Sex Differences in Relationships

Psychological Sex Differences

Thousands of studies on psychological sex differences have been conducted, so many that a number of meta-analyses of these analyses have been conducted. A meta-analysis is a quantitative summary of the results from multiple quantitative studies on a given topic (e.g., sex differences in self-disclosure). First, an effect size, which measures the magnitude of sex differences, is calculated for each study. A typical measure of effect size is d, which represents the difference between the means for women and men in standard deviation units. Then the average effect size across studies (weighted by sample size) is calculated and represents the difference between women and men across studies. Finally, a homogeneity statistic is used to determine whether the effect size varies across studies more than one would expect as a result of sampling error, in which case a search for moderator variables (e.g., variables that affect the size of effect size) may be conducted.

Jacob Cohen offered the following guidelines for interpreting d: d = 0.20 is small, d = 0.50 is moderate, and d = 0.80 is large. An effect size of d = 0.2 indicates that sex accounts for 1 percent of the variance in the dependent variable and represents an 85 percent overlap in the distributions of women and men on the dependent variable. Cohen labeled a value of d = 0.5 as a medium effect size. When d = 0.5, there is a 67 percent overlap in the distribution of women and men on the dependent variable. Cohen labeled a value of d = 0.8 as a large effect size. A large effect accounts for 14 percent of the variance in the dependent variable. When d = 0.8, there is a 53 percent overlap in the scores of women and men on the dependent variable. These associations are depicted in Figure 1.

Recently, Janet Hyde advanced the gender similarities hypothesis, which states that women and men are similar on most, but not all, psychological variables. Hyde reviewed 46 meta-analyses on cognitive variables (e.g., cognitive abilities), verbal and nonverbal communication (e.g., self-disclosure), social or personality variables (e.g., aggression, leadership), psychological well-being (e.g., self-esteem), and miscellaneous constructs (e.g., moral reasoning). The results of her review support the gender similarities hypothesis. She provided evidence that most psychological gender differences are in the close-to-zero or small range (78 percent), the major exceptions were aggression and some aspects of sexuality. Hyde also found that gender differences vary substantially in magnitude at different ages and depend on the context in which measurement occurs.

...