Journal of Behavioral Medicine
© Plenum Publishing Corporation 1996. All Rights Reserved. Volume 19(3), June 1996, pp 289-305
Parenting Style and Adolescent Depressive Symptoms, Smoking, and Academic Achievement: Ethnic, Gender, and SES Differences
[Article]

Radziszewska, Barbara1,4; Richardson, Jean L.2; Dent, Clyde W.2; Flay, Brian R.3

1Prevention Research Branch, Division of Epidemiology, Statistics, and Prevention Research, NICHD, NIH, 6100 Executive Boulevard, Room 7B05, Bethesda, Maryland 20892.
2Department of Preventive Medicine, School of Medicine, University of Southern California, Los Angeles, California.
3School of Public Health, University of Illinois at Chicago, Chicago, Illinois.
Accepted for publication: January 25, 1996
4To whom correspondence should be addressed.

Outline

Graphics

Abstract^

This paper examines whether the relationship between parenting style and adolescent depressive symptoms, smoking, and academic grades varies according to ethnicity, gender, and socioeconomic status. Four parenting styles are distinguished, based on patterns of parent-adolescent decision making: autocratic (parents decide), authoritative (joint process but parents decide), permissive (joint process but adolescent decides), and unengaged (adolescent decides). The sample included 3993 15-year-old White, Hispanic, African-American, and Asian adolescents. Results are generally consistent with previous findings: adolescents with authoritative parents had the best outcomes and those with unengaged parents were least well adjusted, while the permissive and the autocratic styles produced intermediate results. For the most part, this pattern held across ethnic and sociodemographic subgroups. There was one exception, suggesting that the relationship between parenting styles, especially the unengaged style, and depressive symptoms may vary according to gender and ethnicity. More research is needed to replicate and explain this pattern in terms of ecological factors, cultural norms, and socialization goals and practices.



INTRODUCTION^

This study examines whether demographic variables such as ethnicity, gender, and socioeconomic status affect the relationship between parenting style and adolescent adjustment. Three adolescent outcomes are considered in this study: depressive symptoms, smoking, and academic grades.

The results of this study are based on a survey, originally designed to examine the correlates of substance use, administered to a multiethnic sample of adolescents in the ninth grade. Although the study was not specifically designed to investigate the role of parenting style in adolescent adjustment, these data nevertheless are a useful source for an exploratory analysis in that area.

Parenting style is usually defined on the basis of two dimensions of parental behavior: control (strictness, demandingness) and warmth (affection, responsiveness) (Baumrind, 1971; Lamborn et al., 1991; Maccoby and Martin, 1983). Based on the two dimensions, control and warmth, four parenting styles have been differentiated (Lamborn et al., 1991; Maccoby and Martin, 1983). Parents who are warm and at the same time exercise moderate levels of control are classified as authoritative. These parents employ inductive reasoning and explain reasons for their decisions as a means of controlling their children's behavior. Parents who impose a high level of control and are not warm are classified as autocratic; those who are warm but exercise little control over the child are considered permissive (or indulgent); finally, parents who are not warm and who do not supervise and control their children are called unengaged (or neglectful).

There is consistent evidence that the authoritative style (warm, moderate control) is associated with the best outcomes in many domains of child development, including psychosocial functioning, school competence, emotional well-being, and behavioral adjustment, while the unengaged style (low on both warmth and control) correlates with poor adjustment and problem behavior (like drug and alcohol use, school misconduct, delinquency) and is therefore considered a risk factor in child development (Lamborn et al., 1991; Maccoby and Martin, 1983; Steinberg, 1990; Richardson et al., 1993). Studies of the effects of the autocratic and the permissive parenting styles indicate that these styles may also impede certain aspects of development (Maccoby and Martin, 1983).

To date, few studies of the effects of parenting practices on child outcomes have examined whether these effects vary by demographic characteristics, and the results have been inconsistent. Lamborn et al. (1991) reported an absence of interactions between parenting style and adolescent gender, ethnicity, and socioeconomic status, concluding that the effects of parenting style transcend demographic groups. However, other researchers (Baldwin et al., 1990; Dornbusch et al., 1987; Steinberg et al., 1992a, b) found evidence suggesting that demographic variables moderate the impact of parenting styles in at least one domain of development: intellectual competence and academic performance. These studies found either that parenting style was a useful predictor of outcomes only for some demographic subgroups or that the effects of certain styles were opposite for different subgroups.

