Submission Version (short) Thesis Version (long)

Running Head: FIRST IMPRESSIONS

What's In A Face: Correlations Of Self, Spouse, And Stranger Ratings Of Personality

John Hubbard
Macalester College (Student)

Faculty Advisor:
Jack Rossmann
Macalester College

Abstract

One reason for the reported correlations of self and stranger ratings of personality may be that people's facial expressions, over time, form lasting features on their faces. The accuracy of impressions from facial features could, therefore, increase with the age of the people being judged. We attempted to assess this hypothesis by asking subjects (25 married couples) to rate themselves and their spouses on the 20 bipolar questions used by Watson (1989). Facial photographs were taken of each subject and shown to 42 students, who rated the photographed subjects on the same 20 questions. Ratings by self, spouse, and strangers were compared. Although the correlations of stranger with the other two ratings were minimal, they tender to be larger for older targets. Given the small number of subjects, further research is necessary.

What's In A Face: Correlations of Self, Spouse, and Stranger Ratings of Personality

First impressions are formed when people observe others for the first time and then ascribe personality traits based on those observations. First impressions play an important role in human interaction because they affect the ways in which people anticipate reactions from others. It is important to try to increase our understanding of how first impressions are formed so that we may be aware of the possible inaccuracy of our judgments of others. An experiment was designed to test an hypothesis regarding one possible cause of the accuracy of first impressions from facial appearance: the molding of the face by personality over time.

An accurate judgment is defined as one which is agreed upon by others and corresponds to a criterion (Funder, 1987); however, there is no purely objective measure of personality (Zebrowitz, 1990), and any model of personality is often disputed. The personality test used in this experiment was a version of a five-factor model, measured on five scales of Neuroticism, Extroversion, Conscientiousness, Agreeableness, and Culture. For a review of the evidence relating to the comprehensiveness of the five-factor model of personality, see McCrae & John (1992).

Watson (1989), using a version of this model, administered 20 bipolar questions to obtain ratings of stranger's personalities. Groups of 5 to 10 unacquainted students rated themselves and each other. A significant correlation was found between the self and stranger ratings on all scales except Neuroticism. The ratings on Extroversion were significantly correlated (r = .41, p < .05). Other researchers have found similar correlations of such ratings of personality (Albright, Kenny, & Malloy, 1988; Borkenau & Liebler, 1992).

Facial appearance alone has been shown to be an accurate predictor of personality. For example, ratings on Dominance of facial photographs of graduating West Point cadets were significantly correlated (r = .54, p < .001) with cadets' rank in senior year (Mazur, Mazur, & Keating, 1984). Based on facial photographs, strangers' ratings on Honesty correlated negatively with the targets' likelihood of volunteering to participate in a deceptive experiment (Bond, Berry, & Omar, 1992). Subjects' self-ratings on personality tests have been correlated with ratings of facial maturity (Berry & Bronlow, 1989), and, from facial photographs with neutral expressions, with strangers' rated impressions on scales of Power and Warmth (Berry, 1990, 1991).

One basis for these correlations may be that personality influences facial development (cf. Malatesta, Fiore, & Messina, 1987). Emotional expressions over time may make people's faces descriptive of their personalities. To examine this hypothesis, the effects of the age of the subject on the correspondence of self and stranger ratings of personality traits were measured. Correspondence should hypothetically increase with the age of the subjects. Since other studies only used college-age subjects, a larger range of ages was studied. To avoid other aspects of appearance which might influence judgments of personality, photographs of subjects' faces were used. A neutral expression was used to rule out effects of different facial expressions.

Method

Materials

The questionnaire used was very similar to the inventory used by Watson (1989). The version used has 20 questions. Subjects were asked to rate different personality traits anchored by a bipolar pair of adjectives (e. g., talkative/silent) on a scale anchored by 1 (e. g., very talkative) and 5 (e. g., very silent). Four questions were developed for each of the five scales (see Table 1). The same 20 bipolar adjectives and 5-point scales as used by Watson (1989) were given to obtain scores for the five scales. The 20 questions have been shown to be reliable measures (Watson, 1989) of the scales called Extroversion, Agreeableness, Conscientiousness, Neuroticism, and Culture.

