-being (Burroughs and Rindfleisch, 2002) and act asidentity markers (Micken and Roberts, 1999). Browne andKaldenberg (1997) argue that materialism is a cluster of valuesrelated to possessions. While these definitions each describematerialism in slightly different ways, they have much incommon. They all suggest that consumers seek more in theconsumption process than the utility, or instrumental value, ofthe goods themselves and that the context of consumption isimportant.

Materialism is a value structure through which individualsseek more than instrumental value from the goods they acquire.They seek relationships with the objects of consumption thatform their identity and enhance their subjective well-being.Thus, materialism is a multi-faceted construct relating indivi-duals to the goods they possess. The institutionalized characterof materialism in Western societies has both individual andsocial consequences (e.g., see Ahuvia and Wong, 2002; Belk,1985; Richins and Dawson, 1992).

Ger and Belk (1996) argue that the materialistic lifestyle isexpanding on a global scale. Conversely, the Inglehart (1981)

thesis argues that materialism will decline as cultures developeconomically, and that materialistic values will diminish inimportance as economic stability improves. The literature doesnot support this, however. Both Feather (1998) and Ger and Belk(1996) demonstrate in cross-cultural studies that post-materialistvalues have not developed in advanced economic cultures and, atthe same time, materialistic values are growing in the lessdeveloped economies. The materialistic lifestyle is becoming aglobal phenomenon, and the number of individuals pursuing sucha lifestyle is increasing exponentially. This in turn, has thepotential to accelerate the associated negative consequences.

The present article focuses on the environmental conse-quences of materialism. Few in the environmental arena arguethat positive environmental consequences follow from materi-alistic behavior. The exception to this paucity of arguments iswithin the neoclassical economic literature where Bhagwati(1993), for example, argues that economic growth solves ratherthan exacerbates environmental problems. Lofdahl (2002)concludes, however, that the economic argument is speciousin failing to consider trade related economic growth.

The focal concern here is that the collective consequences ofindividual consumption behaviors have negative environmentalconsequences. These consequences emanate from social process-es that are characteristic of market based societies, and they arecategorized as a social trap (Dawes, 1980) and as a “commonsdilemma” (Shultz and Holbrook, 1999). Each suggests that thesum of individual behaviors produces a negative collective resultthat was unintended by any individual actor. While any singleindividual's actions have virtually no environmental conse-quences, the sum of all similarly disposed individuals' actionsdamages the environment of the collective, including theindividual actor (Dawes, 1980). The damage results because, inmarket societies, self-interest governs behavior, and the payoff forbehavior is higher if the individual acts in his/her self-interestregardless of what others do (Shultz and Holbrook, 1999).

Porritt (1984) argues that materialism found in marketsocieties is one of the root causes of environmental decline. InWestern industrial societies, materialism proffers a one-dimen-sional model of the “good life,” and its achievement is a primarysocietal objective (Schmookler, 1991). Jones (1987) argues that,because materialism is deeply embedded in the institutionalstructures of industrial societies, the institutions themselves mustbe examined. As a result, admonishing consumers to consumeless for personal, social, or environmental reasons is likely to beineffective in changing behavior. Culturally embedded institu-tions continuously reinforce and reward materialism as a mode ofconsumption. Consequently, one who is materialistic would findlittle reason for altering consumption behaviors to be moreaccommodating to the environment.

Critiques of materialism in the environmental context haveaddressed many relevant issues including resource depletion,pollution, and waste among others. While examining thesevariables is necessary, the approach is not sufficient becausethese variables are better characterized as symptoms of theproblem than as the problem itself. These symptoms can havevery negative consequences, but eliminating them does noteliminate the problem because the root causes of environmental

886 W. Kilbourne, G. Pickett / Journal of Business Research 61 (2008) 885–893


degradation remain unexamined. Solutions to current wastedisposal requirements, for example, cannot keep pace with theprofligate materialistic lifestyle of industrial societies. Stiglitz(2002) and Kilbourne (2004) argue that the globalization processis accelerating the spread of free market liberalism that proffersmaterialism as the solution to the environmental problem, and, asa result, environmental degradation is continuously acceleratingthrough the consequent increase in trade related economic growth(Lofdahl, 2002). Because of this, a more thorough examination ofmaterialism and its role in environmental decline is necessary.

3. Proposed environmental behavior model

Studies of ERB yield inconsistent results for a number ofreasons. One of these is the issue of measurement specificity.Ajzen and Fishbein (1980) argue that measurement of relation-ships between attitudes and behavior should be at the same levelof abstraction. Within the environmental domain, measurementspecificity problems attenuate the relationship between differentenvironmental constructs and ERB (Alwitt and Pitts, 1996;Fransson and Garling, 1999). This methodological issue makesexamining the relationship between materialism and ERBproblematic as well. The direct assessment of materialism as anabstract construct and specific environmental behaviors mightnot yield consistent results because the levels of abstraction arequite different. To attenuate the problem, the hypothesizedmodel has multiple levels with each becoming less abstract. Thecausal sequence proceeds from the most abstract level tobehaviors as suggested in the model proposed by Stern et al.(1995). Their model suggests beginning with the values leveland proceeding to general beliefs, specific beliefs and concerns,and ending with behavioral intentions and behaviors. Thus, theproposed hierarchical model descends from the abstract to thespecific. The next section explains the model and its constituentconstructs.

