Beliefs about Men’s and Women’s Jealousy Responses

To examine what people believe would make women jealous, we applied a 2 (participant sex) × 2 (sexual orientation) ANCOVA with age as covariate to control for age differences across the four groups.Footnote 3 The model was significant, F(4,1198) = 9.99, and accounted for 2.9% of the variance in beliefs. As evident from Fig. 1, the effect of sex was significant, F(1,1198) = 15.08, p < 0.001, with men scoring slightly higher than women (d = 0.22). Further, the effect of sexual orientation was significant, F(1,1198) = 5.85, p = 0.016, d = 0.14 (minority higher). The sex × sexual orientation interaction effect was not significant, F(1,1198) < 1, and the beliefs were not significantly affected by age, F(1,1198) = 3.10, p = 0.079, r = 0.067.

Fig. 1

People’s beliefs about women’s responses to their partner’s infidelity. Darker bars = heterosexual respondents. A score of 0.5 indicates equally upset by the sexual and the emotional aspect of the infidelity. Lower scores indicate being relatively more upset by the emotional aspect

When we applied the same analysis to beliefs about men’s jealousy responses, the model was not significant, F(4,1199) < 1. Neither participant sex, sexual orientation, their interaction, nor age reached significance. Figure 2 illustrates the lack of group differences.

Fig. 2

People’s beliefs about men’s responses to their partner’s infidelity. Darker bars = heterosexual respondents. A score of 0.5 indicates equally upset by the sexual and the emotional aspect of the infidelity. Lower scores indicate being relatively more upset by the emotional aspect

When we performed pairwise comparisons (t-tests) of the participants’ beliefs about men vs. women’s jealousy reactions, we found that heterosexual women (d = 1.06), heterosexual men (d = 0.78), and sexual minority women (d = 0.89) all believed that men markedly more than women would be upset by the sexual aspect of their partner’s infidelity. In contrast, sexual minority men (n = 30) differentiated less strongly between what aspect of the infidelity that would make men and women jealous (d = 0.38). Notably, the participants’ beliefs about jealousy reactions in men and women showed small correlations, r = 0.13 (rwomen = 0.10; rmen = 0.16).

Own Jealousy Responses

Looking next at the participant’s own reports, the model F(4,1200) = 40.94, accounted for 11.7% of the variance in jealousy responses. The effect of participant sex was significant, F(1,1200) = 25.87, p < 0.001, with men being relatively more upset by the sexual aspect than women. There was also a main effect of sexual orientation: F(1,1200) = 6.35, p = 0.012, suggesting that heterosexuals scored higher than sexual minority persons, and a sex × sexual orientation interaction effect, F(1,1200) = 6.71, p = 0.010. The latter suggests that heterosexual men scored higher than any of the other groups (see Fig. 3). A t-test showed that the sex difference in heterosexual participants was large, t(1054) = 12.44, p < 0.001, d = 0.83 (men, MM = 0.469; women, MM = 0.200). In contrast, the sex difference among sexual minority participants was small and nonsignificant, t(147) = 1.35, p = 0.18, d = 0.28 (men, MM = 0.290; women, MM = 0.203). There was no effect of age on own jealousy responses.

Fig. 3

Women and men’s own jealousy responses. Dark gray = heterosexual respondents

How well do beliefs about other men’s and women’s jealousy reactions concur to men’s and women’s own jealousy responses? Are their beliefs accurate at the group level or are they biased or stereotypical (accentuated) relative to their own responses? To examine this, we ran paired-samples t-tests for heterosexual men and women and for sexual minority men and women separately. Heterosexual men’s beliefs about other men were moderately accentuated in the male stereotypical direction, t(329) = 8.34, p < 0.001, d = 0.45. Also, sexual minority men’s beliefs were accentuated in the male typical direction, t(29) = 3.43, p = 0.002, d = 0.62. Heterosexual women reported beliefs about other women being slightly more upset by the emotional aspect of the infidelity than themselves, i.e., accentuated in the female stereotypical typical direction, t(725) = − 4.38, p < 0.001, d = − 0.16, while sexual minority women’s beliefs about other women did not differ from their own jealousy reactions, t(118) = 0.05.

Perceptions of Infidelity Cues

Men and women highly agreed on what types of behavior would serve as the most distinct cues to infidelity (see Table 1). The participants were most sensitive to behaviors involving having a sexual relationship with somebody, kissing, creating a dating profile, staying overnight with somebody else, saying that one wants to be in another relationship or that one loves somebody else, and having dinner with an ex. More women than men found several of these to “definite” cues to infidelity. Behaviors that very few (or nobody) found to be clearly indicative of infidelity included watching online porn, helping somebody of the opposite sex with renovation of their house, suggesting novel sexual activities, and paying extensive attention to somebody of the opposite sex. Across all 23 possible cues to infidelity, women reported slightly higher scale scores than men (d = − 0.22), which suggest women have a somewhat lower threshold for perceiving these acts as infidelity than men. Relative to the difference among heterosexual men and women (d = − 0.25), sexual minority participants did not report any sex difference in their perception of cues to infidelity (d = − 0.03).

