The Consequences of Friend-Related Stress Across Early Adolescence

Abstract

The current study investigated early adolescents’ experiences of friend-related stress across middle school and its developmental consequences following the transition to high school. Using a sample of approximately 1,000 middle school students, four unique friend-related stress trajectories were observed across middle school: consistently low friend-related stress (57% of the sample), consistently high friend-related stress (7%), moderate and increasing friend-related stress (22%), and moderate but decreasing friend-related stress (14%). Groups characterized by higher levels of friend-related stress across middle school were linked to subsequent poorer socioemotional well-being, lower academic engagement, and greater involvement in and expectancies around risky behaviors following the transition to high school. Increased friend-related stress across the high school transition was also linked to poorer outcomes, even after taking into account earlier stress trajectories. Gender differences highlighted the particular struggles girls experience both in friend stress and in the links between friend stress and subsequent well-being.

Keywords

friend stress school transitions socioemotional well-being engagement alcohol use

The Consequences of Friend-Related Stress Across Early Adolescence

Across the life course, individuals have a fundamental human need to belong, and thus forming and maintaining relationships with close others is a persistent drive (Baumeister & Leary, 1995). Challenges in interpersonal relationships, however, can result in psychological stress, which “involves a particular relationship between the person and the environment that is appraised by the person as taxing or exceeding his or her resources and endangering his or her well-being” (Lazarus & Folkman, 1984, p. 19). During childhood, parents play the primary socialization role in their children’s lives, but as young people enter and move through adolescence, interpersonal relationships shift. Peer relationships take on added significance, and these shifts in interpersonal ties can be a source of stress in adolescents’ lives (Smetana, Campione-Barr, & Metzger, 2006). In the current study, we focus on friend-related stress, investigating the extent to which stress in relationships with close friends accumulates (or not) across early adolescence and its subsequent impact on young people’s health and well-being.

Friendship and Early Adolescents’ Development

As young people move from childhood into adolescence, friends play an increasingly important role in their daily lives. The majority of adolescents’ waking hours are spent in schools surrounded by same-age peers (Rutter, Maughan, Mortimore, Ouston, & Smith, 1979), and their after-school time also is frequently spent in the company of other adolescents (Ferrar, Chang, Li, & Olds, 2013). As compared with childhood, the friendships of early adolescents are characterized by greater intimacy (Newcomb & Bagwell, 1995). Thus, it is unsurprising that a multitude of studies have examined the links between friendship and adolescents’ well-being.

Such scholarship has focused on the importance of friendship quality and support as well as the challenges of friendship conflict. In general, friendship quality is positively linked to mental health and well-being. For example, young people who report greater friendship quality and support are more likely to experience fewer internalizing behaviors, less loneliness, and higher levels of self-worth (Lodder, Scholte, Goossens, & Verhagen, 2017; Rubin et al., 2004). Conflict within friendships, in contrast, is related to lower levels of prosocial behavior (Ehrlich, Dykas, & Cassidy, 2012) and higher levels of aggression in boys (Salvas et al., 2014). Importantly, declines in conflict with reciprocated friends over time are linked to lower levels of subsequent antisocial behavior (Ciairano, Rabaglietti, Roggero, Bonino, & Beyers, 2007), suggesting that adolescents reap benefits from positive changes within the peer realm.

Although friends can be a source of support, conflict and general challenges managing relationships with friends and other important others can also be a stressor, straining adolescent functioning. The larger literature on life stressors during adolescence suggests that stress exposure is linked to greater internalizing and externalizing symptoms and higher substance use (Cardoso, Goldbach, Cervantes, & Swank, 2016; Grant, Compas, Thurm, McMahon, & Gipson, 2004; Hamilton et al., 2016). Although less commonly investigated, there is evidence that greater peer- and friend-related stress is associated with more negative relational self-views and greater social disengagement (Caldwell, Rudolph, Troop-Gordon, & Kim, 2004). In addition, higher levels of peer stress, like general stress, also are related to more symptoms of anxiety and depression (Rudolph, Flynn, & Abaied, 2008; Sontag, Graber, & Clemans, 2011), although there is evidence that this link may be stronger for girls (Conley, Rudolph, & Bryant, 2012).

Recognizing the importance of peers during adolescence, scholars have highlighted the dearth of research on the effects of peer stress (Hankin et al., 2015). In the current study, we focus on friend-related stress, reflecting stress within close peer relationships. This empirical need for a greater focus on the stress within the peer domain runs parallel to theoretical motivations for examining the links between friend-related stress and adolescent well-being. Life course theory suggests that interactions with social convoys—important others in adolescents’ lives—drive development, with positive relationships and interactions promoting positive growth and negative interactions serving a deleterious role (Elder, 1998). Moreover, life course theory suggests that the interactions young people have with social convoys can accumulate advantages or disadvantages for development and well-being across time. These empirical and theoretical motivations informed our examination of potential changes in friend-related stress across early adolescence, with a particular focus on whether certain adolescents are accumulating disadvantages (experiencing stable and high or increasingly high friend-related stress) or advantages (experiencing stably low levels of friend-related stress) that influence their subsequent physical and mental health.

Prior research focused on general stress in an African American sample observed distinct clusters of stress across late adolescence with differential links to anxiety, depression, and antisocial behaviors (Schmeelk-Cone & Zimmerman, 2003). We seek to build on this work by focusing specifically on friend-related stress during an earlier stage of adolescence using a more diverse sample and more sophisticated statistical methods. Moreover, we seek to capture friend-related stress in two key ways: (a) stress that accumulates (or not) across the middle school years, capturing friend-related stress in early adolescence, and (b) the change in stress that occurs across students’ transition from middle to high school, capturing friend-related stress that may be predicated by a change in context that often shifts peer networks and peer relations (Benner, 2011).

The Current Study

In the current study, we used data from a school-based sample of more than 1,000 students attending six different middle schools in the northeast United States. The study had three primary aims. First, we sought to identify how students’ friend-related stress changed across early adolescence (i.e., seventh and eighth grades). Here, we placed particular attention on whether certain groups of students experienced similar stability or change in friend-related stress across time—that is, we used mixture modeling to examine potential profiles of friend-related stress trajectories. This was informed by prior empirical research on trajectories of general stress during adolescence (Schmeelk-Cone & Zimmerman, 2003) as well as life course theory and its focus on accumulating advantage/disadvantage.

Our second research aim examined whether changes in friend-related stress, experienced across two periods of time, influenced adolescents’ subsequent well-being following the transition to high school. We focused on three central domains of adjustment—socioemotional well-being (i.e., anxiety, negative affect), school performance (i.e., school engagement), and risky health behaviors (i.e., alcohol expectancies, alcohol consumption)—because life course theory suggests that development across domains is interconnected. In capturing change in friend-related stress over time, we first investigated the association between the unique trajectories of friend-related stress across early adolescence and later well-being. Consistent with life course theory and prior work on peer stress accumulation in late childhood (Agoston & Rudolph, 2016) and late adolescence (Schmeelk-Cone & Zimmerman, 2003), we expected that trajectories of friend-related stress that suggested accumulating disadvantage—that is, trajectories of high or increasing levels of friend-related stress—to be most disadvantageous for adolescents’ subsequent adjustment.

