Social desirability, sympathy, and attribution ratings were submitted to two-way (Target under Evaluation X Direction of Comparison) analyses of variance (ANOVAs). Means and standard deviations in all conditions are displayed in Table 4.

The ANOVA revealed a marginally significant main effect of target under evaluation, F(1, 718) = 3.59, p < .06, and a significant main effect of direction of comparison, F(1, 718) = 37.85, p < .001, upon social desirability. Participants considered verbal statements about the self as somewhat more undesirable (M = 2.64) than statements about the partner (M = 2.55), and public evaluations in an enhancing manner were seen as less undesirable (M = 2.79) than deprecating evaluations (M = 2.39). As expected, however, these main effects were qualified by a significant two-way interaction, F(1, 718) = 162.05, p < .001. This reliable effect reflects the fact that enhancing statements about the partner were rated as less socially undesirable than deprecating statements, whereas the reverse appeared when the statements concerned statements about the self (see Table 4).

Next, the effects of target and direction of comparison on sympathy ratings were examined in a two-way ANOVA. Participants reported more sympathy for a discussant who publicly evaluated his or her intimate relationship in an enhancing fashion (M = 2.33) than with a discussant who made deprecating evaluations (M = 2.17), F(1, 718) = 6.29, p < .02. Target under evaluation did not yield a significant effect, F < 1, ns, but interacted significantly with direction of comparison, F(1, 718) = 175.41, p < .001. As shown in Table 4, partner-enhancing statements evoked more feelings of sympathy than partner-deprecating statements, whereas self-enhancement elicited less sympathy than self-deprecation.

The ANOVA's performed on the attribution ratings yielded, not surprisingly, a significant main effect of direction of comparison on self-enhancement attributions, F(1, 718) = 85.68, p < .001. The means showed that relative to the deprecating evaluation condition, enhancing evaluations were attributed more strongly to a desire of self-enhancement (Ms = 2.79 vs. 3.30). Finally, the predicted interaction of target under evaluation and direction of comparison was significant only for attributions to self-enhancement, F(1, 718) = 39.55, p < .001. As revealed in Table 4, self-enhancement was more strongly attributed to a motive to enhance the self than partner-enhancement, whereas the reverse was true in the deprecation conditions. No other main effects or interaction effects approached significance.

One of the reasons that people may consider self-enhancing evaluations and partner-deprecating evaluations as more socially undesirable than partner-enhancing and self-deprecating evaluations, respectively, is that they felt an antipathy to the discussant in the former conditions. Put in other words, participants's ratings of sympathy for the discussant could conceivably explain why people consider both former evaluations as more socially undesirable than the both latter evaluations. Also, the reverse may be true, namely that participants's feelings of sympathy could have been influenced by their social desirability ratings. Social desirability, in fact, was highly correlated with the sympathy ratings, r(722) = .72, p < .001.

To determine whether participants' social desirability ratings did account for the effects on sympathy, a 2 (Target under Evaluation) by 2 (Direction of Comparison) analysis of covariance (ANCOVA) was performed with including sympathy as a covariate. Also, to investigate the reverse possibility, the foregoing ANCOVA was repeated treating social desirability as a covariate.⁸ Both analyses revealed that the main effects of direction of comparison and the interactive effects of target under evaluation and direction of comparison remained significant.