Revision of Manuscript # 17-02-88
Running head: ATTRACTIVENESS AND SEXUAL HARASSMENT

Effects of Physical Attractiveness When Evaluating a Male Employee’s Allegation of Sexual Harassment by His Female Employer

Karl L. Wuensch and Charles H. Moore

East Carolina University

Abstract
College students (N = 324) served as mock jurors in a simulated civil case in which a male plaintiff accused a female defendant of sexual harassment. The physical attractiveness of the litigants was experimentally manipulated. Mock jurors were asked to decide whether the defendant was guilty or not and to rate their certainty of belief in the defendant's guilt. Jurors were more certain of the guilt of the defendant when the plaintiff was attractive than when he was unattractive. Plaintiff attractiveness significantly affected female jurors’ verdicts when the defendant was unattractive, but not when she was attractive. With male jurors, plaintiff attractiveness significantly affected their verdicts when the defendant was attractive, but not when she was unattractive. Female jurors were more likely than male jurors to conclude that sexual harassment did take place, but only when the litigants were different in attractiveness.
Effects of Physical Attractiveness When Evaluating a Male Employee’s Allegation of Sexual Harassment by his Female Employer Physically attractive people are perceived in a more positive fashion than are the physically unattractive (Dion, Berscheid, & Walster, 1972). This stereotype has been shown to lead to more lenient sentencing of attractive defendants than of unattractive defendants, in both simulated and actual trials (Mazella & Feingold, 1994; Stewart, 1980). In simulated rape trials, attractive defendants are sentenced more leniently than unattractive defendants, and defendants accused of raping an unattractive victim are less likely to be seen as guilty than are those accused of raping an attractive victim (Jacobson, 1981; Jacobson & Popovich, 1983). In a simulated sexual harassment case, physically attractive litigants were more favorably treated than were unattractive litigants (Castellow, Wuensch, & Moore, 1990). Odds ratios are a convenient way to express the magnitude of effects in research like this. Unattractive defendants were 2.5 times more likely to be found guilty than were attractive defendants. Likewise, guilty verdicts were 2.7 times more likely when the plaintiff was attractive than when she was unattractive. Why do jurors consider sexual harassment more likely when the defendant is unattractive and/or the plaintiff is attractive? Mock jurors do seem to assume that physically attractive persons also have other socially desirable characteristics, rating them higher than physically unattractive persons on attributes such as sincerity, kindness, and intelligence (Castellow et al., 1990). It may be these inferences about other characteristics that make jurors more likely to believe the physically attractive litigant. When mock jurors in a sexual harassment case were given character witness testimony designed to manipulate the perception of the social desirability of the litigants, social desirability significantly affected jurors’ decisions and the effect of physical attractiveness fell short of statistical significance (Moore, Wuensch, Hedges, & Castellow, 1994). When the association between litigant physical attractiveness and the verdict (Castellow et al., 1990) was presented to students at the university where this research took place, and the students were asked to speculate on why jurors most favored the attractive plaintiff, the answers most frequently given were: Attractive men do not need to harass a woman sexually (they can get all the sex they need without resorting to harassment), sexual advances by attractive men are unlikely to be taken as harassment, men will not be motivated to harass sexually an unattractive woman, and an unattractive woman is unlikely to take a sexual advance as being harassment. Clearly these students thought that sexual harassment is motivated, at least in part, by sexual desire. Of course, the actual motivation may well be similar to that in rape, which may be sexual/reproductive (Thornhill & Palmer, 2000) or may be related to hostility and domination (see Ellis, 1989). If jurors are like these students, and assume that sexual harassment is sexually motivated, then one would expect the physical attractiveness of the litigants to have an effect on the jurors’ decisions. Earlier studies have found that female jurors are more likely than male jurors to find a male defendant guilty of sexual harassment of a female plaintiff (Egbert, Moore, Wuensch, & Castellow, 1992; Moore et al., 1994; Wuensch, Campbell, Kesler, & Moore, 2002). This may have resulted from greater identification with the same gender litigant and/or from greater familiarity with the offense (sexual harassment). Calhoun and Townsley (1991, p. 66) noted that women usually show more empathy towards victims of acquaintance rape than do men, possibly because women are more likely to have been victimized themselves. From another perspective, one might argue that men are less likely than women to believe an allegation of sexual harassment if men have been falsely accused of sexual harassment or have worried about being falsely accused. The purpose of the present research was to investigate the effects of gender of the juror and physical attractiveness of the litigants in a simulated civil case where sexual harassment of a male employee by his female employer was alleged. All previously published research involving simulated litigation over alleged sexual harassment has involved male defendants and female plaintiffs. We had, a priori, no theoretical or other reason to expect that the results of our physical attractiveness manipulations would be any different than they were in the earlier research involving male defendants and female plaintiffs. That is, we expected that guilty verdicts would be more likely when the defendant was unattractive and when the plaintiff was attractive. We were unable to feel confident about an expected direction of effect with respect to juror’s gender. If the gender differences observed in earlier studies (female jurors being more likely to find male defendants guilty) resulted from a bias in favor of same-gender litigants, then one would expect that women would be less likely than men to find a female defendant guilty of harassing a male plaintiff. On the other hand, if the gender differences observed earlier resulted from women being more likely than men to have experienced sexual harassment themselves, then one might expect women to be more likely than men to find in favor of the plaintiff, even when the defendant is female and the plaintiff male. We hoped that the current research would provide results that discriminated between these two interpretations of the past research.