We also examined potential changes in friend-related stress as adolescents moved from middle to high school and their link with adolescent well-being. The high school transition involves a developmental shift from early to middle adolescence, and school transitions, by their very nature, also require a shift in physical location. Most school transitions involve integrating students from various schools into a new middle or high school, which can shift peer relations and dynamics (Benner, 2011), disrupting existing friendships (Poulin & Chan, 2010) and potentially creating challenges. Moreover, life course theory suggests that life course transitions (such as transitioning to a new school context) can serve as turning points in adolescents’ lives, shifting developmental trajectories in potentially negative or positive ways (Elder, 1998). As such, in the current study, we were specifically interested in the developmental implications of each of these early trajectories of friend-related stress (across middle school and across the high school transition) for youth.

In our third and final research aim, we examined potential variation by gender. Prior research has shown that girls report greater interpersonal and friend-related stress than boys (Hampel & Petermann, 2006; Hankin, Mermelstein, & Roesch, 2007; Rose & Rudolph, 2006), and there is evidence that girls seem to be more sensitive to interpersonal stress in relation to effects on both anxiety and depression (Rudolph, 2002). Indeed, theorists suggest that gender differences in internalizing symptoms are likely due, in part, to an intensification in the need for close affiliations that girls experience across the transition to adolescence (Cyranowski, Frank, Young, & Shear, 2000). Based on this prior empirical and theoretical research, we expected that girls would be more likely to be members of trajectory groups characterized by high or increasing levels of friend-related stress. Similarly, we hypothesized that the links between friend-related stress and young people’s subsequent adjustment would be stronger for girls than boys, particularly in relation to negative affect and anxiety.

Method

Participants and Procedures

The sample for the current study was drawn from an ongoing prospective cohort-sequential study on alcohol initiation and progression among early adolescents (see Jackson, Colby, Barnett, & Abar, 2015). Early adolescents (sixth, seventh, and eighth graders) were recruited in five cohorts from six schools in the northeast United States. Study information was distributed via mail and through homeroom teachers; interested youth (initial response rate of 38% of those informed of the study) and for whom written parental consent was obtained (65% of those who returned consent forms) were invited to attend a 2-hour in-person group orientation session; 88% of these youth were enrolled into the study. A Certificate of Confidentiality was obtained from National Institute on Alcohol Abuse and Alcoholism to preserve participant confidentiality.

Following the orientation, participants completed a 45-minute computerized baseline survey and then four subsequent web-based surveys every 6 months. A final web-based survey was completed 1 year after the last of the 6-month surveys. Due to the cohort-sequential nature of the sample and our study design, the dataset contains planned missingness such that varying numbers of students completed the assessments for fall and spring semesters of each academic year; thus, participants contributed six waves of data across fall and spring of sixth grade through 11th grade, depending on time of enrollment. The sample was comprised of roughly equal numbers of sixth, seventh, and eighth graders at baseline (33%, 32%, and 35%, respectively) and was 52% female and 36% non-White (12% Hispanic, 5% Black, 8% multiracial, 11% Other race/ethnicity). Slightly more than half of the adolescents (53%) had at least one parent with a college education, and one third of the sample reported receiving free- or reduced-price lunch. Response rates were 90% at T2, 86% at T3, 83% at T4, 81% at T5, and 82% at T6. Compared with the characteristics of the full sample at the time of enrollment, nonresponse at any subsequent time point was not associated with age, t(1,021) = −.67, p = .50, but was associated with being male, χ²(1, N = 1,023) = 7.29, p < .007; non-White, χ²(1, N = 1,023) = 8.12, p < .004; of Hispanic ethnicity, χ²(1, N = 1,023) = 12.74, p < .002; and having lower family economic status (socioeconomic status), χ²(2, N = 898) = 19.33, p < .001. Participants were compensated for participating in the orientation and for their survey participation. The Brown University Institutional Review Board approved all project procedures.

For the first research aim, we included participants who answered the friend-related stress item from the seventh-grade fall assessment to eighth-grade spring assessment (four waves total; N = 1,006). Due to the cohort-sequential design, a subset of participants were not surveyed in ninth grade due to planned missingness, and thus, analyses for the second research aim included the 787 participants who had data on all exogenous variables, who provided data for at least one outcome variable during ninth grade, and who had valid feeder pattern information. The feeder pattern is a composite cluster/nesting variable that accounted for both the middle and high school the student attended; this allowed us to take into account the fact that students within these feeder patterns (clusters) potentially would be more similar than students who did not share the same feeder pattern. Participants who were included (N = 787) versus excluded (N = 221) for the second research aim did not significantly differ on friend-related stress group membership, gender, or parental education. Compared with those excluded, those included were less likely to be Hispanic or “Other” race/ethnicity—versus White, χ²(1) = 6.16, p = .02, and χ²(1) = 9.90, p < .01, respectively—and less likely to receive free- or reduced-price lunch, χ²(1) = 4.54, p = .03.

Measures

Friend-related stress was measured across fall and spring semesters of each academic year. In the current study, adjustment outcomes (i.e., anxiety, negative affect, school engagement, alcohol expectancies, alcohol use) were assessed in ninth grade; due to the panel design, some participants were assessed in the fall, some in the spring, and some were assessed in both semesters (for which, the mean of the two ninth-grade assessments was used). Descriptive statistics and bivariate correlations for the central model constructs appear in Table 1.

Table 1.

Descriptive Statistics and Zero-Order Correlations Among Key Study Constructs.