4. Model conceptualization

Consider the following brief examination of the constructsfound in the proposed model. The discussion of the relation-ships between the constructs leads to the hypotheses. Theconstructs examined include materialistic values, environmentalbeliefs, environmental concern, and environmental behaviors.

4.1. Values

The question posed here is how materialistic values contributeto the environmental problem. Rokeach (1973) and Schwartz andBilsky (1987) argue that values precede beliefs and attitudes andguide their formation. Values transcend situations and arerelatively enduring. Stern et al. (1995) view value orientationsas general predispositions influencing specific environmentalbeliefs held by individuals. Numerous other researchers haveexamined values in the environmental context and foundrelationships (e.g., see Fransson and Garling, 1999; Schultz,2001) between consumption, environmental attitudes, recycling,political behavior, and several others.

Thus, considerable support exists for the role of values in thedevelopment of environmental beliefs, attitudes, and behavior.This refers almost exclusively to general value orientations, andfew studies address values as they relate to modes of con-sumption. Kilbourne et al. (2005) add a necessary dimension tothe values problem, however. In a multi-national study of univ-ersity students, they demonstrate that general values orienta-tions are related to materialism. Specifically, they demonstratean inverse relationship between transcendent values andmaterialism and a direct relationship between enhancementvalues and materialism using Richins and Dawson's (1992)conceptualization of materialism. Kilbourne et al. (2005) didnot examine the relationship between materialism and environ-mental beliefs. Thus, while Fisk (1973) and Porritt (1984) bothsuggest that consumption is an important factor in environ-mental degradation, little empirical evidence relating thevarious types of consumption to environmental behaviorsexists.

4.2. Environmental beliefs

Stern et al. (1995) and Dietz et al. (1998) propose a model inwhich environmental beliefs are subsequent to values. Generalbeliefs are about the human–environmental relationship andrefer to folk wisdom about the environment (Stern, 2000). Mostresearch uses Dunlap and Van Liere's (1978) new environmen-tal paradigm (NEP) to measure general beliefs. Specific beliefsare beliefs about the existence of environmental problems suchas water shortages, ozone depletion, and global warming.

The focus in this study is on specific beliefs. Logically,concern would not arise unless preceded by the belief thatenvironmental problems exist. However, an individual maybelieve in a human–environmental relationship characterizedby ecological folk wisdom without ever being concerned thatproblems actually do exist. In both the Stern et al. (1995) andthe Dietz et al. (1998) models, specific beliefs and attitudesprecede intentions and behaviors. Stern (2000) further arguesthat the link from values to environmentalism mediates beliefsbecause perceived consequences to whatever the individualvalues activate norms. Thus, measures of beliefs in this studyrefer to specific adverse consequences for valued environmentalobjects. This suggests that individuals might be in a state ofcognitive dissonance (Festinger, 1957) if their desired con-sumption behavior is perceived to cause negative environmentalconsequences. As evidence of negative environmental con-sequences emanating from increasing consumption accumu-lates, cognitive dissonance should increase in materialisticindividuals. How this dissonance is resolved depends upon thecircumstances relating to the relative attractiveness of thealternatives (Festinger and Carlsmith, 1959). Because materi-alistic values have been institutionalized in American culture(Wachtel, 1983), while environmentalism is relatively new andless integrated into cognitive structures, the conflict betweensuch disparate values would be resolved in favor of materialism.That is, the individual would likely distort environmentalinformation conflicting with materialistic values. This leads tothe first set of hypotheses for the model.

887W. Kilbourne, G. Pickett / Journal of Business Research 61 (2008) 885–893


H1a. As beliefs in materialism as a measure of success increase,belief in the existence of environmental problems decreases.

H1b. Asbeliefs inmaterialism as ameasure of happiness increase,belief in the existence of environmental problems decreases.

H1c. As the centrality of materialism increases, belief in theexistence of environmental problems decreases.

If the individual perceives that the environment as a valuedobject is threatened, environmental concern will increase. Thiswill then increase the likelihood of more environmentallyfriendly consumption behavior. For this to happen, however, theindividual must exhibit some level of concern.

4.3. Environmental concern

Alwitt and Pitts (1996) state that a gap between environ-mental attitudes and behaviors exists within the US, and thatthis is relevant for policy makers and marketers alike. This isbecause social marketers and public policy makers seek toreduce environmentally negative behavior, and if their inten-tions are not successful, then more market regulation may benecessary. Thus, a more thorough understanding of the relation-ship between concern and behavior is necessary.

