Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Jun 1.
Published in final edited form as: J Res Adolesc. 2022 Oct 28;33(2):389–403. doi: 10.1111/jora.12809

Characterizing Emotional Support Development: From Adolescent Best Friendships to Young Adult Romantic Relationships

Meghan A Costello 1, Joseph P Allen 1, Sean R Womack 1, Emily L Loeb 1, Jessica A Stern 1, Corey Pettit 1
PMCID: PMC10140188  NIHMSID: NIHMS1843469  PMID: 36305166

Abstract

This study examined development of emotional support competence within close friendships across adolescence. A sample of 184 adolescents (53% girls, 47% boys; 58% White, 29% Black, 14% other identity groups) participated in seven waves of multimethod assessments with their best friends and romantic partners from age 13 to 24. Latent change score models identified coupled predictions over time from emotional support competence to increasing friendship quality and decreasing support received from friends. Friend-rated emotional support competence in adolescence predicted supportiveness in adult romantic relationships, over and above supportiveness in adolescent romantic relationships. Teen friendships may set the stage for developing emotional support capacities that progress across time and relationships into adulthood.

The adolescent transition from parents to peers and romantic partners as primary sources of support has been a focus of study for decades (Allen & Land, 1999; Mikulincer & Shaver, 2020; Oudekerk et al., 2015; Waters & Sroufe, 1983). The ability to establish and maintain supportive, high-quality relationships is a key skill across the lifespan (Allen et al., 2016; Allen et al., 2020; Allen et al., under review). With the adolescent focal shift toward peer relationships, a significant proportion of teens’ learning about supportive relationships takes place with friends. In particular, the “best friend” relationship has been identified as key for adolescents’ developing capacities to manage close interpersonal relationships (Narr et al., 2017; Sullivan, 1953/2013), but not much is known about the specific qualities within best friendships that scaffold teens’ burgeoning ability to navigate emotional support exchanges. Given links from adolescent social experiences to later functioning (Oudekerk et al., 2015), the present study capitalizes on repeated assessments in a multi-reporter, twelve-year longitudinal study to provide a detailed exploration of the close friend contexts that foster adolescents’ emotional support competence. In particular, the study investigates potential inter-relations between the quality of a teen’s best friendship and the amount of emotional support they receive from that friend as potential drivers of their emotional support competence.

Developing emotional support competence could not occur without social experiences, processes which ramp up in intensity during adolescence (Bauminger et al. 2008; Sullivan, 1953/2013). Interpersonal theories of social development emphasize that social interaction in adolescence contributes to identity consolidation and understanding of oneself and others (Sullivan, 1953/2013). “Chumships”—non-romantic, trusting, dyadic relationships with peers—grow in importance during the teenage years, setting the stage for interpersonal processes in romantic relationships in adulthood (Sullivan, 2013). Best friends (particularly close “chums”) offer intimacy that builds teens’ ability to give and receive emotional support, improves relationship quality, and ultimately cascades forward into higher-quality and supportive future relationships with romantic partners (Allen et al., 2020; Ruben et al., 2010; Waters & Sroufe, 1983). Similarly, attachment theory proposes that experiences in close relationships shape individual differences in the development of social-emotional competence, including the ability to provide care and support to others (Bowlby, 1969/1982; Mikulincer & Shaver, 2007).

This study tested the hypothesis that the emotional intimacy of close relationships provides the foundation for emotional support development, with close friendships as key contexts in which teens learn to give and receive support (Loeb et al., 2021; Roisman et al., 2004). Receiving support from a friend reinforces support-seeking and support-receiving interactions, and friendships build mutual support “momentum” over time. The processes of repeatedly giving and receiving support proceed better within high-quality relationships (Demaray & Malecki, 2002), which may become iteratively safer and more secure as these support interactions proceed in reinforcing ways. Within a given relationship, teens’ ability to engage in vulnerable conversations, friends’ responsiveness, and the overall quality and tone of the friendship (i.e., “Can I ask this friend for help?”) all play a part in whether and how emotional support skills develop (Mikulincer & Shaver, 2016). Although only parts of this hypothesized process have been examined, the positive findings to date suggest the value in moving toward a unified description of the inputs that promote emotional support competence. The present study does this by simultaneously considering: 1) the teen’s report of the quality of their best friendship, 2) the best friend’s perception of the teen’s competence in providing emotional support, and 3) the observed emotional support that a teen receives from their peer, with the goal of understanding the relational contexts in which teens develop emotional support competence that is meaningful to their best friend and may impact long-term relationship quality.