Subjects

Subjects were recruited through personal acquaintances. To obtain a specific type of peer rating, only married couples were recruited. Couples were asked if they would volunteer fifteen minutes to participate in a research study. After signing a general consent form, they were asked to complete a questionnaire describing themselves and then one describing their spouses. Subjects were next asked if they would agree to be photographed for future research. Consenting subjects were asked to remove all visible jewelry and a black cloth was placed around the shoulders to conceal clothing. Pictures of head and shoulders were then taken against a black background. Subjects were asked to maintain a neutral facial expression while being photographed. Color slides were prepared of each of the pictures of each of the subjects. The ages of the subjects and how long each couple had been married were then recorded.

Subjects included 25 married couples who filled out the questionnaires completely and whose pictures developed clearly. Couples were divided into two groups based on the length of time they had been married. In Group 1 were 12 couples who had been married from 2 to 16 years, averaging 9.8 years (SD = 4.2). In Group 2 were 13 couples who had been married an average of 33.8 years (SD = 8.4), with a range of 22 to 46 years. Individual subjects were divided into two age groups: 26 in a younger group from 28 to 49 years old, with an average age of 39.8 years (SD = 5.8), and 24 in an older group from 50 to 74 years old, with an average age of 62.1 years (SD = 6.6).

Students in psychology classes received course credit for volunteering to serve as the strangers. A total of 42 strangers rated personalities of the subjects from their pictures. In each of three sessions, with 14 strangers participating in each session, the strangers viewed a randomly chosen group of 16 or 17 slides of the pictures. The order of presentation was randomized according to a computer output of 50 randomized numbers. Strangers were given three minutes to complete a questionnaire for each subject. In this manner, ratings by fourteen strangers were obtained for each target.

Test sessions lasted approximately one hour. None of the strangers admitted to being acquainted with any of the pictures subjects. Because most photographed subjects lived several states away from the college, it can be assumed that none of the strangers and subjects were acquainted. At the end of the experiment, strangers were asked to write their age and gender on their packet of questionnaires. Of the 42 strangers, there were 11 males and 31 females whose average age was 19.6 years (SD = 2.0). Nineteen of the total of 14,000 questions the strangers were given were skipped (.14 %); these were scored as 3.

Results and Discussion

Operational Definitions

"Accuracy of first impressions" was defined as the degree of correspondence between self and strangers' ratings of personality. The following four assumptions were made: (1) the rating scale used can measure personality; (2) the subjects' responses were honest; (3) the self-ratings were an accurate measure of personality; and (4) facial photographs were appropriate stimuli for the study of the influence of facial appearance on first impressions.

Reliability

Cronbach coefficients alpha for ratings on each scale were .52 for Extroversion, .31 for Agreeableness, .33 for Conscientiousness, .44 for Neuroticism, and .28 for Culture. The reliability scores obtained for the strangers' ratings were relatively low in comparison with previous research, indicating that there was high variance of strangers' ratings of individual targets. Such poor consensus lends credence to the idea that different strangers view the same pictures differently. There are several possible sources for the individual differences among strangers, including gender, culture, individual biases, beliefs (Dion & Dion, 1987), expectations, and moods. For each subject, strangers' ratings were averaged for comparison with spouse and self ratings.

Correlations of Self and Spouses' Ratings (See Table 2)

Pearson correlations of self and spouse ratings for all fifty subjects were highest for Extroversion (r = .76, p < .05) and lowest for Agreeableness (r = .27, n. s.). The mean correlation of self and spouse ratings for all subjects across each of the five scales was .44. It can be concluded that self and spouse ratings of personality were associated.

On an average of the five scales, the Pearson correlations of self-ratings of men and the ratings by their wives were higher than the correlations of self ratings of women and ratings by their husbands (r = .52 vs. r = .35, p < .05). Correlations of ratings of couples in Group 2 (the longer married) were only slightly higher than those of couples in Group 1 (r = .48 vs. r = .42, n. s.). It could be concluded that subjects in Group 1 rated their spouses' personalities as accurately as couples who had been married for longer periods of time. Also, self-ratings of older targets corresponded more closely to their spouses' ratings of them than did those of younger targets (r = .60 vs. r = .28, p < .05).

