Marpsy.net

Child Development, September/October 2004, Volume 75, Number 5, Pages 1418 – 1434 Longitudinal Development of Family Decision Making: Defining Healthy Behavioral Autonomy for Middle-Class African American Adolescents Judith G. Smetana, Nicole Campione-Barr, and Christopher Daddis The development of decision-making autonomy was examined in 76 middle-class African American earlyadolescents (M 5 13 years) and their mothers, who were followed longitudinally for 5 years. Adolescent deci-sion-making autonomy over conventional, prudential, multifaceted, and personal issues increased over time butat different rates. Mothers viewed prudential and conventional issues as parent decisions, but adolescents in-creasingly viewed them as joint. Adolescents viewed multifaceted and personal issues as increasingly decidedby adolescents (with parental input), whereas mothers viewed them as joint. Greater autonomy over multi-faceted issues in early adolescence was associated with poorer adjustment. Controlling for background variablesand earlier adjustment, increased autonomy over personal and multifaceted issues predicted less depressionand better self-worth in late adolescence.
A great deal of theorizing and research on adolescent and interdependence rather than independence development has focused on the development of (Shweder et al., 1998). Similar distinctions also have autonomy, or the process of becoming a self-gov- been drawn between majority and ethnic minority erning person (Steinberg, 1990, 2002; Zimmer-Gem- families in the United States, with ethnic minorities beck & Collins, 2003). Although different aspects of such as African Americans considered more com- autonomy have been identified, the development of munalistic and as valuing interdependence more behavioral autonomy, or the ability to make inde- than do European American families (Garcia Coll pendent decisions, has been of particular interest et al., 1996; Parke & Buriel, 1998). A recent meta- (Collins, Gleason, & Sesma, 1997; Hill & Holmbeck, analysis has indicated, however, that African Ameri- 1986; Silverberg & Gondoli, 1996; Steinberg, 1990; cans are more individualistic than are European Zimmer-Gembeck & Collins, 2003). There has been Americans (Oyserman, Coon, & Kemmelmeier, considerable conceptual confusion, however, about 2002). More recent research has suggested the im- how to define healthy autonomy, particularly in portance of defining adolescent autonomyFfor minority youth. The aim of the present study was to majority and ethnic minority youth alikeFin terms identify healthy behavioral autonomy across ado- of interdependence and relatedness (Collins et al., lescence in middle-class African American families.
1997; Silverberg & Gondoli, 1996, Zimmer-Gembeck Researchers have disagreed as to whether auton- & Collins, 2003). According to Hill and Holmbeck omy development constitutes a culturally specific (1986), behavioral autonomy ‘‘pertains not to free- socialization goal. Research from cultural psycholo- dom from others (e.g., parents), but freedom to carry gy has proposed that the focus on autonomy as a out actions on one’s own behalf while maintaining positive developmental outcome reflects the indi- appropriate connections to others’’ (Collins et al., vidualism of American culture and that in more 1997, p. 78). Thus, recent definitions of autonomy collectivist cultures, individuals value relatedness have focused on self-governance of behavior in thecontext of supportive guidance, relational ties, andsocial commitments. Moreover, recent research hasindicated that a balance between autonomy and Judith Smetana, Nicole Campione-Barr, and Christopher Dad- relatedness is optimally related to adolescent ad- dis, Department of Clinical and Social Sciences in Psychology,University of Rochester.
justment (Allen, Hauser, Bell, & O’Connor, 1994; Christopher Daddis is currently at the Ohio State University at Several recent studies have examined adolescents’ We are grateful to the many families who participated in this self-governance in the family in terms of adolescents’ research and to the William T. Grant Foundation for their support participation in family decision making (Dornbusch, Correspondence concerning this article should be addressed to Ritter, Mont-Reynaud, & Chen, 1990; Fuligni & Eccles, Judith Smetana, Department of Clinical and Social Sciences inPsychology, Meliora Hall, RC 270266, University of Rochester,Rochester, NY 14627. Electronic mail may be sent to smetana@ r 2004 by the Society for Research in Child Development, Inc.
All rights reserved. 0009-3920/2004/7505-0008 1993; Lamborn, Dornbusch, & Steinberg, 1996). Us- decision making (Lamborn et al., 1996) all have been ing a large and diverse sample of 14- to 18-year- old associated with healthy adjustment for African high school students, Dornbusch et al. (1990) exam- American adolescents. Although some researchers ined the impact of youth-alone (where adolescents have asserted that parent-unilateral decision making make decisions without any parental input), parent- protects adolescents against the dangerous neigh- unilateral (where parents make decisions without borhoods in which poor African Americans often any adolescent input), and joint decision making on live (Baldwin, Baldwin, & Cole, 1990), Lamborn et al.
adolescents’ school performance. They found that (1996) found that parent-unilateral decision making youth-alone decision making was associated with also was positively associated with adjustment for poorer academic performance, whereas joint deci- middle-class African American adolescents. Thus, sion making was associated with more positive ac- more research focusing on variation in African ademic outcomes. The findings also varied by family American families is needed and would address the structure, gender, and ethnic background, however.
call for more research on normative developmental Relationships between family decision making and processes among well-functioning minority families academic performance were strongest for European (Garcia Coll et al., 1996; Graham, 1992; McLoyd, Americans, and few significant associations were 1998; Spencer & Dornbusch, 1990).
In most of the research on family decision making A study by Lamborn et al. (1996), which included (Brody et al., 1994; Dornbusch et al., 1990; Lamborn another large sample of high school students varying et al., 1996), there has been little attempt to specify in ethnicity and socioeconomic status, expanded on theoretically the different types of behaviors where these findings by examining the impact of youth- self-governance is sought. Autonomy has been ope- alone, parent-unilateral, and joint decision making rationalized in terms of expectations for family de- on a broader set of adjustment variables, including cision making or youth involvement regarding such psychosocial development (including self-reliance, issues as how to spend money, which friends to go self-esteem, and work orientation), deviance (in- out with, when to do homework, how late to stay cluding drug and alcohol use, school misconduct, out, and what clothes to buy. These items appear to and antisocial behavior), and academic competence have been chosen on an ad hoc basis, but from the (including grade point average, time spent on perspective of social domain theory (Nucci, 2001; homework, and academic expectations). In general, Smetana, 1995a; Turiel, 1998), they can be seen as Lamborn et al. found that across different ethnic primarily personal issues. Personal issues have been groups, joint decision making consistently predicted defined as comprising the private aspects of one’s less deviance over a 1-year period, whereas youth- life and as entailing issues of personal discretion or alone decision making was consistently associated choice pertaining to friends or activities, the state of with negative outcomes. However, the findings were one’s body, and privacy (Nucci, 1996, 2001).
moderated by ethnicity and community context. For Other research examining the desired pacing of instance, Lamborn et al. found that the negative behavioral autonomy has indicated that across cul- impact of youth-alone decision making on psycho- tural and ethnic groups, autonomy is expected earlier social development was greater for African Ameri- (e.g., in early to middle adolescence) for personal is- can youth growing up in predominantly White sues such as deciding what books and magazines to communities than for African American youth read, how to spend money, and when to do home- growing up in ethnically mixed communities. This work than for responsibilities such as doing chores was in contrast to other ethnic groups, where only (Feldman & Quatman, 1988; Feldman & Rosenthal, joint decision making predicted positive outcomes.