	1	2	3	4	5	6	7	8	9	10	11	12
1. Friend-related stress (Fall seventh)	—	—	—	—	—	—	—	—	—	—	—	—
2. Friend-related stress (Spring seventh)	.50***	—	—	—	—	—	—	—	—	—	—	—
3. Friend-related stress (Fall eighth)	.47***	.46***	—	—	—	—	—	—	—	—	—	—
4. Friend-related stress (Spring eighth)	.39***	.39***	.47***	—	—	—	—	—	—	—	—	—
5. Friend-related stress (Fall ninth)	.35***	.28***	.39***	.46***	—	—	—	—	—	—	—	—
6. Friend-related stress (Spring ninth)	.19	.30***	.35***	.40***	.43***	—	—	—	—	—	—	—
7. Social anxiety (ninth grade)	.16***	.14**	.22***	.22***	.26***	.28***	—	—	—	—	—	—
8. Negative affect (ninth grade)	.23***	.25***	.34***	.34***	.26***	.39***	.43***	—	—	—	—	—
9. Cognitive school engagement (ninth grade)	.04	−.05	.00	−.09*	−.06	−.21***	−.07*	−.15***	—	—	—	—
10. Behavioral school engagement (ninth grade)	−.04	−.10*	−.09*	−.15***	−.10**	−.16***	−.01	−.22***	.31***	—	—	—
11. Positive alcohol expectancies (ninth grade)	.09	.04	.13***	.12**	.10**	.04	.21***	.14***	.01	−.05	—	—
12. Consumed alcohol (ninth grade)	.06	.12**	.14***	.13***	.08*	.18***	.03	.13***	−.13***	−.23***	.12***	—
$\bar{X}$	0.79	0.87	0.89	0.98	0.88	0.97	1.06	1.07	3.45	4.18	2.60	0.59
SD	0.91	0.89	0.92	1.00	0.92	0.92	0.82	0.92	0.84	0.58	0.82	0.49
n	532	593	831	696	769	476	840	872	875	874	864	875

p < .05. **p < .01. ***p < .001.

Friend-related stress

Adolescents reported on their friend-related stress using a single item: “During the past month or two, I have had stress in my life due to my friends.” Ratings ranged from 0 (no) to 3 (large/major stress).

Adolescent adjustment

Social anxiety was measured by seven items adapted from the Social Anxiety subscale of the Multidimensional Anxiety Scale for Children (MASC; for example, “I worry about other people laughing at me”; March, Parker, Sullivan, Stallings, & Conners, 1997). The rating scale ranged from 0 (never true about me) to 3 (often true about me), with higher mean scores indicating higher social anxiety (α = .90). Negative affect was measured by five items adopted from a child version of the Positive and Negative Affect Schedule (PANAS-C, Laurent et al., 1999). Adolescents indicated to what extent they felt sad, upset, scared, miserable, and lonely during the past few weeks. Ratings ranged from 1 (very slightly or not at all) to 5 (extremely), with higher mean scores indicating higher levels of negative affect (α = .89).

School engagement was measured by eight items adapted from the Monitoring the Future Study (Bachman, Johnston, & O’Malley, 2005). Three items assessed adolescents’ cognitive school engagement (e.g., “How often do you feel that the school work you are assigned is meaningful and important?”), and five items assessed behavioral school engagement (e.g., “How often do you fail to complete or turn in your assignments?”). Reflecting on the past 6 months, adolescents rated each item from 1 (never) to 5 (always). Higher mean scores indicated higher levels of cognitive (α = .77) and behavioral (α = .72) school engagement.

Positive alcohol expectancies were measured by nine items adopted from Schell, Martino, Ellickson, Collins, and McCaffrey (2005). Adolescents indicated how likely certain things (e.g., “forget their problems,” “have fun”) would happen to people of their age if they had one or more drinks of alcohol. Ratings ranged from 1 (very unlikely) to 4 (very likely), with higher mean scores indicating higher levels of positive alcohol expectancies (α = .94). For alcohol use, adolescents reported whether they had ever consumed alcohol (“at least a sip”; 0 = no, 1 = yes). For adolescents who had survey data in the fall and spring of ninth grade, adolescents received a score of 1 on the alcohol use item if they reported having had consumed alcohol in either the fall or the spring semester.

Demographic variables

Adolescents reported their gender (0 = female, 1 = male), race/ethnicity (recoded into White, non-Hispanic [reference group]; Black, non-Hispanic; Hispanic; Other, non-Hispanic), parental education (whether at least one parent had a college degree), and family economic status (whether the participant received free- or reduced-price lunch). Study cohort was also included as a covariate in analyses.

Analyses

Analyses were conducted in Mplus 7.31 (Muthén & Muthén, 1998-2015). We used the CLUSTER function in Mplus to account for students nested within middle schools (when examining gender as a predictor of trajectory group membership) and feeder pattern (when predicting adolescent well-being in high school). Based on adolescents’ reports of their school name at eighth and ninth grade, we created a feeder pattern variable that identified both the middle school and high school that the student attended (n = 18 clusters; see Benner & Graham, 2009, for a similar nesting strategy). Mplus uses the full information maximum likelihood estimation method to handle missing data, which enables full usage of all available data in the analyses.

Results

Documenting Friend-Related Stress During Early Adolescence

For the first research question examining changes in friend-related stress across middle school and whether students exhibited distinct patterns of change in stress over time, growth mixture modeling was used. A series of models were specified (i.e., one to five classes) using the four time points spanning seventh and eighth grades. Several model fit indices were used to determine class enumeration in addition to examining whether the classes appeared substantively and conceptually meaningful and qualitatively unique from one another (Nylund, Asparouhov, & Muthén, 2007; Tofighi & Enders, 2008). Specifically, Bayesian information criterion (BIC), sample size–adjusted Bayesian information criterion (ABIC), and the Lo–Mendell–Rubin likelihood ratio test (LMR) were examined. Smaller BIC and ABIC values indicate a better fitting model, and a significant p value on the LMR test indicates that the k-class model fits the data better than the k – 1 class model.

Based on the model fit indices (see Table 2) and the identification of conceptually meaningful and interpretable classes, the four-class solution was identified as optimal. Specifically, the four-class solution had the second smallest ABIC, a BIC value virtually identical to the five-class solution, and an LMR p value smaller than .05. The four friend-related stress trajectory groups are depicted in Figure 1. The largest group displayed low friend-related stress at all time points (labeled “low stress”; n = 577, 57% of the sample), whereas the smallest group reported high levels of friend-related stress across seventh and eighth grades (labeled “high stress”; n = 70, 7% of sample). The third group displayed a moderate level of stress at the beginning of seventh grade and increased in friend-related stress across middle school (labeled “moderate-increasing stress”; n = 223, 22% of sample), and the final group reported a moderate level of friend-related stress at the beginning of seventh grade but decreased in friend-related stress across seventh and eighth grades (labeled “moderate-decreasing stress”; n = 136, 14% of sample).

Table 2.

Model Fit Indices Examining Trajectories of Friend-Related Stress.

	One class	Two classes	Three classes	Four classes	Five classes
Log likelihood	−3,325.30	−3,258.35	−3,242.43	−3,198.03	−3,187.35
Number of parameters	9	10	12	15	18
BIC	6,712.83	6,585.83	6,567.82	6,499.77	6,499.15
ABIC	6,684.24	6,554.07	6,529.71	6,452.13	6,441.98
Entropy	NA	.48	.69	.66	.67
p value of LMR	NA	.00	.01	.03	.32

Note. BIC = Bayesian information criterion; ABIC = adjusted Bayesian information criterion; LMR = Lo–Mendell–Rubin.

Figure 1.

Trajectories of friend-related stress across early adolescence.

