Consumers' need for uniqueness: Scale development and validation;
Tian, Kelly Tepper; Bearden, William O; Hunter, Gary L
Journal of Consumer Research   06-01-2001

Consumers' need for uniqueness: Scale development and validation

Consumers acquire and display material possessions for the purpose of feeling differentiated from other people and, thus, are targeted with a variety of marketing stimuli that attempt to enhance self-perceptions of uniqueness. Because the pursuit of differentness (or counterconformity motivation) varies across individuals to influence consumer responses, we develop and validate a trait measure of consumers' need for uniqueness. Consumers' need for uniqueness is defined as an individual's pursuit of differentness relative to others that is achieved through the acquisition, utilization, and disposition of consumer goods for the purpose of developing and enhancing one's personal and social identity. Following assessments of the scale's latent structure, a series of validation studies examines the scale's validity. The presentation of empirical work is followed by a discussion of how consumers' need for uniqueness could be used in better understanding consumer behavior and the role consumption plays in people's expression of identity.

Being different from others or becoming distinctive among a larger group often results from signals conveyed by the material objects that consumers choose to display. Conceptual models of social nonconformity recognize that behaviors that render a person different relative to other people may reflect several motivational processes (Nail 1986; Tepper 1997). The display of differentiating material objects can be incidental or secondary outcomes from attempts to satisfy various motivations or drives. For example, a person driven by independence motivation adheres to internal tastes when making consumer choices. Although such a choice may at times differentiate the decision maker from others, this outcome is incidental to acting consistent with personal standards (Nail 1986).

In contrast, the display of differentiating consumer goods can be the primary, intended outcome of a person's actions that are driven by the need to feel different from other people. This need, which is labeled "counterconformity motivation" (Nail 1986), arises when individuals feel a threat to their identity, as occurs when they perceive that they are highly similar to others (Snyder and Fromkin 1977). Unlike an individual driven by independence motivation, the counterconforming individual is influenced by the norm behavior of others but behaves so as to be in noncongruence with the norm (Nail 1986). In such instances, an individual may derive satisfaction from differentiating consumer possessions because these alleviate the threat to identity (Snyder and Fromkin 1977). Products and their uses or displays that become classified as being outside of the norm may serve as recognizable symbols of uniqueness or specialness. The self-concept of an individual seeking to be different from others will be "sustained and buoyed if he believes the good he has purchased is recognized publicly and classified in a manner that matches and supports his self-concept" (Grubb and Grathwohl 1967, p. 25).

Relative to other sources of differentiation, differentness that results from counterconformity motivation has broader importance for understanding consumer behavior, particularly reactions to the commercialization and popularization of product offerings. To appeal to individuals' desire to be different from others, marketers develop advertising messages that employ product-scarcity appeals, uniqueness appeals, and appeals to breaking the rules of one's reference group (Frank 1997; Lynn and Harris 1997; Snyder 1992; Thompson and Haytko 1997). Snyder (1992) suggests that marketers exploit individuals' counterconformity motivation by stimulating a consumer catch-22: marketers advertise that a product, brand, or style enhances one's uniqueness; consumers purchase the advertised product in order to express their specialness; the marketer's success from these purchases stimulates more advertising; and many consumers respond similarly to the advertising appeal such that each consumer's expectation of specialness is not achieved. However, people are able to resist this force toward conformity through "a million ineluctable, unfinalizable, individualistic devices" (Frank 1997, p. 17).

Illustrative of consumers' efforts to resist or counter the acceptance of popularized goods that symbolically convey conformity, consumers may dispose of goods that become popular and repeat the cycle described above in search of new and special products, innovations, and emerging fashion trends (Snyder 1992; Tepper 1997). In phenomenological interviews, Thompson and Haytko (1997, p. 22) found that attempts "to stay ahead in the realm of fashion trends" by discarding fashions that catch on and seeking emerging innovations are interpreted as acts of resisting conformity. Consumers also resist conformity with respect to their consumer-product displays by purchasing novelty goods, handcrafted goods, and personalized items. These choices extend the time that products maintain their uniqueness, as does purchasing vintage goods or antique goods that are not available in mass but that may be purchased from nontraditional outlets (e.g., antique stores, garage sales, second-hand stores, swap meets; Tepper 1997). Included among these nontraditional outlets are internet Web sites through which consumers may search and bid in an international marketplace for goods that are customized, rare, or no longer being manufactured. Augmenting these options, mass customization has arisen from marketers' use of computer-facilitated flexible manufacturing such that consumers may create and customize product designs to their own personal specifications. Consumers have also been able to extend the uniqueness benefits of more commonplace products by creatively altering or using them (Tepper 1997), assembling them into collections that taken as a whole are unique (see Belk, Wallendorf, and Sherry 1989), and acquiring and displaying in-depth knowledge of these products (Holt 1998). Finally, consumers may attempt to extend the uniqueness of a consumer display by selecting one that few are willing to copy. In illustration, one 18-year-old, who was asked to write a story about an incident in which she felt different from other consumers, stated that after all her friends copied her by double-piercing one of their ears, she pierced the same ear five more times to make sure that they did not imitate her again (Tepper 1997). Thus, consumers driven by counterconformity motivation may engage in a variety of uniqueness-seeking behaviors in response to situations that heighten perceptions of similarity to others.

While counterconformity motivation may be prompted by situations in which individuals perceive that they are highly similar to others (e.g., possessing a formerly little-known product or displaying a rare style that subsequently becomes popular), some individuals experience greater threats to their identity than do others as a result of the same situation. These individuals are characterized by an enduring tendency to seek differentness relative to others. This tendency is exemplified in the comments of one of Thompson and Haytko's (1997, p. 21) interview participants: "Usually if something is hot, I'll go out of my way to stay away from it. Even if I like it at first, if everyone's wearing it, I don't want to be wearing it." Conceptual marketing models depict counterconformity motivation or the pursuit of differentness relative to other people as a trait or personality characteristic that determines important consumer phenomena, such as consumers' responses to innovative exterior designs of a product (Bloch 1995), the fashion decision process, style selection, style replacement (Miller, McIntyre, and Mantrala 1993), and variety-seeking behavior (McAlister and Pessemier 1982).