1990). Household responsibilities can be conceptual- This research, as well as the broader research on ized as primarily social-conventional issues or as adolescent autonomy, leaves several questions un- arbitrary and agreed-on behavioral norms that answered. First, although the previous research has structure social interactions (such as etiquette and consistently demonstrated that teen-unilateral deci- manners) in different social contexts (Turiel, 1998).
sion making is associated with negative outcomes Research indicates that autonomy is typically granted (Lamborn et al., 1996), there is a need for more spe- later for conventional issues than for personal issues.
cificity in defining the types of family decision Furthermore, autonomy is expected even laterFin making associated with positive outcomes for Afri- late adolescence or young adulthood, if at allFfor can American youth. Youth involvement in decision issues that have negative consequences for adoles- making (Brody, Moore, & Glei, 1994), joint decision cents’ health and safety, such as when to drink alco- making (Lamborn et al., 1996), and parent-unilateral hol, smoke cigarettes, or have sex (referred to as prudential issues; Daddis & Smetana, 2004). These Eccles (1993) found that sixth graders who reported findings suggest that parent decision making over few opportunities for involvement in decision mak- conventional and prudential issues may be associat- ing with parents, as well as no changes over time in ed with healthy adjustment throughout adolescence.
these opportunities, sought less advice from parents Adolescents and parents of different ethnicities and become more peer oriented, choosing activities have been found to agree that parents legitimately with peers over more developmentally positive ac- have the authority to regulate some conventional tivities, such as doing homework. Longitudinal and prudential issues and that adolescents legiti- analyses based on a nationally representative sample mately should be able to decide some personal issues of 11- to 16- year- olds (Brody et al., 1994) indicated (Fuligni, 1998; Smetana, 1988, 2000; Smetana & As- that more adolescent involvement in decision mak- quith, 1994), but research also has shown that ado- ing during these years predicted greater value sim- lescents and parents do not always agree on where ilarity between parents and young adults 6 years the boundaries of adolescents’ personal jurisdiction later. These studies suggest that youth involvement should be drawn. Parents treat some issues (such as in family decision making in early adolescence, at when the adolescent’s bedroom should be cleaned, least for personal issues, may be associated with whether the adolescent should get a tattoo or get his or her body pierced, whether the adolescent should Furthermore, recent research suggests that per- see undesirable friends, or when the adolescent ceived parental overcontrol of the personal domain is should date) as social-conventional (or prudential), associated with psychopathology. Hasabe, Nucci, and whereas adolescents treat them as personal (Sme- Nucci (2004) examined Japanese and U.S. (primarily tana, 1989). Furthermore, research has shown that European American) high school students’ judg- these multifaceted issues, which entail overlaps be- ments of idealized control (who should control or tween conventional and personal (as well as psy- determine things) and perceived control (who actu- chological and prudential) concerns, have been ally controls or makes decisions) over prudential/ found to lead to adolescent – parent conflict, and conventional, overlapping, and personal issues in conflict in turn has been proposed to expand the relation to psychological symptoms. (Although con- boundaries of adolescents’ personal domains (Sme- ventional and prudential issues differ conceptually, tana & Asquith, 1994; Smetana & Gaines, 1999). The Hasabe et al. found that judgments of parental control research on family decision making (Dornbusch for these two types of issues were not differentiated, et al., 1990; Lamborn et al., 1996) suggests that there and thus they were combined in their analyses.) may be corresponding shifts during these years to- Judgments of idealized control were not significantly ward more adolescent input into family decision associated with psychological symptoms in either making, at least for personal issues.
sample, but as hypothesized, more internalizing It is also unclear whether the patterns of family symptoms were associated with greater perceived decision making associated with better adjustment parental control over personal issues in both Japanese shift over the course of adolescence. Maccoby (1984) and U.S. students and with greater perceived paren- has asserted that healthy autonomy development is a tal control over overlapping issues in Japanese youth.
gradual process entailing parental and child coreg- Thus, as has been found previously (Smetana & ulation in childhood and adolescence, with more Daddis, 2002), Hasabe et al. demonstrated that ado- independent decision making occurring in young lescents viewed greater perceived parental control of adulthood. The previous findings (Dornbusch et al., the personal domain as psychologically intrusive and 1990; Lamborn et al., 1996) are consistent with this that overcontrol of the personal domain had delete- proposition in that youth-alone decision making in- rious psychological effects. However, more research creased with age but was associated with negative is needed to examine how different domains of de- adjustment. However, these studies are limited in cision making relate to both healthy and unhealthy that they included only high school students and focused primarily on family decision making re- The present study examined family decision-mak- garding predominantly personal issues.
ing autonomy for conventional, prudential, multifac- Although there has been little research examining eted, and personal issues in relation to healthy and family decision making associated with early ado- unhealthy adjustment in a sample of middle-class lescent adjustment, there is some evidence that ad- African American families with early adolescents, olescents’ supportive involvement in family decision who were followed longitudinally for 5 years. Al- making also is beneficial in early adolescence. For though most of the research on adolescents’ partici- instance, in a 1-year longitudinal study, Fuligni and pation in family decision making (Dornbusch et al., 1990; Fuligni & Eccles, 1998; Lamborn et al., 1996) has years, SD 5 1.29), nearly evenly divided between focused only on adolescents’ reports, we examined boys and girls, and their mothers (n 5 93). Families both African American adolescents’ and mothers’ were followed for 5 years, with Time 2 occurring 2 perceptions of family decision-making autonomy to years after Time 1 and Time 3 occurring 3 years later examine congruence in different informants’ views, as (5 years after the initial assessment). Attrition be- well as the relation of each to adjustment.
tween Time 1 and Time 2 was 10%, resulting in a Based on previous research (Dornbusch et al., Time 2 sample of 85 families with middle adoles- 1990; Lamborn et al., 1996), we expected that ado- cents, including 83 families with mothers and ado- lescent decision-making autonomy would increase lescents (M 5 15.05 years, SD 5 1.28; 41 males and 42 with age, but we also hypothesized that increases females). Attrition over the 5 years of the study was would vary according to the domain of the issue.
17%, resulting in a Time 3 sample of 76 adolescents Based on the previous research on conceptions of (M 5 18.43 years, SD 5 1.39; 38 males and 38 females) parental authority (Fuligni, 1998; Smetana, 1988, and their mothers. The analyses reported here are 2000) and the desired pacing of behavioral autonomy based on 76 adolescents and their mothers for whom (Feldman & Quatman, 1988; Feldman & Rosenthal, data were available at Time 3. (Five families partici- 1990), we expected that African American adoles- pated at Time 1 and Time 3 but not at Time 2, and cents’ decision-making autonomy would be greater their scores for Time 2 were imputed so as not to for personal issues than for all other issues and for multifaceted than for prudential and conventional In all participating families, both parents were issues that, based on Hasabe et al. (2004), were not Black and nearly all (more than 95%) were African American. At Time 1, mothers were, on average, 40.63 Based on previous research (Hasabe et al., 2004; years of age (SD 5 6.33 ) and had 14.86 years of for- Smetana & Daddis, 2002), we hypothesized that how mal education (SD 5 2.27). Marital status was stable adolescents and mothers draw boundaries over the over the study period for 80% of the sample; 51% of personal domainFand changes in those boundaries the families (39) were stable, two-biological-parent with ageFwould influence adjustment in late ado- families; 10% (7) were stable, stepparent families; and lescence. Previous research (Smetana, 1995b) has 20% (15) were stable, single-parent families (either shown that authoritative parents draw clearer divorced or never married). The frequency of boys boundaries among the domains than do other par- and girls living in stable versus changing family ents; authoritative parents granted adolescents per- structures did not differ significantly.
