A Closer Look At the Influence of Age on Consumer Ethics

Barry J. Babin, University of Southern Mississippi
Mitch Griffin, Bradley University
ABSTRACT - Shoplifting is among the most serious and common aberrant consumer behaviors (ACB), yet has been largely ignored in the consumer behavior literature. This study empirically examines this topic, treating shoplifting as a behavior with ethical consequences, thus bringing together research on ACB with emerging research on ethical decision making. This research makes two important contributions to the current literature: (1) multiple age groups are considered to identify the differences between adolescents, young adults, and fully mature adults, and (2) individual difference characteristics are tested as mediators of the ageCshoplifting relationship. The results support the position that adolescents are most prone to shoplifting behavior, but illustrate that this relationship is mediated by other factors.
[ to cite ]:
Barry J. Babin and Mitch Griffin (1995) ,"A Closer Look At the Influence of Age on Consumer Ethics", in NA - Advances in Consumer Research Volume 22, eds. Frank R. Kardes and Mita Sujan, Provo, UT : Association for Consumer Research, Pages: 668-673.

Advances in Consumer Research Volume 22, 1995      Pages 668-673


Barry J. Babin, University of Southern Mississippi

Mitch Griffin, Bradley University


Shoplifting is among the most serious and common aberrant consumer behaviors (ACB), yet has been largely ignored in the consumer behavior literature. This study empirically examines this topic, treating shoplifting as a behavior with ethical consequences, thus bringing together research on ACB with emerging research on ethical decision making. This research makes two important contributions to the current literature: (1) multiple age groups are considered to identify the differences between adolescents, young adults, and fully mature adults, and (2) individual difference characteristics are tested as mediators of the ageCshoplifting relationship. The results support the position that adolescents are most prone to shoplifting behavior, but illustrate that this relationship is mediated by other factors.


Shoplifting is among the most serious and common aberrant consumer behaviors (ACB) (Fullerton and Punj 1993). Statistics reflecting the magnitude of shoplifting are staggering (see Solomon 1994 for a detailed review). Most prominent among these statistics are figures representing shoplifting's pervasive impact on industry and society. For example, the total dollar value of goods shoplifted every year exceeds $30 billion (Fullerton and Punj 1993), over half of all U.S. consumers have shoplifted at some time (Klemke 1982), and shoplifting is a significant factor in many retail failures (Cole 1989). Considering these figures, shoplifting would classify as one of the developed world's leading industries.

Despite the pervasiveness of shoplifting as a consumer behavior and its profound impact on our society, consumer researchers have given very little attention to shoplifting behavior or, more broadly, ethical decision making among consumers (Tsalikis and Fritzsche 1989). An exception to this apparent oversight is recent research illustrating the prevalence of shoplifting among adolescent consumers and pointing out some potential contributory factors (Cox, Cox, and Moschis 1990). Specifically, this study reports that the percentage of the population who report shoplifting peaks during high school (see Cox et al. 1990, Figure 1).

Since shoplifting is purported to be most prevalent among adolescents, it could be argued that developing consumers become socialized over time and actually "learn" that this behavior is unacceptable. However, researchers have not generally examined multiple age groups or considered individual difference traits and their potential effects on shoplifting (Cox, Cox, Anderson, and Moschis 1993). Thus, it is difficult to determine the extent to which age actually causes shoplifting to be considered acceptable behavior.

The broad purposes of this paper are two-fold. First, the study treats shoplifting as a behavior with ethical consequences and thus ties in research on ACB with emerging research on ethical decision making. Second, the relationship between age (consumers varying in consumer socialization) and shoplifting is examined more closely. As Fullerton and Punj (1993, p. 570) point out, "The challenge for researchers is to identify those factors or interactions of factors which are likely to lead some consumers to misbehave some of the time." In the present study, we simultaneously consider several relevant personal traits of individuals. Thus, we examine the relationship between age and shoplifting compared to other traits that consumers develop over time.


Shoplifting as an Ethical Decision

Ethical decision making usually involves consideration of potential personal, cultural, and/or legal sanctions that may be transgressed as a result of the decision (Robin and Reidenbach 1987). Ethical judgments can be explained along three dimensions rooted in moral philosophy (Reidenbach, Robin, and Dawson 1991): (1) moral equity captures an individual's perceptions of an act's "justness" or "fairness;" (2) contractualism represents perceptions of implied or explicit contracts that would be violated by performing a questionable act; and (3) relativism represents how culturally acceptable an act is or has become. Thus, these three dimensions are thought to underlie decisions and judgments involving moral or ethical dilemmas.