Context plays a role in adolescents’ experience of and practice with emotional support conversations. Frequency of opportunities to engage in these conversations vary based on the composition of the dyad. For example, gendered expectations that encourage girls’ expression of emotion, but only reinforce some types of sharing for adolescent boys, may contribute to different frequency of emotional support and vulnerability within friendships (Gillespie et al., 2015; Seiffge-Krenke et al., 2014). Societal and cultural expectations significantly shape not only the opportunities that adolescents have to practice vulnerable conversations, but also the valence and meaning of those conversations that may attenuate or promote learning of emotional support skills (Conger et al., 2010; Consedine et al., 2007; Hamid, 1994). Emotionally vulnerable conversations may also have very different meaning for adolescents experiencing socioeconomic disadvantage; their cognitive and emotional energy may be expended elsewhere if they face challenges such as housing or food insecurity (Evans & Kim, 2013; White & Gager, 2007). Therefore, the current study also considers the role of gender and socioeconomic status as predictors of emotional support development across adolescence.

Adolescents’ beliefs about themselves and others (i.e., attachment style) may also influence social learning that occurs in emotional support interactions. Repeated experiences with close others shape and reinforce these beliefs and expectations about whether others are safe sources of support, as well as whether they are someone who is capable of providing that support to others (Mikulincer & Shaver, 2020). For instance, if a teen receives support from a friend at one point in time, they may be more likely to expect to receive support from friends again in the future. The inverse may also be true; a teen exposed to an unresponsive friend may refrain from seeking emotional support from friends (Loeb et al., 2018; Loeb et al., 2020). Not only do smooth support exchanges offer a chance for teens to practice their emotional support skills, responsive supporters also enhance seekers’ relationship satisfaction, attachment security, and psychological well-being, promoting future engagement with responsive individuals (Haydon et al., 2012; Reis, 2014). Given that some of these processes are primarily intrapsychic, the emotional support skills learned within one relationship may be carried forward into other relationships across adolescence and into adulthood, cascading into more positive relationships, as well.

In the long-term, research suggests that interpersonal experiences in adolescence lay the groundwork for developing support capacities that persist across the lifespan (Capaldi et al., Conger et al., 2012; 2010; Furman, 2018; Rubin et al., 2010). Support in adult relationships has been identified as a strong correlate of relationship satisfaction, as well as a buffer against depressive symptoms, chronic stress, and the long-term development of physical health difficulties (Cramer, 2010; Meuwly et al., 2012). Ultimately, the development of supportive relationships may serve individuals’ mental, physical, and relational health across adolescence and into adulthood. Thus, the current study aims to better understand how this development proceeds.

The current study was a long-term, exploratory investigation of the development of the capacity to give and receive support as it evolves within close relationships over time. The study used prospective multi-reporter data in a socio-demographically diverse community sample, followed from age 13 to 24 to pursue the following goals regarding understanding friendship quality, support competence, and support received and their relation to emotional support functioning in young adult romantic relationships:

Study Goals and Hypotheses:

  1. To better understand univariate developmental trajectories of friendship quality, emotional support competence, and emotional support received from a peer. We hypothesized that friendship quality, emotional support competence, and emotional support received from a close friend would each increase across adolescence (indicated by a positive constant change term in latent change analyses). Additionally, we expected that after accounting for general increases across adolescence, individuals with relatively high amounts of friendship quality, emotional support competence, or emotional support received from a close friend at any time point would go on to demonstrate even greater gains going forward (as indicated by a positive proportional change term).