Method

Participants Participants were 324 college students (160 female, 164 male) who were enrolled in undergraduate psychology classes at a southern, public university whose student body consists mostly of white students. The mean age of participants was 19.5 (SD = 2.88). The majority of participants were Caucasian (78%). Other ethnic groups represented included African American (19%), native American (2%), Asian American (1%), and Hispanic (< 0.5%). Participants did grant informed consent, were offered extra credit in their Psychology class as compensation for their time, and were informed of the results of the study after its completion. It was made clear to participants that the case materials were fictional.
Design
Three independent variables were used in a 2 x 2 x 2 design: Gender of participant, physical attractiveness of the plaintiff, and physical attractiveness of the defendant. Verdict and certainty of guilt were used as dependent variables. All statistical tests employed nondirectional hypotheses and a .05 criterion of significance.

Physical Attractiveness Manipulation

The photographs of the litigants were the same as those used in earlier research in our lab (Castellow et al., 1990; Moore et al., 1994). As part of our initial study (Castellow et al., 1990), we showed 120 photographs of men and women to 149 college students and asked them to rate each photo on a 9-point scale of attractiveness. The photographs were taken from college yearbooks and popular magazines and were black and white portrait-style. The most attractive (M = 7.7) and least attractive (M = 2.6) male photographs were selected for use in the study. Likewise, the most attractive (M = 7.6) and least attractive (M = 1.9) female photographs were selected.
Experimental Materials and Procedure Each participant was provided with a packet containing the following items: Instructions, case background, summaries of the litigants’ testimony, summaries of the litigants’ character witness testimony, a photograph of each litigant, a trial summary, and a participant response form. These materials (available in electronic format upon request from the first author) were nearly identical to those used by Moore et al. (1994), except for changes involved in reversing the genders of the litigants. In the plaintiff’s testimony, it was explained that in the three years that he had been working for the defendant as an assistant manager of her apartment complex, her harassing behaviors had escalated. Initially she made inappropriate and unwanted remarks about his clothing and physique. She stared at his crotch and made remarks about the size of his genitals. She tried to touch and caress his body. When he objected to these behaviors, she told him that he must like it or he would have found other employment. He alleged that in recent months she had tried to put her hand under his clothing, had chased him around the office trying to kiss him, and repeatedly tried to fondle his buttocks and genitals. She even went so far as to describe to him the sexual acts in which she desired to engage him. Since few people would argue that the alleged behaviors do not constitute sexual harassment, we assumed that any juror who believed the plaintiff’s testimony would find in his favor. In the defendant’s testimony, all of the plaintiff’s accusations were denied. The defendant stated that she considered herself to be a friend of the plaintiff, and that she had a genuine concern for his safety and well being. She expressed a willingness to help him find other employment or continue his employment at her firm. The character witness testimonies were altered from those used by Moore et al. (1994). In that earlier study, the character witness testimonies portrayed each litigant as either socially desirable or socially undesirable. The socially desirable litigants were described as friendly, helpful, pleasant, professional, honest, active in civic organizations, high achieving in school, and well dressed. Socially undesirable litigants were described in opposite terms: Grumpy, unhelpful, unpleasant, unprofessional, sloppy in appearance, and so on. Participants in a pilot study rated subjects on a 18-item adjective scale. Responses for each item ranged from 1 (strongly endorsing the undesirable adjective, such as “cold”) to 9 (strongly endorsing the desirable adjective, such as “warm”). The mean across items and respondents was 7.8 for the socially desirable description and 3.2 for the socially undesirable description. In the current study the character witness testimony attempted to portray each litigant as moderate in social desirability. They were described as being somewhat reserved and only occasionally being helpful and pleasant. They were said to have C averages in their evening classes. Sometimes their performance on the job was acceptable, sometimes not. They were involved in civic organizations, but not intensely. Their dress was described as neither neat nor disorganized. A pilot study employing the same adjective scale used in the study by Moore et al. (1994) produced means of 5.1 for the plaintiff and 5.5 for the defendant. The participant response form asked the participants to list their age, gender, race, and academic classification. Participants were asked whether they thought the defendant was guilty of sexual harassment (“yes” or “no”), and were asked to indicate how certain they were of her guilt, on a 9-point scale from “definitely not guilty” to “definitely guilty.” After responding to these two questions, respondents were also asked to rate, on a 9-point scale, the physical attractiveness of each litigant. Participants were run in small groups, ranging in size from 10 to 40.