For example, Dornbusch et al. (1987), who examined the influence of authoritative, autocratic, and permissive styles on academic achievement, found that these three parenting styles were predictive of grades mostly among White adolescents and less so for adolescents from other ethnic groups. Steinberg and his colleagues (1992a, b) found that the academic performance of African-American and Asian adolescents, in contrast to that of White and Hispanic adolescents, did not benefit from the authoritative parenting style. In addition, parental involvement and parental encouragement were significant predictors of school performance and school engagement among several ethnic groups but not among African-American youth.

Baldwin et al. (1990) found that among children living in high-risk families (defined by low SES, minority status, and absence of a father), in contrast to those from low-risk families, those who had more restrictive and autocratic parents had higher IQ scores, higher scores on an arithmetic achievement test, and higher overall achievement. This pattern is consistent with Baumrind's (1972) finding that self-assertiveness and independence among African-American preschool girls was enhanced by a parenting style that, by White middle-class definitions, would be classified as autocratic.

Based on Kohn's (1977) work, Steinberg et al. (1991) postulate that authoritative parenting, with its emphasis on democratic parent-child relations and encouragement of autonomy may benefit mostly middle-class children and less so lower-class children. Autocratic parenting, on the other hand, may be detrimental to children living in middle-class families but may have less negative impact and may even serve as a protective factor, for children living in poor homes and dangerous environments.

At the earlier stages of research on parenting style, only three parenting styles were examined: authoritative, autocratic, and permissive. The unengaged parents were not considered separately from permissive parents, and therefore, the relative impact of that style in different sociodemographic groups could not be determined. For example, it was not possible to examine whether the relative costs of unengaged parenting style are greater for adolescents living in poor, high-risk neighborhoods. In the present work, the exploration of the relationship between parenting style and adolescent adjustment involved all four parenting styles.

The questionnaire that was the data source for this study did not measure parental warmth, but it included a question regarding students' perception of the balance between parental and adolescent's involvement in everyday decision making that affects the adolescent (on issues like spending money, leisure activities, or how late the adolescent can stay out). Parental involvement in decision making is a useful indicator of parental control (Rollins and Thomas, 1979). The responses to this question provided us with the basis for the operational definition of parenting style, similar to that used by several other researchers (Elder, 1962; Kandel and Lesser, 1969; Kandel and Davies, 1982). Parents who make most of the decisions by themselves were classified as autocratic, while parents who ask the adolescent's opinion but retain final authority were considered to be authoritative. Parents who contribute to the decision-making process but allow adolescents to make the final decision were classified as permissive. Finally, parents of adolescents who reported making decisions mostly on their own were classified as unengaged.

In the present study we examined whether the relationship between parenting style and adolescent outcomes varies according to ethnicity, gender, and income. The following issues were addressed.

A. We explored whether the negative impact of the unengaged and the autocratic parenting styles transcends all demographic groups or whether the negative effects of these styles are especially high for some subgroups.
B. We examined whether the autocratic style is more effective than the authoritative style for any ethnic or SES group, especially African-Americans and lower-income students.
C. In addition, we asked whether the moderating role of demographics is confined to academic achievement or whether other aspects of development are likely to be affected as well.
METHODS^
Subjects^

The data were gathered in February 1988, as a part of a larger survey study of adolescent smoking and drug use. A detailed description of that study is provided elsewhere (Flay et al., 1988). The sample consisted of 3993 ninth-grade students in Los Angeles County and San Diego County, representing public schools in six school districts. Trained data collectors provided instructions and administered the questionnaires, which took approximately 45 min to complete, to entire classes of students. Letters describing the study were sent to the parents of children in these classrooms. Less than 8% of the children did not participate due to parental or student refusal.

Demographic Characteristics of the Sample. The sample consisted of 1930 males and 2063 females. Students were classified into ethnic categories according to their responses on the questionnaire. SES was determined on the basis of median income from the 1986 intercensal data for a given zip code area. Table I summarizes the demographic characteristics of the sample.



Table I. Demographic Characteristics of the Sample

The information in Table I indicates that in terms of income, the majority of Asians and Whites were in the upper-income category, while the opposite was true of African-Americans. Only Hispanics were roughly equally represented in both income categories. The cell sizes for low income Asians and higher income African-Americans were small, indicating that the results of the analyses reported below, especially those involving higher-order interactions, have to be interpreted cautiously.