sonal jurisdiction over personal issues but treated Parent-reported family income was highly stable multifaceted issues as legitimately subject to parental across time, from Time 1 to Time 2, r(70) 5 .88 (alpha authority. In contrast, authoritarian parents were was set at po.05, and thus all results reported as more likely than other parents to view multifaceted significant are po.05 or better), and from Time 1 to and personal issues as legitimately controlled by Time 3, r(69) 5 .79. At Time 3, 16% of the sample parents, whereas permissive parents treated both earned between $25,000 and $40,000/year, 28% multifaceted and personal issues as under adoles- earned between $40,000 and $70,000/year, and 55% cents’ legitimate authority. These findings led us to earned more than $70,000/year, with 30% reporting expect that more parent input into decisions about annual incomes of more than $100,000/year. Fami- multifaceted issues, particularly in early adoles- lies lived in a moderately sized Eastern city and were cence, would predict better adjustment in late ado- nearly evenly split between urban and suburban lescence. We also expected that more adolescent dwellers. Analyses of census tract data indicated that autonomy over personal issues, particularly in late neighborhood composition varied widely for par- adolescence, would be associated with better ad- ticipating families; the percentage of White families justment in late adolescence. Sex differences in in their neighborhoods ranged from 2% to 98%. So- family decision making also were examined, but no cioeconomic status, as assessed by maternal educa- tion and family annual income, was moderatelyassociated with neighborhood composition, rs(73) 5.25 and .27, respectively; families with more educated mothers and higher annual incomes lived in neigh-borhoods with higher percentages of White families.
Adolescents who did not participate at Time 3 did The original sample at Time 1 consisted of 95 not differ significantly at Time 1 from those who African American early adolescents (M 5 13.10 continued their participation on any of the adjust- ment variables (described later), and there were no on the adolescent’s most recent grade point average.
sex differences between adolescents continuing and Their scores were highly correlated, r(71) 5 .63, and discontinuing study participation. Mothers’ age, edu- they were averaged for the analyses. Scores were cational attainment, and family income also did not reverse-scored so that higher scores indicated better differ significantly between attrited and nonattrited Deviance. At each time, adolescents rated their involvement in problem behaviors using the Prob-lem Behavior Survey (PBS; Mason, Cauce, Gonzales, & Hiraga, 1996), a 19-item report of problem be- Using procedures developed by Dornbusch et al.
havior adapted from Jessor and Jessor (1977). In (1985; Dornbusch et al., 1990), adolescents and addition, at Time 1 and Time 2, adolescents also mothers rated family decision making for each of 20 rated how often their friends participated in each issues. There were four conventional items (whether behavior, assessing association with an antisocial to do assigned chores, how to talk to parents, peer group. Adolescents rated items focusing on whether to use manners, and what type of language drug and alcohol use, gang activity, vandalism, to use), four prudential items (whether to smoke stealing, truancy, precocious sexual activity, and cigarettes, drink alcohol, do drugs, and have sex), fighting with or without a weapon on a 7-point scale and four personal items (what time to get up, what ranging from 1 (never happens) to 7 (happens very of- clothes to wear, how to spend free time, and how to ten). Alphas for own behavior were .97 for Time 1, .55 spend allowance money). We combined two types of for Time 2, and .72 for Time 3. The low alpha at Time overlapping items (multifaceted and multifaceted 2 was partly due to lack of variability, as adolescents friendship issues) to form the multifaceted category, in our sample indicated that they never engaged in which included eight items (choosing whether to four of the behaviors. Alphas for their friends’ in- clean bedroom, what TV shows to watch, what volvement in problem behaviors were .93 for Time 1 music to listen to, how late at night to stay out, who and .92 for Time 2. Ratings of own and peers’ in- should be friends, how much time to spend with volvement in problem behavior were positively and significantly correlated, rs(72) 5 .45 and .54 at Time 1 Each issue was rated on a 5-point scale ranging from whether parents decide each issue without At Time 2, adolescents also were administered a discussing it (coded as 5), ask the adolescent’s 20-item inventory assessing the frequency and types opinion but retain the final say (coded as 4), make of alcohol and drug use, including beer, wine, and the decision together (coded as 3), leave the decision different soft and hard drugs (Winters & Henley, up to the child after discussing it (coded as 2), or 1989). Thus, adolescents rated the frequency of use of leave it entirely up to the adolescent (coded as 1).
each of 10 items on a 4-point scale ranging from 0 Responses were reverse-scored for the current anal- (never) to 3 (often) and then indicated how many yses so that higher scores indicated more adolescent times each item was used on a scale from 0 (0) to 3 decision-making autonomy. Mothers’ and adoles- (11 – 19 times). Winters and Henley (1989) originally cents’ mean ratings of family decision making for the proposed a 6-point scale for number of times of use, total set of items were uncorrelated at Time 1 and but we employed a 4-point scale here, as adolescents Time 3, rs(75) 5 .17, .19, ns, and moderately corre- in our sample never endorsed higher values. The lated at Time 2, r(75) 5 .29. Given these findings, we alpha was .91. Adolescents’ reports of alcohol and examined mothers’ and adolescents’ ratings of fam- drug use were significantly associated with their ily decision making separately for conventional, reports of their own and peers’ involvement in prudential, personal, and multifaceted issues. Mean problem behavior, rs(72) 5 .71 and .34 , respectively.
decision-making responses for each rater and for Scores for these measures were standardized, and a mean composite score for deviance was obtained. AtTime 3, only the mean score for the PBS was used.
Means for these measures are in Table 1.
Self-worth. At Time 1 and Time 2, the seven-item Academic performance. At Time 1 and Time 2, global self-worth scale from Harter’s (1982, 1985) parents reported on their adolescents’ grade point Rating Scales of Perceived Competence and at Time average during the current semester on a 5-point 3, the global self-worth scale from Harter’s (1988) scale ranging from 1 (mostly As) to 5 (mostly Fs). At Perceived Competence Scales for Adolescents were Time 3, adolescents and parents separately reported used to assess self-worth. For each item, adolescents Table 1Means and Standard Deviations for Decision-Making Autonomy Note. Decision-making autonomy was scored on a 5-point scale, where higher scores indicate greater autonomy. Subscripts (a vs. b)indicate means that differ significantly.
were first asked to decide which of two statements ministered the questionnaires in separate rooms. To describing other children are more like them, and ensure full comprehension, the interviewers read all then they were asked to indicate whether the state- questionnaire instructions to adolescents and were ment is really true or just sort of true for them. Fol- available to answer their questions.
lowing Harter (1982), responses were scored on a Families were recontacted 2 years later and in- scale ranging from 1 (least competent) to 4 (most vited to participate again in the project. At Time 2, competent). Alphas were .78 for Time 1, .77 for Time 2, families were sent the same set of measures as part of and .80 for Time 3. We obtained a mean score for a larger packet of questionnaires to complete; they were collected during a home (or university) visit.
Depressed mood. At Time 3, adolescents completed Incomplete questionnaires were finished at that time.
the Center for Epidemiologic Studies Depression Parents were contacted by phone to solicit their Scale (CES – D; Radloff, 1977), a well-known 20-item participation at Time 3. In addition, at Time 3, all measure of depressed mood. For each item, adoles- adolescents were contacted individually to invite cents responded about how they felt during the past them to participate. Because many adolescents were week on a scale ranging from 1 (rarely or none) to 4 living away from home and several families had (most or all of the time). Cronbach’s alpha was .89.
moved out of the area, adolescents and parents weremailed questionnaires with instructions for comple-tion and a return envelope to mail them back.