Empirical tests of propositions derived from these models would be enhanced by a trait measure that captures consumers' counterconformity motivation with respect to possession acquisition and display, or consumers' need for uniqueness. In particular, such a measure would enhance the study of individual differences in consumer behavior and allow the study of how the enactment of consumers' need for uniqueness is prompted by different situations to influence responses to design elements of products and advertising messages. Such a measure would further allow examination of how consumers' need for uniqueness influences product acquisition behaviors that are tied to attaining differentness but that avoid the commercialized counterculture cycles that are manufactured by marketers. Notably, consumer counterconformity is not simply the antithesis of trait measures tied to conformity that have been employed in prior consumer research (e.g., attention to social comparison information and consumers' susceptibility to interpersonal influence; see Bearden, Netemeyer, and Teel 1989). Reverse scoring the latter measures would reflect not conforming, but these measures would not capture the underlying motivation; therefore, using them would confound counterconformity motivation with independence motivation. Further, the need for a measure of consumers' need for uniqueness is supported by the largely disappointing and inconsistent results from prior consumer investigations that have employed general trait measures of need for uniqueness (Snyder and Fromkin 1977) and individuation Maslach, Stapp, and Santee (1985) borrowed from psychology (see the review by Tepper [1996]). The limited empirical support for the influence of these measured constructs on consumer behavior may be partially attributed to their content that also confounds motivations for nonconformity (Tepper 1996) and to the lack of empirical support for their respective hypothesized latent structures (Tepper 1998; Tepper and Hoyle 1996). Accordingly, the current research first defines consumers' need for uniqueness and the dimensions that compose the concept. A series of studies is then reported that develops a measure of consumers' need for uniqueness and assesses the new measure's latent structure, reliability, and validity.

CONSUMERS' NEED FOR UNIQUENESS

Theoretical Origins

The concept of consumers' need for uniqueness derives from Snyder and Fromkin's (1977) theory of uniqueness. According to this theory, the need to see oneself as being different from other persons is aroused and competes with other motives in situations that threaten the self-perception of uniqueness (i.e., situations in which individuals see themselves as highly similar to others in their social environment). Individuals attempt to reclaim their self-esteem and reduce negative affect through self-distinguishing behaviors. These expressions of uniqueness are sought in different forms and outlets where the social penalties for being different are not severe. Material expressions of one's differentness from others are particularly valued because they satisfy the need for uniqueness without risking severe social penalties (Snyder 1992). Snyder and Fromkin (1977) recognize that different individuals evidence varying degrees of uniqueness motivation. Because individuals may fulfill their desire to be unique in a variety of ways (e.g., through possession displays [see Belk 1988], style of interpersonal interaction [see Maslach, Stapp, and Santee 1985], or the domains of knowledge in which they establish expertise [see Holt 1995]), they are likely to vary in their tendency to satisfy their uniqueness motivation through consumer behaviors and possessions.

Following from uniqueness theory, consumers' need for uniqueness should reflect individual differences in consumer counterconformity motivation-a motivation for differentiating the self via consumer goods and the visual display of these goods that involves the volitional or willful pursuit of differentness relative to others as an end goal. As such, consumers' need for uniqueness is more specific in nature than willingness to be individuated (i.e., willingness to stand out as different among others), which may serve various motivations (Maslach et al. 1985). Consumers' need for uniqueness is also distinct from independence, a motivation that may inadvertently manifest in social differentness as a result of adhering to one's personal taste (Nail 1986). Further, as elaborated in Grubb and Grathwohl's (1967) work on how consumer goods serve individuals in forming and maintaining desired self-concepts, consumers' need for uniqueness should reflect both self-image and social image enhancement processes. A unique product may be sought out to restore a person's self-view as one who is different from others, such as when an anonymous art collector bids via the internet or telephone for a rare painting she wants to display in her bedroom. Self-image enhancement, which occurs via the transference of symbolic meaning from a purchased product to the self, results from an internal, personal process. However, the effect on the individual is ultimately dependent on the consumer good being a publicly recognized symbol. Because of its recognized meaning, a unique product can be used to gain desired evaluations from others (i.e., a social image as one who is different) that further enhances self-image.

Conceptual Definition

Accordingly, consumers' need for uniqueness is defined as the trait of pursuing differentness relative to others through the acquisition, utilization, and disposition of consumer goods for the purpose of developing and enhancing one's self-image and social image. Consumer goods used for satisfying counterconformity motivations refer to product categories, brands, and versions or styles. Based on needfor-uniqueness theory, nonconformity research, and the consumer behavior literature, consumers' need for uniqueness is conceptualized as subsuming three behavioral manifestations or dimensions.

Creative Choice Counterconformity. In Western culture, expressing one's differentness from others, individuality, or unique identity requires creating a personal style via material goods that represent the self (Kron 1983). Reflecting one's personal style in material displays is accomplished through the purchase of original, novel, or unique consumer goods (Kron 1983) or via the decorative collection, arrangement, and display of goods (Belk et al. 1989; Kron 1983). This goal-directed consumer behavior reflects creative choice counterconformity. Creative choice counterconformity reflects that the consumer seeks social differentness from most others but that this consumer makes selections that are likely to be considered good choices by these others. The notion that these creative consumer activities are undertaken to establish one's uniqueness parallels suggestions in the consumer literature and the popular press. McAlister and Pessemier (1982) suggest that a desire for social distinction via unusual products influences new product adoption and variety-seeking behavior. Belk (1988) suggests that a traumatic lessening of one's uniqueness and sense of self accompanies the loss of one's accumulated personal possessions. Consumer magazines commonly feature instructional articles on "how to make your home say `you"' and "what you can do to develop your personal style" (Kron 1983, p. 67). While creative consumer choices involve some risk (Kron 1983), these acts also potentially elicit positive social evaluations of the consumer as being one who is unique (Snyder and Fromkin 1977).

Unpopular Choice Counterconformity. Unpopular choice counterconformity refers to the selection or use of products and brands that deviate from group norms and thus risk social disapproval that consumers withstand in order to establish their differentness from others. Ziller (1964) suggests that if individuals fail to see a means of differentiating themselves from others in a socially appropriate manner, they may prefer acts that negatively distinguish them over more subtle distinctions that are available within the domain of positively valued acts. Breaking rules or customs or challenging existing consumer norms risks social disapproval, including evaluations that one exhibits poor taste. However, similar to the dimension of creative choice counterconformity, unpopular counterconformity may also result in an enhanced self-image and social image. People who break rules and risk social disapproval in the service of asserting their differentness often affirm good character and, thus, enhance their self-image (Gross 1977). In addition, initially unpopular consumer choices may later gain social acceptance and thereby positively distinguish the consumer as an innovator or fashion leader (Heckert 1989).