We believe shoplifting fits the requirements of an ethical judgment. There are multiple barriers-personal, cultural, and legal-that must be overcome before a shoplifting act can be performed. Further, recent research indicates that the three dimensions described above (i.e., moral equity, contractualism, and relativism) may underlie consumers' shoplifting judgments. More specifically, survey results suggest that consumers' ethical judgments are related to both attitudes toward shoplifting and intentions to shoplift positively, and that highly ethical judgments of shoplifting may represent consumers' attempts to rationalize the behavior (Babin, Robin, and Pike 1994). As consumers judge a shoplifting act as more ethical (on each dimension), he/she is predisposed more favorably toward it and becomes a more likely participant in that behavior. These findings are consistent with other research addressing ethical questions in nonconsumer contexts (Reidenbach et al. 1991).

Age and Shoplifting Judgements

Substantial evidence suggests that age is related to ethical judgments of shoplifting. For example, thirty-seven percent of high school students studied reported that they had shoplifted in the previous year (Cox et al. 1990), while nearly forty percent of shoplifters apprehended are adolescents (Baumer and Rosenbaum 1984). In speculating on potential reasons for this phenomenon, Cox et al. (1990) suggest that adolescents may lack the moral development necessary to deter stealing from retailers. As consumers become socialized, they gain the development necessary to judge such acts differently. Thus, the following basic relationship is suggested:

H1: Age is related negatively to moral equity, contractualism, and relativism ethical judgments of shoplifting.

Individual Difference Traits and Shoplifting Judgements

A growing body of evidence suggests that most ACB results from a combination of situational and individual variables (Fullerton and Punj 1993). When investigating shoplifting behavior, previous empirical research has focused primarily on situational factors such as a product's economic value, the type of product being considered, the number of people present, and the chance of being apprehended (Babin et al. 1994; Cole 1989; Cox et al. 1990). Apart from age, individual difference traits have received little attention in comparison.

Although establishing a relationship between age and ethical judgments of shoplifting (H1) is useful descriptively, it fails to identify specific reasons for shoplifting. Interestingly, many individual traits that could be related to shoplifting attitudes and behavior tend to covary with age. Thus, the present study examines the partial contribution of age in predicting ethical judgments of shoplifting in light of these additional individual difference characteristics. The result is a test of mediation of the age-shoplifting judgment perceptions relationship by the individual difference characteristics considered. While space precludes a detailed analysis of each variable considered here, a brief discussion of each with respect to its potential mediating role follows.

Risk Aversion. Given the inherent risk associated with stealing, high levels of risk aversion may discourage someone from shoplifting. As analogous empirical evidence, Cox et al. (1990) found that shoplifting adolescents scored higher in measures of "rule breaking" than nonshoplifting adolescents. Although varying across all age groups, adolescents tend to be particularly low in risk aversion (Moschis 1987). Thus, there is the potential that part of the effect of age on shoplifting is due to risk aversion. Consequently, risk aversion should relate to ethical perceptions of shoplifting while controlling for the effect of age.

Self-Esteem. Shoplifters' self-esteem also may play a role in shoplifting decisions. Adolescents may view shoplifting as a way of fitting in or gaining acceptance among desired referents, a characteristic that may indicate low self-esteem (Brown and Lohr 1987). Self-esteem may also be a factor in adult shoplifting. Both adolescent and adult shoplifters tend to steal luxury items capable of enhancing self-esteem as opposed to necessity items whose need is more easily rationalized (Cameron 1964). This effect transcends social classes as well as age groups (Cox et al. 1990). Thus, self-esteem may be negatively related to shoplifting judgments. That is, consumers with low self-esteem will judge shoplifting acts to be relatively more ethical.

Susceptibility to Interpersonal Influence. Individual differences in susceptibility to peer influence can be captured along two dimensions (Bearden, Netemeyer, and Teel 1989). Normative influence refers to compliance with peers' expectations to help gain rewards or avoid punishment (e.g., buying the brand my friends approve of). Peer influence is also exerted through informational influence, where peer behavior is observed for information that is potentially useful in decision making (e.g., buying a brand because its what others are using).