Studies of environmental concern in marketing began in theearly 1970s with a series of studies relating to various di-mensions of the problem. Anderson and Cunningham (1972),Kinnear et al. (1974), and many others sought to characterizeenvironmentally concerned consumers. Since that time, envi-ronmental concern has never declined as an environmentallyrelated construct within the marketing literature. Fransson andGarling (1999) argue that, while concern plays a role in be-havior change, research uses the construct inconsistently. Boththe definition and measurement of concern vary in research. Intheir review of research on environmental concern, theymaintain a definition that includes attitudes, values, and folkwisdom measures. Their review posited that low correlationsbetween concern and behavior are usually attributable to ameasurement issue relating to the specificity of themeasures used.Two conclusions drawn by Fransson and Garling (1999) are thatenvironmental concern should include both narrow and moregeneral measures and that increasing environmental concern as amatter of policy can result in desired behavior changes.

Following Fransson and Garling (1999), the present articleassesses concern at both individual and social levels. The firstlevel relates to abuse of the environment by individual con-sumers, and the second includes perceived need for social,political, and legal changes to protect the environment.

H2. As environmental beliefs increase, environmental concernincreases.

Drawing from cognitive consistency theory again, to theextent that one is concerned about environmental problems,consistency suggests they should be motivated to act on them.When concern increases, individuals should be more willing tomake changes in their own behavior. The discussion of thedifferent forms of behavior is next.

4.4. Environmental behavior

Like concern, research uses a number of different independentvariables to examine environmental behaviors. These haveincluded such individual factors as locus of control (McCartyand Shrum, 2001) and gender (Stern et al., 1993) and such socialfactors as collectivism (McCarty and Shrum, 2001) and politicalattitudes (Blake, 2001) among others. The relationship betweenbehavior and antecedent conditions has been inconsistent acrossall the different independent variables, and the inconsistencycould be the result of measurement deficiencies and definitionalproblems (Fransson and Garling, 1999; Stern, 2000).

Stern (2000) argues that one of the measurement problems isthat environmental behavior is not a one-dimensional construct,and recent evidence points to different types of behaviors thathave different causal factors. One way to classify behaviors is todistinguish between public and private spheres. Public spherebehavior refers to activities such as petition signing, contribu-tions, and joining environmental groups. These differ fromdirect behaviors and activism (Dietz et al., 1998). They affectthe environment only indirectly by influencing public policy.Private sphere behaviors have a direct effect on the environmentand include such behaviors as green consumption, recycling,and purchasing organic foods. Such behaviors, while they di-rectly affect the environment, are only efficacious in the ag-gregate, that is, when many people perform the same behaviors.Dietz et al. (1998) describe three types of ERBs they refer to asconsumer behaviors, environmental citizenship, and policysupport. The first two relate to private and public sphere be-haviors respectively, and the third reflects willingness to sacrificeeconomically through such things as higher taxes, higher prices,and consuming less. The present study takes the approach tobehavior measurement Stern and his colleagues suggest andexamines two different types of behavior. The behaviors reflectdirect and indirect effects on the environment, and these form thebasis for the last set of hypotheses to be tested.

H3a. As one becomes more concerned about the environment,self-reported direct ERBs such as reducing waste, buying envi-ronmentally friendly products, and buying organic will increase.

H3b. As one becomes more concerned about the environment,self-reported indirect ERBs such as joining or contributing toenvironmental organizations and contacting political represen-tatives will increase.

The causal structure presented in Fig. 1 summarizes this setof relationships. The model specified provides a summary of theconstructs and their relationship to each other. The positiveand negative paths indicate the hypothesized direction of therelationships.

5. Methodology

5.1. Sample

The data collection procedure for the study was a telephonesurvey using a sampling frame provided by Scientific Telephone



Samples, a research service that provides random samples oftelephone numbers throughout the US. From this sampling frame,interviewers called respondents. The callers eliminated refusalsand replaced no-answers into the database where they could bedrawn again on a successive selection. A no-answer was replaced9 times before removed. Interviewers repeated the process untilthey finished the required number of interviews.

The final number of calls completed was 337. The researcherseliminated 34 of these because of incomplete data. Respondentswere required to be 18 years old ormore to participate in the survey.The final sample consisted of 303 respondents of which 44%weremale. The median and average age of the respondents were both48 years which is slightly higher than the distribution of 18+individuals in the population. The median of this age group in theUS is approximately 44 years. The education level indicated that55% had some college and that 20% had completed a four-yeardegree. The median family income for the sample was approx-imately $45,000. Thus, the sample was a reasonable representationof the US population for all of the demographics measured.