Avoidance of Similarity. The third manifestation of consumers' need for uniqueness, avoiding similarity, refers to the loss of interest in, or discontinued use of, possessions that become commonplace in order to move away from the norm and reestablish one's differentness. Because those individuals who possess a high need for consumer uniqueness should monitor others' ownership of goods in product categories where replacement is expected, avoiding similarity also refers to devaluing and avoiding the purchase of products or brands that are perceived to be commonplace. Disposition and discontinued product use or purchase to avoid similarity to others occurs because consumers' success in creating distinctive self images and social images is often short lived. Because consumer choices, particularly creative choices, may establish one's uniqueness, such choices are likely to attract followers who also seek to develop their specialness or share a common link with early adopter groups (Fisher and Price 1992). And, as previously noted, even initially unpopular choices can gain widespread acceptance over time (Heckert 1989).

Consumers who were asked to write narratives about being different as a consumer reported experiencing the diminution of their uniqueness via the popularization of both creative choices and unpopular choices, to which they responded by disposing of the formerly valued possession (Tepper 1997). As an illustration of the creative choice, one man reported that he purchased a jacket and bought patches from a supply store in order to create his "own unique leather fighter pilot jacket." However, before he completed assembling his jacket, the movie Top Gun popularized fighter pilot jackets. He reports that "after seeing everyone in my high school wearing one, it no longer had the same appeal . . . I never did complete the jacket" (Tepper 1997, p. 233). In illustration of the unpopular choice, a high school student, who arrived at school to find the "queen of trends" wearing the same shoes as she had, reports, "I groaned inside when I realized that my beautiful brown suede shoes-the ones that were so big I looked almost like a clown-were in style! I had prided myself on these shoes because they were so distinctive and added to my image as `that weird smart girl' . . . my nifty suede shoes were no longer special, and I didn't feel special either. I put the shoes in a dark corner of my closet and haven't worn them since" (Tepper 1997, p. 233). Hence, pursuing social differentness requires a willingness to change past consumer behaviors and preferences (via avoidance, disposition, or devaluation). It should be noted that changing from an initially preferred choice to a new one in order to avoid similarity is a criterion for distinguishing counterconformity from other motivations that incidentally result in being different (Nail 1986).

SCALE DEVELOPMENT

Item Development

An initial pool of 93 items was generated to reflect the three facets of consumers' need for uniqueness. Item generation relied on gleaning published, popular, and theoretical conceptions of the consumer behavior of individuals desiring to be different (e.g., Snyder and Fromkin 1977), examining qualitative data gathered in an exploratory investigation, and converting frequently mentioned descriptions of unique consumers into items. The content validity of the items was assessed in two stages (Bearden et al. 1989). First, five judges were given the definition of each dimension, a related explanation, and an example item. The judges were then asked to allocate the statements to one of the three dimensions or to a "not applicable" category. After eliminating items that did not receive the appropriate categorization by at least four of the five judges, 74 items remained, and these were submitted to four other judges. The second panel of judges was given the definition for each dimension, and each judge was asked to rate each statement as being clearly representative, somewhat representative, or not representative of the dimension. Items evaluated as clearly representative by three judges and as no worse than somewhat representative by a fourth judge were retained. This process eliminated 12 items, leaving 62 items. Redundant items were eliminated at this point such that 15 items for each dimension remained. Each item was formatted into a five-point (strongly agree to strongly disagree) Likert-type response scale. Items for the three dimensions were interspersed in all subsequent questionnaires.

Samples for Scale Development and Assessment of the Latent Structure

The first sample consisted of 273 undergraduate business students (119 females and 154 males). The students were recruited via an opportunity to win a raffle for one of several gift certificates for a music compact disc (to be awarded in the proportion of one per 15 respondents). The second, more heterogeneous, sample was obtained through a mail survey. Using the telephone directory of a Midwestern city as a sampling frame, 1,650 households were randomly selected and precontacted by phone to request a family member's participation in the mail survey (213 numbers were later identified as businesses). Participants were promised confidentiality of their responses. Of 1,320 completed calls, members from 909 of these households agreed to complete the survey. These contacts were mailed a copy of the questionnaire, a stamped envelope for return of the questionnaire, and (two weeks later) a reminder postcard. Of 621 returned and usable questionnaires (a 43 percent response rate), 341 were completed by women; the average age for the sample was approximately 45.

Item Refinement

Using data obtained from the student sample, items that did not have corrected item-to-total subscale correlations above .50 were deleted. Items that did not have statistically higher correlations with the dimension to which they were hypothesized to belong in comparison with item correlations with remaining dimensions' total scores were also deleted (Bearden et al. 1989). These analyses resulted in a reduced scale of 31 items-11 items each for creative choice counterconformity and unpopular choice counterconformity and nine items for avoiding similarity (see Table 1).

ASSESSMENT OF THE LATENT STRUCTURE, SCALE RELIABILITY, AND SCALE NORMS

Evaluation of the Latent Structure

Corresponding with its theoretical basis, the new scale should exhibit the latent structure of a higher-order factor model in which each of the three dimensions are first-order factors that collectively are accounted for by a higher-order factor. Statistically, however, such a model is the equivalent of a three-factor correlated model. Thus, a three-factor correlated measurement model was examined that hypothesized the existence of the three facets in which consumers' need for uniqueness is manifested. Each item was modeled to reflect a nonzero loading on its respective factor and zero loadings on the other two factors. Tests of model fit were based on sample covariance matrices.

Distribution of Item Responses. The assumption of a normal distribution associated with the maximum likelihood method of estimation did not hold for the responses to the scale in either sample. Hence, a scaling procedure, in which the Satorra-Bentler chi-square (1988) (S-BX') and robust standard errors were obtained with the maximum likelihood procedure in the EQS program, was used to estimate the fit of the measurement models to the sample covariance matrix.

The Relative Adequacy of Competing Measurement Models. Table 1 provides the standardized factor loadings associated with the three-factor oblique model. Table 2 presents the model fit indices for the two samples. The threefactor oblique model provided a better fit relative to five more restricted competing models in both samples: a onefactor model in which all 31 items loaded on a single factor, a three-factor orthogonal model in which items load on three uncorrelated factors, and three two-factor oblique models in which all possible pairs of factors were combined to form a single factor that was correlated with the remaining factor. The hypothesized model was the only model to exhibit acceptable fit, as indicated by its CFI^sub S-BX2^ and TLI^sub S-BX2^, which both exceeded .90, and RMSEA of .053 in the student sample and .057 in the mail survey sample. Each indicator zvalue exceeded 6.78 (p < .001), and all standardized factor loadings were larger than .40. Measurement invariance tests conducted across the two samples supported the replicability of the three-factor oblique structure and the equivalence of the factor variances and covariances, which suggests that the measure is similarly perceived by students and the more heterogeneous consumer sample.