Peer pressure has been hypothesized as a major influence on adolescent shoplifting (Klemke 1982). Adolescents with relatively high numbers of friends that shoplift report fewer moral objections to shoplifting and more frequent shoplifting behavior (Cox et al. 1993). One explanation for this observation is that adolescents are susceptible to interpersonal influence concerning shoplifting because it provides information allowing them to rationalize the behavior (i.e., "my friends are doing it, so it must be o.k. for me to do it too"). One might also argue that shoplifting is a way of gaining acceptance among peers. The former reason suggests a relationship between informational influence and shoplifting judgments, while the latter suggests a relationship between normative influence and shoplifting judgments. Based on the preceding discussion, we offer our second research hypothesis:

H2: Risk aversion, self-esteem, and susceptibility to interpersonal influence relate negatively to moral equity, contractualism, and relativism ethical judgments of shoplifting.

Individual Difference Traits as Mediators

We believe that an investigation of the personality variables presented above will extend our present knowledge of shoplifting. Furthermore, since these traits tend to covary with age, analyzing them simultaneously with age will provide a better understanding of the interrelationships among all variables. Specifically, we propose that these personality traits mediate the relationship between age and shoplifting judgements:

H3: The relationships between age and ethical judgments of shoplifting is attenuated when controlling for risk aversion, self-esteem, and susceptibility to interpersonal influence.


This study investigates three related hypotheses to test for mediation of the ageCshoplifting relationship. In this case, three requirements must be met to establish mediation: (1) a direct relationship between age and ethical judgements of shoplifting (H1) must be shown, (2) a direct relationship between risk aversion, self-esteem, and susceptibility to interpersonal influence and shoplifting judgements (H2) must exist when controlling for age, and (3) the magnitude of the relationship between age and ethical judgments of shoplifting must be reduced by controlling for the individual difference traits (H3) considered here. The following section describes a study examining these hypotheses.



A total of 168 respondents from the university community participated in the study in return for a small incentive. Variation in respondent age is required to test the hypotheses stated above. Thus, the study included three distinct age groups. Adolescent respondents were included as shoplifting has been considered particularly common among this age group. In the present study, local high school sophomores and juniors were interviewed at their school. Young adult respondents were students of junior and senior standing at the local university. This age group was selected to examine how rapidly consumers become socialized. Finally, a group of mature consumers was obtained from employers at a local manufacturing company. These respondents represented a wide range of demographic variables and included both blue and white collar workers. Forty-nine respondents were high school students (mage=16.0), 65 were college students (mage=21.7), and 54 were adult consumers (mage=40.1). The demographics of the sample were representative of the community with respect to family income levels and ethnic background.

Surveys were administered in group settings (15-25 respondents) within a single week. Given the potential sensitive nature of the questionnaire, several precautions were taken to help ensure respondent anonymity and confidentiality. Data collection closely followed the procedure described by Cox et al. (1990).

Data Collection and Measures

Respondents were handed a questionnaire booklet containing a brief shoplifting scenario and relevant measures. In the scenario, an ambiguously described consumer (Pat) is examining merchandise at a retail store in the local mall. When the clerk is distracted, the consumer shoplifts a small item worth about $25. Respondents then turned the page and were asked not to turn back to this page to respond to the items that followed.

Survey items that followed were assessed using a six point Likert format. Respondent perceptions of the ethicality of the shoplifting situation were collected using an ethical judgment scale validated in numerous marketing and consumer contexts (e.g., Babin et al. 1994; Bearden, Netemeyer, and Mobley 1993; Reidenbach et al. 1991). Five items assessed moral equity, three items contractualism, and two items assessed relativism.



Following these judgments and some filler items, several individual difference characteristics were assessed. Six items assessed respondent risk aversion and were patterned after items found in Zuckerman (1979) that reflected tolerance of risk (e.g., "I consider myself a risk-taker"). Respondent self-esteem was assessed using six items taken from Heatherton and Polivy (1991; Bagozzi and Heatherton 1994). Finally, susceptibility to interpersonal influence was assessed along two dimensions (informational and normative) using twelve items developed and tested previously in a consumer context (Bearden et al. 1989).