Knowledge from Social Influences and Personal Experiences

Of the possible sources of social influence, family of origin and education were the least likely sources of knowledge of what may be (1) cues to infidelity on the one hand and (2) typical reactions to infidelity on the other with mean scores between 1.65 and 2.08 (on a scale from 1 (not at all) to 4 (much)). The most likely source was peers (Mcues = 3.01 and Mreactions = 3.00). Knowledge from personal experience (Mcues = 2.48 and Mreactions = 2.40) and from various media (Mcues = 2.64 and Mreactions = 2.66) was more often reported as a source than education.

Predictors of Beliefs About Men’s and Women’s Jealousy Responses

Before predicting the respondents’ beliefs about what makes men and women more jealous, we examined separately for men and women all possible and relevant factors that could be associated with such beliefs. These included lifetime number of committed relationships, age, perceptions of infidelity cues, peer influences, media influences, personal experiences including being cheated on by current or former partner, and finally own jealousy responses. We also examined the bivariate associations among all possible predictors (Tables 2 and 3). Across all sources of knowledge, only the media items (“cues” and “typical reactions”) showed any significant associations with jealousy beliefs, and only so for female respondents’ beliefs about women’s reactions. As shown in Tables 2 and 3, equivalent items measuring knowledge of “cues” and “typical reactions” were largely overlapping. As could be expected, having experienced being cheated upon was clearly associated with number of life-time committed relationships and with both items on knowledge from “personal experience” for both men and women. However, having experienced being cheated upon was not associated with own jealousy responses. Further, knowledge of cues and typical reactions from personal experience did not show any correlations with the other knowledge indicators, except for peers (r’s ranging from 0.06 to 0.23), a result that probably reflects some sharing of personal experiences with friends.

Furthermore, perceptions of cues to infidelity showed inconsistent associations with sources to knowledge of cues to infidelity for both men and women. For men, the media scale on cues correlated r = 0.15 with perceptions of cues to infidelity (r = 0.05 for women). Knowledge from own experiences and having been cheated upon showed some association with perception of infidelity cues for women, but not for men. The lack of strength and regularity of these associations suggests that knowledge from external sources and personal experiences has very little bearing on people’s perception of infidelity cues. Number of committed lifetime relationships was moderately associated with having been cheated upon and stronger so than age, but the association with own jealousy reaction and beliefs about other’s jealousy reactions was absent except for women’s beliefs about men’s reactions (r = 0.08).

For men, own jealousy responses correlated strongly with beliefs about other men’s jealousy responses, r = 0.67. For women, the association between own and other women’s responses was r = 0.43. In comparison, the associations between own responses and beliefs about opposite-sex responses were markedly smaller (r = 0.21 for men and r = 0.18 for women).

Informed by the bivariate associations above, we ran a path model with sex as a grouping variable to predict two belief outcomes: same-sex and opposite-sex (see Fig. 4). We included only age, perceptions of infidelity cues, media knowledge of typical reactions to infidelity, and being cheated on as primary predictors in the model. We added own jealousy response as a secondary predictor to the model. Parameter invariance (Wald) tests were performed to compare if men’s and women’s path coefficients differed significantly.

Fig. 4

Path model for predicting beliefs about jealousy reactions in same- and opposite-sex persons. Coefficients for each path are presented as women/men. Solid bold lines indicate significant effects for both men and women. Dashed lines indicate effects for one sex only. Dotted lines indicate no effect. *p < .05; **p < .01, ***p < .001

Further, women, but not men, who experienced cheating from their partner were slightly more prone to identify cues to infidelity. However, participants who had been cheated upon reported somewhat lower levels of knowledge of typical reactions from the media.

As would be expected from the bivariate correlations, the paths running from perception of infidelity cues and knowledge of reaction from media and cheating experiences to own jealousy reactions on the one hand and to beliefs about same- and opposite-sex reactions were very weak. Basically, the only paths of significance ran from own jealousy reactions to same- and opposite-sex beliefs for both men and women. The parameter invariance (Wald) test strongly suggests that the association between own jealousy and same-sex jealousy belief was significantly stronger for men than for women (χ2 = 25.59, p < 0.001). Importantly, when accounting for the effect of own jealousy responses, there was no association between the beliefs the participants held toward same-sex and opposite-sex jealousy reactions, suggesting that these beliefs were highly independent constructs.