Results

The Physical Attractiveness Manipulation

For the defendant, the attractive photograph was rated significantly more attractive (M = 7.86, SD = 1.29) than the unattractive photograph (M = 2.90, SD = 1.74), t(322) = 29.15, p = .000, d = 3.24. For the plaintiff, the attractive photograph was rated significantly more attractive (M = 6.72, SD = 1.78) than the unattractive photograph (M = 2.09, SD = 1.25), t(322) = 27.19, p = .000, d = 3.02. Within-gender comparisons showed that the physical attractiveness manipulations remained highly effective (t > 14 and p = .000 for each litigant).

The Verdict

A logit analysis was employed to evaluate the effects of gender of juror and physical attractiveness of plaintiff and defendant upon the verdict rendered. A reduced model was obtained by removing from the saturated model all effects which were not significant at the .10 level. All effects deleted were so trivial in magnitude that each had a z < .65 and a p > .50. The resulting model included a significant main effect of plaintiff attractiveness, z = 2.65, p = .008, and a significant interaction between juror’s gender, defendant attractiveness, and plaintiff attractiveness, z = 2.19, p = .029. The effect of juror’s gender fell short of the traditional standard of statistical significance, z = 1.68, p = .093. The reduced model fit the data adequately, as indicated by a nonsignificant goodness of fit test, G 2(4, N = 324) = 1.08, p = .897. When the main effect of gender was dropped from the model, there was a noticeable, but not statistically significant, reduction in the model’s goodness of fit, G2(5, N = 324) = 3.94, p = .558. Each of the effects in the reduced model was also evaluated with marginal likelihood ratio chi-square tests, as detailed below. Guilty verdicts were significantly more frequent (77%) when the plaintiff was attractive than when he was unattractive (63%), G2(1, N = 324) = 7.05, p = .008, odds ratio = 1.92. This effect was modified by its participation in the significant interaction. Female jurors returned guilty verdicts more frequently (74%) than did male jurors (66%), but this difference fell short of statistical significance by the marginal test, G2(1, N = 161) = 2.81, p = .094, odds ratio = 1.50. Given the significant main effect of plaintiff attractiveness, we first investigated the interaction effect by testing the effect of plaintiff attractiveness for each combination of gender and defendant attractiveness (see the odds ratios in Figure 1). When the defendant was unattractive, female jurors were significantly more likely to render a guilty verdict when the plaintiff was attractive (85%) than when he was unattractive (58%), G2(1, N = 80) = 7.62, p = .006, odds ratio = 4.19, but when the defendant was attractive, the physical attractiveness of the plaintiff was without effect (78% guilty verdicts for both levels of plaintiff attractiveness), G2(1, N = 80) = 0.00, p = 1.000, odds ratio =1.00. With the male jurors, the effect of plaintiff attractiveness fell short of statistical significance when the defendant was unattractive (68% guilty verdicts with the attractive plaintiff, 63% with the unattractive), G2(1, N = 83) = 0.20, p = .653, odds ratio = 1.23. When the defendant was attractive, male jurors were significantly more likely to find her guilty when the plaintiff was attractive (78%) than when he was unattractive (56%), G2(1, N = 81) = 4.24, p = .040, odds ratio = 2.70. Because of our a priori interest in the effect of gender of juror, we also investigated the interaction by testing the effect of gender for each combination of plaintiff attractiveness and defendant attractiveness (see the odds ratios in Figure 2). When the defendant was attractive and the plaintiff was not, female jurors were significantly more likely to render a guilty verdict (78%) than were male jurors (56%), G2(1, N = 81) = 4.24, p = .040, odds ratio = 2.70. When the plaintiff was attractive and the defendant was not, female jurors were more likely to render a guilty verdict (85%) than were male jurors (68%), but this difference fell just short of statistical significance, G2(1, N = 80) = 3.45, p = .063, odds ratio = 2.73. When both litigants were unattractive, the gender difference fell well short of statistical significance (58% guilty verdicts from female jurors, 63% from male jurors), G2(1, N = 83) = 0.24, p = .623, odds ratio = 0.80. When both litigants were attractive, there was absolutely no difference between female and male jurors (78% guilty verdicts for both genders), G2(1, N = 80) = 0.00, p = 1.000, odds ratio = 1.00.

