Ninety-eight young U.S. children (mean age = 48 months) with either European, Latin American, or multiple ethnic backgrounds were videotaped with their mothers and their fathers on separate occasions in their families' homes. Parent-child pairs played for 8 min each with a feminine-stereotyped toy set (foods and plates) and a masculine-stereotyped toy set (track and cars). Levels of affiliation (engaging vs. distancing) and assertion (direct vs. nondirect) were rated on 7-point scales every 5 s from the videotapes for both parent and child. Overall, the play activity accounted for a large proportion of the variance in parents' and children's mean affiliation and assertion ratings. Some hypothesized gender-related differences in behavior were also observed. In addition, exploratory analyses revealed some differences between the different ethnic groups. The results highlight the importance of role modeling and activity settings in the socialization and social construction of gender.

Many contemporary theories of social development are similar and/or share complementary constructs. Yet, there have been relatively few efforts toward theoretical integration. The present chapter represents a call for increased theory bridging. The problem of theoretical fragmentation in psychology is reviewed. Seven highlighted reasons for this predicament include differences between behavioral sciences and other sciences, theoretical paradigms as social identities, the uniqueness assumption, information overload, field fixation, linguistic fragmentation, and few incentives for theoretical integration. Afterward, the feasibility of theoretical synthesis is considered. Finally, some possible directions are proposed for theoretical integration among five contemporary theories of social and gender development: social cognitive theory, expectancy-value theory, cognitive-developmental theory, gender schema theory, and self-categorization theory.

Children's discourse with peers was examined in relation to speaker gender, partner gender, and age level. 138 children were matched with either a same- or an other-gender peer at the early childhood and middle childhood age levels (median ages = 5 and 7 years) and asked to play with puppets for 10 min. Speech acts were coded as either collaborative (affiliative involvement and direct influence), controlling (distancing involvement and direct influence), obliging (affiliative involvement and nondirect influence), or withdrawing (distancing involvement and nondirect influence). Girls' and boys' communication patterns were more similar than different. However, gender-related differences with medium to large effect sizes were found. Gender-typed communications were more likely at the middle childhood than the early childhood age level and in same-gender than mixed-gender dyads. The findings are interpreted in terms of developmental and contextual accounts of gender and social behavior. Recommendations for future research are offered.

Recent historical trends and current directions in the study of children's gender and relationships are reviewed using Maccoby and Jacklin's (1974) The Psychology of Sex Differences as a reference point. Since the publication of Maccoby and Jacklin's review, researchers have questioned the extent and the magnitude of gender differences in social behaviour as well as the degree to which parents play a primary role in gender development. More attention is now paid to the impact of gender-segregated peer groups and other social relationships (e.g., friendships, romantic relationships) as well as cognitive-motivational and biological processes. Furthermore, the role of the larger social-structural context is addressed in studies of sexism and gender bias during childhood and adolescence. Recommendations for future research are offered.

The hypothesis that the terms masculine and feminine would not have the same connotations as the personality traits that are typically associated with them was investigated. Two hundred five undergraduates (116 women and 89 men) at a US university rated how much they expected to like various hypothetical women and men, each of whom was described by a single trait adjective. Ten instrumental-trait and 10 socioemotional-trait adjectives, including the terms masculine and feminine, were associated with a targeted female or male character (e.g., "an independent woman" or "an understanding man"). Although "a masculine woman" and "a feminine man" both received low ratings, female targets described by instrumental adjectives and male targets described by socioemotional adjectives received high ratings. Nonstereotypic characters were rated more positively by the female respondents than they were by the male respondents. Stereotypic targets were rated more positively by other-gender respondents than by same-gender respondents.

Decision-making processes were examined in conversations between same-gender and cross-gender friends. Participants were university students (mean age = 19 years) from mostly middle-class, European-American backgrounds. Each pair of friends was asked to participate in two decision-making topics for 5 minutes each. Transcripts of the taped conversations were coded for suggestions, agreement, disagreement, and abstentions (i.e., neither agreement nor disagreement). There were no significant differences between either the women or the men friendship pairs or between the women and men partners within the mixed-gender pairs in any of the observed behaviors. However, when speaker gender and partner gender interaction effects were analyzed, it was found that women with a woman friend were more likely to receive supportive responses and less likely to receive negative responses to their suggestions than were women with a man friend. There were no partner gender effects on responses to men 's suggestions. The results highlight ways in which women and men may handle joint decision-making with friends depending on both the speaker's gender and the partner's gender.

Agency, communion, and gender were compared as predictors of communication style in the conversations of 76 unacquainted pairs of male-female college students. Each dyad was given a popular issue to discuss for 5 min. Agency and communion were measured using Spence and Helmreich's (Masculinity and Femininity: Their Psychological Dimensions, Correlates, and Antecedents, Austin, Texas: University of Texas Press, 1978) Personal Attributes Questionnaire. Self-perceived agency was significantly associated with verbal assertiveness: High-agency persons used fewer indirect statements and fewer passive self-references than low-agency persons. Nonsignificant trends indicated that self-perceived communication tended to be associated with interpersonal involvement: High-communion persons used fewer impersonal references and were liked more than low-communion persons. © 1987 Plenum Publishing Corporation.