Children's Susceptibility to Peer Group Purchase Influence: an Exploratory Investigation

Gwen Rae Bachmann, University of Minnesota
Deborah Roedder John, University of Minnesota
Akshay R. Rao, University of Minnesota
ABSTRACT - The purpose of this paper is to examine how children of different ages factor peer group influence into their purchase decisions. In contrast to the prevailing view that peer influence becomes greater as children progress through early childhood into adolescence, our findings indicate that peer influence operates in a more subtle fashion. As children grow older, they begin to recognize that peer influence is important in some product situations, such as those involving publicly consumed items, but not in others, such as those involving privately consumed items. These results are discussed in terms of their importance for understanding the development of peer group influence in children's purchase decisions and for designing consumer education programs for increasing the awareness of such influence in children of all ages.
[ to cite ]:
Gwen Rae Bachmann, Deborah Roedder John, and Akshay R. Rao (1993) ,"Children's Susceptibility to Peer Group Purchase Influence: an Exploratory Investigation", in NA - Advances in Consumer Research Volume 20, eds. Leigh McAlister and Michael L. Rothschild, Provo, UT : Association for Consumer Research, Pages: 463-468.

Advances in Consumer Research Volume 20, 1993      Pages 463-468


Gwen Rae Bachmann, University of Minnesota

Deborah Roedder John, University of Minnesota

Akshay R. Rao, University of Minnesota


The purpose of this paper is to examine how children of different ages factor peer group influence into their purchase decisions. In contrast to the prevailing view that peer influence becomes greater as children progress through early childhood into adolescence, our findings indicate that peer influence operates in a more subtle fashion. As children grow older, they begin to recognize that peer influence is important in some product situations, such as those involving publicly consumed items, but not in others, such as those involving privately consumed items. These results are discussed in terms of their importance for understanding the development of peer group influence in children's purchase decisions and for designing consumer education programs for increasing the awareness of such influence in children of all ages.


Questions surrounding the role of media, specifically television advertising, in influencing aspects of children's consumer behavior have been the focus of much academic research and public policy debate. Starting with the Federal Trade Commission's proposal in the late 1970's to ban television advertising to young children, academic researchers have focused much of their attention on the influence of advertising on children's product preferences (Goldberg 1990; Goldberg, Gorn, and Gibson 1978), purchase requests to parents (Galst & White 1976; Robertson and Rossiter 1976), and actual product choices (Goldberg and Gorn 1982; Roedder, Sternthal, and Calder 1983). This body of research clearly documents the impact of television advertising on young children.

Less clear, however, is the role other influence agents play in children's product preferences and purchase decisions. In particular, there has been a surprising lack of research on the topic of peer influence as it impacts children's consumer behavior and decision making. Anecdotal evidence would suggest that peer influence becomes very important as children enter their teenage years and that peer influence exerts more impact for some products (e.g., athletic shoes) than others (e.g., gloves). And, some evidence from studies of general consumer socialization patterns would support these age trends in susceptibility to peer influence (e.g., Churchill and Moschis 1979). Apart from these preliminary indications, however, little research evidence exists on how peer groups influence children of different ages.

Accordingly, the purpose of this paper is to examine the emergence of peer group influence in children's purchase decisions. While there are many different sources and types of influence, this study focuses on the normative component of peer influence. [Normative influences are based on an individual's need to identify with group norms, standards and values.] Specifically, we investigate the degree to which children of different ages factor peer group opinions into their product decisions, depending on the nature of the product class involved in the decision. Evidence to this effect is important not only for a more complete understanding of children's consumer behavior, but also as a basis for consumer education programs that seek to develop children's critical thinking skills about the purchase decisions they make.


While our focus is peer influence on children's decision making, an existing body of research, reference group influence, is used to frame our question and draw theoretical arguments. Reference group influence is discussed next.

Reference Group Influence Defined

A reference group can be defined as "...a group of people that significantly influences an individual's behavior" (Bearden and Etzel 1982, p. 184). In consumer behavior contexts, reference groups are typically comprised of significant others from the individuals' social network, including family members, co-workers, peers and friends as well as inspirational figures such as sports heroes, movie stars, and fictional characters. For children, family members and peers are undoubtedly the most dominant reference groups, followed by more distant figures such as sports heroes and movie stars (Shaffer 1988).

