Perry, David G.

This thesis examined maternal caregiving as a function of maternal depression. Subjects were 184 fourth through seventh grade children and their mothers. Mothers' depression classification was determined by scores on a depression inventory. Both children and mothers responded to questionnaires developed to measure parenting style. Children also completed a peer nomination inventory to measure adjustment in the peer group. It was hypothesized that depressed mothers would be perceived as more likely to engage in negative parenting behaviors and less likely to exhibit positive caregiving. Similarly, it was expected that children with depressed mothers would exhibit greater internalizing/externalizing behavior problems in the peer group. Results were the reverse of those expected, with children of depressed mothers perceiving less use of aversive caregiving behaviors, and more positive interaction. Maternal depression was unrelated to measures of internalizing and externalizing behaviors.

The purpose of this study was to explore relations between maternal childrearing styles and children's coping styles during mother-child conflict. Mothers reported on six parenting style dimensions: coercive control, overprotective control, emotional control, constructive control, responsiveness, and positive interaction. Children (mean age 12.01 years) reported on seven dimensions of coping during conflict with their mothers: idealization of parent, endorsement of submission, endorsement of aggression, self-blame, self-efficacy, anger reaction, and fear reaction. Few relations between the maternal variables and the child variables were significant, but those that were made sense. For example, positive interaction with the mother predicted low endorsement of aggression toward her, and maternal emotional control predicted self-blaming coping. Further research should examine whether the effects of maternal styles on children's behavior problems (e.g., aggression, depression) are mediated by children's coping styles.

Children nominated same-sex, same-grade peers on aggression, victimization, friendship, and behavioral characteristics. Results showed that aggressive children did not have fewer friends, but did have more enemies than nonaggressive children. Victimized children had fewer friends and more enemies than nonvictimized children. It was also shown that aggressive children's friends are also aggressive and victimized children's friends are other victims. Children who were both aggressive and victimized were perceived as the most disruptive and least liked of all children. It was found that children's friends are indeed similar to them while their enemies are dissimilar.

One hundred ninety-four children in the fourth through eighth grades were administered self-report questionnaires to measure hyperactivating and deactivating coping strategies with their parents and best friends. Subjects also completed a peer nomination inventory designed to assess the degree to which peers are aggressive and/or victimized. It was hypothesized that children who scored higher on hyperactivating strategies would be rated as higher in victimization and that children who scored higher on deactivating strategies would be rated as higher in aggression. It was also hypothesized that as children get older, other relationships besides that with the mother begin to play a major role in predicting children's behavior with peers. Results confirmed these hypotheses.

This study explored the social-cognitive functioning of aggressive and victimized elementary school children. Fourth- through seventh-grade male and female children (mean age 11.8 years) responded to four self-report questionnaires which assessed: (a) hostile attributional bias; (b) outcome expectancies for aggressive behavior; (c) outcome values for aggressive behavior; and (d) self-perceived efficacy for regulating three states of emotional arousal (anger, fear, and euphoria). Aggressive children, compared to nonaggressive children, expected aggression to result in control over their victims, did not expect retaliation from their victims, and did not care whether their victims tried to retaliate or not. Victimized children were more likely to expect retaliation for aggressing.

The stability of victimization (and of related variables, such as aggression and rejection) was explored over a two-year period. Subjects were 135 elementary school children (third through sixth graders) who were given the Victimization and Aggression Inventory (VAI) as well as a sociometric status measure. Two years later the measures were readministered. All three variables (victimization, aggression, and rejection) were found to be moderately stable over the two-year period, but stability varied with measure and with cohort (grade of child at first testing). Aggression was stable for all four cohorts whereas victimization was stable only for the older cohorts. Partial correlations revealed that the stability of victimization, but not aggression, was dependent to a large degree on children's rejection scores. Hierarchical regression analyses yielded evidence consistent with the hypothesis that rejection causes victimization.

Based on a peer nomination device measuring victimization and aggression (VAI), 172 children from grades four through seven were classified into four subgroups: aggressive victims, nonaggressive victims, aggressive nonvictims, and nonaggressive nonvictims. Another peer nomination inventory measuring 13 behavioral attributes (BAI) was used to assess behaviors correlated with the subject classifications. Distinctive behavioral profiles for the four subgroups were found. Of particular importance were findings supporting the hypothesis that two distinct types of victims exist: aggressive or "provocative" victims and nonaggressive or "passive" victims. Both types of victims lack prosocial skills and reinforce aggressive attacks by crying, but the two types of victims differ in how they elicit aggression. The provocative victim evidences disruptive behavior, blames others, has difficulty managing conflict, and is perceived as dishonest. The passive victim is withdrawn but expresses anxiety and depression, signalling vulnerability. Implications for conceptualization of peer problem behavior and for intervention are discussed.

This study assessed the perceived fairness of vignette characters who
had contributed either more or less than a coworker in a task and had
subsequently divided the rewards either equitably or equally. The
objective was to explore the development of sex differences in
distributive justice between preadolescence and adulthood in light of
two competing explanations of those differences. The "normative"
explanation states that males and females vary in their respective
preferences for the norms of equity and equality. The "motivational"
explanation states that the sexes vary their norm preference according
to self-favoring (males) or generous (females) motives . Results
provided no support for the former explanation, but support in the
fifth grade and college groups for the latter explanation. The eighth
grade group was unique in that those males rated generous allocations
more fair than the females did. A developmental pattern of cognitive
changes in sex-role concepts was suggested.

A gender self-socialization model was conceptualized, wherein gender identity and idiographic gender stereotypes conjointly influence children's adoption of gendered behavior (i.e., gender typing). Further, children differ in their beliefs of sex differences as immutable versus fluid (entity vs. incremental theory); and it was hypothesized that entity beliefs would moderate the self-socialization process. Children (N=305, M age 10.8 years) responded to gender identity, gender stereotype, and self-efficacy measures. Two kinds of gender typing were computed. Personal gender typing was the correlation between personal stereotypes and self-efficacy; consensus gender typing was the correlation between the same-sex peer stereotypes and self-efficacy. Results indicated that gender typicality and gender contentedness were associated with personal gender typing, and felt pressure against other-gender behavior was related to consensus gender typing. Entity theory strengthened the relation between gender identity and gender typing. Results support the self-socialization model.

This thesis examined the possibility of meaningful associations between children's attachment styles in middle childhood and children's perceptions of the parent. Participants were 199 students (94 males, 105 females) in grades three through eight (mean age = 11.03 years) from a Florida university school. The children were administered self-report measures and peer-report nomination measures. Five attachment coping strategies (preoccupied, indecisive, avoidant, coercive, and caregiving) and four aspects of perceived maternal behavior (reliable support, overprotection, harassment, and fear induction) were assessed and numerous and meaningful associations were found. For example, perceived maternal overprotection was positively associated with preoccupied coping. Significant associations were also found between our avoidant, coercive, indecisive, and caregiving coping measures and perceived maternal reliable support, harassment, and fear induction. Our numerous and significant findings lend further support for the usefulness and value of our concurrent correlational self-report measures and to justify future longitudinal research to compare alternative models.