For the change of friend-related stress across the high school transition, we calculated the stress change score by subtracting friend-related stress in eighth grade from stress in ninth grade. Across the transition to high school, on average, adolescents reported a small decrease in friend-related stress ( $\bar{X}$ = −0.03, SD = 0.88). To better understand the links between these two periods of friend-related stress, we conducted an ANOVA to compare mean-level differences in shifts in friend-related stress across the high school transition by middle school stress trajectory class membership. There were significant differences across groups, F(3, 844) = 58.00, p < .001. Specifically, Tukey’s post hoc tests revealed that the high-stress group ( $\bar{X}$ = −0.76, SD = 0.94) and the moderate-increasing group ( $\bar{X}$ = −0.52, SD = 0.94) experienced greater declines in friend stress across the transition than the moderate-decreasing group ( $\bar{X}$ = −0.04, SD = 0.85). The low-stress group actually experienced a small increase in stress across the high school transition ( $\bar{X}$ = 0.24, SD = 0.70), and this mean difference was significantly different than the three other friend-related stress trajectory groups.

Linking Friend-Related Stress to Adolescent Well-Being

For our second research aim, we examined whether adolescents’ middle school friend-related stress trajectory group membership and change of friend-related stress across the transition to high school were related to subsequent socioemotional, academic, and alcohol outcomes following the transition to high school. To do so, we conducted structural regression analyses. We tested three path models with the reference group for friend-related stress trajectory group being rotated to obtain all possible comparisons among stress profiles. In each model, four friend-related stress variables (i.e., three dummy variables representing stress trajectory group and one continuous variable of stress change across high school transition, measured as the difference between ninth- and eighth-grade stress) were included as independent variables. All models included all outcomes and control variables (i.e., gender, race/ethnicity, parental education, family economic status, and cohort).

The results demonstrated that both friend-related trajectories across middle school and changes in friend-related stress from eighth to ninth grade had unique effects on well-being indicators in ninth grade above and beyond each other. Specifically, adolescents displaying different trajectories of friend-related stress during middle school varied in their socioemotional (social anxiety, negative affect), academic (cognitive and behavioral school engagement), and alcohol outcomes (positive alcohol expectancies, alcohol use) following the transition to high school. Mean levels of each outcome for each trajectory group are shown in Table 3. Path parameters of the regression analyses are presented in Table 4.

Table 3.

Descriptive Statistics of Adolescent Outcomes at Ninth Grade for Each Group.

	Low		Moderate-decreasing		Moderate-increasing		High
	$\bar{X}$	SD	$\bar{X}$	SD	$\bar{X}$	SD	$\bar{X}$	SD
Social anxiety	0.93	0.76	1.26	0.84	1.24	0.86	1.25	0.92
Negative affect	0.82	0.76	1.31	1.02	1.41	1.00	1.57	0.91
Cognitive school engagement	3.50	0.82	3.46	0.75	3.35	0.86	3.54	0.86
Behavioral school engagement	4.23	0.59	4.23	0.48	4.05	0.58	4.16	0.65
Positive alcohol expectancy	2.53	0.86	2.58	0.79	2.71	0.76	2.96	0.66
Consumed alcohol (%)	55		59		67		76

Table 4.

Path Parameters of the Relations Between Friend-Related Stress and Adolescent Outcomes Following the High School Transition.

Predictors in each model	Social anxiety			Negative affect			Cognitive school engagement			Behavioral school engagement			Positive alcohol expectancies			Consumed alcohol
Predictors in each model	β	SE	p	β	SE	p	β	SE	p	β	SE	p	β	SE	p	β	SE	p
High-stress group versus . . .
Low	−.23	.06	.00	−.59	.05	.00	.08	.07	.25	.28	.06	.00	–.21	.11	.05	–.40	.09	.00
Mod-dec	−.02	.04	.67	−.20	.03	.00	.03	.07	.65	.14	.05	.01	−.14	.09	.11	–.20	.07	.01
Mod-inc	.00	.06	.95	−.12	.03	.00	−.05	.05	.40	.02	.06	.72	−.12	.10	.24	–.20	.10	.04
Change	.16	.03	.00	.28	.04	.00	–.13	.02	.00	–.14	.02	.00	.00	.03	.97	.05	.04	.15
Mod-inc group versus . . .
Low	−.24	.03	.00	–.45	.03	.00	.13	.06	.02	.26	.03	.00	−.07	.07	.32	–.17	.08	.04
Mod-dec	−.02	.04	.55	–.11	.01	.00	.07	.07	.32	.12	.03	.00	−.04	.04	.32	−.04	.04	.29
High	.00	.03	.95	.07	.02	.00	.03	.03	.40	−.01	.03	.72	.07	.06	.24	.11	.05	.04
Change	.16	.03	.00	.28	.04	.00	–.13	.02	.00	–.14	.02	.00	.00	.03	.97	.05	.04	.15
Mod-dec group versus . . .
Low	−.21	.05	.00	–.30	.04	.00	.03	.06	.57	.08	.04	.04	−.12	.07	.08	.14	.03	.00
Mod-inc	.03	.05	.55	.13	.02	.00	−.08	.08	.32	–.15	.03	.00	.05	.04	.30	.20	.03	.00
High	.01	.03	.67	.14	.02	.00	−.02	.05	.65	–.10	.04	.01	.14	.05	.01	.11	.03	.00
Change	.16	.03	.00	.28	.04	.00	–.13	.02	.00	–.14	.02	.00	.05	.04	.15	.16	.03	.00

Note. Significant results are in bold. mod-dec = moderate-decreasing; mod-inc = moderate-increasing; change = stress change across transition to high school.

Overall, the high-stress group struggled most following the high school transition. These students had higher negative affect and a higher likelihood of having consumed at least a sip of alcohol in ninth grade compared with all other groups. The high-stress group also reported lower behavioral school engagement than both the low-stress and moderate-decreasing friend-related stress groups as well as higher social anxiety and higher positive alcohol expectancies compared with the low-stress group. Not surprisingly, the low-stress group exhibited the most positive outcomes following the high school transition. In addition to the better functioning they exhibited relative to the high-stress group, the low-stress group also reported lower social anxiety, lower negative affect, higher behavioral school engagement, and a lower likelihood of having ever consumed alcohol compared with moderate-decreasing and moderate-increasing stress groups. Differences were also found for cognitive and behavioral school engagement, with the low-stress group favored over the moderate-increasing friend-related stress group. Finally, the moderate-increasing group reported higher negative affect and lower behavioral school engagement compared with the moderate-decreasing friend-related stress group.

Above and beyond the effects of friend-related stress trajectories across middle school, changes in friend-related stress from eighth to ninth grade were significantly related to all well-being indicators except positive alcohol expectancies. Specifically, increases in friend-related stress across the transition to high school were related to higher social anxiety, higher negative affect, lower cognitive and behavioral school engagement, and a higher likelihood of having ever consumed alcohol.

