The influence of gender-stereotyped perfumes on leadership attribution

The classification of perfumes as ‘women’s’ and ‘men’s’ fragrances is based on certain gender stereotypes. In two experiments, female and male participants were asked to assume the role of a manager. In Experiment 1, they read an application for the position of a junior manager written by a male or female job applicant. Application papers were prepared with a typically masculine perfume, a typically feminine perfume or no perfume at all (control group). In Experiment 2, participants conducted a job interview with a female or male applicant (a confederate) who had applied the respective perfume or no perfume. Persons with a typically masculine perfume were ‘employed’ with a higher degree of certainty compared to persons with a typically feminine perfume. Copyright # 2002 John Wiley & Sons, Ltd. In their ‘cultural history of smell’ Classen, Howes, and Synnott (1994) describe ‘ . . . how olfactory codes create and inform power relations between social classes, ethnic groups, and women and men in the contemporary West’ (p. 161). Accordingly, women are the ‘perfumed sex’, and the use of fragrances is recommended and promoted to enhance women’s desirability in the eyes of men. In contrast, men are expected to disdain the use of perfume, there is ‘no particular need for them to enhance their attractiveness. It is women’s duty . . . ’ (p. 163). The gender-stereotyped role prescriptions for heterosexual behaviors which are described here have been changing in recent years. These days, both sexes use perfume for the purpose of managing impressions of personality in social interactions. Perfumes are even classified as ‘women’s’ and ‘men’s’ fragrances, with ‘flowery’ or ‘fruity’ fragrances being offered for women, ‘spicy’ and ‘strong (tangy)’ fragrances for men. While ‘odor has been almost totally neglected by social psychologists’ (Levine & McBurney, 1986, p. 180), the few existing studies found that olfaction plays an important role in person perception. An influence of olfactory cues on social perception (for an overview see Levine & McBurney, 1986), particularly an influence of perfume on impression-formation processes, was found in several studies conducted by Baron (1981, 1983, 1986) since the 1980s: The use of fragrance by informally dressed female confederates positively influenced male participants’ impressions of their attractiveness and personal traits. When the female confederates were dressed in a more formal attire, perfume seemed to Received 13 September 2001 Copyright # 2002 John Wiley & Sons, Ltd. Accepted 19 February 2002 *Correspondence to: Dr Sabine Sczesny, Universität Mannheim, Lehrstuhl für Sozialpsychologie, D 68131 Mannheim, Germany. E-mail: [email protected] reduce the attraction (Baron, 1981). Fragrance also had an impact on evaluations of job applicants (female confederates wearing a popular and pleasant ‘women’s’ fragrance or no fragrance and male confederates wearing a ‘men’s’ fragrance or no fragrance (Baron, 1983, 1986). In the respective studies, participants evaluated applicants on job-related and personal dimensions in simulated employment interviews. Male participants rated perfumed job applicants lower than non-perfumed applicants on these dimensions; female participants showed the opposite pattern (Baron, 1983). In another study, only female applicants were evaluated. Here, their use of a ‘women’s’ perfume as a tactic of self-presentation in a job interview increased the ratings they received from female as well as from male participants on job-related and personal dimensions (Baron, 1986). The same was true for another tactic of self-presentation: the emission of numerous positive non-verbal cues (e.g. smiling, high level of eye contact, friendly posture). When the two tactics were employed in combination, however, they reduced the ratings of male interviewers, an effect that was interpreted by the author as ‘too much of a good thing’. In the reported studies, the impact of fragrance use was thus moderated by other factors: mode of dress (Baron, 1981), sex of participant/perceiver (Baron, 1983), and nonverbal cues (Baron, 1986). Only one of Baron’s studies included the sex of stimulus person as an independent variable. This variable failed to exert any significant influence on the evaluation of perfumed job applicants (Baron, 1983). In sum, the results of Baron’s work reflect the complex interrelations of olfactory and other social cues in the impression formation process. The application of perfume had neither consistently positive nor consistently negative effects. It is important to consider the perceptual quality of odors in order to predict their effects on person perception (McArthur & Baron, 1983). Research on olfactory stimuli conducted by Fiore (1992) showed that specific components of fragrances influence impressions of personality in specific ways. Participants (all female in this study) were asked to imagine fictitious persons wearing three different commercial women’s fragrances (floral, oriental or chypre) and to evaluate the personality of these persons. The three different fragrances were presented on scent strips. Compared to oriental and chypre fragrances, the floral fragrance triggered lower ratings on traditionally male traits (e.g. competent, professional, career-orientated, confident, assertive). Rather than depending on the mere presence or absence of perfume, the impression formation process is thus apparently influenced by finer distinctions of the cues which result from the expressive quality of aesthetic components: ‘If the stimuli for Baron’s study were oriental fragrances, associated with the Traditional Male factor and other traits compatible with a professional image, the studies may have resulted in higher evaluations on job-related dimensions’ (Fiore, 1992, p. 159). Consider the classification of perfumes as ‘women’s’ and ‘men’s’ fragrances that was mentioned above. The association of ‘flowery’ or ‘fruity’ fragrances with women and of ‘spicy’ and ‘strong (tangy)’ fragrances with men is based on gender stereotypes