5.2. Measurement instrument

The questionnaire for the study consisted of eight sectionswith seven measuring different constructs and the last,

demographics. Because of the large number of constructsmeasured and the time limitations imposed on telephonesurveys, the scales used for some constructs were reducedfrom their original form. The four scales used in the presentstudy were a materialism scale (9 items), an environmentalbelief scale (6 items), an environmental concern scale (6 items),and an environmental behavior scale (8 items). All the items inthe first three scales were Likert type with 1 indicating StronglyDisagree and 7 indicating Strongly Agree. The eight behavioralquestions were yes/no regarding the specific behavior. Appen-dix A provides the scale items and their statistics.

5.2.1. MaterialismThe materialism scale was a subset of items taken from the

Richins and Dawson (1992) materialism scale. Materialismresearchers use this scale extensively, and social acceptabilitybias is low. While the original scale was composed of 18items, the present study used only nine. The nine items chosen(see Appendix A) are those for which the factor loadings werehighest for their factors in the original Richins and Dawson(1992) study. The use of such short forms of the scale hasrecently been justified by Richins (2004). The original scaleprovided a three-factor solution with success, happiness, andcentrality as the three independent dimensions of the con-struct. Kilbourne et al. (2005) demonstrate that the constructsused here are consistent with the original dimensionalityacross three countries. Because of the limited use of thisreduced Material Values Scale, the psychometric propertieswere examined.

While scale development was not the purpose of this study,data were available through which external criteria (Stantonet al., 2002) for the reduced scale could be reasonably tested. Areduced form of Schwartz's Value Inventory (SVI) wasincluded in the test instrument and used as a test for therelationship between the reduced materialism scale and therequisite constructs in the SVI. In addition, a measure of thedominant social paradigm, the components of which have beenshown to relate to consumption intentions (Kilbourne, 2002),was included. These two scales provide an opportunity toexamine the relationship of the short Material Values Scale toexternal criteria. In this assessment, the reduced Material ValuesScale correlated positively with the constructs of the dominantsocial paradigm and the self-transcendence and opennessdimensions of the SVI as predicted. All correlations betweenthe reduced Material Values Scale and the external criteria werestatistically significant at pb0.05.

Confirmatory factor analysis (CFA) was used to assess thevalidity of the scales in this context. The fit statistics and cutoffcriteria used included the goodness of fit index (GFI) at 0.9, theTucker–Lewis Index (TLI) at 0.9, the comparative fit index(CFI) at 0.9, and the root mean square error of approximation(RMSEA) at 0.08. These are the criteria recommended by Hairet al. (1998). CFA confirmed the original three-factor structurewith the appropriate three items corresponding to each constructand acceptable fit statistics. Table 1 presents the fit statistics forall constructs. In addition, all the fit statistics for the reducedMaterial Values Scale were within 0.001 of the same fit criteria

Fig. 1. Proposed causal model.



reported by Richins (2004) in assessing a similar (four itemswere common to both forms) reduced Material Values Scale. Inassessing the reliability of the reduced Material Values Scale,composite reliability and variance extracted for the constructswere success (.83 and .62), centrality (.71 and .46), andhappiness (.82 and .60). Only the variance extracted forcentrality was below the desired .50 level. Thus, the reducedMaterial Values Scale used in this study appears to satisfy bothinternal and external criteria and maintained the properdimensionality.

5.2.2. Environmental belief scaleThe environmental belief scale consisted of six items

derived from Kilbourne, et al. (2002) and Cotgrove (1982).Each item reflected belief in the existence of environmentalproblems such as ozone depletion, global warming, pollution,damage from chemicals, resource shortages, and speciesextinction. As seen in Table 1, the scale met the fit criteriaestablished, and the factor loadings were all significant at lessthan 0.01.

5.2.3. Environmental concern scaleThe third scale used in the analysis was a measure of

environmental concern indicating that the respondent wasconcerned about the environment and believed that individual,social, and political changes were necessary to reduce damageto the environment. The intention was for the scale to containthree items for individual concern and three for social concern,but exploratory factor analysis indicated only one factorexplaining 55% of the variance. The six items reflected con-cern, environmental abuse, importance of limiting consump-tion, political and social change, and stricter enforcement ofenvironmental laws.

A subsequent EFA combining the items representing beliefsand concern yielded two factors explaining 56% of the varianceinsuring that beliefs and concerns represented two distinctfactors. All loadings on the appropriate factors were above .6,and all cross loadings were below .3 except for shortages, whichloaded on concern at 0.36. This indicates clearly that the twomeasures are distinct.

5.2.4. Environmental behavior scaleThe present study used two relevant types of ERBs, direct

and indirect. The behavior measurement model includes both ofthese types. Four items measured direct actions perceived to

have immediate, positive effects on the environment if manypeople elicit them. The items relate to purchasing environmen-tally friendly products, organic products, products that reducehousehold waste, and products that contain recycled material.Four items measured the indirect effects of joining environ-mental organizations, contributing money to environmentalorganizations, subscribing to environmental magazines, andcontacting a legislative policy maker. Each of the eight itemswas a yes or no question indicating the behavior and clearlyreflects self-reported behavior. Exploratory factor analysis wasperformed on the items to determine if they reflected differenttypes of behaviors. The results of the analysis indicated twofactors explaining 54% of the variance. The two factorsseparated the items as intended with the first factor containingdirect actions and the second containing indirect actions. Asargued by Stern (2000) and Dietz et al. (1998), these types ofbehavior should be distinguished from each other. In the modeltested, they were considered to be two separate dependentvariables.