The influence that reference groups exert on the types of products and brands an individual purchases is diverse, with referents exercising three forms of influence on decision making: information, utilitarian, and value-expressive (Deutsch and Gerard 1955; Kelman 1961). Seeking out information from a referent, feeling that a purchase would enhance one's image with a reference group, and allowing one's liking of the reference group to influence one's decision to purchase a product would be examples of information, utilitarian, and value-expressive components of reference group influence, respectively. Empirically and conceptually, the utilitarian and value-expressive component are difficult to distinguish and are often combined, into a "normative" component of influence (Bearden, Netemeyer & Teel, 1989). It is this normative component of peer influence that we address.

Age Differences in Reference Group Influence

In order for reference groups, such as peer groups, to exert influence on children's product decisions, children must have developed certain social sensitivities and cognitive skills. First, a child must be able to take another person's perspective and realize that another person's preferences may be different from one's own preferences. Second, a child must understand and/or believe that people draw inferences about each other based on product choices and possessions. And, third, other people's opinions must be important to the child in forming his or her own self concept. Without one of these "building blocks," reference group influence of any kind may be weak if not altogether absent.

These three building blocks can provide a basis for predicting and explaining when peer influence might emerge and how it might develop as children mature. Several frameworks from child psychology that discuss the emergence of these key aspects of social-cognitive development may be applicable when examining peer influence at different ages. These frameworks are diagrammed in Table 1.



Selman (1980) suggests that, in order to role play a child must be able to take on another person's perspective, understand their thoughts and feelings, and anticipate the other person's reactions versus their own. In the first stage, ego-centric role-taking, young children are unaware of any perspective other than their own. By the second stage, social-informational role-taking, a child may realize that others may have different perspectives but has difficulty anticipating what they might be. In the third stage, the self-reflective stage, a child can anticipate and consider another's person's viewpoint but is unable to simultaneously consider his/her own perspective and that of another. The final stage, social and conventional system role-taking, involves not only mutual role taking but a consideration of the situation or context.

A second framework of impression formation, Barenboim (1981), is a three-step developmental sequence describing the changes in children's impressions of others during the grade school years. In the first stage, the behavioral comparisons phase, children's impressions involve comparisons in concrete behavioral terms (e.g., "He's the best climber"). In the second stage, the psychological constructs stage, children begin to base their impressions on regularities in other's behavior and describe others in terms of abstract psychological attributes (e.g., "She's stubborn"). In the final stage, the psychological comparisons phase, children begin to compare others on important psychological dimensions (e.g., "Sara is more generous than Mary").

Erikson's (1972) theory of psychosocial development is based on the belief that humans are adaptive beings who go through eight social-conflict stages in a lifetime. The eight stages or crises are based on the development of trust, autonomy, initiative, industry, identity, intimacy, generativity (productivity in work and family responsibilities), and ego integrity. The key feature of these stages for our purpose is the fact that the primary social agents involved in resolving each stage's social conflicts differ by stage. In the first three stages, parents and family are the primary social agents. For children in the fourth stage, teachers and peers are the primary social agents. In the fifth stage, peers once again are the primary social agents. Spouses and lovers are added to the list in the sixth stage. In the seventh stage, primary social agents are spouses, children, and cultural norms. In the final stage, social agents have not been specified.

Since these three frameworks address different social-cognitive skills, the number of stages and the specific ages at which they occur varies between frameworks. However, these frameworks are remarkably similar in identifying major breaks in children's development across domains: ages 5 and under, ages 6 to 8, ages 9 to 11, ages 11/12 and older. Because each of these frameworks was selected to provide evidence of the prerequisites necessary for peer influence, such as perspective (role) taking and social impression formation, these major age breaks appear to indicate relevant differences between age groups in their use of peer influence.

Specifically, based upon the developmental theories just reviewed, children 5 years of age and under would seem to be the least susceptible to referent group influence. Because children of this age have greater difficulty in acknowledging and taking another person's perspective, they most likely view other's preferences to be similar to their own, leaving little room for external influences. Children between the ages of 6 and 8 are more likely to be influenced by their peers. Yet, their impression formation skills are such that they are not prone to the type of sophisticated inferences involved in recognizing that other people not only have their own preferences but also that these preferences are used as a basis of comparison in judging other people. Children 9 to 11 years of age would appear to be even more susceptible to reference group influence than their younger counterparts by virtue of the fact that they can anticipate others' reactions to their opinions and behavior, can consider their own preferences in conjunction with others' opinions, and have rather well-developed person impression skills. In short, they are beginning to understand external influences and are open to influences from their peers. Finally, children 12 years of age and older are probably the ultimate purveyors and receivers of reference group influence. Not only do they now recognize the intricacies of social interactions with others, but are also painfully aware of the tendency for psychological impressions to be formed on the basis of consumption preferences and choices.