Families were initially recruited through Black churches and Black social and professional organi- zations as part of a larger project on adolescent – Domain Differences and Age-Related Changes in parent relationships. Prospective families were identified through these settings, and their partici-pation was solicited. Because the present study Means and standard deviations for adolescents’ focused on middle-class families, family income and mothers’ ratings of decision-making autonomy (minimum of $25,000/year as assessed in 1995) was over conventional, prudential, multifaceted, and employed as a criterion for participation.
personal issues are shown in Table 1. In the first At Time 1, all questionnaires were completed analysis, we examined hypothesized changes over during a home visit (or in some cases, a visit to the time in adolescents’ and mothers’ ratings of adoles- university) conducted by two African American in- cent decision-making autonomy in the different terviewers. The research was briefly explained to the domains. A 2 (adolescents’ gender)  3 (time)  2 participants, and participating families were ad- (generation: mothers vs. adolescents)  4 (domain: conventional, prudential, multifaceted, and person- tonomy over multifaceted and personal issues al) repeated measures analysis of variance (ANOVA) increased significantly from Time 1 to Time 3. Nev- with gender as the between-subject factor, and time, ertheless, at each wave (and except for ratings of domain, and generation as within-subjects variables conventional issues at Time 1), adolescents rated were performed on the mean ratings of decision- themselves as having more autonomy over each type making autonomy for conventional, prudential, The analyses revealed significant main effects for Associations Between Decision-Making Autonomy in generation, F(1, 73) 5 56.35; time, F(2, 146) 5 48.37; Different Domains and Adolescent Adjustment and domain, F(3, 219) 5 320.11. Adolescents viewedthemselves as having more decision-making auton- The means for the adjustment measures are pre- omy than did mothers. Post hoc Bonferroni t tests sented in Table 2. As adolescents’ and mothers’ rat- indicated that overall, adolescents’ decision-making ings of decision-making autonomy did not differ autonomy increased significantly from Time 1 and significantly for conventional and prudential issues Time 2 to Time 3. Participants rated adolescents as for five of the six comparisons, scores for these issues more autonomous over personal than all other issues were combined for the next set of analyses. Based on and over multifaceted than prudential or conven- Lamborn et al. (1996), we examined concurrent as- tional issues (see Table 1 for means). These findings sociations between family decision-making autonomy were qualified, however, by significant Time  Do- for (combined) prudential/conventional, multifacet- main, F(6, 438) 5 16.92; Generation  Domain, F(3, ed, and personal issues and the different measures of 219) 5 9.09; and Time  Generation  Domain , F(6, adjustment. Bivariate correlations between the four 438) 5 5.42, interactions. Gender was not significant adjustment measures and the ratings of decision- as either a main effect or in interaction with other making autonomy in the different domains were tested separately for mothers and adolescents at each Like Hasabe et al. (2004) and consistent with hy- wave. The results are presented in Table 3.
potheses, we found that adolescents’ and mothers’ As can be seen, at Time 1, African American early ratings of decision-making autonomy over conven- adolescents’ ratings of greater decision-making au- tional and prudential issues did not differ signifi- tonomy over the combined conventional and pru- cantly at each wave except at Time 3, when dential issues were significantly associated with adolescents viewed themselves as more autonomous poorer self-worth, whereas early adolescents’ ratings over prudential than conventional issues. Further- of greater autonomy over multifaceted issues were more, adolescents’ ratings of their decision-making significantly associated with poorer academic per- autonomy over conventional issues and adolescents’ formance, lower self-worth, and greater deviance.
and mothers’ ratings of adolescents’ autonomy over Mothers’ ratings of family decision making were not prudential issues increased significantly from Time 1 significantly associated with any of the adjustment and Time 2 to Time 3, but mothers’ ratings of ado- lescents’ decision-making autonomy over conven- At Time 2, African American middle adolescents tional issues did not increase significantly over time.
who rated themselves as having more autonomy In contrast, adolescents’ and mothers’ ratings of au- over conventional and prudential issues also reported Table 2Means and Standard Deviations for Adjustment Variables Note. GPA was scored on a 5-point scale where 5 5 mostly As. At Time 3, current GPA 5 parents’ report. Problem behavior was scored on a7-point scale where 7 5 happens very often. Drug and alcohol use were rated on a 5-point scale where 0 5 never and 4 5 used often.
Table 3Associations Among Adolescent Adjustment, Decision-Making Autonomy, and Sociodemographic Background Note. Worth 5 self-worth, neighborhood 5 neighborhood composition (% White).
more deviance. However, when mothers’ ratings tional (or prudential issues) and personal issues, we were considered, deviance was greater among ado- expect multifaceted issues to be significantly associ- lescents whose mothers viewed them as having more ated with both the combined prudential and con- autonomy over multifaceted and personal issues.
ventional issues and personal issues, but ratings of At Time 3, adolescents’ ratings of more decision- the combined prudential and conventional issues making autonomy over multifaceted issues were should not be strongly associated with personal is- significantly associated with greater deviance and sues. Table 4 shows this expected pattern.
more depressed mood. More autonomy over per- The final set of analyses focused on predicting sonal issues was marginally associated with better African American adolescents’ adjustment at Time 3 academic performance and less depressed mood.
from adolescents’ (and mothers’) earlier ratings of Mothers’ ratings of family decision making were not their decision-making autonomy. We used hierar- significantly associated with any of the adjustment chical regressions to examine the hypothesis that adolescents’ and mothers’ boundaries over the per-sonal domainFand changes in those boundaries (asassessed in terms of family decision-making autono- Longitudinal Influence of Autonomy on African my)Fwould influence adjustment at Time 3. All analyses controlled for the effects of adolescents’ sex First, we examined stability in domain ratings (girls coded as 1 and boys coded as 0). As in previous over time and cross-domain associations. Because studies (Dornbusch et al., 1990; Lamborn et al., 1996), few associations were obtained between mothers’ we controlled for the effects of adolescents’ age and ratings of adolescents’ decision-making autonomy family socioeconomic status, using mothers’ educa- and adolescent adjustment, only the correlations for tion as a proxy for socioeconomic status. Because adolescents’ ratings of decision-making autonomy Lamborn et al. (1996) found that adjustment for Af- are presented in Table 4. However, analyses of the rican American youth varied as a function of influence of maternal reports of adolescents’ deci- whether they were growing up in predominantly sion-making autonomy on adjustment also were White communities , we controlled for the effects of performed; the results are summarized briefly , and neighborhood composition, defined here as the per- interested readers may contact the first author for a centage of White families in their census tract. In addition, Collins, Maccoby, Steinberg, Hetherington, As expected, adolescents’ ratings of decision- and Bornstein (2000) have asserted that in examining making autonomy in each domain were moderately outcomes, it is important to control for preexisting to strongly stable over time. Because multifaceted differences among adolescents. Therefore, in each of items by definition entail overlaps between conven- the analyses of Time 3 adjustment (except depressed Table 4Associations Among Adolescents’ Ratings of Decision-Making Autonomy wpo.10. Ãpo.05. ÃÃpo.01. ÃÃÃpo.001.
mood, which was not assessed before Time 3), we the analyses of maternal reports of adolescent deci- controlled for Time 1 levels of the variable so that the sion-making autonomy; mothers’ ratings of adoles- analyses focused on change in the dependent varia- cents’ decision-making autonomy over multifaceted ble over time. All control variables were included in and personal issues did not significantly predict the first step of the analyses. Then, decision-making autonomy at Time 1 was added in the second step.
Academic performance. Adolescent gender and Two successive steps followed, first adding Time 2 neighborhood composition significantly influenced and then adding Time 3 decision-making autonomy.
late adolescents’ academic performance. Academic These latter steps assessed the influence of change performance was higher in girls than in boys, and over time in autonomy on later adjustment.
controlling for socioeconomic status, early adoles- The analyses were performed separately for each cents living in neighborhoods with more White of the four adjustment measures (deviance, academic families had better academic performance 5 years performance, self-worth, and depression). The anal- later than did adolescents living in more integrated yses also were performed separately on decision- neighborhoods. Not surprising, better academic making autonomy over multifaceted, personal, and performance in late adolescence was also predicted (combined) conventional and prudential issues. (Al- by better academic performance in early adoles- though the bivariate correlations indicated that au- cence. Together, these variables accounted for 23% of tonomy over personal and multifaceted issues were the variance. After controlling for these variables, significantly and positively associated, we examined neither adolescents’ reports of decision-making au- them separately to determine whether they had dif- tonomy over personal or multifaceted issues nor ferential effects on adjustment.) As the analyses changes in adolescents’ ratings of autonomy over yielded no significant effects for decision-making these issues significantly influenced African Ameri- autonomy over (combined) prudential and conven- can adolescents’ academic performance at Time 3.
tional issues, these results are not discussed further.