6. Results

Each of the individual measurement models satisfied the fitcriteria set out. The causal model in Fig. 1 was then tested.Environmental beliefs and environmental concern were bothfirst-order constructs as indicated above.

The analysis confirmed most of the hypotheses. Hypothesis1a–1c stated that the paths from success, happiness, andcentrality in the materialism construct to environmental beliefswould be negative. This was true for each of the dimensions ofmaterialism. The final standardized path coefficients forsuccess, happiness, and centrality were −0.26, −0.24, and−0.26 respectively, and all of the parameter estimates weresignificant at less than 0.003. Hypothesis 2 indicated that thepath from environmental beliefs to environmental concernwould be positive. The path coefficient was 0.73 and wassignificant at less than .001. This confirmed Hypothesis 2.Hypotheses 3a–b suggested that the paths from environmentalconcern to both direct and indirect behaviors would be positive.The standardized path coefficient from concern to directbehavior was 0.35 and was significant at less than 0.001. Thestandardized path coefficient from concern to indirect behaviorswas 0.36 and was significant at less than 0.001. These resultsconfirm both H3a and H3b. For the entire model, CFI= .953,GFI= .901, TLI= .957, and RMSEA=0.037. Thus, the pro-posed model of materialism and ERBs fit the data extremelywell.

7. Discussion

Researchers in the past few decades examined materialismfrom many perspectives. The conclusion frequently drawn hasbeen that materialism has negative consequences in both theindividual and social domains. Material consumption alsorelates to identity formation and maintenance, and to theextent that consumption provides for this human requirement,it serves a useful function. Because of the frequent critiques,

Table 1Statistical results for analyses

RMSEA GFI CFI TLI

Material values .056 .97 .98 .96Environmental beliefs .055 .97 .98 .96Environmental concern .075 .98 .98 .96Environmental behavior .054 .97 .96 .95Direct .027 .99 .99 .99Indirect .001 .99 .99 .99

Structural model .037 .90 .95 .95



however, Mick (1996) refers to materialism as a “dark side”variable. One aspect of the dark side of materialism, and thesubject of this article, is the relationship between materialismand the natural environment. The consensus among envir-onmentalists is that the environment is in a precipitousdecline, and a better understanding of the “root causes” ofthe decline is necessary (Porritt, 1984). The materialisticlifestyle of Western industrial societies is among these rootcauses.

The present study hypothesized a negative relationshipbetween materialism, as measured by the Richins and Dawson(1992) Material Values Scale, and environmental beliefs. Thepositive relationship between environmental beliefs andenvironmental concerns found in previous research (e.g., seeKilbourne et al., 2002) is hypothesized here as well. Finally, thisstudy tested the important link between environmental concernand ERBs.

The key argument underlying the model tested is that whenconfronted with the negative environmental consequences oftheir behavior, consumers experience cognitive dissonance thatmust be resolved to preserve their self-image. Individualssimply do not like to see themselves as profligate consumerswhose desire for material goods is destroying the environmentat accelerating rates. The consequent dissonance can only bereconciled if they change their views of the value of materialismor change their views about the consequences of their behavior.Materialism is institutionalized in American society and iscontinuously rewarded and reinforced through interactions withsociety. As a result, it has the power to influence perceptionsindividuals hold of themselves and their environment. Envi-ronmentalism, or the desire to take action with proenviron-mental intent (Stern, 2000), is not such a potent force inAmerican society. Because of the ubiquity of reinforcements formaterialism and the relative paucity of such reinforcements forenvironmentalism, the dissonance would logically be resolvedin favor of materialism. Thus, as beliefs in materialism increase,perceptions of environmental problems would decrease throughprocesses of selective perception and distortion. This implies anegative relationship between materialism and environmentalbeliefs. The data indicated that this negative relationship didmaterialize as hypothesized.

The study also confirmed some previously demonstratedrelationships. The path from environmental beliefs to concernwas positive as has been shown in previous research. Thisindicates that, as one's belief in the existence of environmen-tal problems increases, their level of concern also increases.The present study also supports previous research that linkselevated levels of concern with the desire for behaviorchanges.

Stern (2000) argues that both direct and indirect environ-mental behavior are important. These relate to behaviors thathave immediate effects on the environment and those that leadto later effects through the policy process. The hypothesizedpositive relationship for both direct and indirect behaviorssuggests that raising levels of environmental concern increasesthe likelihood of proenvironmental behaviors. Alwitt and Pitts(1996) found no direct relationship between general environ-

mental concern and purchase intentions, but product attitudeand importance mediate the effect of concern. This is consistentwith the measurement issues discussed earlier. General concernstypically do not link to specific behaviors. However, in thepresent study, the measure of concern relates to specificenvironmental problems believed to exist. Environmentalconcern is more specific here than a general concern measuresuch as the NEP scale that does not relate to specific problems.Thus, the present study shows a close relationship betweenconcern and behaviors.