Measures^

Since the main use of the questionnaire was to study drug abuse, we present only these subscales or individual items which are relevant to the present study. Due to constraints on the length of the questionnaire imposed by time limitations, our subscales were very brief. Below, we describe our measures. The contents of the items, response categories, and score ranges are presented in the Appendix.

Depressive Symptoms. The Children's Depression Inventory (CDI), developed by Kovacs and Beck (1977), was modeled after the Beck Depression Inventory for adults. It is a widely used, 27-item self-report measure. We used an abbreviated form of the CDI, based on the work by Saylor et al. (1984), who factor analyzed the CDI using data from normal and from emotionally disturbed children. They found that three items were common on Factor 1 and reflected a feeling of not being loved by others, hating oneself, and feeling responsible for all bad things. The second factor had four common items and reflect feelings of being tired all the time, feelings of doing everything wrong, feelings of an ominous nature, and feelings of being alone all the time. Additional factors were composed of only a few items each and the dimensions of these factors were less clear. In this study, we used seven items of the CDI, corresponding to the first two factors isolated by Saylor et al. (1984). No specific cutoff score was used to classify adolescents as either depressed or nondepressed; instead, we used a sum score to gauge the level of depressive symptoms. The scores ranged from 0 to 2 per item and 0 to 14 for the entire scale. Cronbach's alpha for the seven-item scale was.76.

Academic Grades. Students categorized their grades as fitting one of nine categories, ranging from “mostly F's” (score 0) to “mostly A's” (score 8).

Cigarette Use. Cigarette smoking was categorized based on a combination of scores on current and lifetime use. We distinguished eight levels of cigarette use: (1) never used (39.6%), (2) puffs or 1 cigarette but not in the past 12 months (20.9%), (3) more than 2 cigarettes but not in the past 12 months (14.5%), (4) a few cigarettes in the past 12 months (12.1%), (5) 1 cigarette in the past month (2.4%), (6) a few cigarettes in the past month (3.6%), (7) 1 or a few cigarettes per week (3.5%), and (8) several cigarettes per day or more (3.4%).

Family-Adolescent Conflict. The purpose of measuring the extent of conflict between adolescents and their family was to provide a validity check for our parenting style measure (see below). Conflict was measured by three items, reflecting the extent to which the adolescent feels being nagged, not being understood, and having frequent arguments (Cronbach's alpha =.84).

Parenting Style. This variable was measured by a single item, asking about the extent to which parents exercise control over the process of making decisions relevant to adolescent's daily life (see Appendix). Based on the adolescent's response, we categorized the adolescent as having parents who are either Autocratic (parents make all decisions), Authoritative (joint decision making, but parents have the final say), Permissive (the adolescent has greater influence on decision making than the parents), or Unengaged (the adolescent makes all decisions by him-/herself).

The validity of the parenting style measure used in this study was verified by checking the patterns of the relationship between the degree of parental control exercised in decision making and the extent of family-adolescent conflict and depressive symptoms in adolescents. Studies have consistently shown a curvilinear relationship between the extent of parental control, on one hand, and family-adolescent conflict and adolescent depressive symptoms, on the other: the level of conflict and depressive symptoms is lowest when parents exercise a moderate amount of control and highest in families with extreme levels (either low or high) of control (Elder, 1962; Ellis, 1986; Kandel and Davies, 1982; Rollins and Thomas, 1979). In order to provide information on the validity of the parenting style measure, an examination of the effect of parenting style on parent-adolescent conflict and adolescent depressive mood was carried out using one-way ANOVAs. As expected, the relationship found between parental control and family-adolescent conflict and depressive symptoms in this study conformed to the curvilinear pattern (see results below).

Plan of Analysis^

Chi-square analyses were performed to determine whether there were ethnic, gender, or income level differences in parenting styles. To determine whether there were significant parenting style, ethnicity, gender, or income main effects or any interactions between these variables, we performed three four-way ANOVAs (one for each dependent variable): parenting style (4 categories) × ethnic group (4 categories) × gender (M/F) × income (low/high), with depressive symptoms, grades, and cigarette use as dependent variables. Any significant interactions involving parenting style and the demographic variables would suggest that the influence of parenting style on adolescent outcomes varies according to the demographic characteristics of adolescents. All ANOVAs performed in this study used procedure GLM (SAS), which is recommended for unbalanced designs.