The Appendix briefly describes each scale and gives the measurement results. Coefficient a ranged from .65 (relativism) to .92 (normative influence) and averaged .78. Thus, individual scale items were summed forming measures used in further analyses. The three ethical judgment scales were coded so that respondents with higher scores perceived the shoplifting act as more ethical. The four individual difference traits were scored positively so that higher scores represented higher levels of the construct.

The correlations between summated constructs were assessed to examine the extent of colinearity. The correlations were modest with relativism and moral equity being the most highly correlated scales (r=.49). Informational and normative susceptibility to interpersonal influence were the most highly correlated predictor variables used in the study (r=.39) with only two other significant correlations. Thus, the colinearity among predictors does not appear high enough to hamper interpretation of effects.


A series of ANOVA and ANCOVA models were used to test the research hypotheses. Age was treated as a three level predictor variable based upon group (adolescents, college students, and adults). Table 1 shows the mean scores on each interval level variable included in the study by age group. Although not directly pertinent to the hypotheses, age predicted risk aversion, self-esteem, and informational influence significantly (p<.05). Normative influence was not affected by age.

Hypothesis 1

Given the potential intercorrelation among ethical judgment dimensions (Reidenbach et al. 1991), a multivariate analysis of variance was computed prior to analyzing individual effects. The results indicate that age affects ethical judgments significantly (L= .72; F6,304=8.90; p<.0001); thus, further analysis of each dependent variable is warranted.

Table 2 displays individual ANOVA results for each ethical judgment dimension. Moral equity (p<.001), contractualism (p<.05), and relativism (p<.001) all varied significantly with age. By comparing effect sizes, we can see that age has the strongest impact on perceived ethicality (w2=.25). Further, the means by age group suggest the relationship is in the hypothesized direction (see Table 1). In other words, high school respondents perceived the shoplifting act as significantly more ethical on each dimension than did either college students or adult respondents, providing support for H1. Interestingly, scores among college students and older adults did not differ significantly.

Hypothesis 2

Hypothesis Two examines the effect of several individual difference characteristics on respondents ethical judgments of the shoplifting episode. Table 3 shows the results of ANCOVA models testing each trait's effect while controlling for age. Individual parameter estimates are shown in Table 4.

The results show that self-esteem relates to perceived moral equity negatively (b=-.17; F=8.50; p<.01) while controlling for age. Contractualism is affected significantly by respondent risk aversion (p<.05) and susceptibility to interpersonal informational influence (p<.05). Increased risk aversion (b=-.11) and informational influence (b=-.14) decrease contractualism judgments. Finally, in addition to the effect of age, both risk aversion (p<.05) and self-esteem (p<.01) affect relativism significantly. Specifically, increased risk aversion (b=-.10) and high self-esteem (b=-.13) lower the perception that shoplifting behavior is culturally and/or traditionally acceptable. Thus, although not every potential relationship is significant, these five relationships are consistent with H2.

Hypothesis 3

Hypothesis Three requires a comparison of the strength of the relationship between age and ethical judgments of the shoplifting act before and after consideration of mediators (covariates). Effect sizes (w2) corresponding to these relationships are shown in Tables 2 and 4.

In all three cases, the effect size representing the direct relationship between age and perceived ethicality drops when the individual difference traits are added to the model. Specifically, the ageCmoral equity effect size drops from .25 to .15, the ageCcontractualism effect size drops from .05 to .02, and the ageCrelativism effect size drops from .08 to .03. The F-values representing the corresponding relationships indicate similar results. Likewise, the multivariate F drops to 4.82 (df14,288) when the four individual difference traits are entered into the model.








The data reveal that five of twelve potential relationships between individual difference traits and perceived ethical judgments of the shoplifting act are significant. None of the relationships representing normative influence are significant. Thus, risk aversion, self-esteem, and susceptibility to informational interpersonal influence all play a role in predicting ethical perceptions of shoplifting.

Results also address the extent to which these individual difference traits mediate the relationship between age and perceived ethicality. Full mediation would require that age's direct effect become nonsignificant when the mediators are entered into the model. In the current case, since the levels of significance and effect sizes are only reduced, evidence of partial mediation is presented (Baron and Kenny 1986).