The overall structural model fit the data very well in thisstudy. Materialism was negatively associated with environ-mental beliefs as predicted. From a policy standpoint, thisresult is very significant. Many environmental policy initiativesincrease awareness and concern for environmental problems.This assumes that when concern is increased, consumptionbehaviors would become more environmentally responsible,and consumers would shift their purchase patterns by buyinggreen products or by reducing their overall level of consump-tion. While the relationship between concern and behaviorsproposed seems reasonable, neither of these results hasemerged in practice except in an ephemeral fashion (i.e., theexistence of the attitude–behavior gap described earlier). Theresults of this study offer a suggestion as to why this may be thecase.

Stern et al. (1995) argues that the process through whichchanges might occur is more complex than that assumed bypolicy makers. The present study confirms the more complexmodel and has significant policy implications. Because thevalues associated with materialism are ubiquitous in Americansociety, they become the evaluative filter for many aspectsof daily life. This is important from a policy perspectivebecause the information received and how such information iscategorized are among the factors on which the values filteroperates (Festinger, 1957). Information that is inconsistent withone's primary values or that creates dissonance within the self-image, is selectively removed and/or distorted to conformto that image. Consequently, because of the centrality of val-ues defining the self, individuals' beliefs in the detrimentalconsequences of their consumption behavior are attenuated.Environmentally relevant information rejected or distortedby this filtering process can reduce the efficacy of environ-mental initiatives. This has significant implications for policyconstruction.

If the objective of environmental or consumer policy is tochange environmentally damaging consumption behavior intoERB, then an expanded approach to consumer policy isnecessary. Because materialistic values militate againstconsumer policy aimed at lower levels of the model, policyshould be directed at those values themselves. This meansconsumer policies aimed at materialism might be necessary.This, of course, entails a long-term approach because ma-terialism is so deeply entrenched in American economicpolicy. This would involve changes in both consumer andproduction policy relating to both the quality and quantity ofgoods produced and consumed. Promotion of alternativelifestyles, such as voluntary simplicity (Elgin, 1981) and



sustainable production alternatives would both become a partof consumer policy seeking to alter unsustainable behavior.The problems with such transformations are clear, and oneneed only recall the US rejection of the environmental policyspecified in the Rio Accord to see this. In his rejection of theaccord, then President Bush stated that the “American way oflife” was not up for negotiation. Still today, economic growthunderpins the American way of life, and that requiresmaterialism as a dominant value. Boulding (1949) arguedagainst such a policy more than fifty years ago, and Daly(1996) continues the position into the present.

7.1. Limitations and future research

The present study is limited in several ways. The use oftelephone surveys limited that amount of interaction betweenresearchers and the randomly selected respondents. Thus, noqualitative information was collected that might have amplifiedresponses. In addition, the amount of nonresponse bias isunknown. The response rate was approximately 53%. Themethods of analysis also precluded the possibility of determin-ing causal relationships in the data. While the theory citedsupports the causal sequence tested, only correlational methodswere available to analyze the data.

The study does provide some direction for future re-search. The development of policy tools ought to take intoconsideration the materialistic lifestyle. While outside thescope of the present study, future research exploring the linkthe educational system, business models, and social systemshave to the materialistic lifestyle and its relationship tosustainability, particularly on a global scale, is necessary. Eachsuccessive generation will understand consumption from aunique cultural perspective. Hence, learning more about theextent to which the next generation will be more or lessmaterialistic and what impact this perspective may have onenvironmental beliefs is important. Similarly, as generationalvalues are shaped by scientific and societal forces (forinstance, the acceptance of global warming and the main-streaming of media coverage imparting this knowledge), onemight expect change in generally held beliefs regarding theimpact of individual versus societal action. The progression ofenvironmental decline now understood by a broader cross-section of individuals in many Western societies may producenew and interesting research results in this field for years tocome.

Marketers might revisit Kotler's (2002) social marketingconcept and examine its importance in the sustainability debatethat will surely intensify in the coming years. The socialmarketing concept raises the question regarding the efficacy ofmarkets in a regime of limited resources. Policy must addressthe consequences of resource allocation at both systemic leveland individual levels. Otherwise, severe limits with the long-term effectiveness of any future policies will occur. Many haveargued that materialism has negative consequences for bothindividuals and society. Recent empirical support substantiatesthe conclusion. Addressing the larger issues that are growing inimportance is past due.