RESULTS^
Validity of the Parenting Style Measure^

The results confirm the hypothesized curvilinear relationship pattern between parenting style and family-adolescent conflict and depressive symptoms (see Table II). There was a significant parenting style effect on family-adolescent conflict [F(3,2751) = 15.6, p <.0001], and on depressive symptoms [F(3,2671) = 18.5, p <.0001]. The highest levels of family-adolescent conflict and depressive symptoms were found in families with unengaged parents, followed by those with autocratic parents; those families with permissive and authoritative parents (especially the latter) had relatively low levels of conflict and adolescent depressive symptoms.



Table II. Main Effects of Parenting Style on Conflict, Depressive Symptoms, Smoking, and Academic Grades

Ethnic, Gender, and SES Differences in Proportions of Parenting Styles^

There were small but significant ethnic differences in the proportions of different parenting styles (see Table I) ([chi]2 = 22.7, df = 9, p =.007). The differences were roughly similar to those found by other researchers (e.g., Dornbusch et al., 1987; Steinberg et al., 1991). Whites were somewhat more likely to be authoritative than the other groups, particularly African-Americans and Asians. African-Americans and Hispanics were somewhat more autocratic than Whites and Asians. Hispanics had proportionately fewer unengaged parents than the other groups.

There was a significant relationship between gender and parenting style ([chi]2 = 39.7, df = 3, p =.0001). Among females, 7.7% had unengaged parents (vs. 14.8% for males), 20.2% had permissive parents (20.7% for males), 40.3% had authoritative parents (34.3% for males), and 31.2% had autocratic parents (30.2% for males). These figures show that boys were more likely to have unengaged parents and less likely to have authoritative parents than girls.

Finally, there was also a significant relationship between income level and parenting style ([chi]2 = 10.5, df = 3, p =.015). Among lower-income students, 10.3% had unengaged parents (vs. 11.5 for higher-income families), 18.3% had permissive parents (21.7% for higher income), 37.1% had authoritative parents (37.8% for higher income), and 34.3% had autocratic parents (29.0% in higher-income families). Consistent with Kohn's (1977) hypothesis, lower-income parents were more likely to be autocratic than higher-income families.

Main Effects of Parenting Style^

There was a significant main effect of parenting style on depressive symptoms, [F(3,2446) = 9.9, p <.0001]. The authoritative style was related to lower levels of depressive symptoms, followed in turn by permissive, autocratic, and the highest level for unengaged (see Table II). There was a significant main effect of parenting style on smoking [F(3,2622) = 10.9, p <.0001]. Smoking was not significantly different between students with permissive, authoritative, or autocratic parents. However, those with unengaged parents were most likely to smoke. There was also a main effect of parenting style on academic grades [F(3,2639) = 5.0, p =.0018]. The pattern was the same as for depressive symptoms, with best outcomes for those with authoritative parents and worst for those with unengaged parents.

Significant Interactions Involving Parenting Style^

There was only one significant interaction involving parenting style: an ethnicity × gender × parenting style interaction effect on depressive symptoms [F(9,2446) = 2.0, p =.04], which is presented graphically in Figs. 1 and 2. Among girls with unengaged parents, Asians had the highest depressive symptoms mean (4.1), followed by Whites (3.0), Hispanics (2.9), and African-Americans (1.8). Duncan's multiple-range test (used for all pairwise comparisons in this study) indicated that the difference between the mean for Asian and that for African-American girls was significant (p <.05). Among boys with unengaged parents, the pattern was different: the highest depressive symptoms mean was found for African-Americans (4.1), followed by Asians (2.6), Hispanics (2.6), and Whites (2.4). The mean for African-American boys was significantly higher (p <.05) than all of the remaining three means.



Fig. 1. Depressive symptoms by parenting style and ethnicity: females.



Fig. 2. Depressive symptoms by parenting style and ethnicity: males.

There were no significant ethnic differences among males with autocratic parents (means range from 2.0 to 2.8). However, among females with autocratic parents, Asians had the highest depression mean (2.8), followed by Whites (2.7), Hispanics (2.2), and African-Americans (1.6). The difference between Asians and African-Americans was significant (p <.05).

There were no significant ethnic differences among means for girls with authoritative parents (means ranged from 1.7 to 2.1), but among males, the mean was the highest for African-Americans (2.2), followed by Hispanics (1.8), Whites (1.6), and Asians (1.3). However, it is important to note that the means for all adolescents with authoritative parents, including the mean for African-American boys, were relatively low.