This characterization of age differences is quite consistent with the prevailing notion that reference group influence, particularly peer group influence, increases as children progress through childhood to adolescence. But, the increasing sophistication in impression formation skills also suggests that reference group influence probably varies by the product context and situation as children become more socially adept. An important factor in this regard, suggested by studies of reference group influence with adult consumers, is the type of product involved in the purchase decision. It is to this topic that we now turn.

Product Class Differences in Reference Group Influence

Recent investigations into reference group influence in consumer decision making have identified the need to consider the conspicuousness of the product or brand of interest. Product conspicuousness is a function of two dimensions (see Bearden and Etzel 1982). First, the extent to which the product is not owned by everybody makes it exclusive, and thus, conspicuous. Thus, luxuries, which are more exclusive than necessities, tend to be relatively more conspicuous. The second dimension refers to the degree to which product usage is performed in public versus in private. Publicly consumed items are, of course, more conspicuous than privately consumed products. Four types of products emerge from these two dimensions: publicly consumed luxuries, publicly consumed necessities, privately consumed luxuries, and privately consumed necessities.

Extant research indicates that the degree of reference group influence varies by product conspicuousness. For peer group influence, both Bearden and Etzel (1982) and Childers and Rao (1992) find that public luxuries and private necessities make up the ends of the conspicuousness continuum, with public luxuries being subject to significantly more influence than private necessities. In addition, there is a tendency for public products of all types, regardless of whether they are luxuries or necessities, to be subject to more reference group influence than private products of all types.

In light of these findings, peer influence in children's decision making should also be linked to product conspicuousness, at least for some age groups. Recall though, that this study is only addressing the normative component of peer influence, not all types of reference group influence. Specifically, it would seem that peer influence should vary by product type for children with the requisite social-cognitive skills to understand the consequences of ignoring others' opinions for products that are conspicuously displayed and used versus those that are not. It is, in fact, a fairly sophisticated view of social interaction that would need to be held in order to understand that other people's opinions matter only for those products that are in public view or are exclusive.

Following this line of reasoning, the ability to make distinctions between product types in terms of how much peer influence should be considered should develop with age. Older children, with a more sophisticated view that allows them to view peer influence in different situational contexts, will be more likely to factor peer opinions into some types of product decisions (e.g., those involving publicly consumed items) but not others (e.g., those involving privately consumed items). Younger children will be more likely to either dismiss peer influence altogether, or simply believe that their friends' opinions are somewhat important no matter what type of product is being considered. Therefore, consistent with the age groups identified earlier, and their corresponding levels of social-cognitive skills, the following hypotheses are advanced:

H1: Differences in peer influence as a function of product conspicuous will be larger for older than younger children.

H1a: Product conspicuousness will make no difference in the degree of peer influence for children in the early elementary school years (ages 6-8).

H1b: Product conspicuousness will make a small to moderate difference in the degree of peer influence for children in the middle elementary school years (ages 9-11). If a difference emerges, it will be a distinction between the most extreme ends of the product conspicuous continuum, public luxuries versus private necessities.

H1c: Product conspicuousness will make a moderate to large difference in the degree of peer influence for children in the later elementary school years (ages 12-14). Distinctions are likely to be made not only at the extreme ends of the product conspicuous continuum, but also for public versus private products and for luxuries versus necessities.



One hundred forty one children from an elementary school in a major metropolitan area were recruited for the study. In order to obtain subjects in the appropriate age groups, consent forms were sent home to children in the first, second, fourth, fifth, seventh, and eighth grades. Of those returning consent forms and participating in the study, 54 children were 6-8 years of age, 49 children were 9-11 years of age, and 38 children were 12-14 years of age. In return for their participation, children received a small prize and the school received a small donation for their cooperation.