Similar results were obtained in the analyses of ma- The results of the analyses for adolescent-rated au- ternal reports; background variables were significant tonomy over multifaceted issues are presented in in the analyses, but maternal reports of decision- Table 5, and the results for adolescent-rated autono- my over personal issues are presented in Table 6.
Self-worth. In the analysis of adolescent-reported Deviance. As can be seen in Tables 5 and 6, neither decision-making autonomy over personal issues, adolescent-reported decision-making autonomy nor adolescents’ age and, in both analyses, Time 1 self- changes in adolescent-reported autonomy signifi- worth had significant influences on self-worth in late cantly influenced deviance in late adolescence. Fur- adolescence. Younger adolescents and adolescents thermore, deviance in early adolescence did not who reported better self-worth at Time 1 reported significantly predict later deviance. Only adoles- greater self-worth at Time 3. In both analyses there cents’ gender was significant in these analyses; in were trends indicating that African American ado- late adolescence, there was more deviance among lescents who lived in neighborhoods with fewer boys than girls. The same results were obtained in White families and who had more educated mothers Table 5Regressions of Late Adolescent Adjustment on Decision-Making About Multifaceted Issues Note. AutM 5 autonomy over multifaceted issues. Betas are for the final step in the model.
wpo.10. Ãpo.05. ÃÃpo.01. ÃÃÃpo.001.
had better self-worth at Time 3. Controlling for these In the analyses using maternal reports, neither effects, the analyses revealed that increased decision- mothers’ ratings of adolescents’ decision-making making autonomy over personal and multifaceted autonomy over multifaceted issues nor changes in issues from Time 2 to Time 3 led to better self-worth those ratings over time had significant effects on in late adolescence. Increases in autonomy from adolescents’ self-worth. However, with background Time 2 to Time 3 accounted for 6% of the variance variables and Time 1 self-worth controlled, increas- (total model R2 5 .19) in the analysis of multifaceted es between Time 1 and Time 2 in mothers’ ratings issues and 13% of the variance (total model R2 5 .21) of adolescents’ decision-making autonomy over personal issues led to better self-worth in late Table 6Regressions of Late Adolescent Adjustment on Decision-Making About Personal Issues Note. AutP 5 autonomy over personal issues. Betas are for the final step in the model.
wpo.10. Ãpo.05. ÃÃpo.01. ÃÃÃpo.001.
adolescence, b 5 .32, F(8, 66) 5 2.01 (total model cused on what we considered here to be multifaceted and personal issues. Adolescents also must become Depression. Finally, none of the demographic autonomous regarding conventional and prudential background variables had a significant effect on late (as well as moral) rules. Indeed, understanding the adolescent depression. Adolescents’ ratings of au- bases for social rules and expectations, having a set tonomy over personal issues at Time 1 had a signif- of principles regarding right and wrong, and resist- icant influence on depression at Time 3, with more ing pressures to go along with others’ demands have decision-making autonomy over personal issues in been defined as the hallmarks of autonomous func- early adolescence predicting greater depression 5 tioning in the moral and conventional domains years later. A similar trend was found in the analysis (Steinberg, 2002). However, the analyses indicated of maternal reports. Mothers’ ratings of adolescents’ that parent involvement in decision making is con- autonomy over personal issues at Time 1 had a sistently greater across adolescence for conventional marginally significant influence on adolescent de- and prudential issues than for other issues and, pression at Time 3; mothers who viewed their early conversely, that adolescent autonomy is greater for adolescents as having more decision-making auton- omy over personal issues had adolescents whotended to report more depression 5 years later, Domain-Differentiated Autonomy and African American Furthermore, in the analyses of adolescents’ rat- ings of both multifaceted and personal issues, in- It has been proposed that parent-unilateral deci- creases in decision-making autonomy from Time 2 to sion making is beneficial primarily because it protects Time 3 led to less depression at Time 3. Increases in poor African American youth from the high-risk en- autonomy from Time 2 to Time 3 accounted for 14% vironments and the dangerous neighborhoods where of the variance in depressed mood (total model poor families typically reside (Baldwin et al., 1990; R2 5 .19) in the analysis of multifaceted issues but 8% Kelley, Power, & Wimbush, 1992). Lamborn et al.
of the variance (total model R2 5 .18) in the analysis (1996), however, found that parent-unilateral deci- of personal issues. Similar findings were not ob- sion making (where parents make decisions without tained the analyses of mothers’ ratings, however.
adolescent input) predicted positive adjustmentamong middle-class African American youth as well,perhaps because middle socioeconomic status does not protect African American adolescents from the The present study examined changes from early to pervasive risks of racism and prejudice (Spencer & late adolescence in African American adolescents’ Dupree, 1996). In the present study, parent decision decision-making autonomy regarding prudential, making (either alone or with adolescent input) was conventional, multifaceted, and personal issues.
normative in our African American middle-class Healthy family decision making was identified here sample, but only for conventional and prudential is- by examining concurrent associations and longitu- sues (and was reported more frequently by mothers dinal influences of autonomous decision making on four measures of psychosocial adjustment: deviance; Moreover, our results are consistent with Hasabe academic performance; self-worth; and at Time 3, et al. (2004) in showing that except for late adoles- depression. The results indicated that healthy deci- cents, adolescents and mothers did not differentiate sion-making autonomy differs by domain and shifts between conventional and prudential issues in their ratings of family decision-making autonomy, al- Although there have been significant theoretical though prudential and conventional issues have advances in defining the different components of been found to differ conceptually on a variety of adolescent autonomy, the assessment of autonomy other measures. That late adolescents viewed them- has not kept pace. In the present study, we drew on selves as having more decision-making autonomy social domain theory (Nucci, 2001; Smetana, 1995a, over prudential than conventional issues is consist- 2002; Turiel, 1998) to examine distinctions among ent with some research indicating that high school conventional, prudential, multifaceted, and personal students, and particularly adolescents who engage in issues in family decision-making autonomy. In most more drug use, view prudential issues pertaining to research on adolescent autonomy, the items used to risk behaviors (such as drug and alcohol use) as assess behavioral autonomy have been chosen on an under personal jurisdiction, even though they may ad hoc basis, although most of this research has fo- view such behaviors as foolish or harmful to the self (Nucci, Guerra, & Lee, 1991). Consistent with Lam- The results of the present study indicated that born et al. (1996), we found that adolescents’ reports having limited autonomy (with parental guidance) of more parental decision making over conventional in early adolescence over a clearly specified set of and prudential issues was associated concurrently personal issues (such as choice of music, clothes, and with better self-worth in early adolescence and hairstyles, and how to spend money) was not sig- less deviance in middle adolescence, but in both nificantly associated with adjustment (although the adolescents’ and mothers’ reports, parental deci- correlation between early adolescents’ ratings of sion making over prudential and conventional personal issues and deviance approached signifi- issues was not associated concurrently with adjust- cance). In contrast, mothers’ ratings of middle ado- ment in late adolescence. Thus, the findings sug- lescents’ greater autonomy over personal issues was gested that for middle-class African American associated concurrently with greater deviance. Even youth, healthy adjustment was associated with con- for personal issues, however, decision making was tinued parental decision making, at least through described as youth supported (where adolescents middle adolescence, for prudential and conventional make decisions after seeking parental advice) rather than youth alone (where adolescents make decisions The analyses indicated that there were significant without parental involvement). The longitudinal increases from early to late adolescence in autono- finding (obtained in analyses of both adolescents’ mous decision making over multifaceted issues, but and mothers’ ratings) that greater autonomy over on average, and even in early adolescence, both ad- personal issues in early adolescence was associated olescents and mothers reported that decisions about with more depression in late adolescence is consist- these issues were jointly made rather than decided ent with previous research, which has consistently by parents, with a greater tilt over time toward ad- indicated that too much youth-alone decision mak- olescent decisions with parental guidance. Further- ing has negative effects on adjustment (Dornbusch more, the results of the concurrent analyses with et al., 1990; Lamborn et al., 1996).
