No abstract

Interview data from a sample of 262 poor African American single mothers and their 7th- and 8th-grade children were used to investigate the relations between neighborhood conditions and adolescents’educational values and school effort. The model tested incorporates both subjective and objective assessments of neighborhood quality and controls for several family- and school-related constructs. United States Census data on household incomes served as an objective measure of neighborhood quality. In the full sample, the findings revealed that the percentage of middle-class neighbors and self-perceived academic abilities were significantly linked to adolescents’educational values, which were, in turn, related to school effort. Several theories identifying the processes by which neighborhood characteristics influence adolescents’ educational values are discussed. Gender differences emerged when the model was tested separately for females and males. For African American females, but not for adolescent males, neighborhood conditions were associated with educational values.

Although the word "mentor"has traditionally been used to describe a relationship between an older adult and a younger person, recent work has extended its usage to relationships with peers and groups rather than with individuals and uncoupled the instrumental and affective qualities of the role. This paper examines (a) the extent to which adolescents' relationships with significant others in different social roles are characterized by mentoring and (b) the extent to which mentoring and other relationship functions covary. Adolescents' naturally occurring social relationships are explored in two very different contexts-Japan and the United States-that differ in the norms and patterning of social interactions. College students (N = 365) used questionnaires to describe the extent to which relations with significant others were characterized by mentoring. Results indicate striking similarity in the patterning of results in the two countries and support the traditional view of mentoring. Mentoring is most likely to occur in relationships with adults (especially parents), rather than with peers, and with same-gender, rather than other-gender associates. Mentoring by parents appears to covary with other aspects of positive relationships, but be more independent in relationships with unrelated adults or peers. Although more of the variability in experienced mentoring is attributed to differences between associates than to differences between adolescents in both the United States and Japan, this is especially true of the United States. Results suggest that although "classic" mentoring is most common in both countries, mentoring is somewhat less constrained by social role differences in Japan than is in the United States.

Using data from a sample of 455 African American children (ages 10 to 12 years) and their parents, this study tests a hypothesized model linking (a) maternal work demands to family routines through work-family conflict and depressive symptoms and (b) maternal work demands to children's externalizing and internalizing problems through family routines. Partial support for our hypotheses was found in single-mother families, but not in two-parent families. Work demands decreased family routines in single-mother families by increasing work-family conflict. In addition, higher levels of work-family conflict and maternal depressive symptomatology predicted an increase in children's externalizing problems and these relations were mediated through a decrease in family routines. Contrary to prediction, work demands bore no relation to children's behavior problems either directly or indirectly through family routines. Family structure differences in mediating processes are discussed in relation to mothers' subjective experiences of work and the roles of adults in the household.

No abstract