The study employed a 3 (Age) x 4 (Product Conspicuousness) x 3 (Version) factorial design. The age of the child (AGE) was a between subject factor, whereas product conspicuousness (CONSPIC) was varied on a within-subject basis to represent the four product types of interest (public luxuries, public necessities, private luxuries, private necessities). In order to provide multiple operationalizations of the four product types, three versions of the questionnaire (VERSION) were compiled, with each version containing a different operationalization of each product type. Different versions of the questionnaire were administered to different groups of subjects; therefore version was a third (between subject) factor. The statistical model employed to test the various hypotheses considered the appropriate mean squared error terms in light of the one repeated factor.

Independent Variables

Based on the developmental research reviewed earlier, and the hypotheses of interest, three age groups were included in the study: 6-8 year olds, 9-11 year olds, and 12-14 years olds. Children under 6 were excluded from the study due to problems that surfaced in a pilot study with regard to understanding the instructions and measurement scales. Children over 14 were excluded from the study due to potential problems in identifying products representative of each product category for a large age range.

For the product conspicuousness factor, pretests were conducted to identify a set of products representing the four product types: public luxury, public necessity, private luxury, and private necessity. A large list of possible products that were gender-neutral and familiar to children of all ages was generated. Children were presented with the list of products and asked to classify each product as either public or private and as a necessity or a luxury. Before responding to the list of products, children were provided with a concrete definition of each term and were given two "practice" products as a training task. To record their answers, older subjects simply circled the appropriate term (e.g., "public" or "private"), whereas younger children circled a picture that portrayed the concept of "public" (many children watching the owner), "private" (owner by himself/herself), "luxury" (very few owners), and "necessity" (many owners). [Interested readers may contact the authors for complete pretest instructions and pictures used to represent conspicuousness dimensions.] The results were analyzed by age group to identify a set of 12 products, three for each product type, that were classified correctly by the large majority of children (at least 60%) in each of the three age groups. Most of the products, in fact, were classified correctly by at least 75% of the children across age groups.

These twelve products were then used to create three versions of the questionnaire. Each version included a public luxury, a public necessity, a private luxury, and a private necessity, resulting in the following sets: (1) snow skis, winter coat, TV for own room, pajamas; (2) ice skates, pants, home computer, hair brush; (3) 10-speed bike, shoes, stereo for own room, toothbrush.

Dependent Variable

The measure of peer group influence was developed from similar scales used in previous studies of reference group influence in adults (Bearden and Etzel 1982; Bearden, Netemeyer and Teel 1989; Childers and Rao 1992; Park and Lessig 1977). In modifying these scales for use with children, the number of items was kept at a reasonable number to avoid fatigue, and the scale points were kept at a minimum to make it easier for younger children to respond. The final scale, adapted from the one developed by Bearden et al. (1989), included seven items that children indicated their agreement or disagreement with on a four-point YES-No scale ("YES" "yes" "no" "NO"). Children's responses to the items were summed, based on results indicating that the scale was unidimensional with a more than acceptable degree of inter-item reliability (alpha = .94). [Interested readers may contact the authors for complete influence scale.]


Questionnaires were administered to children in their classrooms during the school day. After receiving general instructions about the study, and specific instructions about the YES-NO scale, subjects were given a training task to ensure that they understood the rating scale. For this purpose, children were asked to respond to a series of test items, such as "I like cookies" and "I like icky medicine", using the YES-NO scale. After the training phase, subjects were given the major task, involving their responses to the YES-NO purchase influence scale for each of four products. The order in which products appeared was randomized to minimize possible order effects. In addition, a distractor task was presented in the middle of the task, to minimize possible fatigue or boredom with the rating task. As a manipulation check, subjects completed some final questions regarding the conspicuousness category of each product. Children were then thanked, given a prize, and dismissed.