Correlations of Spouses' and Strangers' Ratings (See Table 3)

The correlation of spouses' and strangers' ratings for all 50 subjects was highest for Conscientiousness (r = .35, p < .05) and lowest for Neuroticism (r = .05, n. s.). The average correlation of strangers' and spouses' ratings for all targets across each of the five scales was r = .18.

Correlations of Self and Stranger Ratings (See Table 4)

The correlations between self and strangers' ratings for all 50 targets was not significant for any scale. The average correlation of self and strangers' ratings for all targets across each of the five scales was r = .08. The correlations between self and stranger ratings of personality were also mostly insignificant; considering how many correlations were measured, such a number can be expected to be significant solely by chance. These are much lower correlations than reported in previous experiments on the accuracy of facial impressions (Berry, 1990, 1991).

Self ratings of older targets correlated more closely with strangers' ratings than did self ratings of younger targets (r = .18 vs. r = -.06, n. s.). Although the trends are congruent with the hypothesis, due to the low correlations and a limited sample size, this difference was not significant at the five percent level.

Conclusion

The results of this study should be viewed with extreme caution for several reasons. The subjects were not representative of the general population. All subjects volunteered to participate in this experiment, and all strangers were students in psychology courses at one college. Taking the average of strangers' ratings has its limits, as the low reliability scores indicate. it is possible that some strangers were more accurate than others.

This experiment yielded limited support for the hypothesis, so the repetition of this procedure with larger groups seems necessary. Further research could assess individual differences of the strangers in relation to their personality ratings. One could also investigate the effects of specific facial characteristics on strangers' ratings.

References

Albright, L., Kenny, D. A., & Malloy, T. R. (1988). Consensus in personality judgments at zero acquaintance. Journal of Personality and Social Psychology, 55, 387-395.

Berry, D. S. (1990). Taking people at face value: Evidence for the kernel of truth hypothesis. Social Cognition, 8, 343-361.

Berry, D. S. (1991). Accuracy in social perception: Contributions of facial and vocal information. Journal of Personality and Social Psychology, 61, 298­307.

Berry, D. S., & Bronlow, S. (1989). Were the physiognomists right? Personality correlates of facial babishness. Personality and Social Psychology Bulletin, 15, 266-279.

Bond, C. F., Jr., Berry, D. S., & Omar, A. (1992). The 'kernel of truth' in judgments of deception. Manuscript submitted for publication, Texas Christian University and Southern Methodist University.

Borkenau, P., & Liebler, A. (1992). Trait inferences: Sources of validity at zero acquaintance. Journal of Personality and Social Psychology, 62, 645­657.

Dion, K. L., & Dion, K. K. (1987). Belief in a just world and physical attractiveness stereotyping. Journal of Personality and Social Psychology, 52, 775-781.

Funder, D. C. (1987). Errors and mistakes: Evaluating the accuracy of social judgment. Psychological Bulletin, 101, 75-90.

Malatesta, C. Z., Fiore, M. J., & Messina, J. J. (1987). Affect, personality, and facial expressive characteristics of older people. Psychology and Aging, 2, 64-69.

Mazur, A., Mazur, J., & Keating, C. (1984). Military rank attainment of a West Point class: Effects of cadets' physical features. American Journal of Sociology, 90, 125-150.

McCrae, R. R., & John, O. P. (1992). An introduction to the five-factor model and its applications. Journal of Personality, 60, 175-215.

Watson, D. (1989). Strangers' ratings of the five robust personality factors: Evidence of a surprising convergence with self­report. Journal of Personality and Social Psychology, 57, 120-128.

Zebrowitz, L. (1990). Social perception. Pacific Grove, CA: Brooks/Cole.

This report was written by John Hubbard for the Macalester College course "PSYCH 96: Independent Project" in December 1994. It is an abridged version of the full report and has not been significantly altered from the original version; the last modified date shown below indicates when this Webpage was last uploaded in its present form.

Created, maintained and © by John Hubbard (write to me). Disclaimers. Hosted by Dreamhost. Last modified: July-24-2003.