Finally, among adolescents with permissive parents, there were no significant differences either for girls (range 1.8 to 2.4) or for boys (range 2.0 to 3.0).

Other Significant Interactions and Main Effects^

There was a significant ethnicity × gender interaction effect on depressive symptoms [F(3,2446) = 5.8, p =.0006]. This interaction reveals a different pattern of gender differences on depressive symptoms among African-American adolescents than among adolescents from other ethnic groups. While White, Asian, and Hispanic females had slightly (nonsignificantly) higher depressive symptoms scores than their male counterparts, the pattern was opposite for African-Americans: males had higher depressive symptoms means than females [2.7 vs. 1.8; t(315) = 3.6, p <.0004].

There was a significant income by ethnicity interaction effect on smoking [F(3,2622) = 10.9, p =.04]. Among lower-income students, Whites had the highest mean (2.9), followed by Hispanics (2.4), African-Americans (2.1), and Asians (2.0). Pairwise mean differences between Whites vs. Hispanics, Whites, and Asians, and between Hispanics vs. Asians and African-Americans vs. Asians, were significant (p <.05). Among adolescents from higher-income families, Asians had the lowest smoking mean (2.2), which differed significantly (p <.05) from the means for all three remaining ethnic groups (which did not differ significantly among each other, and ranged from 2.8 to 3.0).

There was a significant main ethnicity effect on smoking [F(3,2622) = 3.8, p =.0105]. The means for smoking were 2.1 (Asians), 2.2 (African-Americans), 2.6 (Hispanics), and 2.8 (Whites). The means for Whites and Hispanics did not differ significantly, but each was significantly higher than the means for African-Americans and Asians (the latter two did not differ significantly).

There was a significant gender effect on academic grades [F(1,2639) = 10.5, p =.0012], with females having somewhat higher average grades than males (6.7 vs. 6.4). There was also a significant ethnicity effect on grades [F(3,2639) = 98.8, p =.0001]. Asians had the highest average grades (7.5), followed by Whites (6.7), Hispanics (6.0), and African-Americans (5.8). Each grade mean was significantly different from the other (p <.05).

DISCUSSION^

The results of the present study are generally consistent with earlier findings, such as those of Lamborn et al. (1991): the authoritative style was generally associated with the best outcomes, the unengaged style with the worst outcomes, and the permissive and autocratic styles with intermediate outcomes. Also consistent with Lamborn at al. (1991), for the most part, the effects of parenting styles appear to be similar across ethnicity, gender, and income groups.

There was one exception to that pattern, which suggests the need for further exploration of the hypothesis that sociodemographic variables moderate the relationship between parenting styles and adolescent outcomes. Unlike in the Lamborn et al. (1991) study, where no interactions were found between parenting style and demographics, we found one such interaction. It is important to emphasize, however, that due to small sizes of some sociodemographic subgroups, especially of Asians and African-Americans, the present results must be viewed with caution.

Are the Negative Effects of Unengaged and Autocratic Parenting Styles Uniform Across Demographic Subgroups? We found evidence of moderating effects of ethnicity and gender on the relationship between parenting style and depressive symptoms. The results indicated that among boys with unengaged parents, African-Americans had a significantly higher mean depression score than Whites, Hispanics, and Asians (the latter three did not differ). On the other hand, among girls with unengaged parents, it was Asians who had the highest mean, and African-Americans had the lowest; the latter difference was statistically significant (the means for White and Hispanic girls were intermediate).

Further, among females with autocratic parents, Asians had the highest mean depression score and African-Americans had the lowest mean depression score, a pattern similar to that found for females with unengaged parents. Among boys with autocratic parents, however, no significant differences were found.

In sum, the significant parenting style × ethnicity × gender interaction found in this study points out three subgroups at elevated risk for depressive symptoms: African-American boys with unengaged parents, Asian girls with unengaged parents, and Asian girls with autocratic parents. Interestingly, African-American girls had relatively low depressive symptoms scores, which did not vary with parenting style.