Gender Differences in Friend-Related Stress and Adolescent Well-Being

For our third research aim, we explored potential variation by gender. We first examined adolescent gender as a predictor of the friend-related stress indicators (i.e., stress trajectory group membership, change of stress across the transition to high school) using regression models with all covariates. Then, we tested whether the relations between friend-related stress indicators and adolescent outcomes varied by adolescent gender using multigroup comparisons with Wald tests of parameter constraints. We made six sets of multigroup comparisons (one set for each outcome). When the moderation effect was significant in a given set of constrained paths, each individual path in that set was constrained individually. We used a Bonferroni correction to control the Type I error rate (p = .05 / 6 = .008) to interpret findings from the multiple group comparisons.

The results showed that girls were more likely than boys to be in the high-stress group (B = 0.83, p < .001), moderate-increasing stress group (B = 0.93, p < .001), and moderate-decreasing stress group (B = 0.62, p < .01) relative to the low-stress group. In contrast, boys struggled more with friend-related stress across the high school transition, with boys, on average, reporting increases in friend-related stress across the transition ( ${\bar{X}}_{change}$ = 0.05, SD = 0.82), whereas girls experienced declines in friend-related stress over the same time period on average ( ${\bar{X}}_{change}$ = −0.09, SD = 0.91; B = 0.15, p < .001).

There were also significant adolescent gender differences in the relations between friend-related stress indicators and adolescents’ negative affect, W(4) = 18.12, p = .001, and cognitive school engagement, W(4) = 26.87, p < .001. Specifically, the negative association between being in the low (vs. high) stress group and adolescent negative affect was stronger for girls (B = −0.65, p < .001) than for boys (B = −0.45, p < .001). Similarly, the negative association between being in the low (vs. moderate-decreasing) stress group and adolescent negative affect was stronger for girls (B = −0.37, p < .001) than for boys (B = −0.13, p < .001). Finally, in examining moderation of the link between mean-level changes in friend stress across the high school transition and outcomes, we found that the association between greater increases in friend stress and cognitive school engagement was more than 3 times as strong for boys (B = −0.22, SE = 0.04, p < .001) compared with girls (B = −0.06, SE = 0.05, p = .23).

Discussion

Friends are increasingly important socializing agents as young people move from late childhood into early adolescence (Brown & Larson, 2009). Although much is known about friendship quality and conflict during this time in the life course, much less is known about the role of stress in friend relationships for early adolescents. In the current study, we sought to determine exactly how friend-related stress changed over middle school and the consequences of changes in friend-related stress for students’ well-being following the transition to high school. Particular attention was placed on potential gender differences, as girls tend to be both more sensitive to peer relations and more susceptible to socioemotional difficulties at this time (Hampel & Petermann, 2006; Rudolph, 2002).

Overall, we found that a majority of students (57%) experienced stable and low levels of friend-related stress across middle school, yet almost 30% of the sample reported either stably high or moderate and increasing levels of friend stress across this time. The interpersonal life-stress generation model (Hammen, 2003; Rudolph et al., 2000) suggests that individuals may generate stress within their everyday lives that compromise interpersonal relations, thus maintaining or generating additional stress. Consistent with this model, it could be that early adolescents with high or increasing levels of friend-related stress may be engaging in interactions with friends that maintain these interpersonal stressors, that they may select into friendships that they believe will experience high levels of peer stress, or that they may select high-stress friendships without necessarily being consciously aware that the friendship will bring stress. Alternatively, adolescents in the high or increasing friend-related stress groups may have deficits in interpersonal skills that contribute to stress within their friendships or render them less capable of negotiating challenges that occur as friendships change over time. This is consistent with intervention models that seek to alleviate stress within childhood and adolescence by building coping resources such as interpersonal skill development (Compas, 1987). Future research is necessary to determine whether stress generation or interpersonal skills deficits (or both) are contributing to elevated levels of friend-related stress during early adolescence.

Understanding the antecedents of friend-related stress is particularly important given the clear negative repercussions young people face when experiencing high or increasing levels of friend stress. The stress generation model has most commonly been referenced in relation to the links between stress and depression (R. T. Liu & Alloy, 2010), but whether the stably high or moderate and increasing peer stress groups are also experiencing depression is an open question, as the current study measured negative affect generally but not depression specifically; however, these two groups exhibited the highest levels of negative affect in the current study, which lends some credence to the potential for stress generation to be driving this relation. Indeed, research has shown that depression is influenced by interpersonal stress generation in particular, as opposed to noninterpersonal stress (Flynn & Rudolph, 2011). Similarly, Caldwell and colleagues (2004) observed a transactional link between peer stress and relational self-views, such that peer stress influenced negative relational self-views over time and vice versa. Future research linking depression and friend-related stress over time would provide greater insights into these potential transactional links that are informed by the stress generation perspective.

For those experiencing changes in peer stress, either those who reported increasing or decreasing levels of friend stress across middle school or across the transition to high school, we cannot determine whether this is due to shifts in friend group composition—that is, who the adolescent considers to be a close friend—or if the dynamics of friendships are changing over time. During early adolescence, friendships can be quite unstable, with up to one half of friendships not persisting year over year, and at least some of this instability is attributed to normative school transitions that occur during this time (Poulin & Chan, 2010). To ascertain whether friendship instability or relationship dynamics within stable friendships is driving changes in friend-related stress across middle school, social network data are needed. Although not available in the current study, such data are critical for determining the underlying reasons for friend-related stress, which is important for designing appropriate and effective interventions that support early adolescents’ well-being.

Such work is crucial, as findings from the current study clearly highlight the wide-ranging repercussions of friend-related stress. Those in the high-stress groups reported the poorest socioemotional well-being, lowest levels of school engagement, and the most positive expectancies around and greatest likelihood of alcohol consumption. Such developmental challenges were also demonstrated for the moderate but changing groups in comparison with the low friend-related stress group. These findings are consistent with prior research highlighting the negative toll that general stress can exert during adolescence (Cardoso et al., 2016; Hamilton et al., 2016), and they are also consistent with the two central caveats of life course theory (Elder, 1998). The first, that social convoys play an important role in individuals’ lives, is reflected in the strain that friend-related stress exerted on young people, particularly those experiencing stably high or increasing levels of stress in relationships with their close friends. Parallel to this, we also documented evidence of accumulating advantages and disadvantages, as the low friend-related stress group reaped particular rewards from the lack of stress in their friend relationships across middle school, whereas the accumulating disadvantages young people experienced when facing high or increasing amounts of stress with their friends were subsequently reflected in their adjustment and well-being after making the transition to high school. These findings were above and beyond the proximal stress experienced in the transition to high school.