Certainty of Guilt

A three way factorial ANOVA was employed to evaluate the effects of gender of juror and physical attractiveness of plaintiff and defendant upon certainty of guilt. The only significant effect was that of plaintiff attractiveness, with certainty of guilt being significantly greater when the plaintiff was attractive (M = 6.21, SD = 1.81) than when he was unattractive (M = 5.52, SD = 2.17), F(1, 316) = 9.26, MSE = 4.05, p = .000, d = .34. None of the remaining effects even approached significance, F < 1.00 in every case. The means presented in Table 1 provide only the slightest hint of the sort of interaction found with the verdict variable.
Discussion
The most prominent effect was that of physical attractiveness of the plaintiff. When the male plaintiff was attractive, jurors were nearly twice as likely to find in favor of the plaintiff and were somewhat more certain (d = .34) of the guilt of the female defendant. These results mirror those of found earlier with a female plaintiff and a male defendant (Castellow et al., 1990). Jurors found it difficult to believe that a physically unattractive man would be sexually harassed by his female employer. The effect of plaintiff attractiveness on the recommended verdict was modified by an unanticipated interaction with gender of juror and attractiveness of the defendant. In the present research female jurors favored the physically attractive male plaintiff only when the female defendant was unattractive. Perhaps female jurors are more likely to think of the harassment as being sexual (making physical attractiveness of the male plaintiff relevant) if the female defendant is physically unattractive. The unattractive female defendant may be thought of as being sexually frustrated, in which case it is easier to believe that she would sexually harass an attractive (sexy) man than an unattractive man.
Female jurors may think of harassment in terms of other motives when the female defendant is physically attractive, assuming that a physically attractive woman can get all the sexual attention she needs without resorting to harassment If the motive behind the harassment is not sexual, then there is no strong reason to believe that an attractive man would be more likely targeted then an unattractive man. In earlier research (Castellow et al., 1990), female jurors favored the physically attractive female plaintiff who had accused a male defendant of sexual harassment regardless of the attractiveness of the defendant. Female jurors may think of the harassment as being sexually motivated in all cases where the defendant is male. In the present research male jurors favored the attractive male plaintiff only when the female defendant was physically attractive. As noted by Etcoff (1999), men are much more likely than women to infer that attractive women are high in sex drive and seek variety in their sexual lives. Accordingly, male jurors may think a physically attractive woman more likely to be sexually motivated than an unattractive woman. Having inferred that the attractive defendant’s motivation was sexual, it would be easier to believe that that she would sexually harass an attractive (sexy) man than an unattractive man. If male jurors are less likely to think of harassment in sexual terms when the defendant is not attractive, then there is less reason to expect them to believe that an attractive man would be more likely targeted than an unattractive man. In earlier research (Castellow et al., 1990), male jurors favored the physically attractive female plaintiff who had accused a male defendant regardless of the attractiveness of the defendant, suggesting that male jurors think of harassment as being sexually motivated in all cases where the defendant is male. Our admittedly speculative interpretation of the unanticipated interaction involves gender differences with respect to beliefs about the motives of those who sexually harass and the effect of situational characteristics (attractiveness of the parties) on such attributions of motivation. Future research could test these speculations by asking participants about the motivations of the harasser. While the results of the current study indicate that there will not always be a statistically