Treatment of the Scale in Validation Studies. The intercorrelations among the three factors, which ranged from .52 to .57 in the student sample and from .56 to .67 in the mail survey sample, suggest that the same individuals who indicate a tendency to satisfy counterconformity motivations in consumer contexts through creative responses also tend to differentiate themselves via unpopular choices and through avoiding similar choices. Thus, the three facets were summed to form a composite index for purposes of validation. Summing across facets of a latent construct was deemed conceptually appropriate given that the multifaceted consumers' need for uniqueness construct should relate to a diverse range of outcome measures better than does any one component dimension, rendering it more important than the lower-level information obtained (see Carver 1989). Further, as illuminated in the introduction, it is the higher-level construct of consumers' need for uniqueness, rather than its constitutive dimensions, that is of interest in tests of uniqueness theory and marketing and consumer behavior theories. Thus, the scale validation studies described in this article were designed to validate the overall consumers' need for uniqueness (the measure is denoted CNFU) rather than the individual dimensions that compose it.

Scale Reliability and Scale Norms

The estimates of internal consistency reliability, adjusted for dimensionality (Nunnally 1978), were .94 in the student sample and .95 in the consumer mail survey sample. Using subsamples drawn from the previously described consumer mail survey sample, test-retest reliability estimates of .81 (n = 84) and .73 (n = 346) were obtained from administrations separated by one and two years, respectively. With respect to scale norms, the mean scores for CNFU in the consumer mail survey sample (n = 621) was 2.60 (range = 1.06.55), with a standard deviation of .56. Mean scores on individual items ranged from 2.21 to 3.31, while standard deviations ranged from .56 to 1.08. Using data from the consumer mail survey sample, CNFU was not related to education or gender at the .01 level of significance. However, CNFU exhibited a modest negative correlation with age (r = -.19, p < .01). Also, CNFU differed across income groups (F(6,579) = 2.91, p <.01), which were represented by the following categories: < $5,000, $5,000-$9,999, $10,000-$19,999, $20,000-$34,999, $35,000-$49,000, $50,000-$99,999, and over $100,000. However, only the mean for the lowest income group (< $5,000) differed from means of the other six groups.

CONSTRUCT VALIDATION STUDIES

Beyond assessing the latent structure, internal consistency reliability, test-retest reliability, and lack of confounding by demographic variables, we subjected CNFU to a series of tests established for well-designed measures of personality (see Bearden and Netemeyer 1999). Tables 3 and 4 summarize the various psychometric tests that were conducted using 10 independent samples. We report results of tests of known-groups validity, response bias, and discriminant validity prior to introducing a series of tests of nomological validity.

Known-Groups Validity

Known-groups validity of the scale was assessed by investigating whether the measure could distinguish between groups of people who should score high and low on the trait based on group differences in mean scores. In total, five tests of known-groups differences were conducted with each of five chosen comparison groups representing different types of consumer differentiation. The first three groups, which drew members from a heterogeneous population, were compared with the mean for the heterogeneous mail survey sample (n = 621, M = 2.60). The two remaining groups drew their members from a student population, and bus they were compared against the larger student sample used previously to validate the latent structure (n = 73, M = 2.71).

The groups, the type of consumer differentiation that the group represented, the data collection locales for the selfadministered survey, and the results were as follows: (1) tattoo and body piercing artists, representing the domain of bodily displays of uniqueness, were surveyed at their booths at a regional tattooing convention (M = 3.05, n = 39, t = 3.22, p < .001) ; (2) owners of customized low-rider automobiles, representing the domain of possessing unique objects, were surveyed during one of their weekly meetings held in a parking lot of a local retail establishment (M = 2.99, n = 22, t = 3.22, p < .001); (3) members of the Society for Creative Anachronism (SCA), representing the domain of enacting unique consumer performances (via swordfighting and using medieval-period language and costumes), were surveyed at a regularly scheduled meeting (M = 2.91, n = 21, t = 2.49, p < .01); (4) a drawing class attended by art majors, representing the domain of self-manufacturing unique consumer possessions, were surveyed during a scheduled class (M = 3.06, n = 22, t = 3.15, p < .01); and (5) purchasers of new art posters, representing the domain of buyers of unique but commercially manufactured objects, were surveyed on exiting a university-sponsored art sale (M = 2.83, n = 78, t = 1.89, p < .05).

Tests of Response Bias

The potential confounding of responses to the CNFU scale by social desirability response bias was assessed, as has recently been advocated in consumer research 1996). Such an assessment seemed warranted given that making a consumer choice for the purpose of being different from others often involves changing one's previously displayed preferences. That is, endorsing counterconformity behaviors can be construed as a lack of socially desirable traits of independence or self-consistency in adhering to one's personal taste. This assessment was conducted with a sample of 129 undergraduate students who completed a survey in which the presentation of the new measure was counterbalanced with two measures of socially desirable responding. Results indicated that socially desirable responding was not a threat to the measure's validity (see Table 4). The CNFU did not correlate with the Marlowe-Crowne (MC) socially desirable response scale (Crowne and Marlowe 1960) (r = .10, NS) or impression management (Paulhus 1993) (r = .01, NS).

Discriminant Validity

Discriminant validity was evaluated using responses to two measures that are similar but conceptually distinct from CNFU: optimum stimulation level (OSL; Steenkamp and Baumgartner 1992) and general need for uniqueness (NFU; Snyder and Fromkin 1977). Specifically, whereas CNFU reflects the desire for differentness relative to other people, the conceptually similar but distinct construct of OSL reflects desire for differentness relative to one's prior experiences (i.e., self-counterconformity; Nail 1986) that is driven by the need for sensory and cognitive stimulation rather than social differentiation. Using a sample of 244 undergraduate students, we collected the CNFU scale one week prior to collecting responses to OSL, which is composed of four scales (247 items); these four scales were rotated across the sample (see Steenkamp and Baumgartner 1992). Each of the four scales was summed to form four indicators of the OSL latent variable. Evidencing the new measure's distinctiveness, CNFU exhibited a moderate disattenuated correlation of .54 with OSL (see Bearden and Netemeyer 1999).

Whereas CNFU is a measure of trait counterconformity motivation in a consumer context, NFU was developed to capture the general trait of need for social differentness. The two measured constructs reflect a similar process of social comparison and moving away from a norm response to acquire differentness relative to others, but the two represent different domains of satisfying counterconformity motivation (i.e., consumer contexts vs. verbal interpersonal interactions). Further, NFU reflects both counterconformity and independence motivations (Tepper 1996). Supporting the new measure's discriminant validity, data from 261 students, who were part of the sample used in assessing the latent structure, demonstrated that CNFU possessed a moderate positive correlation with NFU (r = .44, p < .001). Based on t-tests of differences in dependent correlations (i.e., correlations drawn from the same sample), CNFU correlated more strongly with the NFU subscale, which reflects counterconformity, desire not to follow the rules (r = .46, p < .001,) than with the other two NFU subscales, which reflect independence (see Nail 1986), willingness to defend one's beliefs (r = .17, p < .01; t(258) = 4.17, p < .001) and lack of concern for others' reactions (r = .33, p < .001; t(258) = 1.99, p < .01).