Finally, it should be noted that gender did play a role in how friend-related stress was related to subsequent well-being. In terms of change in friend-related stress across middle school, girls were significantly more likely than boys to experience high or increasing levels of friend-related stress. These findings are consistent with prior research highlighting the greater amounts of interpersonal stress and general emotional turmoil facing girls as they move through early and into middle adolescence (Hampel & Petermann, 2006; Hankin et al., 2007; Nolen-Hoeksema & Hilt, 2013). In examining the consequences of friend-related stress across middle school for functioning following the transition to high school, it appeared that girls benefited more psychologically from being in the low-stress group. This is consistent with other research that has documented that positive, interdependent friendships were linked to less depression for girls but not for boys (Selfhout, Branje, & Meeus, 2009). Boys, however, did report greater challenges related to friend-related stress as they moved from middle to high school. We suspect that this was, in part, due to the low levels of friend-related stress boys reported at the end of middle school. It is, however, also consistent with other research that suggests that the high school transition can be particularly challenging to navigate for boys (Finn & Rock, 1997; Roderick, 2003).

Strengths, Caveats, and Limitations

Although friendships play a substantial role in early adolescents’ lives, friend-related stress in adolescence has been overlooked in much of the existing research on friendships (Hankin et al., 2015), which typically focuses on friendship patterns, quality, and conflict (e.g., Bukowski, Newcomb, & Hartup, 1998; Poulin & Chan, 2010). Informed by the life course perspective, we used longitudinal data spanning middle school and the transition to high school to identify four unique trajectories of friend-related stress experienced by early adolescents and the particular challenges faced by those in trajectories characterized by high or increasing levels of stress. Such work fills a critical void and provides insights into which adolescents may be at particular risk for negative developmental outcomes. That said, some limitations must be acknowledged. First, friend-related stress was assessed with a single item rather than a more comprehensive scale. As such, the particular kinds of stress individuals were experiencing within their friendships (e.g., arguments, isolation, relational aggression) or whether they were considering their larger peer network (vs. close friends) cannot be determined. The item, however, had sufficient variation across participants, was measured longitudinally, and clearly clustered into unique trajectories across early adolescence. In addition, although we hypothesize that friendship stress is impacting young people’s subsequent well-being, the data are not experimental, and thus causal conclusions cannot be made. The stress generation model (Hammen, 2003) suggests that the links between stress and well-being are bidirectional in nature, and the interrelationships between friend stress and adjustment should be unpacked in future research.

Finally, because we had low rates of return of the parental consent forms, the enrollment rate was lower than anticipated. We expect this is likely due to the intensive study design, the de-personalized contact through mailed/distributed flyers, and the focus on the sensitive topic of underage drinking. As concluded by Frissell and colleagues (2004), parental nonresponse to traditional active consent procedures may reflect failure of parents to attend to the request (e.g., because of inconvenience) as opposed to explicit refusal. Because active consent requires more effort on the part of parents (C. Liu, Cox, Washburn, Croff, & Crethar, 2017), it is possible that this impacted the characteristics of those who enrolled. Our prior work (Jackson, Colby, Barnett, & Abar, 2015) has demonstrated that our sample is more racially diverse and less socioeconomically disadvantaged than the school populations from which it was drawn, but it is representative with respect to gender and grade. Other work also shows that substance use rates in our sample are lower than those observed in epidemiological data, suggesting that our sample is somewhat “low-risk” in nature, which may have resulted in underestimates of stress in the present study. Our sample also is drawn from a subset of schools in a single geographic region and thus is not representative of the United States.

Implications for Intervention

Friend-related stress can take a negative toll on development across domains, and in the current study, more than 40% of the current sample reported high or moderate levels of friend-related stress. As such, it is clear that early adolescents would likely benefit from supports that teach effective interpersonal skills as well as coping strategies to manage friend-related stress. Promoting active coping strategies, or those that are “intended to achieve some degree of personal control over the stressful aspects of the environment and one’s emotions” (Compas, Connor-Smith, Saltzman, Thomsen, & Wadsworth, 2001, p. 120), may be one strategy for intervening with students facing high levels of stress in interpersonal relations with friends.

Meta-analytic studies have observed that active coping strategies, such as problem-solving or eliciting social support, are significantly and positively associated with greater psychosocial health (Clarke, 2006) and are negatively related to psychopathology during adolescence (Compas et al., 2017); in contrast, motives related to drinking to cope are positively associated with alcohol use and problems (Cooper, Russell, Skinner, & Windle, 1992; Kuntsche, Knibbe, Gmel, & Engels, 2005). Moreover, these more active coping styles appear to buffer the negative links between peer stress and psychopathology, at least for girls (Sontag & Graber, 2010). Yet, in the face of peer stressors, such adaptive coping strategies may be less likely to be implemented. For example, scholarship on relational aggression within girls’ friendships suggests that although girls are most likely to report active coping strategies, when they feel particularly hurt by the relational aggression, they are more likely to adopt more disadvantageous coping mechanisms, such as passive or avoidant coping (Remillard & Lamb, 2005). As such, intervention and prevention efforts may want to consider promoting early adolescents’ use of active coping methods, particularly for those facing higher levels of interpersonal stressors or those who are particularly emotionally sensitive to these stressors. Clearly, friendships are critical for positive growth and development, but young people will also need supports to deal with interpersonal challenges to flourish.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors acknowledge the support of grants from the William T. Grant Foundation to Aprile D. Benner, from National Institute on Alcohol Abuse and Alcoholism (NIAAA) to Kristina Jackson (R01AA016797 and K02AA13938), and from Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) to the Population Research Center at the University of Texas at Austin (P2CHD042849).

Author Biographies

Aprile D. Benner is an associate professor in the Department of Human Development and Family Sciences at the University of Texas at Austin. She received her PhD in education at the University of California, Los Angeles. Her research interests center on the development of low-income and race/ethnic minority youth, investigating how the everyday contexts of children and adolescents’ lives shape their developmental outcomes.

Yang Hou is a postdoctoral fellow at the National Cancer Institute. She received her PhD in human development and family sciences at the University of Texas at Austin. Her research interests focus on how contextual factors (i.e., sociocultural, school, peer, and family factors) influence adolescent socioemotional, behavioral, academic, and health outcomes.

Kristina M. Jackson is a professor (research) at the Center for Alcohol and Addiction Studies at Brown University. She received her PhD in social psychology at Arizona State University. She studies initiation and progression of alcohol use and the influence of motivational and contextual risk factors on substance use.

References

Agoston

A. M.

Rudolph

K. D.

(2016). Interactive contributions of cumulative peer stress and executive function deficits to depression in early adolescence. Journal of Early Adolescence, 36, 1070-1094. doi:10.1177/0272431615593176

Bachman

J. G.

Johnston

L. D.

O’Malley

P. M.

(2005). Monitoring the future: Questionnaire responses from the nation’s high school seniors, 2004. Ann Arbor, MI: Institute for Social Research.