Children's responses to the peer purchase influence scale were first analyzed using a 3 (age) x 4 (product conspicuousness) x 3 (version) repeated measures MANOVA. The results indicate general support for our hypothesis that the degree of peer group influence would vary as a function of age and product conspicuousness, with a significant AGE x CONSPIC interaction emerging from the analysis (F(6,396)=8.66, p<.01). Also of interest, though not predicted, a significant main effect for conspicuousness was observed (F(3,396)=35.04, p<.001). Of less direct interest, there were two significant effects involving the VERSION factor, specifically the main effect for VERSION (F(2,132)=5.86, p<.005) and the VERSION x CONSPIC interaction (F(6,396)=5.92, p<.001). These results indicate, as might be expected, that different operationalizations of the four product types were not always equivalent in terms of how much peer influence was exerted, thus confirming the importance of our inclusion of multiple operationalizations for the conspicuousness factor. The use of multiple operationalizations and the finding of significant version effects are common in this literature (Bearden & Etzel, 1982; Childers & Rao, 1992). Note, however, the important finding that the VERSION x AGE interaction was not significant, indicating that the pattern of results for each of the products was consistent across age groups. Consequently, the remainder of our analysis examined data collapsed across the three versions. Means and standard deviations for the experimental conditions are presented in Table 2.

Further analyses were conducted to examine the predictions regarding differences between product types within each age group. Specifically, contrasts between means for the four product types were conducted within each age group using a Newman-Keuls' procedure. The appropriate error term based on the repeated measures analysis was used. The results, shown in Table 3, indicate general support for our hypotheses by age group. For the youngest age group, none of the contrasts between product types was significant, as predicted. The data suggest that children in this age group view the degree of peer group influence as rather constant across product types. It appears that these children have not yet developed an understanding of the social significance of using products in different contexts. Distinctions with regard to peer influence are not made between publicly consumed and privately consumed goods, nor between luxuries and necessities. Even differences at the ends of the conspicuousness continuum, public luxuries versus private necessities, are not recognized.



Consistent with our predictions, children in the middle age group reacted differently to peer group influence depending on product conspicuousness, distinguishing between products at the extremes of the conspicuousness continuum (public luxuries versus private necessities). But, interestingly, these children also showed an understanding of the social implications of the public versus private dichotomy. An examination of the results indicates that only the (1) public luxury versus public necessity, and (2) private luxury versus private necessity contrasts were not significant. It would appear that the significance of the luxury versus necessity distinction emerges later in children, probably intertwined with an understanding of economic concepts such as relative price (a principal determinant of luxury prices) and relative scarcity.

Finally, as predicted, children in the oldest age group demonstrated the most sophisticated sensitivity to peer group influence. These children exhibited differences in peer group influence for almost every product type tested, with the only nonsignificant contrast being the one between public luxuries and public necessities. Further, when the influence scores for both luxury items were combined and contrasted with the influence scores for both necessity items, a paired T-test of the luxury/necessity comparison was significant for this age group, whereas it was not significant for the two younger age groups. The remaining contrasts, all significant, indicate an ability to adapt to different social contexts, with a full consideration of the public versus private dichotomy and a partial understanding of the luxury versus necessity distinction. In point of fact, it is quite likely that these older children may have exhibited a much more definitive appreciation of the luxury-necessity dichotomy even for publicly consumed items had the price differences between the luxury and necessity products in the public category (e.g., ice skates and pants) been as great as the ones evident in the private category (e.g., computer versus hairbrush).


Our findings indicate that peer group purchase influence emerges slowly as children progress through their elementary school years. In contrast to prevailing wisdom, however, peer group influence does not accelerate with increasing age for a wide range of products. With advancing age, children become more susceptible to peer group influence only for those products that are more conspicuous in nature, such as public luxuries. Older children are more susceptible to influence for some products, such as public luxuries, but are less susceptible to influence for other products, such as private necessities. Distinctions such as these are no doubt fueled by the increasingly sophisticated view of social interaction and impression formation possessed by older children.

These findings suggest that peer group influence, and probably other types of reference group influence as well, is a far more complex phenomenon than previous anecdotal evidence would suggest. As children progress from early childhood to adolescence, what develops is an awareness that peer groups are an important influence in some types of product decisions, but are irrelevant in others. As this "contingency view" develops, children will no doubt begin to incorporate opinions from their peers for some types of decisions but incorporate opinions from different referents for other types of decisions.

Due to the concern over television advertising as a major influence on children, most of the current consumer education efforts have focused on helping children understand the persuasive intent of ads and on providing skills to detect possible puffery or deception. Considering the findings presented here, it would appear to be important to consider peer group influence in these consumer education efforts. Increasing children's awareness of the role that peer influence plays in their purchase decisions, and discussing the positive and negative aspects of this type of influence, would seem to be important in improving the critical thinking skills of young and developing consumers.