adjustment indicated that more adolescent-reported The results also indicated that there were signifi- autonomous decision making regarding multifacet- cant increases over time in adolescents’ reports of ed issues was associated with poorer adjustment, their autonomous decision making (adolescent de- including lower school grades, greater deviance, and cisions with parental input) over personal issues poorer self-worth in early adolescence and with between early and late adolescence. Although more deviance and more depression in late adoles- mothers viewed parents as having more input into cence. Although adolescent-rated autonomy over decision making regarding personal issues than ad- multifaceted issues was not significantly associated olescents acknowledged, mothers likewise viewed with adjustment in middle adolescence, maternal decision making over these issues as shifting from ratings of greater adolescent autonomy over these jointly made to adolescent-made decisions with issues were associated with more deviance in middle parental guidance. Therefore, these findings are consistent with Maccoby’s (1984) theorizing that Previous research with European American fami- parental guidance and parent – adolescent coregula- lies has indicated that authoritative parents are dis- tion may facilitate healthy development, at least until tinguished from other parents primarily in their adolescents have acquired the competence to make belief that parents should retain legitimate parental mature and safe decisions about these issues.
authority over multifaceted issues but grant adoles- It has been proposed that defining a domain of cents personal jurisdiction over personal issues personal discretion is necessary for psychological (Smetana, 1995b). The present findings are consistent well-being and the healthy development of self and with this previous research but extend the findings identity (Nucci, 1996, 2001; Smetana, 2002). Never- by suggesting that granting adolescents too much theless, as with multifaceted issues, the means sug- autonomy over multifaceted issues too early in de- gested that on average, African American families velopment (in early adolescence, according to ado- granted adolescents some personal jurisdiction, but lescent reports, and in middle adolescence, according in the context of parental input and guidance. This to maternal reports) is associated with poorer ad- may seem to contradict the notion that having a justment. The present findings indicate the impor- domain of personal freedom is a psychological need tance for African American adolescents’ psychosocial that is essential for the development of agency and adjustment of continued parental involvement across effectance (Nucci, 1996, 2001), but in our view, it does adolescence (in the context of joint decision making) not. Previous research on the personal domain has in decisions over multifaceted issues.
focused primarily on children’s and adolescents’ assertions of the legitimacy of personal jurisdiction, pression) for African American youth. Some caution whereas the present research focuses on how those in interpreting the findings for depression is war- needs are negotiated in the context of family decision ranted, however. Because depressed mood was as- making. Changes over time in family decision mak- sessed only at Time 3, we could not control for earlier ing over personal issues may be due to the interact- levels, and thus we cannot be certain whether these ing influences of adolescents’ expressed desires for findings reflect changes in family decision making or greater personal jurisdiction over personal issues preexisting differences in depression (Collins et al., (Smetana, 2000), parents’ (culturally influenced) be- liefs about the appropriate timing for granting The present results are consistent with Hasabe autonomy (Daddis & Smetana, 2004), and parents’ et al. (2004) in demonstrating the importance of some assessment of adolescents’ abilities and competence adolescent autonomy over personal issues for ado- to assume more privileges and responsibilities. For lescent development, but our findings extend their African American parents, judgments about adoles- research in two ways. Hasabe et al. found that pa- cents’ decision-making competence may be based on rental overcontrol of the personal domain was as- their assessments of adolescents’ developing matu- sociated with more internalizing symptoms, whereas rity, as well as their awareness of the everyday re- our findings demonstrate that autonomy over per- alities of living in a social environment where racism sonal issues may also influence more positive aspects and discrimination remain pervasive (Spencer & of adjustment (e.g., better self-worth). However, our Dupree, 1996) and, thus, where the consequences of findings also highlight the need for considering these rash or immature decisions may loom large (Boyd- findings within a developmental framework that Franklin & Franklin, 2000). In such a context, pa- specifies the developmental periods when expansion rental involvement and guidance in decision making of the boundaries of the personal domain may be over personal issues may be very important for particularly important. The present study indicated that greater autonomy over multifaceted and per-sonal issues had negative consequences in early ad-olescence, whereas increases in the boundaries of Longitudinal Influence of Decision-Making Autonomy on personal issues in middle to later in adolescence fa- cilitated better adjustment, as assessed using multi- The longitudinal analyses indicated that adoles- ple indexes. Finally, previous theorizing and cents’ developing autonomy over personal and research has described psychological control and multifaceted issues, as well as changes over time in overcontrol as particularly relevant for self and those variables, significantly influenced self-system identity development (Barber, 1996, 2002; Nucci, processes (self-worth and depression) in late ado- 1996; Smetana & Daddis, 2002). Our findings are lescence. Controlling for background factors and consistent with this assertion in that changes in de- prior self-worth, increases in African American ad- cision-making autonomy over personal and multi- olescents’ ratings of autonomy over multifaceted and faceted issues influenced self-worth and depression personal issues from middle to late adolescence (both self-system processes), but not deviance or predicted both better self-worth and less depression in late adolescence. Moreover, similar findings for Although decision-making autonomy (or changes self-worth were obtained in maternal reports of ad- in autonomy over time) did not predict either devi- olescents’ autonomy over personal issues, except ance or academic performance in the longitudinal that significant increases in autonomy were found at analyses, decision-making autonomy was signifi- earlier ages (e.g., between early and middle adoles- cantly associated concurrently with both of these cence). The concurrent analyses indicated that low adjustment outcomes. Like a great deal of other re- levels of (adolescent-reported) autonomy over mul- search, we found that academic performance was tifaceted issues were associated with better self- higher among girls than among boys, and late ado- worth in early adolescence, whereas the longitudinal lescent boys engaged in more deviance than did analyses revealed that more autonomy over multi- girls. Overall, however, the rate of problem behavior faceted (but not personal) issues in early adolescence in this middle-class African American sample was predicted more depressed mood 5 years later. Taken low. This, as well as the lack of stability in deviance together, these findings suggest that less autonomy over time (as indicated by the failure of Time 1 de- in early adolescence, coupled with increasing au- viance to predict Time 3 deviance) may have ac- tonomy in middle to late adolescence, predicts counted for the nonsignificant findings in the optimal adjustment (better self-worth and less de- Furthermore, our findings for neighborhood also included mothers’ perceptions of family deci- composition differed from Lamborn et al. (1996), sion making and found that African American who found that youth-alone decision making had a mothers consistently viewed parents as having more more negative effect for African American youth input into family decision making than did adoles- living in primarily White communities than for Af- cents. Previous research has shown that adolescents rican American youth living in ethnically mixed and parents have differing views of family relation- communities. Controlling for mothers’ education (as ships (Noller & Callan, 1986; Smetana, 1989). Ado- a proxy for socioeconomic status), we found that lescents’ desires for greater autonomy may lead African American adolescents living in neighbor- them to overestimate their input into family decision hoods with more White families had better academic making, whereas parents’ greater investment and performance. This could be because in our study, ‘‘generational stake’’ in the family (Noller, 1994) may White communities were more likely to be in sub- lead to underestimates of adolescents’ role.