Baumeister

R. F.

Leary

M. R.

(1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117, 497-529. doi:10.1037/0033-2909.117.3.497

Benner

A. D.

(2011). The transition to high school: Current knowledge, future directions. Educational Psychology Review, 23, 299-328. doi:10.1007/s10648-011-9152-0

Benner

A. D.

Graham

(2009). The transition to high school as a developmental process among multiethnic urban youth. Child Development, 80, 356-376. doi:10.1111/j.1467-8624.2009.01265.x

Brown

B. B.

Larson

(2009). Peer relationships in adolescence. In Lerner

R. M.

Steinberg

(Eds.), Handbook of adolescent psychology: Contextual influences on adolescent development (3rd ed., Vol. 2, pp. 74-103). Hoboken, NJ: John Wiley.

Bukowski

W. M.

Newcomb

A. F.

Hartup

W. W.

(1998). The company they keep: Friendship in childhood and adolescence. Cambridge, UK: Cambridge University Press.

Caldwell

M. S.

Rudolph

K. D.

Troop-Gordon

Kim

D. Y.

(2004). Reciprocal influences among relational self-views, social disengagement, and peer stress during early adolescence. Child Development, 75, 1140-1154. doi:10.1111/j.1467-8624.2004.00730.x

Cardoso

J. B.

Goldbach

J. T.

Cervantes

R. C.

Swank

(2016). Stress and multiple substance use behaviors among Hispanic adolescents. Prevention Science, 17, 208-217. doi:10.1007/s11121-015-0603-6

10.

Ciairano

Rabaglietti

Roggero

Bonino

Beyers

(2007). Patterns of adolescent friendships, psychological adjustment and antisocial behavior: The moderating role of family stress and friendship reciprocity. International Journal of Behavioral Development, 31, 539-548. doi:10.1177/0165025407080573

11.

Clarke

A. T.

(2006). Coping with interpersonal stress and psychosocial health among children and adolescents: A meta-analysis. Journal of Youth and Adolescence, 35, 10-23. doi:10.1007/s10964-005-9001-x

12.

Compas

B. E.

(1987). Coping with stress during childhood and adolescence. Psychological Bulletin, 101, 393-403.

13.

Compas

B. E.

Connor-Smith

J. K.

Saltzman

Thomsen

A. H.

Wadsworth

M. E.

(2001). Coping with stress during childhood and adolescence: Problems, progress, and potential in theory and research. Psychological Bulletin, 127, 87-127. doi:10.1037/0033-2909.127.1.87

14.

Compas

B. E.

Jaser

S. S.

Bettis

A. H.

Watson

K. H.

Gruhn

M. A.

Dunbar

J. P.

. . . Thigpen

J. C.

(2017). Coping, emotion regulation, and psychopathology in childhood and adolescence: A meta-analysis and narrative review. Psychological Bulletin, 143, 939-991. doi:10.1037/bul0000110

15.

Conley

C. S.

Rudolph

K. D.

Bryant

F. B.

(2012). Explaining the longitudinal association between puberty and depression: Sex differences in the mediating effects of peer stress. Development and Psychopathology, 24, 691-701. doi:10.1017/S0954579412000259

16.

Cooper

M. L.

Russell

Skinner

J. B.

Windle

(1992). Development and validation of a three-dimensional measure of drinking motives. Psychological Assessment, 4, 123-132. doi:10.1037/1040-3590.4.2.123

17.

Cyranowski

J. M.

Frank

Young

Shear

M. K.

(2000). Adolescent onset of the gender difference in lifetime rates of major depression: A theoretical model. Archives of General Psychiatry, 57, 21-27. doi:10.1001/archpsyc.57.1.21

18.

Ehrlich

K. B.

Dykas

M. J.

Cassidy

(2012). Tipping points in adolescent adjustment: Predicting social functioning from adolescents’ conflict with parents and friends. Journal of Family Psychology, 26, 776-783. doi:10.1037/a0029868

19.

Elder

G. H.

Jr. (1998). The life course and human development. In Damon

Lerner

R. M.

(Eds.), Handbook of child psychology (Vol. 1, pp. 939-991). New York, NY: John Wiley.

20.

Ferrar

Chang

Olds

T. S.

(2013). Adolescent time use clusters: A systematic review. Journal of Adolescent Health, 52, 259-270. doi:10.1016/j.jadohealth.2012.06.015

21.

Finn

J. D.

Rock

D. A.

(1997). Academic success among students at risk for school failure. Journal of Applied Psychology, 82, 221-234.

22.

Flynn

Rudolph

K. D.

(2011). Stress generation and adolescent depression: Contribution of interpersonal stress responses. Journal of Abnormal Child Psychology, 39, 1187-1198. doi:10.1007/s10802-011-9527-1

23.

Frissell

K. C.

McCarthy

D. M.

D’Amico

E. J.

Metrik

Ellingstad

T. P.

Brown

S. A.

(2004). Impact of consent procedures on reported levels of adolescent alcohol use. Psychology of Addictive Behaviors, 18, 307-315.

24.

Grant

K. E.

Compas

B. E.

Thurm

A. E.

McMahon

S. D.

Gipson

P. Y.

(2004). Stressors and child and adolescent psychopathology: Measurement issues and prospective effects. Journal of Clinical Child & Adolescent Psychology, 33, 412-425. doi:10.1207/s15374424jccp3302_23

25.

Hamilton

J. L.

Potter

C. M.

Olino

T. M.

Abramson

L. Y.

Heimberg

R. G.

Alloy

L. B.

(2016). The temporal sequence of social anxiety and depressive symptoms following interpersonal stressors during adolescence. Journal of Abnormal Child Psychology, 44, 495-509. doi:10.1007/s10802-015-0049-0

26.

Hammen

(2003). Interpersonal stress and depression in women. Journal of Affective Disorders, 74, 49-57. doi:10.1016/S0165-0327(02)00430-5

27.

Hampel

Petermann

(2006). Perceived stress, coping, and adjustment in adolescents. Journal of Adolescent Health, 38, 409-415. doi:10.1016/j.jadohealth.2005.02.014

28.

Hankin

B. L.

Mermelstein

Roesch

(2007). Sex differences in adolescent depression: Stress exposure and reactivity models. Child Development, 78, 279-295. doi:10.1111/j.1467-8624.2007.00997.x

29.

Hankin

B. L.

Young

J. F.

Abela

J. R. Z.

Smolen

Jenness

J. L.

Gulley

L. D.

. . . Oppenheimer

C. W.

(2015). Depression from childhood into late adolescence: Influence of gender, development, genetic susceptibility, and peer stress. Journal of Abnormal Psychology, 124, 803-816. doi:10.1037/abn0000089

30.

Jackson

K. M.

Colby

S. M.

Barnett

N. P.

Abar

C. C.

(2015). Prevalence and correlates of sipping alcohol in a prospective middle school sample. Psychology of Addictive Behaviors, 29, 766-778.

31.

