Can the impression created by a perfume enter into judgments about a person’s ability to do a job? A study published in 2002 by Sabine Sczesny and Dagmar Stahlberg explored this question in two simulated hiring experiments.
The focus was not simply whether a scent was liked or disliked. The social-psychological question was whether perfumes perceived as masculine or feminine could evoke gender stereotypes and change judgments of suitability for a leadership role. The study did not measure actual hiring outcomes in companies or applicants’ abilities.
People assigned the gender associations of the scents
The researchers obtained 12 commercial perfumes: six marketed by their manufacturers for women and six for men. In a preliminary study, 60 people each smelled one perfume and rated how masculine and feminine they found it.
The researchers then selected the perfume rated most masculine and least feminine, and the one rated most feminine and least masculine. The terms masculine perfume and feminine perfume below refer to this preliminary classification. They do not mean that scents have a biological sex or determine the gender of the people who wear them.
The research was motivated by interest in the stereotype linking leadership with masculinity. If scents could evoke similar associations, the researchers reasoned, evaluations might differ even when applicants provided the same information and had the same background.
First, participants evaluated scented application documents
The first experiment involved 74 students at the University of Mannheim in Germany: 37 women and 37 men, with an average age of 23.7 years. Acting as personnel managers, they read applications for an entry-level position intended for a prospective manager.
The document content was standardized, with the applicant’s name changed to a male or female name. There were three further conditions: masculine perfume applied to the documents and desk pad, feminine perfume applied in the same way, or no perfume. Participants were randomly assigned to conditions.
The outcome score combined the decision to hire or not hire with the strength of confidence in that decision. A score of minus 4 meant strong confidence in rejecting the applicant; plus 4 meant strong confidence in hiring. It was not a hiring percentage.
Mean scores were 2.04 with the masculine perfume, 0.58 with the feminine perfume, and 2.71 without perfume. In planned comparisons, the masculine-perfume condition was rated higher than the feminine-perfume condition, and the no-perfume condition was also rated higher than the feminine-perfume condition. There was no clear difference between masculine perfume and no perfume.
In other words, masculine perfume did not provide an advantage over no scent in this experiment. There was also no clear effect of whether the applicant’s name was male or female.
The comparisons changed in face-to-face mock interviews
Perfumed application documents are somewhat unusual. Considering the possibility that the scent had been too strong or seemed inappropriate for work documents, the researchers conducted a second experiment.
This time, 116 students acted as interviewers and met a research confederate playing an applicant. Two women and two men served as applicants, wore standardized clothing, and gave the same rehearsed answers to five predetermined questions. Perfume was applied to shirts or blouses rather than skin. Again, the two perfumes were compared with no perfume.
Mean hiring-decision scores were 1.95 with the masculine perfume, 1.00 with the feminine perfume, and 0.88 without perfume. However, the overall test of differences among the three perfume conditions had a p value of approximately 0.096, which was not significant at the usual 5% threshold.
In planned comparisons using one-sided tests focused on the predicted direction, however, masculine perfume received higher ratings than both feminine perfume and no perfume. There was no clear difference between feminine perfume and no perfume.
The authors regarded the same directional difference between masculine and feminine perfumes in both experiments as preliminary support for their hypothesis. Nevertheless, the strength of the findings requires qualification, including the inconclusive overall test in the second experiment.
Was it simply a scent people liked?
In the first experiment, participants who noticed the scent were also asked to rate its pleasantness. The feminine perfume was rated as more pleasant than the masculine perfume.
Greater pleasantness was associated with more favorable hiring judgments, but the difference between the two perfumes remained after pleasantness was taken into account statistically. At least in this analysis, liking alone did not explain the result. Too few participants in the second experiment reported noticing a scent to permit the same analysis.
The authors therefore proposed that the scent’s gender associations might have activated stereotypes about suitability for the job. However, they did not directly measure and confirm the pathway from a scent to a particular stereotype and then to a changed judgment.
Scent intensity and how well it fits the setting remain open questions
Comparisons with no perfume differed between the document task and the face-to-face interviews. The authors suggest that scent intensity and a mismatch with what is expected in a setting may affect attention. A more noticeable scent might prompt a more conscious judgment of whether it is appropriate at work.
This explanation remains to be tested. The two experiments did not systematically manipulate matched scent intensities, and the perfume brands and spray quantities were not reported. The study did not fully isolate scent type from all other differences.
Only a minority of participants in the face-to-face interviews reported noticing a scent. The authors discuss possible effects outside awareness, but a later report of not noticing a scent does not itself prove that a scent below the sensory threshold acted unconsciously.
Furthermore, the assessors were mainly young students, not experienced hiring professionals. Findings from Germany reported in 2002 using two particular perfumes cannot be directly generalized to every workplace or culture today.
Summary
Across two simulated hiring experiments, there was a pattern of more favorable hiring judgments for applicants associated with a perfume rated as masculine than for those associated with one rated as feminine. However, comparisons with no perfume were inconsistent, and the overall difference in the face-to-face experiment was not clear.
The important question is not whether perfume can improve ability, but whether sensory cues unrelated to ability can enter into evaluations of people. Testing the possible influence of scents through social associations, while also considering intensity and the naturalness of the setting, can help us understand how we judge others.
References
Sczesny, S., & Stahlberg, D.(2002). The influence of gender-stereotyped perfumes on leadership attribution. European Journal of Social Psychology, 32, 815–828. DOI: 10.1002/ejsp.123. Full article (university-hosted PDF). © 2002 John Wiley & Sons, Ltd.