The study presented here extends earlier research on shoplifting by examining multiple age groups. Our findings are consistent with previous research suggesting adolescents are more prone to shoplift than older consumers. A series of ANOVA models indicates that adolescents view shoplifting as significantly more ethical than do either college students or older adults. However, no significant differences were found between college students (mage=21.7) and older adults (mage=40.1). This is consistent with Cox et al. (1990) who report that the percentage of students shoplifting peaks in tenth grade, then drops off rapidly. They did not, however, extend their study beyond high school aged consumers.

Why do perceptions of shoplifting change substantially between the ages of 16 and 21, but remain relatively constant thereafter? It may be that consumers rapidly become socialized during this period, experience moral development, and/or change their value structure. Each of these suggests that we become more "ethical" as we move from adolescence into adulthood. Alternatively, it is possible that our basic ethical makeup remains constant, but ACB is shifted from shoplifting to other arenas. For example, shoplifting and vandalism are largely the domain of adolescents, while insurance and credit card fraud is dominated by adults (Hirschi and Gottfredson 1983). Future research examining both multiple age groups and forms of ACB simultaneously may provide answers to this question.

While establishing a relationship between age and perceptions of shoplifting, our results also suggest that age may not be as strong a causal factor as the percentages may imply. Rather, there are numerous individual difference characteristics-also influenced by age-that are responsible for consumers' ethical perceptions of shoplifting. Risk aversion, self-esteem, and susceptibility to interpersonal influence were all shown to influence perceived ethicality to varying degrees. Analyzing these effects more specifically, by dependent variable, reveals some interesting results that relate closely to earlier research.

Our data suggests that both respondent self-esteem and age relate negatively to perceived moral equity. Thus, consumers with lower self-esteem see shoplifting as a more "just" or "fair" behavior than do those with higher self-esteem. One reason that adolescents are more prone to shoplifting may be the lower self-esteem that often accompanies this time in one's life (see Table 1). Across all age groups though, stealing luxury items may be a way of temporarily enhancing self-esteem. However, this may set up a vicious circle-much by like that associated with compulsive consumption-where the behavior leading to short-term improvements fails to improve self-esteem in the long term (O'Guinn and Faber 1989).

Both risk aversion and susceptibility to informational interpersonal influence also affect consumer perceptions of shoplifting. As consumers became more risk averse, they rated the shoplifting scenario lower on contractualism, implying increased perceptions of a violation of an implied or explicit rule or contract. As with self-esteem, adolescents exhibited less risk aversion than did college students or adults. Therefore, some of the effect formally attributed to age may more precisely be due to variance in consumer risk aversion.

The negative effect of informational influence is also interesting. It suggests that the more consumers rely on others' behavior for information, the less ethically they perceive shoplifting. The fact that informational, rather than normative, influence relates to contractualism suggests that Cox et al.'s (1993) speculation about the mechanism with which peer influence operates may be correct. They hypothesized that people consider other's shoplifting behavior as information in making judgements about the appropriateness of the behavior, rather than shoplifting in an effort to win approval. Conversely, if shoplifting was done to gain approval, a positive relationship might be expected between these constructs. Under these conditions, the more you intended to seek approval through mimicking others' behavior, the more likely you would be to shoplift. These relationships, however, are dependent on the accepted behaviors of ones peer group.

Relativism, how culturally or traditionally acceptable a behavior is perceived, is affected by both risk aversion and self-esteem significantly. The negative relationship between risk aversion and relativism indicates that as a consumer becomes more risk averse, he/she is more likely to perceive shoplifting as an unacceptable behavior. In contrast, risk takers perceive it as a socially or traditionally acceptable act. The negative self-esteemCrelativism relationship indicates that consumers reporting low self-esteem tend to view shoplifting as socially and/or culturally acceptable.


This study investigated an area of aberrant consumer behavior that has been largely ignored in the consumer research literature. By examining multiple age groups, we have additional evidence that adolescents perceive shoplifting differently than more mature consumers and are more likely to engage in this behavior. However, we also have evidence that the direct affect of age on shoplifting is mediated by individual difference variables. Thus, the factors influencing shoplifting are much more complicated than we sometimes recognize. Certainly we are only building the foundation of knowledge in regard our understanding of shoplifting. As consumer researchers, we must continue to investigate the aberrant as well as the typical elements of consumption behavior.