Additional research is warranted to provide a more complete view of peer group influence than the preliminary one offered here. Several avenues for extension should be considered. First, future research should examine the impact of peer group influence on children's brand-level decisions in addition to the product-level decisions explored in our study. It is quite possible that peer influence may be more pervasive, but may develop later, for brand decisions than for product decisions (see Bearden and Etzel 1982). In pursuing this line of research, investigators will need to be particularly cautious, however, in selecting brands that are familiar to children across a wide range of age groups.

A second direction for future research would involve a more detailed examination of the different components of peer or referent groups influence identified in research with adults. The information component and the normative component may yield different effects (Bearden and Etzel 1982; Bearden et al. 1989). Finally, different methodologies should be pursued in future research to provide convergent evidence of the age differences observed here. Validation of our findings in an experimental setting, in which peer group opinions are manipulated and purchase influence directly measured, would be beneficial in increasing our confidence that the self-report methodology used in this study did indeed capture the nature of the peer influence phenomenon for children of all ages.

An experimental methodology might also be useful in assessing the usefulness of consumer education efforts aimed at developing critical thinking skills about peer group influence. Considering the interest among parents and educators in promoting decision making skills in children, research in this area would serve to further the goal of making children more informed and more knowledgeable consumers. With children exerting more direct and indirect influence in a wide range of individual and household purchase decisions, such research could not be more timely nor more important.


Barenboim, Carl (1981), "The Development of Person Perception in Childhood and Adolescence: From Behavioral Comparisons to Psychological Constructs to Psychological Comparisons," Child Development, 52, 129-244.

Bearden, William O. and Michael J. Etzel (1982), "Reference Group Influence on Product and Brand Purchase Decisions," Journal of Consumer Research, 9 (September), 183-94.

Bearden, William O., Richard G. Netemeyer, and Jesse E. Teel (1989), "Measurement of Consumer Susceptibility to Interpersonal Influence," Journal of Consumer Research, 15 (March), 473-481.

Childers, Terry L. and Akshay R. Rao (1992), "The Influence of Familial and Peer-Based Reference Groups on Consumer Decisions," Journal of Consumer Research, (September), 198-211.

Churchill, Gilbert A., Jr. and George P. Moschis (1979), "Television and Interpersonal Influences on Adolescent Consumer Learning," Journal of Consumer Research, 6 (June), 23-35.

Deutsch, M. and Harold B. Gerard (1955), "A Study of Normative and Informational Social Influences Upon Individual Judgment," Journal of Abnormal and Social Psychology, 51, 624-636.

Erickson, Erik H. (1972), "Eight Stages of Man," in C.S. Lavatelli and F. Stendler (eds.), Readings in Child Behavior and Child Development, San Diego, CA: Harcourt Brace Jovanovich.

Galst, Joann Paley and Mary Alice White (1976), "The Unhealthy Persuader: The Reinforcing Value of Television and Children's Purchase-influencing Attempts at the Supermarket," Child Development, 17, 1089-1096.

Goldberg, Marvin E. (1990), "A Quasi-Experiment Assessing the Effectiveness of TV Advertising to Children," Journal of Marketing Research, 27 (November), 445-54.

Goldberg, Marvin E. and Gerald J. Gorn (1982), "Behavioral Evidence of the Effects of Televised Food Messages on Children," Journal of Consumer Research, 9 (September), 200-205.

Goldberg, Marvin E., Gerald J. Gorn and Wendy Gibson (1978), "TV Messages for Snack and Breakfast Foods: Do They Influence Children's Preferences?" Journal of Consumer Research, 5 (September), 73-81.

Kelman, Herbert C. (1961), "Processes of Opinion Change," Public Opinion Quarterly, 25, 57-78.

Robertson, Thomas S. and John R. Rossiter (1976), "Short-Run Advertising Effects on Children: A Field Study," Journal of Marketing Research, 13 (February), 68-70.

Roedder, Deborah L., Brian Sternthal, and Bobby J. Calder (1983), "Attitude-Behavior Consistency in Children's Responses to Television Advertising," Journal of Marketing Research, 20 (November), 337-49.

Selman, Robert L. (1980), The Growth of Interpersonal Understanding, Orlando, FL: Academic Press.

Shaffer, David R. (1988), Social and Personality Development, Pacific Grove, CA: Brooks/Cole Publishing Company.