and is therefore of special interest for the area of research on leadership attribution reported here. GENDER ROLE STEREOTYPES AND LEADERSHIP ATTRIBUTION Research on gender stereotypes has consistently demonstrated that men are generally seen as more agent and more competent than women, while women are seen as more expressive and communal than men (e.g. Broverman, Vogel, Broverman, Clarkson, & Rosenkrantz, 1972; Williams & Best, 1982). One area in which gender stereotypes manifest themselves is the attribution of achievement performance and leadership: ‘If women in general are believed to be less competent, for example, then a specific woman’s performance is viewed less positively and her success is less likely to be explained by assuming ability’ (Deaux, 1995, p. 13). Regarding the context of leadership, Schein (1973, 1975) examined the social image of successful middle managers and found that the attributes 816 Sabine Sczesny and Dagmar Stahlberg Copyright # 2002 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 32, 815–828 (2002) ascribed to a ‘successful middle manager’ yielded a significantly higher correlation with the description of a typical man than with the description of a typical woman. This phenomenon of ‘think manager—think male’ was confirmed in many subsequent studies (e.g. Brenner, Tomkiewitz, & Schein, 1989; Massengil & DiMarco, 1979; Schein, Mueller, Lituchy, & Liu, 1996; Sczesny, Spreemann, & Stahlberg, presentation at the 12th General Meeting of the European Association of Experimental Social Psychology, Oxford, 1999). Heilman’s (1983) lack-of-fit-model provides an explanation of this phenomenon (see also prototype-matching processes, e.g. Lord & Maher, 1991). According to Heilman, ‘expectations about how successful or unsuccessful an individual will be when working at a particular job are determined by the fit between the perception of an individual’s attributes and the perception of the job’s requirements in terms of skills and abilities’ (p. 278). For organizational jobs that are male-typed, e.g. leadership positions, there is a lack of fit between the perceived requirements of the job and the skills and abilities typically attributed to women as a group. This presumed lack of fit may increase the likelihood of sex-biased judgments or behaviors (see also ‘Role congruity theory of prejudice toward female leaders’, Eagly & Karau, in press). BIOLOGICAL SEX AND OTHER CUES THAT ACTIVATE GENDER STEREOTYPES Not only does biological sex activate gender role stereotypes, but other cues that are part of a global, multi-faceted gender role stereotype can also activate the attribution of leadership. A multidimensional conception extends the formerly trait-based view of gender stereotypes and opens a more differentiated perspective. According to this conception, stereotypes are composed of diverse components, such as traits, role behaviors, occupations and physical appearance (Deaux & Kite, 1993; Deaux & Lewis, 1983, 1984; Freeman, 1987; Jackson & Cash, 1985). While the salience of a person’s biological sex is often considered sufficient for the activation of the corresponding stereotype components, the work of Deaux and Lewis (1984) suggests that gender-stereotyped physical characteristics (e.g. broad shoulders versus dainty physique) can outweigh sex as a basis of judgment. Their results show that the components of gender stereotypes differ in their ability to implicate other components, with physical appearance playing a dominant role (see also Freeman, 1987). Biological sex—in terms of a social category—activates gender stereotypes resulting in different evaluations of women and men. But other physical aspects such as facial features, voice or smell are probably also associated with gender stereotypes: Sczesny, Spreemann, and Stahlberg (under review; Spreemann, 2000) assumed that—because of the close association of typically masculine attributes with attributes of the typical successful manager—masculine features were expected to trigger the contents of gender stereotypes. The results of four experiments indicated that more pronounced leadership qualities were attributed to stimulus persons with typically masculine physical characteristics than to those with typically feminine characteristics, regardless of the person’s biological sex. Since odor can be seen as another facet of physical appearance, it seems worth looking into its influence on person perception. OVERVIEW OF RESEARCH The main aim of the present research was to examine the effects of ‘women’s’ and ‘men’s’ perfumes on person perception. We intended to test whether a typically masculine perfume can activate gender stereotypes, similar to visually perceivable characteristics (see above). For this purpose, we conducted two experiments. Our hypothesis at the onset of the investigations was that gender-stereotyped Perfume 817 Copyright # 2002 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 32, 815–828 (2002) perfumes have an effect on the evaluation of persons who apply for a leading position in a company. We expected stereotypically masculine and stereotypically feminine perfumes to trigger significantly different responses in the perceivers. The applicants associated with a typically masculine perfume should be employed with a higher degree of certainty than applicants associated with a typically feminine perfume because of the activation of the male stereotype and in correspondence with the phenomenon ‘think manager—think male’ (Hypothesis 1). Because the application of perfume had shown both positive and negative effects in previous research we compared gender-stereotyped perfumes with a perfume-free condition: The applicants associated with a typically masculine perfume should be employed with a higher degree of certainty than applicants associated with no perfume whereas applicants associated with a typically feminine perfume should be employed with a lower degree of certainty than applicants associated with no perfume (Hypothesis 2). In order to explore the effects of sex of applicants, this aspect was included as a factor in the experimental design.