NOMOLOGICAL VALIDITY

Evidence of nomological validity is provided by a construct's possession of distinct antecedent causes, consequential effects, or modifying conditions, and quantitative differences in the degree to which a construct is related to antecedents or consequences or varies across conditions in exhibiting consequential effects (Iacobucci, Ostrom, and Grayson 1995). Established models of nonconformity constructs provided the framework from which we developed tests of the nomological validity of CNFU. Specifically, models of nonconformity suggest that counterconform-tity and independence are two concepts tied to distinct motivational sources of social differentness encompassed by a superordinate construct of voluntary social differentness that reflects willingness to stand out as different in the service of various motivations (Nail 1986; Tepper 1997). Correspondingly, the CNFU scale reflects counterconformity motivation, NFU (Snyder and Fromkin 1977) largely reflects displays of differentness that derive from independence motivation but is confounded with counterconformity motivation, and the Individuation scale (Maslach et al. 1985) reflects voluntary social differentness without specific reference to the motivational source (Tepper 1996). We derived and tested hypotheses to assess whether the measure of CNFU, as compared with NFU and Individuation, distinctly operates in the manner posited for counterconformity motivation.

Tests of Trait Antecedents

Individuals driven by counterconformity motivation should seek or pursue differentness as an end goal and, where possible, often choose to do so in nonconfrontational venues (i.e., through choices that do not harm communion with others, including the selection of unique consumer products; Snyder 1992). Thus, it is hypothesized that CNFU should exhibit a positive correlation with collective individualism, which captures the notion that individuals have different goals from members of their in-group (and thus make social comparisons in terms of their differentness from the in-group) but at the same time desire a sense of social cohesion with others (Triandis 1995). Moreover, CNFU should exhibit a positive correlation with desire for unique consumer products (Lynn and Harris 1997). Further, as a result of pursuing differentness through nonconfrontational venues such as the purchase of unique products, individuals driven by counterconformity motivation should not perceive themselves to be similar to others with respect to their consumer choices (see Kilduff 1992; Snyder and Fromkin 1977). Thus, it is hypothesized that CNFU should exhibit a negative correlation with self-perceptions of social similarity with respect to consumer choices. Finally, counterconformity motivation leads individuals to change initial preferences when this is necessary in order to maintain their difference. Continually changing preferences inhibits "selfconcept clarity," defined as "the extent to which self beliefs are clearly and confidently defined, internally consistent, and stable" (Campbell et al. 1996, p. 141). This is because consumer preferences are used to define the self (see Belk 1988; Grubb and Grathwohl 1967). Thompson and Haytko (1997, pp. 21-22) have suggested that, for those who construct their personal identity through a contrast between their perceived fashion orientation and that of others in their social setting, "personal identity does not reflect a stable set of essential features but is negotiated in a dynamic field of social relations." Consumer goods are used as vehicles of "perpetual identity transformation." Following from these conceptual arguments, it is hypothesized that CNFU should exhibit a negative correlation with self-concept clarity (Campbell et al. 1996). Further, all of the previously hypothesized associations with CNFU should be stronger than they are for NFU or Individuation, which are not closely tied to counterconformity motivation.

The Sample, Procedure, and Measures. Data were collected at two separate time periods from 121 students. The first questionnaire requested that the respondent provide the names of 20 fellow students with whom they were familiar, and it then presented measures of CNFU, NFU, and Individuation with the order rotated across the sample. Two days following this administration, the respondents completed a second survey. Respondents were first presented with a measure of social similarity as a consumer, which was adapted from Kilduff (1992). Specifically, each was given the list of 20 names that she or he had provided in the previous survey and were asked to "place a check by the name of each individual who you think is especially similar to yourself in terms of their consumer behavior." Following this task, measures of self-concept clarity (Campbell et al. 1996) and collective individualism (Triandis 1995) were presented. Twelve weeks later, respondents completed the desire for unique consumer products scale (Lynn and Harris 1997), which reflects the behavioral tendency to purchase unique products.

Results. Results are summarized in Table 3, which presents the Pearson correlation coefficients of the three traits with these measures. The hypotheses were tested using ttests of differences between correlations for dependent samples. The internal consistency reliability estimates for CNFU, NFU, and Individuation were .95, .70, and .90, respectively. Similar estimates for the remaining measures ranged from .63 for collective individualism to .89 for selfconcept clarity.

The pattern of correlations provided evidence that CNFU operates in the manner posited for counterconformity motivation, though Individuation and NFU do not. The CNFU exhibited a stronger positive correlation with collective individualism than did NFU (for CNFU, r = .39, p < .001 vs. r = .19, NS; t(118) = 1.87, p < .05) or Individuation (r = .19, NS; t(118) = 2.06, p < .01). As expected, CNFU also exhibited a stronger positive correlation with desire for unique consumer products (for CNFU, r = .65, p < .001 vs. r = .17 for NFU and r = .19 for Individuation, both NS, t(118)'s >: 5.40, p < .001). The CNFU correlated negatively with perceived social similarity as a consumer (r = .17,p <.05). This correlation was not stronger than that for NFU (t = .61,NS), but it was statistically different from Individuation's positive correlation with this trait (t(118) = 2.94, p < .01). Further, and as hypothesized, CNFU exhibited a negative correlation with self-concept clarity (r = .20, p < .05), and NFU exhibited a positive correlation with self-concept clarity (r = .27,p<.01). These correlations were significantly different (t(118) = 4.32, p < .001). Individuation was not correlated with self-concept clarity (r = .16,NS) such that its correlation also statistically differed from CNFU's correlation (t(118) = 3.70, p < .001).

Tests of Consequential Effects

Prior theory of consumers' responses to exterior product designs suggests that individual differences in counterconformity motivation should influence the selection of unique as compared with more commonplace exterior product designs (Block 1995). Choosing commonplace designs that are produced and owned by the masses increases perceptions of similarity to others. Those who are highly driven by counterconformity motivation, as is captured by CNFU, would thus attempt to avoid this similarity and heighten self-perceptions of their differentness through the selection of unique designs. In contrast, someone who asserts independence, as is captured by NFU, may prefer either unique or more commonplace designs, depending on which better satisfies her or his personal aesthetic tastes or standards (Tepper 1997). Similarly, willingness to volunteer to be socially different would not always lead to preference for unique exterior product designs, given that various motivations for becoming individuated, such as leading a group of people, may be better served by other distinguishing behaviors (e.g., displaying novel thoughts or speaking loudly). Thus, we hypothesize that CNFU, as compared with NFU and Individuation, will possess a stronger correlation with preference for unique as compared with common exterior product designs.