significant effect of a litigant’s physical attractiveness, we do not feel that these results warrant modification of the recommendation that litigants should do their best to appear physically attractive in the courthouse. Our results do suggest that the physical appearance of litigants can influence the judgments of jurors. Attorneys or judges who wish to prevent this effect should consider bringing it to the attention of the jurors and suggesting to them that physical attractiveness is an extra-legal dimension that should be ignored. In previously reported research, the person accused of sexual harassment has always been male and the putative victim always female. In this circumstance, female jurors have been more likely to find in favor of the female litigant than are male jurors. This result may be due to jurors tending to identify with litigants of their own gender (as suggested by Popovich et al., 1996), or it may be due to women being more likely to identify with the victim of sexual harassment because women are more likely to have experienced such harassment than are men. The results of the current research are not entirely consistent with either of these interpretations. Our second investigation of the interaction revealed that there was little or no difference between female and male jurors when the litigants were of similar attractiveness, but that female jurors were more likely than male jurors to render guilty verdicts when one litigant was attractive and the other was not. Perhaps women are simply more likely than men to think that the alleged behaviors were consensual or were invited by the plaintiff when the litigants were similar in physical attractiveness. From a biological perspective, this could reflect greater female motivation for positive assortative mating (attraction to others similar to oneself), as might be expected given that women have more invested in each reproduction than do men (see Etcoff, 1999). Perhaps both types of identification (with another member of one’s own gender and with another victim of sexual harassment) are important determinants of gender differences in decisions regarding allegations of sexual harassment. The hypothesis that women are usually more likely to find in favor of female plaintiffs in sexual harassment trials because they are themselves more likely to have experienced sexual harassment themselves could be tested by including in future research a measurement of the extent to which jurors have themselves been sexually harassed. Under the proposed hypothesis one would expect that jurors who have experienced sexual harassment themselves would be more likely to find in favor of the plaintiff whether the plaintiff were female or male. The limitations of the present study are those common to experimental research on jury decision making. Because the participants used in the present study were college students, although eligible for actual jury selection, one might wonder whether their decision making differs from that of the population of persons that typically serve on juries. Because the case is simulated, one might also worry that the mock jurors may not care about the outcome as much as they would if it were an actual case. There is, however, evidence that mock jurors show a nearly equal level of involvement in simulated cases as do actual jurors in real cases (Hastie, Penrod, & Pennington, 1983; Kalven & Zeisel, 1966). Another potential limitation of the external validity of the current study is that it focused on predeliberation verdicts. Might the effects of the predictor variables differed following group delibertation? Again, past research has shown that a high correlation exists between pre- and post-deliberation verdicts (Hastie et al., 1983; Kalven & Zeisel, 1966). It is not impossible to reduce these threats to the external validity of experimental research on jury decision making, but doing so requires resources not commonly available to social psychological researchers. Recent research by Conley, Turnier, and Rose (2000) is an excellent example of experimental research on jury decision making with special attention to reducing problems of external validity. One of the steps that these researchers took to increase external validity was to recruit participants from persons called for jury duty in two North Carolina counties and then to conduct the research in the courthouse. Of course, this required the cooperation of the superior court judges in those counties. Few social psychologists have access to the level of support enjoyed by Conley et al., but this is not to say that the typical study on jury decision making is without value.