Appendix A

Questionnaire items and statistics

Factor⁎

Loading

Material values scaleSuccess (coefficient alpha .83 average loading .79)

I admire people who own expensive homes, cars, and clothes .78Some of the most important achievements in life include acquiringpossessions

.80

I don't placemuch emphasis on the amount ofmaterial objects people own .79As a sign of success (R)

Centrality (coefficient alpha .67 average loading .66)I usually buy only the things I need .73I try to keep my life simple, as far as possessions are concerned (R) .82The things I own aren't really that important to me (R) .42

Happiness (coefficient alpha .81 average loading .77)I have all the things I really need to enjoy life (R) .88My life would be better if I owned certain things I don't have .83I'd be happier if I could afford to buy more things .59

Environmental beliefs (coefficient alpha .86 average loading .70)Many types of pollution are rising to dangerous levels .71Some living things are being threatened with extinction .70Continued use of chemicals in agriculture will damage the environment .74Shortages of some important resources will occur in the near future .76Global warming is becoming a problem .65Ozone depletion is an environmental problem .65The availability of clean water will become a problem in the future

Environmental concern (coefficient alpha .81 average loading .64)I am very concerned about the environment .61Humans are severely abusing the environment .64I would be willing to reduce my consumption to help protect theenvironment

.66

Major political change is necessary to protect the natural environment .61Major social changes are necessary to protect the natural environment .60Anti-pollution laws should be enforced more strongly .69

Environmental behaviorsDirect behavior (coefficient alpha .69 average loading .61)

I buy environmentally friendly products whenever possible .71I reduce household waste whenever possible .61I use products made from recycled material whenever possible .71I buy organic food whenever possible .42

Indirect behavior (coefficient alpha .70 average loading .62)I am a member of an environmental organization .77I contribute money to an environmental organization .71I subscribe to an environmental magazine .62I would contact my political representative about an environmental issue .39

⁎Critical ratios are all above 6 indicating p-values less than .001.

References

Ahuvia AC, Wong NY. Personality and values based materialism: theirrelationship and origins. J Consum Psychol 2002;12(4):389–402.

Ajzen I, Fishbein MA. Understanding attitudes and predicting social behavior.Englewood Cliffs, NJ: Prentice-Hall; 1980.

Alwitt LF, Pitts RE. Predicting purchase intentions for an environmentallysensitive product. J Consum Psychol 1996;5(1):49–64.

Anderson TW, Cunningham WH. The socially conscious consumer. J Mark1972;36(3):23–31.



Bhagwati J. The case for free trade. Sci Am 1993;269:42–9 (November).Belk RW. Materialism: trait aspects of living in the material world. J Consum

Res 1985;12:265–80 (December).Blake D. Contextual effects on environmental attitudes and behavior. Environ

Behav 2001;33(5):708–25.Boulding K. Income or welfare? Rev Econ Stud 1949;17:79.Bredemeier HC, Toby J. Social problems in America: Costs and casualties in an

acquisitive society. New York, NY: Wiley; 1960.Browne BA, Kaldenberg DO. Conceptualizing self-monitoring: links to

materialism and product involvement. J Consum Mark 1997;14(1):31–44.Burroughs JE, Rindfleisch A. Materialism and well-being: a conflicting values

perspective. J Consum Res 2002;29:348–70 (December).Capra F. The turning point. London: Fontana Flamingo Series; 1982.Carson R. Silent spring. Harmondsworth: Penguin; 1962.Cotgrove S. Catastrophe or cornucopia: the environment, politics, and the

future. New York: Wiley; 1982.Daly HE. Beyond growth. Boston, MA: Beacon Press; 1996.Daly HE. Steady-state economics. Washington, DC: Island Press; 1991.Dawes RM. Social dilemmas. Annu Rev Psychol 1980;46:190–203.Dietz T, Stern PC, Guagnano GA. Social structural and social psychological

bases of environmental concern. Environ Behav 1998;30(4):450–71.Dowie M. Losing ground: American environmentalism at the close of the

twentieth century. Cambridge, MA: MIT Press; 1995.Dunlap RE, Van Liere K. The “New Environmental Paradigm”: a proposed

instrument and preliminary results. J Environ Educ 1978;9(1):10–9.Elgin D. Voluntary simplicity: toward a way of life that is outwardly simple,

inwardly rich. New York: Morrow; 1981.Feather NT. Attitudes toward high achievers, self-esteem, and value priorities

for Australian, American, and Canadian students. J Cross-Cult Psychol1998;29(6):749–60.

Festinger L. A theory of cognitive dissonance. Stanford, CA: StanfordUniversity Press; 1957.

Festinger L, Carlsmith JM. Cognitive consequences of forced compliance.J Abnorm Soc Psychol 1959;58:203–10.

Fisk G. Criteria for a theory of responsible consumption. J Mark 1973;37(1):24–31.Fransson N, Garling T. Environmental concern: conceptual definitions, measure-

ment methods, and research findings. J Environ Psychol 1999;19(4):369–82.Ger G, Belk RW. Cross-cultural differences in materialism. J Econ Psychol

1996;17(1):55–77.Hair JF, Anderson RE, Tatham RL, Black WC. Multivariate data analysis.