Babin, Barry J., Donald P. Robin, and Kristi Pike (1994), "To Steal, or not to Steal? Consumer Ethics and Shoplifting," in Marketing Theory and Application, Vol. 5, eds. C. W. Park and D. C. Smith, American Marketing Association: Chicago, IL, 200-205.

Bagozzi, Richard P. and Todd F. Heatherton (1994), "A General Approach to Representing Multifaceted Personality Constructs: Application to State Self-Esteem," Structural Equations Modeling, 1 (1), 35-67.

Baron, Reuben M. and David A. Kenny (1986), "The Moderator-Mediator Variable Distinction in Social Psychological Research: Conceptual, Strategic, and Statistical Considerations," Journal of Personality and Social Psychology, 51 (June), 1173-1182.

Baumer, Terry and Dennis Rosenbaum (1984), Combatting Retail Theft: Programs and Strategies, Stoneham, MA: Butterworth.

Bearden, William O., Richard G. Netemeyer, and Mary F. Mobley (1993), Handbook of Marketing Scales: Multi-Item Measures for Marketing and Consumer Behavior Research, Newbury Park, CA: SAGE Publications.

Bearden, William O., Richard G. Netemeyer, and Jesse E. Teel (1989), "Measurement of Consumer Susceptibility to Interpersonal Influence," Journal of Consumer Research, 473-481.

Brown, B. Bradford and Mary Jane Lohr (1987), "Peer-Group Affiliation and Adolescent Self-Esteem: An Integration of Ego-Identity and Symbolic-Interaction Theories," Journal of Personality and Social Psychology, 52 (January), 47-55.

Cameron, M. O. (1964), The Booster and the Snitch: Department Store Shoplifting, New York: Free Press.

Cole, Catherine A. (1989), "Research Note: Deterrence and Consumer Fraud," Journal of Retailing, 65 (Spring), 107-120.

Cox, Dena, Anthony D. Cox, and George P. Moschis (1990), "When Consumer Behavior Goes Bad: An Investigation of Adolescent Shoplifting," Journal of Consumer Research, 17 (September), 149-159.

Cox, Anthony D., Dena Cox, Ronald D. Anderson, and George P. Moschis (1993), "Social Influences on Adolescent ShopliftingCTheory, Evidence, and Implications for the Retail Industry," Journal of Retailing, 69 (Summer), 234-246.

Fullerton, Ronald A. and Girish Punj (1993), "Choosing to Misbehave: A Structural Model of Aberrant Consumer Behavior," in Advances in Consumer Research, Leigh McAlister and Michael L. Rothschild, eds., Vol 20, Provo, UT: Association for Consumer Research, 570-574.

Heatherton, Todd F. and Janet Polivy (1991), "Development and Validation of a Scale for Measuring State Self-Esteem," Journal of Personality and Social Psychology, 60 (June), 895-910.

Hirschi, Travis and Michael Gottfredson (1983), "Age and the Explanation of Crime," American Journal of Sociology, 89 No. 3, 552-584.

Klemke, Lloyd W. (1982), "Exploring Adolescent Shoplifting," Sociology and Social Research, 67 (1), 59-75.

Moschis, George P. (1987), Consumer Socialization: A Life Cycle Perspective, Lexington, MA: Lexington.

O'Guinn, Thomas C. and Ronald J. Faber (1989), "Compulsive Buying: A Phenomenological Approach," Journal of Consumer Research, 16 (September), 147-157.

Reidenbach, R. Eric, Donald P. Robin, and Lyndon Dawson (1991), "An Application and Extension of a Multidimensional Ethics Scale to Selected Marketing Practices," Journal of the Academy of Marketing Science, 19 (Spring), 83-92.

Robin, Donald P. and R. Eric Reidenbach (1987), "Some Initial Steps Toward Improving the Measurement of Ethical Evaluations of Marketing Activities," Journal of Marketing, 51 (Winter), 44-58.

Solomon, Michael R. (1994), Consumer Behavior, Allyn and Bacon: Needham Heights, MA.

Tsalikis, John and David J. Fritzsche (1989), "Business Ethics: A Literature Review with a Focus on Marketing Ethics," Journal of Business Ethics, 8 (September), 695-743.

Zuckerman, Marvin (1979), Sensation Seeking: Beyond the Optimum Stimulation Level of Arousal, Hillsdale, NJ: Lawrence Erlbaum.