References
Calhoun, K. S., & Townsley, R. M. (1991). Attributions of responsibility for acquaintance rape. In A. Parrot, & L. Bechhofer (Eds.), Acquaintance rape: The hidden crime (pp. 57 - 69). New York: Wiley.
Castellow, W. A., Wuensch, K. L., & Moore, C. H. (1990). Effects of physical attractiveness of plaintiff and defendant in sexual harassment judgments. Journal of Social Behavior and Personality, 16, 39 - 50.
Conley, J. M., Turnier, W. J., & Rose, M. R. (2000). The racial ecology of the courtroom: An experimental study of juror response to the race of criminal defendants. Wisconsin Law Review, 2000, 1185 - 1220.
Dion, K. K., Berscheid, E. & Walster E. (1972). What is beautiful is good. Journal of Personality and Social Psychology, 24, 285 - 290.
Egbert, J. M., Moore, C. H., Wuensch, K. L., & Castellow, W. A. (1992). The effect of litigant social desirability on judgments regarding a sexual harassment case. Journal of Social Behavior and Personality, 7, 569 - 579.
Ellis, L. (1989). Theories of rape: Inquiries into the causes of sexual aggression. New York: Hemisphere.
Etcoff, N. (1999). Survival of the prettiest: The science of beauty. New York: Doubleday.
Hastie, R., Penrod, S.D., & Pennington, N. (1983). Inside the jury. Cambridge, MA: Harvard University Press.
Jacobson, M. B. (1981). Effects of victim's and defendant's physical attractiveness on subjects' judgments in a rape case. Sex Roles, 7, 247 - 255.
Jacobson, M. B., & Popovich, P. M. (1983). Victim attractiveness and perceptions of responsibility in an ambiguous rape case. Psychology of Women Quarterly, 8, 100 - 104.
Kalven, H. Jr., & Zeisel, H. (1966). The American jury. Boston: Little, Brown.
Mazella, R., & Feingold, A. (1994). The effects of physical attractiveness, race, socioeconomic status, and gender of defendants and victims on judgments of mock jurors: A meta-analysis. Journal of Applied Social Psychology, 24, 1315 - 1344.
Moore, C. H., Wuensch, K. L., Hedges, R. M., & Castellow, W. A. (1994). The effects of physical attractiveness and social desirability on judgments regarding a sexual harassment case. Journal of Social Behavior and Personality, 9, 715 - 730.
Popovich , P. M., Gehlauf, D. N., Jolton, J. A., Everton, W. J., Godinho, R. M., Mastrangelo, P. M., & Somers, J. M. (1996). Physical attractiveness and sexual harassment: Does every picture tell a story or every story draw a picture? Journal of Applied Social Psychology, 26, 520 - 542.
Stewart. J. E. II (1980). Defendant’s attractiveness as a factor in the outcome of criminal trials: An observational study. Journal of Applied Social Psychology, 10, 348 - 361.
Thornhill, R., & Palmer, C. T. (2000). A natural history of rape: Biological bases of sexual coercion. Cambridge, MA: MIT Press.
Wuensch, K. L., Campbell, M. W., Kesler, F. C., & Moore, C. H. (2002). Racial bias in decisions made by mock jurors evaluating a case of sexual harassment. Journal of Social Psychology, 142, 587 - 600.
Author Note Karl L. Wuensch, Department of Psychology; Charles H. Moore, Department of Psychology. Correspondence concerning this article should be addressed to Karl L. Wuensch, Department of Psychology, East Carolina University, Greenville, NC 27858-4353. Electronic mail may be sent via the Internet to WuenschK@ECU.edu.
Table 1
Mean Certainty of Guilt by Gender of Juror and Attractiveness of Litigants

| | | |Attractiveness of Plaintiff |
|Gender of Juror |Attractiveness of Defendant | |Attractive |Unattractive |
|Female |Attractive | |6.28 |5.72 |
|Female |Unattractive | |6.38 |5.52 |
|Male |Attractive | |6.18 |5.32 |
|Male |Unattractive | |6.00 |5.53 |

[pic] Figure 1. Odds ratios expressing the advantage given to attractive plaintiffs.
[pic]
Figure 2. Odds ratios illustrating female jurors being more likely than male jurors to find the defendant guilty.