Kuntsche

Knibbe

Gmel

Engels

(2005). Why do young people drink? A review of drinking motives. Clinical Psychology Review, 25, 841-861. doi:10.1016/j.cpr.2005.06.002

32.

Laurent

Catanzaro

S. J.

Joiner

T. E.

Jr. Rudolph

K. D.

Potter

K. I.

Lambert

. . . Gathright

(1999). A measure of positive and negative affect for children: Scale development and preliminary validation. Psychological Assessment, 11, 326-338. doi:10.1037/1040-3590.11.3.326

33.

Lazarus

R. S.

Folkman

(1984). Stress, appraisal, and coping. New York, NY: Springer.

34.

Liu

Cox

R. B.

Jr. Washburn

I. J.

Croff

J. M.

Crethar

H. C.

(2017). The effects of requiring parental consent for research on adolescents’ risk behaviors: A meta-analysis. Journal of Adolescent Health, 61, 45-52.

35.

Liu

R. T.

Alloy

L. B.

(2010). Stress generation in depression: A systematic review of the empirical literature and recommendations for future study. Clinical Psychology Review, 30, 582-593.

36.

Lodder

G. M. A.

Scholte

R. H. J.

Goossens

Verhagen

(2017). Loneliness in early adolescence: Friendship quantity, friendship quality, and dyadic processes. Journal of Clinical Child & Adolescent Psychology, 46, 709-720. doi:10.1080/15374416.2015.1070352

37.

March

J. S.

Parker

J. D. A.

Sullivan

Stallings

Conners

C. K.

(1997). The Multidimensional Anxiety Scale for Children (MASC): Factor structure, reliability, and validity. Journal of the American Academy of Child & Adolescent Psychiatry, 36, 554-565. doi:10.1097/00004583-199704000-00019

38.

Muthén

L. K.

Muthén

B. O.

(1998-2015). Mplus user’s guide (7th ed.). Los Angeles, CA: Author.

39.

Newcomb

A. F.

Bagwell

C. L.

(1995). Children’s friendship relations: A meta-analytic review. Psychological Bulletin, 117, 306-347. doi:10.1037/0033-2909.117.2.306

40.

Nolen-Hoeksema

Hilt

L. M.

(2013). The emergence of gender differences in depression in adolescence. In Nolen-Hoeksema

Hilt

L. M.

(Eds.), Handbook of depression in adolescents (pp. 127-152). New York, NY: Routledge.

41.

Nylund

K. L.

Asparouhov

Muthén

B. O.

(2007). Deciding on the number of classes in latent class analysis and growth mixture modeling: A Monte Carlo simulation study. Structural Equation Modeling: A Multidisciplinary Journal, 14, 535-569. doi:10.1080/10705510701575396

42.

Poulin

Chan

(2010). Friendship stability and change in childhood and adolescence. Developmental Review, 30, 257-272. doi:10.1016/j.dr.2009.01.001

43.

Remillard

A. M.

Lamb

(2005). Adolescent girls’ coping with relational aggression. Sex Roles, 53, 221-229. doi:10.1007/s11199-005-5680-8

44.

Roderick

(2003). What’s happening to the boys? Early high school experiences and school outcomes among African American male adolescents in Chicago. Urban Education, 38, 538-607.

45.

Rose

A. J.

Rudolph

K. D.

(2006). A review of sex differences in peer relationship processes: Potential trade-offs for the emotional and behavioral development of girls and boys. Psychological Bulletin, 132, 98-131.

46.

Rubin

K. H.

Dwyer

K. M.

Kim

A. H.

Burgess

K. B.

Booth-Laforce

Rose-Krasnor

(2004). Attachment, friendship, and psychosocial functioning in early adolescence. Journal of Early Adolescence, 24, 326-356. doi:10.1177/0272431604268530

47.

Rudolph

K. D.

(2002). Gender differences in emotional responses to interpersonal stress during adolescence. Journal of Adolescent Health, 30, 3-13. doi:10.1016/S1054-139X(01)00383-4

48.

Rudolph

K. D.

Flynn

Abaied

J. L.

(2008). A developmental perspective on interpersonal theories of youth depression. In Abela

J. R. Z.

Hankin

B. L.

(Eds.), Handbook of depression in children and adolescents (pp. 79-102). New York, NY: Guilford Press.

49.

Rudolph

K. D.

Hammen

Burge

Lindberg

Herzberg

Daley

S. E.

(2000). Toward an interpersonal life-stress model of depression: The developmental context of stress generation. Development and Psychopathology, 12, 215-234.

50.

Rutter

Maughan

Mortimore

Ouston

Smith

(1979). Fifteen thousand hours: Secondary schools and their effects on children. Cambridge, MA: Harvard University Press.

51.

Salvas

M.-C.

Vitaro

Brendgen

Dionne

Tremblay

R. E.

Boivin

(2014). Friendship conflict and the development of generalized physical aggression in the early school years: A genetically informed study of potential moderators. Developmental Psychology, 50, 1794-1807. doi:10.1037/a0036419

52.

Schell

T. L.

Martino

S. C.

Ellickson

P. L.

Collins

R. L.

McCaffrey

(2005). Measuring developmental changes in alcohol expectancies. Psychology of Addictive Behaviors, 19, 217-220. doi:10.1037/0893-164X.19.2.217

53.

Schmeelk-Cone

K. H.

Zimmerman

M. A.

(2003). A longitudinal analysis of stress in African American youth: Predictors and outcomes of stress trajectories. Journal of Youth and Adolescence, 32, 419-430. doi:10.1023/a:1025934301468

54.

Selfhout

M. H. W.

Branje

S. J. T.

Meeus

W. H. J.

(2009). Developmental trajectories of perceived friendship intimacy, constructive problem solving, and depression from early to late adolescence. Journal of Abnormal Child Psychology, 37, 251-264. doi:10.1007/s10802-008-9273-1

55.

Smetana

J. G.

Campione-Barr

Metzger

(2006). Adolescent development in interpersonal and societal contexts. Annual Review of Psychology, 57, 255-284. doi:10.1146/annurev.psych.57.102904.190124

56.

Sontag

L. M.

Graber

J. A.

(2010). Coping with perceived peer stress: Gender-specific and common pathways to symptoms of psychopathology. Developmental Psychology, 46, 1605-1620. doi:10.1037/a0020617

57.

Sontag

L. M.

Graber

J. A.

Clemans

K. H.

(2011). The role of peer stress and pubertal timing on symptoms of psychopathology during early adolescence. Journal of Youth and Adolescence, 40, 1371-1382. doi:10.1007/s10964-010-9620-8

58.

Tofighi

Enders

C. K.

(2008). Identifying the correct number of classes in growth mixture models. In Hancock

G. R.

Samuelsen

K. M.

(Eds.), Advances in latent variable mixture models (pp. 317-341). Greenwich, CT: Information Age.