The Sample and Method Design. Undergraduate students from two universities (n = 235) were recruited to participate in two ostensibly separate investigations. The first investigation, presented as one being conducted by researchers at an out-of-state university, contained measures of CNFU, NFU, and Individuation, with the order of presentation of the three trait measures rotated. All questionnaires were collected prior to introducing the second investigation.

The second investigation was described as a study conducted by the students' own university for the purpose of creating a new university subject pool awards program. We described plans for a new subject pool program in which students could accrue points for participation. The points could then be redeemed for products presented in an awards catalog. To make sure that they included the most enticing incentives in the awards catalog, the program administrators ostensibly were conducting the survey to obtain students' preferences for one of two versions for each of 20 product types. Participation was encouraged by announcing that respondents would receive the opportunity to participate in a raffle for $20 gift certificates from national mail order companies (awarded in the proportion of one to 20). The participation incentive was offered at this stage rather than prior to the first survey to reinforce that the two studies were unrelated.

In evaluating the 20 photographed products pairs, students viewed one pair at a time and indicated a preference. Each product pair was depicted on a separate page and depicted a different product category. The product categories were drawn from three domains: recreational goods, practical goods, and home furnishings. Relative to other possession types, products within these domains could be identified that are relevant to most students and to both sexes. An initial collection of 101 photographed products was compiled that had won various innovative design awards, as judged by product design experts. Aided by the input of two students, we screened products within this collection based on their relevance to students in terms of use and expense. For products that remained, photographs of common exterior design counterparts were sought from retail catalogs targeting the general population. In a final screening, a collection of objects that varied in value were compiled (i.e., participants would expect that they could purchase merchandise of varying values depending on their number of accumulated points). Of the 20 products evaluated by the student participants, 15 depicted a unique exterior design version alongside a common design counterpart. The presentation of the unique design version was alternated throughout the booklet to appear on the left and then on the right of the common version. Five filler pairs were included to prevent respondents from guessing the study's purpose. These were presented in the first, fifth, tenth, fifteenth, and twentieth positions and presented unique exterior designs for both products in the choice set. The photograph booklet was professionally developed by a computer graphic designer at a northeastern advertising agency. Photographs of common and unique product design versions were scanned into a computer in order to adjust each pair to a monochrome print of uniform size and background and to remove brand names.

Outcome Measures. Outcome measures were created by coding preferences for a common design, coded zero, and preferences for a unique design, coded one, then summing across product categories to create a single choice measure of preference for unique exterior design (the range was 0-15). Respondents also allocated 10 points across the two design versions to indicate how much they liked one version relative to the other. Points assigned to unique designs were summed across product categories to create a ratings measure of preference for unique exterior design.

Results. Results are summarized in Table 3. Cronbach's alpha coefficients were .95 for the new scale, .85 for NFU, and .88 for Individuation. The results of t-tests of differences between dependent correlations (i.e., correlations drawn from the same sample) demonstrate that the new scale does a better job of predicting consumer personal preferences for unique exterior product designs than do either of the competing measures. The CNFU correlated positively and significantly with the choice measure of personal preference for unique exterior product designs (r = .36, p < .001), and it exhibited a stronger correlation with this outcome than either NFU (r = .17, p < .05; t(232) = 2.96, p < .01) or Individuation (r = .08,NS, t(232) = 3.59, p < .001). The same pattern of relationships emerged with the ratings measure of personal preference. The CNFU correlated positively and significantly with personal preference for unique exterior product designs (r = .34, p < .001), and it exhibited a stronger correlation with this outcome than either NFU (r = .19, p < .05; t(232) = 2.32, p < .01) or Individuation (r = .11, NS; t(232) = 2.93, p < .001). A regression test indicated that CNFU contributed to explained variance beyond the effects of the competing measures. For the choice preference measure, the three-predictor adjusted R-squared was .13 (F(3,234) = 11.35,p<.001); the change in Rsquared resulting from adding CNFU as the last predictor was .10. For the ratings-preference measure, the three-predictor R-squared was .12 (F(234,3) = 10.29,p<.001); and the change in R-squared resulting from the addition of CNFU was .09.

Tests of Situational Moderators of Consequential Effects

Individuals who are driven by consumer counterconformity motivation are influenced by situations that shape individuals' perceptions of the potential for the unique offering to gain popularity (Snyder 1992; Snyder and Fromkin 1977). This is because a unique offering that gains popularity loses its ability to distinguish the owner (see Grubb and Grathwohl 1967). Thus, when individuals who are motivated to act in a counterconforming manner encounter a unique offering that they perceive is likely to become popular or commonplace, they are less likely to purchase it than when the object's potential to become popular is low (see Tepper 1997). The influence of CNFU on the selection of differentiating consumer offerings should be diminished when there is higher rather than lower potential for these goods to become popular. Because NFU does not solely reflect counterconformity motivation but confounds this with independence motivation and Individuation contains neither motivation, individuals scoring high on these two scales should be less sensitive to a unique object's future popularity and the need to adjust behavior in anticipation of such popularity. Therefore, we hypothesize that the effect of CNFU on individuals' selection of differentiating offerings should be diminished, whereas the effects of NFU and Individuation should not be diminished, when the differentiating product possesses higher rather than lower potential to become popular or commonplace.

Individuals possessing a high need to feel different relative to others via consumer goods are willing to pay for distinctive designs even when there are significant costs involved (Bloch 1995). Higher prices contribute to the perceived exclusivity of the object beyond that achieved by the physical design. Further, distinctive designs may represent "artworks that are sacralized by a high purchase price" (i.e., made special or unique; Belk et al. 1989, p. 24). Thus, counterconforming individuals' tendency to select a differentiating offering over a commonplace one should be enhanced when price is higher as compared with lower. Consistent with this expectation, in analyzing critical incidents from consumers who described a real-life experience of being different from other consumers, Tepper (1997) found that higher-priced goods were viewed as offering greater opportunity for social differentness by excluding ownership of the object by the majority. Thus, CNFU's effect on the selection of differentiating consumer offerings should be enhanced when the differentiating product relative to its nondifferentiating counterpart possesses higher rather than lower price. Both NFU's and Individuation's effects should not be enhanced by higher prices because they do not emphasize counterconformity motivation.