(5 ed.). Upper Saddle River, NJ: Prentice Hall; 1998.Inglehart R. Post-materialism in an environment of insecurity. Am Polit Sci Rev

1981;75:880–900.Jones A. The violence of materialism in advanced industrial society: an eco-

sociological approach. Sociol Rev 1987;35(1):19–47.Kilbourne W. Globalization and development: an expanded macromarketing

view. J Macromark 2004;24:122–35 (December).Kilbourne W, McDonagh P, Prothero A. Sustainable consumption and the

quality of life: a macromarketing challenge to the dominant social paradigm.J Macromark 1997;17(1):4–24.

Kilbourne WE. A multi-national examination of the role of the DSP inenvironmental attitudes: a structural equation modeling approach. In: WilsonJW, editor. FromArt to Technology: Opportunities in Marketing Research andEducation. Savannah, GA: Atlantic Marketing Association; 2002. p. 102–6.

Kilbourne WE, Beckmann SC. Review and critical assessment of research onmarketing and the environment. J Market Manag 1998;14(6):513–32.

Kilbourne WE, Beckmann SC, Thelen E. The role of the dominant socialparadigm in environmental attitudes: a multi-national examination. J BusRes 2002;55(3):193–204.

Kilbourne WE, Grünhagen M, Foley JA. Cross-cultural examination of therelationship between materialism and individual values. J Econ Psychol2005;26(5):624–41.

Kinnear TC, Taylor JR, Ahmed SA. Ecologically concerned consumers: who arethey? J Mark 1974;38(2):20–4.

Kotler P. Marketing management. (11 ed.). Upper Saddle River, NJ: Prentice-Hall; 2002.

Leonard-Barton D. Voluntary simplicity lifestyles and energy conservation.J Consum Res 1981;8(3):243–52.

Lofdahl CL. Environmental impacts of globalization and trade. Cambridge,MA: The MIT Press; 2002.

McCarty JA, Shrum LJ. The influence of individualism, collectivism, and locusof control on environmental beliefs and behavior. J Public Policy Mark2001;20(1):93–104.

Menon A, Menon A. Enviropreneurial marketing strategy: the emergence ofcorporate environmentalism as market strategy. J Mark 1997;61(1):51–67.

Mick DG. Are studies of dark side variables confounded by socially desirableresponding? The case of materialism. J Consum Res 1996;23:106–19(September).

Micken KS, Roberts SD. Desperately seeking certainty: narrowing thematerialism construct. In: Arnould E, Scott L, editors. Association forConsumer Research, vol. 26. ACR; 1999. p. 513–8.

Porritt J. Seeing green: the politics of ecology explained. Oxford, UK: BasilBlackwell; 1984.

Richins ML. The material values scale: measurement properties anddevelopment of a short form. J Consum Res 2004;31(1):209–19.

Richins ML, Dawson S. A consumer values orientation for materialism and itsmeasurement: Scale development and validation. JConsumRes 1992;19:303–16(December).

Rokeach M. The nature of human values. New York: Free Press; 1973.Schmookler AB. The insatiable society: materialistic values and human needs.

Futurist 1991;25(4):17–20.Schultz WP. The structure of environmental concern: concern for self, other

people, and the biosphere. J Environ Psychol 2001;21(4):327–39.Schwartz S, Bilsky W. Toward a psychological structure of human values. J Pers

Soc Psychol 1987;53:550–62.Shultz CJ, Holbrook MB. Marketing and the tragedy of the commons: a

synthesis, commentary, and analysis for action. J Public Policy Mark1999;18(2):218–29.

Smith TM. The myth of green marketing: tending our goats at the edge ofapocalypse. Toronto: University of Toronto Press; 1999.

Stanton JM, Sinar EF, Balzar WK, Smith PC. Issues and strategies for reducingthe length of self report scales. Pers Psychol 2002;55:167–94 (Spring).

Stern PC. Toward a coherent theory of environmentally significant behavior.J Soc Issues 2000;56(3):407–24.

Stern PC, Dietz T, Guagnano GA. The new ecological paradigm in social–psychological context. Environ Behav 1995;27(6):723–43.

Stern PC, Dietz T, Kalof L. Value orientations, gender, and environmentalconcern. Environ Behav 1993;25(3):322–48.

Stern PC, Dietz T, Kaloff L, Guagnano GA. Values, beliefs, and proenviron-mental action: attitude formation toward emergent attitude objects. J ApplSoc Psychol 1995;25(18):1611–36.

Stiglitz JE. Globalization and its discontents. New York, NY: W. W. Norton &Company; 2002.

Trainer T. Abandon affluence! London: Zed Books; 1985.Wachtel PL. The poverty of affluence: a psychological portrait of the American

way of life. New York: The Free Press; 1983.


How materialism affects environmental beliefs, concern, and environmentally responsible behavior

Documents