urbs, with better quality schools. Our findings also Although the present data cannot shed light on could reflect peer group influences on academic whose views of family decision making were more achievement. Steinberg, Dornbusch, and Brown accurate, it is possible that accuracy varies according (1992) reported that peer support for academic suc- to social-cognitive domain. Although parents may cess is very limited for high achieving African set clear expectations for adolescents’ behavior re- American youth. African American adolescents who garding prudential issues of risk (such as drug and go to primarily White schools may escape the con- alcohol use or having sex), adolescents spend in- flict that high-achieving African American students creasing amounts of time away from parents and in often feel between doing well in school and being the company of peers. As adolescents may not fully popular with their African American peers (Ford- disclose their involvement in risk behaviors to par- ham & Ogbu, 1986). However, it is notable that the ents (Kerr & Stattin, 2000), they may have greater analyses for self-worth yielded trends in the opposite decision-making autonomy over these issues than direction. Better self-worth was associated with liv- parents are willing to acknowledge. Likewise, par- ing in neighborhoods with fewer White families.
ents also may overestimate their involvement in Establishing a strong ethnic identity may be easier decisions over personal issues, as many of these for African American adolescents living in more in- decisions (such as how adolescents spend their al- tegrated neighborhoods, and being more ethnically lowance) may occur beyond parental purview. Ac- identified consistently has been linked to better self- curacy may be greater over multifaceted issues, esteem (Martinez & Dukes, 1997; Phinney & Chavira, which are frequent sources of conflict and disagree- ment in parent – adolescent relationships (Fuligni,1998; Smetana, 1989; Smetana & Gaines, 1999) andthus may be more actively negotiated and debated in Study Limitations and Directions for Future Research the family context. A strength of the present study In assessing decision making as a continuous was that we obtained both adolescents’ and mothers’ variable, the present study differed from previous ratings of family decision making. Associations be- research, which has examined decision making in tween the informants were weak, but similar pat- terms of the percentages of responses endorsing terns of results for adolescents’ and mothers’ ratings parent-unilateral, joint, and youth-alone decision were found in the longitudinal analyses, although making (Dornbusch et al., 1990; Lamborn et al., the findings were more robust for adolescents’ than 1996). Although the method used here has some mothers’ ratings of decision-making autonomy.
limitations in that we could not determine the exact These findings should not be dismissed as merely percentages of these different types of family deci- reflecting an informant bias, as the assessment of sion making, our strategy provided a parsimonious academic performance was based on parent, not approach that eliminates statistical dependencies in adolescent, report. It is interesting that significant the assessment of different decision making styles.
concurrent associations between mothers’ ratings of Although family decision making reveals how family decision making and adjustment occurred families communicate, interact, and solve problems only for middle-adolescent deviance, as deviance together (Grotevant & Cooper, 1986; Lamborn et al., normatively increases in middle adolescence (Jessor 1996), previous research has assessed family decision & Jessor, 1977; Mason et al., 1996). Although in- making almost exclusively in terms of adolescents’ cluding mothers was a significant advance over reports (Dornbusch et al., 1990; Fuligni & Eccles, previous research, future research should include 1993; Lamborn et al., 1996). In the present study, we Although our study was limited by a relatively Dornbusch, S. M., Carlsmith, J. M., Bushwall, S. J., Ritter, small sample size, our findings extend previous re- P. L., Leiderman, H., Hastorf, A. H., & Gross, R. T. (1985).
search (Dornbusch et al., 1990; Fuligni & Eccles, 1993; Single-parents, extended households, and control of Lamborn et al., 1996) by examining family decision adolescents. Child Development, 56, 326 – 341.
making longitudinally over 5 years, from early to late Dornbusch, S. M., Ritter, P. L., Mont-Reynaud, R., & Chen, adolescence, with relatively low levels of attrition.
Z. (1990). Family decision-making and academic per-formance in a diverse high school population. Journal of Furthermore, our longitudinal analyses provided a Adolescent Research, 5, 143 – 160.
stringent test of our hypotheses in that we controlled Feldman, S., & Quatman, T. (1988). Factors influencing age for several background variables. The results indi- expectations for adolescent autonomy: A study of early cated that adolescents’ ratings of family decision adolescents and parents. Journal of Early Adolescence, 8, making over personal and multifaceted issues had robust longitudinal effects on adjustment. Future Feldman, S. S., & Rosenthal, D. A. (1990). The acculturation research should examine shifts in family decision of autonomy expectations in Chinese high schoolers making in the transition to young adulthood to ex- residing in two Western nations. International Journal of amine how healthy behavioral autonomy in different Fordham, S., & Ogbu, J. U. (1986). Black students’ school success: Coping with the burden of ‘‘acting White’’.
Urban Review, 18, 176 – 206.
Fuligni, A. J. (1998). Authority, autonomy, and parent – adolescent conflict and cohesion: A study of adolescentsfrom Mexican, Chinese, Filipino, and European back- Allen, J., Hauser, S. T., Bell, K., & O’Connor, T. (1994).
grounds. Developmental Psychology, 34, 782 – 792.
Longitudinal assessment of autonomy and relatedness Fuligni, A. J., & Eccles, J. (1993). Perceived parent – child in adolescent – family interactions as predictors of ado- relationships and early adolescents’ orientations toward lescent ego development and self-esteem. Child Devel- peers. Developmental Psychology, 29, 622 – 632.
Garcia Coll, C. G., Lamberty, G., Jenkins, R., McAdoo, Baldwin, C., Baldwin, A., & Cole, R. (1990). Stress-resistant H. P., Crnic, K., Wasik, B. H., et al. (1996). An inte- families and stress-resistant children. In J. Rolf, A.
grative model for the study of developmental compe- Masten, D. Cicchetti, K. Neuchtherlin, & S. Weintraub tencies in minority children. Child Development, 67, (Eds.), Risk and protective factors in the development of psychopathology (pp. 257 – 280). New York: Cambridge Graham, S. (1992). Most of the subjects were White and middle class: Trends in published research on African- Barber, B. K. (1996). Parental psychological control: Revis- Americans in selected APA journals, 1970 – 1989. Amer- iting a neglected construct. Child Development, 67, Grotevant, H. D., & Cooper, C. R. (1986). Individuation in Barber, B. K. (Ed.). (2002). Intrusive parenting: How psycho- family relationships. Human Development, 29, 82 – 100.
logical control affects children and adolescents. Washington, Harter, S. (1982). The Perceived Competence Scale for DC: American Psychological Association.
Children. Child Development, 53, 87 – 97.
Boyd-Franklin, N., & Franklin, A. J. (2000). Boys into men: Harter, S. (1985). Manual for the Self-Perception Profile for Raising our African American teenage sons. New York: Children. Unpublished manual, University of Denver.