Participants. Participants were the same as those who participated in the study of trait antecedents (n = 121). They attended a single, one-hour computer laboratory session that was conducted two weeks following the completion of the trait measures. Following the procedure of Steenkamp and Baumgartner (1992), the single lab session incorporated multiple investigations.

The Context. The context of the study disguised two experiments testing the effects of the hypothesized moderators as an investigation of the user-friendliness of computer software. At the beginning of the computer lab session, subjects were informed that the survey was sponsored by Mon Group Incorporated, a company that provides computer software design and consulting services to various clients. Mon was conducting the survey to obtain feedback on the ease of use of different on-line ordering programs for consumers. Subjects were further informed that the programs differed in some of their instructions, answering formats, types of product options, and the information that the computer provides back to the consumer. In addition, they were told that, in some instances, options had been limited to a smaller number in order to shorten the time to complete the study. Instructions stressed that in order for the results to be useful, participants should enter their purchase orders as they would if they were actually buying from the product category, because Won also intended to pass on the product choices to their clients as potentially useful market information.

The Procedure. The procedure began when, at a separate, appointed time, each subject entered the computer lab. On logging onto the computers, subjects first read background information on IKon. After completing a question regarding their prior computer experience (a task to enhance the cover story), participants were exposed to a series of on-line product-ordering programs that ultimately would be used as Web sites for IKon's clients. Each Web site began with a screen that provided the company name, logo, and the products or services it provided. Completing the orders required the subject to make multiple choices between differentiating and nondifferentiating product offerings.

The Method Design. The method design incorporated features aimed at minimizing potential confounds. To avoid order effects, the presentation of the differentiating good was alternated across all choices within the computer laboratory session to appear on the left and then on the right of the nondifferentiating good. Differentiating and nondifferentiating visual stimuli for each choice situation were scanned into the computer and altered to create uniform sizes across alternatives and to eliminate brand names. Other choices between attributes or features that were equivalent in their uniqueness were included to enhance realism and minimize demand characteristics. To enhance the believability of the cover story, each program ended with Likerttype questions regarding the user friendliness of the system. Each program was professionally developed using Authorware Software that was selected for its ability to (1) present quality visual stimuli, (2) randomly assign subjects to a treatment group for each experiment independent of treatment assignments in the other experiment, and (3) track participants' choices and compile these in a data file. The Web sites were arranged to avoid carryover effects, which, for instance, could have occurred if the Web sites in which participants were promised feedback from the computer had been presented prior to choice situations in which no feedback was offered.

The Experiment Manipulating the Potential Popularization of a Differentiating Offering. The experiment manipulating the potential popularization of a differentiating offering was administered during a single on-line ordering program, entitled "Creative Lighting and Lamps." This online ordering program allowed consumers to design a lamp by selecting several component parts. The potential for a differentiating offering to become popular was manipulated between subjects via textual information provided to subjects at the beginning of the Web site. In the low condition, the computer screen informed participants that the product that they designed would be compared with the computer's database. They would then receive feedback from the computer as to whether their design was acceptable. The high condition was constructed to heighten perceptions that differentiating offerings would become popular or commonplace. The introductory computer screen informed participants that comparisons of designs with the computer's database were for the purpose of determining the most creative designs. The designs identified as unique would be printed in the company's mail brochure and displayed on their Web site, thus potentially popularizing the design.

The outcome measure of selection of a differentiating offering was constructed from the choices participants made for each of four component parts: the lamp shade shape, the lamp shade color, the lamp shade materials, and the lamp shade base. Four options (half differentiating) were provided separately for the shade shape, color, and base; two options (one differentiating) were provided for the shade material. Each choice appeared on a separate screen. The nondifferentiating and differentiating alternatives were selected with the aid of a professional interior designer based on her perceptions of what appeared commonplace in her clients' homes. Professional artists were hired to produce black and white sketches of the various shapes of lamp shades and bases. Selection of the differentiating consumer offering was measured as the sum of four choices (all scored 0 = selection of nondifferentiating offering, and 1 = selection of the differentiating offering; range = 0-4).

Results regarding trait correlations with selection of differentiating offerings under different conditions of situational variables are presented in Table 3. Tests of hypotheses were conducted by comparing correlations across levels of the manipulation for each trait using z-tests of differences in independent correlations (i.e., correlations drawn from two different samples). The moderating role of potential popularization of a differentiating offering in the relationship between CNFU and selection of a differentiating offering was supported. Specifically, the effect of CNFU on selection of differentiating offerings was diminished when there was a high as compared with low potential that the differentiating selections would become popularized. In the high potential for popularization of the differentiating design, the correlation coefficient of .13 was nonsignificant, whereas the low condition exhibited a correlation of .42 (p < .001; z = 9.06; p < .01). A similar pattern across the conditions emerged for Individuation, which exhibited correlations of .04 and .33 (z = 8.66,p < .01) in the high and low conditions, and for NFU, which exhibited correlations of .19 and .28 (z = 2.74,p < .01). Notably, relative to the other traits, the magnitude of CNFU's correlation with selection of differentiating offerings was larger under the condition of low potential for popularization.

The Experiment Manipulating Price. The experiment manipulating price was carried out between subjects across four of the on-line ordering programs via textual information provided to subjects along with the visual representation of the products. Price for the differentiating offering was presented as either 20 percent higher or lower for the differentiating offering as compared with the nondifferentiating offering. Participants made four separate choices between nondifferentiating and differentiating alternatives, with one choice made in each of the first four on-line ordering programs. The nondifferentiating/differentiating alternatives were operationalized in terms of (1) brand, (2) color, (3) materials, and (4) overall novelty of the item across four product categories as follows: running sneakers (Adidas and Nike/Diadora and Mizuno), umbrellas (solid color dome/ multicolor dome), cameras (plastic and metal/cherry wood), and bike accessories (a water bottle/novelty water bottle), respectively. Selection of the differentiating offering was measured as the sum of four choices (scored 0 = selection of nondifferentiating offering and 1 = selection of the differentiating offering; range = 0-4).

The pattern of correlations supports the hypothesis that the effect of CNFU on individuals' selection of differentiating offerings should be enhanced when the differentiating product relative to its nondifferentiating counterpart possesses higher rather than lower price (in the higher price condition, r = .24, p < .05; in the lower price condition, r = .05, NS; z = 5.57, p < .01). As further hypothesized, Individuation's and NFU's correlations with selection of differentiating offerings were nonsignificant across both price treatments.