Perhaps the above findings could be accounted for in terms of a combination of ethnic and gender differences in socialization goals and practices. There have been reports that suggest African-American mothers have higher educational and career aspirations for their daughters than for their sons and place particular emphasis on the girls' early independence, self-reliance, and assertiveness (Bock, 1969; Lewis, 1975; Kunkel and Kennard, 1971; Reid, 1972). Consistent with these observations, young African-American females reach independence and assume adult responsibility at an early age, and are more successful in academic, career, and psychological development, than their male counterparts (Smith, 1982). This gender difference in socialization practices in African-American families may result in greater coping skills and resilience in girls than in boys. The unengaged parenting style during adolescence may thus have no impact on the girls, who are already fairly independent, but may have a devastating effect on the boys, who are often exposed to the most high-risk environments. More research is needed to explore in greater depth the reasons for the elevated depression levels among African-American boys as this relates to their family structure, environmental demands, and parental influence.

The finding that Asian girls are likely to have elevated level of depressive symptoms if their parents are either unengaged or autocratic also deserves a follow-up investigation. There is relatively little information available about the patterns of socialization in Asian-American families. The existing literature indicates a strong emphasis on respect for authority, obligation to parents and family, and strict control (e.g., Sung, 1985; Suzuki, 1980). There is no information currently available to speculate as to why Asian girls are particularly likely to experience depressive symptoms if their parents exercise either excessive or minimal control.

Does Autocratic Style Benefit Any Demographic Subgroups? No evidence was found that the autocratic style produced better outcomes than the authoritative style for any of the demographic subgroups.

Are There Domains of Functioning that Consistently Reveal the Moderating Impact of Demographics? So far, most of the evidence that the relationship between parenting practices and child outcomes varies by demographic variables has been documented for academic and cognitive functioning (Baldwin et al., 1990; Dornbusch et al., 1987; Steinberg et al., 1992a, b). In contrast, in the present study, we found some evidence suggesting a moderating effect for depressive symptoms, but not for academic grades.

Limitations of the Study. There are at least four types of problems in this study which may limit the validity of its findings. First, the sample sizes in some ethnicity/gender/income cells were small. That was especially the case for higher-income African-Americans and lower-income Asians. This problem may have limited the power to detect higher-order interaction effects. Second, except for income, all data utilized in this study were based on students' self-report measures, without independent validation. It is possible that at least part of the associations between unengaged or autocratic parenting styles and poor adolescent adjustment comes as a result of bias in adolescent reports of parental behaviors. Third, some of the measures used in the study were limited. For example, parenting style was determined on the basis of students' responses to one item in the questionnaire. However, the measure of parenting style utilized in this study was similar to those used by other researchers (e.g., Elder, 1962; Kandel and Davies, 1982), and the pattern of relationships between the current parenting style measure and related variables was as hypothesized, lending support to the validity of the parenting style measure.

Finally, perhaps the most serious limitation of the study is its cross-sectional rather than longitudinal design, precluding examination of reciprocal relationships between parent and child behavior. There is general consensus that the relationships between parental and child behavior patterns are complex, dynamic, and bidirectional (Maccoby and Martin, 1983; Rollins and Thomas, 1979). The coercion model developed by Patterson (1982) illustrates this complexity. Patterson and his colleagues (Dishion and Patterson, 1992; Dishion et al., 1991; Patterson et al., 1992; Patterson and Stouthamer-Loeber, 1984) found that ineffective parenting methods involving a high level of coercion and inconsistent discipline (sometimes ignoring antisocial behavior and at other times using harsh punishment) lead to noncompliant, aggressive behavior in children, which in turn leads to social rejection, school failure, and ultimately an association with antisocial peers and antisocial, delinquent behavior and substance use. Attempts to monitor children's behavior by coercive parents meet with child's resistance and become increasingly ineffective, to the point that parents may give up attempts to control their children's behavior.

Thus, it is possible that the group of parents classified as “unengaged” in the present study contains a subgroup of parents who may have initially tried to control their children's behavior but did that in an ineffective way and may have eventually given up attempts to monitor their children. These parents may continue to use intermittent but harsh discipline. The observed negative effects of the unengaged style may thus be a result of a combination of problematic practices: limited involvement in the adolescent's life, poor monitoring, and harsh, inconsistent discipline. Other unengaged parents may simply be uninvolved in the adolescent's life. Similarly, the group of autocratic parents may include parents who are harsh and consistent as well as those who are harsh and inconsistent.

To sort out these complexities, future studies of parenting styles and their impact on child and adolescent development should use longitudinal designs to account for reciprocal parent-child influences and should include a variety of measures of parenting behavior. In addition to assessing the traditional dimensions of parenting style—control and warmth—measures of monitoring, coercion, and consistency in the use of discipline would permit more refined classification of parenting styles. A better-differentiated measurement of parent (and child) behavior, utilized in a longitudinal study, may help explain ethnic and cultural differences in socialization goals and practices, and their outcomes.