Harter, S. (1988). The Self-Perception Profile Scale for Adoles- Brody, G. H., Moore, K., & Glei, D. (1994). Family processes cents. Unpublished manual, University of Denver.
during adolescence as predictors of parent – young adult Hasabe, Y., Nucci, L., & Nucci, M. S. (2004). Parental con- attitude similarity. Family Relations, 43, 369 – 373.
trol of the personal domain and adolescent symptoms of Collins, W. A., Gleason, T., & Sesma, A. Jr. (1997). Inter- psychopathology: A cross-national study in the United nalization, autonomy, and relationships: Development States and Japan. Child Development, 75, 815 – 828.
during adolescence. In J. E. Grusec & L. Kuczynski Hill, J. P., & Holmbeck, G. N. (1986). Attachment and (Eds.), Parenting and the internalization of values (pp. 78 – autonomy during adolescence. Child Development, 57, Collins, W. A., Maccoby, E. E., Steinberg, L., Hetherington, Jessor, R., & Jessor, S. L. (1977). Problem behavior and psy- E. M., & Bornstein, M. H. (2000). Contemporary research chosocial development: A longitudinal study of youth. New on parenting: The case for nature and nurture. American Kelley, M. L., Power, T. G., & Wimbush, D. D. (1992). De- Daddis, C., & Smetana, J. G. (2004). Middle class African terminants of disciplinary practices in low-income Black American families’ expectations for adolescents’ behavioral mothers. Child Development, 63, 573 – 582.
autonomy. Unpublished manuscript, University of Kerr, M., & Stattin, H. (2000). What parents know, how they know it, and several forms of adolescent adjustment: Further support for a reinterpretation of monitoring.
of development: One mind, many mentalities. In R. M.
Developmental Psychology, 36, 366 – 380.
Lerner (Vol. Ed.) & W. Damon (Series Ed.), Handbook of Lamborn, S. D., Dornbusch, S. M., & Steinberg, L. (1996).
child psychology. Vol. 1: Theoretical models of human devel- Ethnicity and community context as moderators of the opment (5th ed., pp. 865 – 937). New York: Wiley.
relations between family decision making and adoles- Silverberg, S. B., & Gondoli, D. M. (1996). Autonomy in ad- cent adjustment. Child Development, 67, 283 – 301.
olescence: A contextualized perspective. In G. R. Adams, Maccoby, E. E. (1984). Middle childhood in the context of R. Montemayor, & T. P. Gullota (Eds.), Psychosocial de- the family. In W. A. Collins (Ed.), Development during velopment during adolescence: Progress in developmental middle childhood: The years from six to twelve (pp. 184 – contextualism (pp. 12 – 61). Thousand Oaks, CA: Sage.
239). Washington, DC: National Academy Press.
Smetana, J. G. (1988). Adolescents’ and parents’ con- Martinez, R., & Dukes, R. (1997). The effects of ethnic ceptions of parental authority. Child Development, 59, identity, ethnicity, and gender on adolescent well-being.
Journal of Youth and Adolescence, 26, 503 – 516.
Smetana, J. G. (1989). Adolescents’ and parents’ reasoning Mason, C. A., Cauce, A. N., Gonzales, N., & Hiraga, Y.
about actual family conflict. Child Development, 60, (1996). Neither too sweet nor too sour: Problem peers, maternal control, and problem behavior in Afri- Smetana, J. G. (1995a). Morality in context: Abstractions, can American adolescents. Child Development, 67, ambiguities, and applications. In R. Vasta (Ed.), Annals of child development, Vol. 10 (pp. 83 – 130). London: McLoyd, V. (1998). Changing demographics in the Amer- ican population: Implications for research on minority Smetana, J. G. (1995b). Parenting styles and conceptions of children and adolescents. In V. C. McLoyd & L. Stein- parental authority during adolescence. Child Develop- berg (Eds.), Studying minority adolescents: Conceptual, methodological, and theoretical issues (pp. 3 – 28). Mahwah, Smetana, J. G. (2000). Middle-class African American ad- olescents’ and parents’ conceptions of parental authority Noller, P. (1994). Relationships with parents in adolescence: and parenting practices: A longitudinal investigation.
Process and outcome. In R. Montemayor, G. R. Adams, Child Development, 71, 1672 – 1686.
& T. P. Gullotta (Eds.). Advances in adolescent development.
Smetana, J. G. (2002). Culture, autonomy, and personal Vol. 6: Personal relationships during adolescence (pp. 37 – jurisdiction in adolescent – parent relationships. In H. W. Reese & R. Kail (Eds.), Advances in child develop- Noller, P., & Callan, V. J. (1986). Adolescent and parent ment and behavior, Vol. 29 (pp. 51 – 87). New York: perceptions of family cohesion and adaptability. Journal Smetana, J. G., & Asquith, P. (1994). Adolescents’ and Nucci, L. P. (1996). Morality and personal freedom. In E. S.
parents’ conceptions of parental authority and adoles- Reed, E. Turiel, & T. Brown (Eds.), Values and knowledge cent autonomy. Child Development, 65, 1147 – 1162.
(pp. 41 – 60). Mahwah, NJ: Erlbaum.
Smetana, J. G., & Daddis, C. (2002). Domain-specific an- Nucci, L. P. (2001). Education in the moral domain. Cam- tecedents of psychological control and parental moni- bridge, England: Cambridge University Press.
toring: The role of parenting beliefs and practices. Child Nucci, L. P., Guerra, N., & Lee, J. (1991). Adolescent judgments of the personal, prudential, and normative Smetana, J. G., & Gaines, C. (1999). Adolescent – parent aspects of drug usage. Developmental Psychology, 27, conflict in middle-class African American families. Child Oyserman, D. M., Coon, H. M., & Kemmelmeier, M. (2002).
Spencer, M., & Dornbusch, S. M. (1990). Challenges in Rethinking individualism and collectivism: Evaluation studying minority youth. In S. S. Feldman & G. R. Elliot of theoretical assumptions and meta-analyses. Psycho- (Eds.), At the threshold: The developing adolescent (pp. 123 – 146). Cambridge, MA: Harvard University Press.
Parke, R. D., & Buriel, R. (1998). Socialization in the family: Spencer, M., & Dupree, D. (1996). African American Ethnic and ecological perspectives. In N. Eisenberg (Vol.
youths’ ecocultural challenges and psychosocial oppor- Ed.) & W. Damon (Series Ed.), Handbook of child psy- tunities: An alternative analysis of problem behavior chology. Vol. 3: Social, emotional, and personality develop- outcomes. In D. Cicchetti & S. L. Toth (Eds.), Adolescence: ment (5th ed., pp. 463 – 552). New York: Wiley.
Opportunities and challenges (pp. 259 – 282). Rochester, Phinney, J., & Chavira, V. (1992). Ethnic identity and self- esteem: An exploratory longitudinal study. Journal of Steinberg, L. (1990). Interdependency in the family: Au- tonomy, conflict, and harmony in the parent – adolescent Radloff, L. (1977). The CES – D Scale: A self-report de- relationship. In S. S. Feldman & G. R. Elliot (Eds.), At the pression scale for research in the general population.
threshold: The developing adolescent (pp. 255 – 276). Cam- Applied Psychological Measurement, 1, 385 – 401.
bridge, MA: Harvard University Press.
Shweder, R. A., Goodnow, J. J., Hatano, G., LeVine, R. A., Steinberg, L. (2002). Adolescence (6th ed.), Boston: McGraw- Markus, H., & Miller, P. (1998). The cultural psychology Steinberg, L., Dornbusch, S. M., & Brown, B. (1992). Ethnic Winters, K. C., & Henley, G. G. (1989). Personal Experience differences in academic achievement. American Psychol- Inventory Test and manual. Los Angeles: Western Psy- Turiel, E. (1998). Moral development. In N. Eisenberg (Vol.
Zimmer-Gembeck, M. J., & Collins, W. A. (2003). Autono- Ed.) & W. Damon (Series Ed.), Handbook of child psy- my development during adolescence. In G. R. Adams & chology, Vol. 3: Social, emotional, and personality develop- M. Berzonsky (Eds.), Blackwell handbook of adolescence ment (5th ed., pp. 863 – 932). New York: Wiley.
(pp. 175 – 204). Oxford, England: Blackwell.

Source: http://www.marpsy.net/bio/chris/pubs/defining%20healthy%20autonomy.pdf