Psychological mediators of sex differences in emotional support: A reflection on the mosaic

Abstract

* The papers in this issue begin to answer some questions about factors that help to explain sex differences in emotional support processes. More important than the tentative answers these papers provide are the enduring questions they pose. In particular, these papers raise significant questions concerning (a) the frequency, size, and meaningfulness of sex differences in emotional support, (b) the circumstances that lead such differences to be more and less prominent, and (c) the reasons why such differences exist and how we can best understand them. In concluding this special issue of Communication Reports (CR), I want to reflect briefly on each of these questions. Sex Differences in Emotional Support: How Many, How Big, and How Important? All of the studies appearing in this issue of CR report statistically significant sex differences in varied aspects of emotional support. But one should not conclude from this that sex differences in emotional support are generally widespread or large. Because the editorial focus of this special issue centered on psychological mediators of sex differences in emotional support, a prerequisite for a paper's inclusion was the existence of some statistically significant sex difference in a relevant dependent variable. After all, it doesn't make sense to examine potential mediators when there is no relationship to mediate! Although this selection principle makes sense on editorial grounds, it necessarily precluded consideration of whether sex differences in emotional support are either typical or sizeable. Importantly, many studies have not found statistically reliable sex differences in key emotional support variables (e.g., Cutrona & Suhr, 1994; Goldsmith & Dun, 1997), even when research designs have permitted detection of differences that might exist (e.g., large samples providing high levels of statistical power; highly reliable assessments minimizing measurement error). Indeed, there is good reason for believing that, at least with respect to emotional support, men and women exhibit more, and larger, similarities than differences (see the discussions by Kunkel & Burleson, 1998; Vangelisti, 1997). Finding no discernable sex differences in specific aspects of emotional support may be as theoretically and practically significant as the detection of large, reliable differences. As just one example of this, Susanne Jones and I recently found that the effectiveness of comforting messages exhibiting different levels of person centeredness, as delivered by peer helpers in actual interactions, was not qualified by the sex of either the helper or recipient (Jones & Burleson, in preparation). This finding, in conjunction with others (e.g., Kunkel & Burleson, 1999), suggests that both men and women are most comforted by messages that help them explicitly articulate and elaborate their distressed feelings, and this remains true regardless of the sex of the helper. Such findings have obvious practical and theoretical import, precisely because there are no reliable sex differences. For example, they indicate that men and women should not be given intercultural communication training in how to comfort members of the opposite sex because both sexes are most comforted by the same types of messages. We should, then, avoid becoming so enamored with sex differences that we overlook stronger, more interesting, and more important similarities (Vangelisti, 1997; Wood & Dindia, 1998). However, it is also the case that when sex differences in emotional support are detected they are quite consistent in their direction. Specifically, sex differences for varied aspects of emotional support consistently indicate that women are more supportive than men. For example, the five studies included in this special issue of CR indicate that, compared to men, women: (a) produce comforting messages exhibiting higher levels of person centeredness (Samter; MacGeorge et al. …