REFERENCES^

Baldwin, A. L., Baldwin, C., and Cole, R. E. (1990). Stress-resistant families and stress-resistant children. In Rolf, J., Masten, A. S., Cicchetti, D., Nuechterlein, K. H., and Weintraub, S. (eds.). Risk and Protective Factors in the Development of Psychopathology, Cambridge University Press, Cambridge. [Context Link]

Baumrind, D. (1971). Current patterns of parental authority. Dev. Psychol. Monogr. 4(1; Pt. 2). [Context Link]

Baumrind, D. (1972). An exploratory study of socialization effects on black children: Some black-white comparisons. Child Dev. 43(1): 261-267. UIC Options Bibliographic Links Library Holdings [Context Link]

Bock, W. E. (1969). Farmer's daughter effect: The case of the Negro female professionals: Phylon 30: 17-26. UIC Options Library Holdings [Context Link]

Dishion, T. J., and Patterson, G. R. (1992). Age effects in parent training outcome. Behav. Ther. 23: 719-729. UIC Options Bibliographic Links Library Holdings [Context Link]

Dishion, T. J., Patterson, G. R., Stoolmiller, M., and Skinner, M. L. (1991). Family, school, and behavioral antecedents to early adolescent involvement with antisocial peers. Dev. Psychol. 27(1): 172-180. UIC Options Bibliographic Links Library Holdings [Context Link]

Dornbusch, S. M., Ritter, P. L., Leiderman, P. H., Roberts, D. F., and Fraleigh, M. J. (1987). The Relation of Parenting Style to Adolescent School Performance. Child Dev. 58: 1244-1257. UIC Options Bibliographic Links Library Holdings [Context Link]

Elder, G. H. (1962). Structural variations in the child rearing relationship. Sociometry 25: 241-262. UIC Options Library Holdings [Context Link]

Ellis, G. J. (1986). Societal and parental predictors of parent-adolescent conflict. In Leigh, G. K., and Peterson, G. W. (eds), Adolescents in Families, South-Western, Cincinnati, Ohio. [Context Link]

Flay, B. R., Brannon, B. R., Johnson, C. A., et al. (1988). The television school and family prevention/cessation project. I. Theoretical basis and television program development. Prev. Med. 17: 585-607. [Context Link]

Kandel, D. B., and Davies, M. (1982). Epidemiology of depressive mood in adolescents. Arch. Gen. Psychiat. 39: 1205-1212. UIC Options Bibliographic Links Library Holdings [Context Link]

Kandel, D., and Lesser, G. S. (1969). Parent-adolescent relationships and adolescent independence in the United States and Denmark. J. Marriage Family 31: 348-358. [Context Link]

Kohn, M. (1977). Class and Conformity, 2nd ed., University of Chicago Press, Chicago. [Context Link]

Kovacs, M., and Beck, A. T. (1977). An empirical-clinical approach toward a definition of childhood depression. In Schulterbrandtt, J. G., and Raskin, A. (eds.), Depression in Childhood: Diagnosis, and Conceptual Models, Raven Press, New York, pp. 1-25. [Context Link]

Kunkel, P., and Kennard, S. S. (1971). Spout Spring: A Black Community, Holt, Rinehart and Winston, New York. [Context Link]

Lamborn, S. D., Mounts, N. S., Steinberg, L., and Dornbusch, S. M. (1991). Patterns of competence and adjustment among adolescents from authoritative, authoritarian, indulgent, and neglectful families. Child Dev. 62: 1049-1065. UIC Options Bibliographic Links Library Holdings [Context Link]

Lewis, D. (1975). The Black family: Socialization and sex roles. Phylon 36: 221-237. [Context Link]

Maccoby, E., and Martin, J. (1983). Socialization in the context of the family: Parent-child interaction. In Hetherington, E. M., (ed.), Mussen, P. H. (series ed.), Handbook of Child Psychology, Vol. 4. Socialization, Personality, and Social Development, Wiley, New York, pp. 1-101. [Context Link]

Patterson, G. R. (1982). Coercive Family Processes, Castalia, Eugene, OR. [Context Link]