SUMMARY

General Discussion

This article reports on the development and validation of an instrument that measures consumers' need for uniqueness. Reflecting a latent construct at a high level of abstraction, the new measure that captures trait counterconformity with respect to consumer behaviors is reflected by three intercorrelated dimensions: creative choice counterconformity, unpopular choice counterconformity, and avoidance of similarity. The summed index exhibits internal consistency reliability, test-retest reliability over a two-year period, known-groups validity, minimal threat from social desirability response bias, discriminant validity, and nomological validity. The CNFU is a product-oriented scale that is devised to correspond with conceptual marketing models of consumers' responses to exterior product designs (Block 1995), product fashion cycles (Miller et al. 1993), and variety-seeking behavior (McAlister and Pessemier 1982). As such, the scale emphasizes visual rather than verbal communications of differentness. Future research might examine the reinforcement of visual displays of consumer uniqueness via products with verbal communications of consumer differentness.

Theoretical Applications

Consumers' need for uniqueness may fit into a broader theory of consumption as an extension of self (Belk 1988). As such, CNFU should be useful in empirical tests of theories regarding the origins of counterconformity motivation that manifests in consumer possession acquisitions and their display, as well as of theories that model the manner in which individuals consume products, experience and resolve conflicts that result from consumer counterconformity motivation, respond to marketing stimuli that employ uniqueness appeals, and make product replacement decisions.

Consumption as an Extension of Self. Consumers' need for uniqueness may fit into an overarching theory of the role that consuming material goods plays in people's sense of identity (cf. Belk 1988). Posited antecedents of consumers' need for uniqueness may suggest distal antecedents in a theory of consumption as an extension of self. Specifically, uniqueness theory suggests that individual differences in motivation to seek differentness arise from early childhood socialization that either emphasizes obedience and following norms or emphasizes creativity and individuality (Snyder and Fromkin 1977). This emphasis is thought to be determined, in part, by family size, order of birth, and number of same-sex siblings (Chrenka 1983). These predictions have not been investigated by consumer researchers and warrant additional research.

The Ways Individuals Consume Products. Following from Holt's (1995) theoretical work, CNFU should be a determinant of how individuals consume. It would be expected that individuals with a high consumer need for uniqueness would more often engage in the consumption of products for purposes of classification rather than experience, integration, or play. Additional research that we have undertaken found support for the ability of consumers' need for uniqueness to predict the valuation of important material possessions for reasons of enhancing social differentness. Future research might examine the extent to which individuals express differentness via material goods relative to other forms of communication, attempt to move goods from the realm of the profane to that of the sacred in order to foster recognition of their unique attributes, and make tangible fleeting experiences via the collection of souvenirs that reflect the participant's uniqueness (see Belk et al. 1989; Holt 1995).

Conflicts Arising from Consumer Counterconformity Motivation. Consumers' need for uniqueness could be examined as a trait influencing processes whereby situations that elicit consumer counterconformity motivation lead to conflict and acts of conflict resolution (Tepper 1997). For example, commercial products marketed for their uniqueness, but which later gain popularity, elicit conflict in the form of an on-going struggle to maintain or assert uniqueness. Among acts of conflict resolution, individuals may adopt prepurchase strategies for combating copycatting of unique products, such as combining fashion objects from multiple retail outlets rather than purchasing ensembles displayed in the store, importing from larger cities that offer a greater selection of merchandise, importing from regions where fashions emerge earlier than in the consumer's place of residence, and searching for discontinued goods in antique shops and at garage sales. As a second example, some buying contexts introduce consumer goals that conflict or compete with counterconformity motivation. Consumer counterconformity goals may compete with family oriented goals (e.g., buying a differentiating sports car vs. a sports utility vehicle that better meets family needs) or the desire for privacy (e.g., a car of a distinctive color in a small town renders one different but also makes one's actions known to others). Consumers attempt to resolve the conflicts by either postponing purchases that satisfy the desire to be different until after purchases that satisfy the competing goal (e.g., putting family needs first) or restricting counterconformity purchases to product categories for which there are no competing goals (e.g., preserving privacy by restricting use of unique goods to private contexts). The CNFU should moderate the influence of perceived conflicts on the endorsement of different methods of conflict resolution. Further, because consumers' methods of resolving conflict suggest interventions that marketers might devise for use in promotions and product offerings to assist consumers in resolving these conflicts, CNFU could be used to test the effectiveness of these interventions among groups that vary in their consumer counterconformity motivation.

Responses to Marketing Stimuli. Future research might also employ CNFU to empirically test other theoretical propositions that elaborate how individual differences in uniqueness motivation determine selections of product styles. Bloch (1995) proposes that individual differences in the need for uniqueness influences consumers' product selections through its effect on affective and cognitive responses to the exterior design. Distinctive designs are more likely to be incongruent with the taste preferences of consumers with low uniqueness needs, leading to negative affective reactions (e.g., a design is perceived to be in poor taste), negative cognitive responses (e.g., undesirable product-related beliefs), and the rejection of the product. Thus, future research might test Bloch's (1995) conceptual model by examining the mediating role of affective and cognitive responses to designs in the relationship between CNFU and selection of products with distinctive versus common exterior designs.

Product Replacement. The theoretical model of the fashion process (Miller et al. 1993) suggests that, after making product selections, consumers continue to check the styles adopted by others. Uniqueness theory and counterconformity research suggest that, for individuals with a high consumer need for uniqueness, satisfaction with self-expressive products diminishes faster, resulting in changes in previously adopted styles or designs and a higher rate of replacement of self-expressive products (i.e., shorter interpurchase cycles). Longitudinal studies might examine the moderating role of CNFU on product satisfaction, replacement, and disposition.

Individual differences in consumers' need for uniqueness might also influence product or style replacement behavior through its effect on consumers' optimum stimulation level. McAlister and Pessemier's (1982) theoretical model of variety-seeking behavior suggests that individuals' desire to be distinct from others influences variety seeking (e.g., style or design switching) through its relationship with consumers' ideal level of stimulation (i.e., novelty, complexity, change). Consumers possessing a high need for uniqueness relative to others will seek to avoid popular consumer preferences, will become more familiar with product offerings during their search for unique goods, and, thus, will require greater novelty or complexity to achieve optimal levels of arousal. Future studies should assess the mediating role of optimum stimulation level in the relationship between CNFU and variety-seeking behavior.

Finally, CNFU should be related to consumers' modification of products that may extend product replacement cycles. Because goods modified by consumers in order to obtain uniqueness may be less susceptible to commercial duplication, and hence popularization, the need for replacement may be suspended beyond typical fashion cycles. Future studies might examine such processes to facilitate understanding of how individuals combat threats to an identity that is constructed via unique goods.