Patterson, G. R., and Stouthamer-Loeber, M. (1984). The correlation of family management practices and delinquency. Child Dev. 55: 1299-1307. UIC Options Bibliographic Links Library Holdings [Context Link]

Patterson, G. R., Reid, J., and Dishion, T. (1992). Antisocial Boys: A Social Interactional Approach, Vol. 4, Castalia, Eugene, OR. [Context Link]

Reid, I. (1972). Together Black Women, Third Press, New York. [Context Link]

Richardson, J. L., Radziszewska, B., Dent, C. W., and Flay, B. R. (1993). Relationship between after-school care of adolescents and substance use, risk taking, depressed mood, and academic achievement. Pediatrics 92(1): 32-38. Ovid Full Text Bibliographic Links Library Holdings [Context Link]

Rollins, B. C., and Thomas, D. L. (1979). Parental support, power, and control techniques in the socialization of children. In Burr, W. R., Hill, R., Nye, F. I., and Reiss, I. L. (eds.), Contemporary Theories About the Family: Research-Based Theories. Vol. 1, Free Press, New York. [Context Link]

Saylor, C. F., Finch, A. J., Spirito, A., and Bennett, B. (1984). The Children's Depression Inventory: A systematic evaluation of psychometric properties. J. Consult. Clin. Psychol. 52: 955-967. [Context Link]

Smith, E. J. (1982). The Black female adolescent: A review of the educational, career, and psychological literature. Psychol. Women Q. 6(3): 261-288. UIC Options Library Holdings [Context Link]

Steinberg, L. (1990). Autonomy, conflict, and harmony in the family relationship. In Feldman, S. S., and Elliott, S. R. (eds.), At the Threshold: The Developing Adolescent, Harvard University Press, Cambridge, MA. [Context Link]

Steinberg, L., Mounts, N. S., Lamborn, S. D., and Dornbusch, S. M. (1991). J. Res. Adolesc. 1(1): 19-36. [Context Link]

Steinberg, L., Dornbusch, S. M., and Brown, B. B. (1992a). Ethnic differences in adolescent achievement: An ecological perspective. Am. Psychol. 47(6): 723-729. UIC Options Bibliographic Links Library Holdings [Context Link]

Steinberg, L., Lamborn, S. D., Dornbusch, S. M., and Darling, N. (1992b). Impact of parenting practices on adolescent achievement: Authoritative parenting, school involvement, and encouragement to succeed. Child Dev. 63: 1266-1281. UIC Options Bibliographic Links Library Holdings [Context Link]

Sung, B. L. (1985). Bicultural conflicts in Chinese immigrant children. J. Comp. Family Studies 16(2): 255-269. [Context Link]

Suzuki, B. H. (1980). The Asian-American family. In Fantini, M. D., and Cardenas, R. (eds.), Parenting in a Multicultural Society, Longman, New York. [Context Link]

APPENDIX: ITEMS FOR SCALES^
Depressive Symptoms^
1. a. I do most things okay.b. I do many things wrong.c. I do everything wrong.
2. a. I think about bad things happening to me once in a while.b. I worry that bad things will happen to me.c. I am sure that terrible things will happen to me.
3. a. I hate myself.b. I do not like myself.c. I like myself.
4. a. All bad things are my fault.b. Many bad things are my fault.c. Bad things are not usually my fault.
5. a. I am tired once in a while.b. I am tired many days.c. I am tired all the time.
6. a. I do not feel alone.b. I feel alone many times.c. I feel alone all the time.
7. a. Nobody really loves me.b. I am not sure if anybody loves me.c. I am sure that somebody loves me.

The scores ranged from 0 to 2 per item and from 0 to 14 for the entire subscale.

Parenting Style^

In general, when it comes to issues such as spending money, free-time leisure activities, and how late you can stay out, which of these statements most closely describes how you and your parents make decisions? Circle one.

A. My parents generally make these decisions.
B. My parents ask my opinion but generally make the decisions.
C. I ask my parents' opinions but generally make the decisions.
D. I generally make the decisions.

Parenting styles: a = autocratic, b = authoritative, c = permissive, d = unengaged.

Family-Adolescent Conflict^
1. My family looks for things to nag me about.
2. My family doesn't understand me.
3. I have a lot of arguments with my family.

Score range, 1-5 for each item and 1-5 for the entire subscale. [Context Link]

KEY WORDS: parenting style; adolescent adjustment; ethnicity; gender; socioeconomic status



Accession Number: 00005020-199606000-00005