  2. To investigate potential inter-relations among friendship quality, support competence, and support received. We hypothesized that displaying relatively high friendship quality, emotional support competence, or support received at any given time would predict relative increases in the other constructs at the following time point (as indicated by positive coupling parameters in bivariate models).

  3. To predict from adolescent friendship support processes to functioning in adult romantic relationships. We hypothesized that age 18 friendship quality, emotional support competence, and support received would predict age 24 romantic attachment, supportiveness in romantic relationships, and support received from a partner even after accounting for age 18 romantic partner variables.

Method

Participants

This study was part of a longitudinal investigation of social development across adolescence and young adulthood. Participants included 184 seventh- and eighth-graders (86 boys and 98 girls). Participants were assessed annually; this study utilizes the age 13 (Mage = 13.35, SD = .64) through age 18 (Mage = 18.38, SD = 1.04) assessments with friends as well as the age 18 (Mage = 18.43, SD = 1.00) and age 24 (Mage = 23.99, SD = 1.12) assessments with romantic partners (see Supplemental Table 1 for detailed age breakdown by assessment). Participants were originally recruited from the seventh and eighth grades of a public middle school drawing from suburban and urban populations in the Southeastern United States. Students were recruited via an initial mailing to all parents of students in the school along with follow-up contact efforts at school lunches. Families of adolescents who indicated interest in the study were contacted by telephone. Of all students eligible for participation, 63% agreed to participate either as target participants or as friends who participated in interactions tasks with the target teen. Once a student participated as a friend, they were no longer eligible to be a primary participant. Participants were not required to list distinct friends from one another, leading to some repeated participation of peers, ranging from a high of 12 individuals who had a close peer who had already served as a peer to another participant (at age 13), to a low of 2 individuals who had peers who had already served as a peer to another participant (at age 18; see Supplemental Table 2 for detailed rates).

The sample was racially and socioeconomically diverse: 107 adolescents (58%) identified themselves as White, 53 (29%) as Black, 15 (8%) as Multiracial, and 9 (5%) as being from other identity groups, which approximately mirrors the distribution of the catchment area for the school from which the sample was drawn. Adolescents’ parents reported an annual median family income in the $40,000–$59,999 range, relative to a national median household income of approximately $39,000 at the time (US Census Bureau, 1999).

Each year, target adolescents nominated their closest friend to be included in survey and observational measures with them in the study. The same peer participated in subsequent years, on average, 42.3% of the time (min = 35.9%, max 45.5%). All best friend dyads identified as same-gender, although this was not a requirement to participate in the study.

At two assessment time points (covering a three-year timespan, first when participants were 16–18 and then at ages 23–25), participants’ romantic partners were invited to participate in survey and observational measures after reporting a relationship of at least 3 months in duration (Mduration = 2.67 years, SD = 2.28 years). This age range was chosen for follow-up data collection to increase the likelihood that participants had established relationships of at least this duration and were more likely to be living independently (after traditional-aged students completed college). At age 16–18, 139 participants (79.4% of the sample) indicated that they were in a relationship of at least 3 months’ duration. At age 24, 137 participants (78.3% of the sample) indicated that they were in a relationship of at least 3 months’ duration. Relationship status and participation with a romantic partner at both time points was not associated with demographic variables, baseline support variables, or concurrent support variables. All participating couples identified as heterosexual, although this was not a requirement to participate in the study.

Procedures

In the initial introduction and throughout each session, confidentiality was explained to all participants, and adolescents were told that their parents would not be informed of any of the answers they provided. A Confidentiality Certificate issued by the U.S. Department of Health and Human Services protected all data from subpoena by federal, state, and local courts.

Attrition Analysis

On average, individuals participated at 6.34 of the 7 study time points (SD = 1.16). Attrition at each study time point was also low, with participation rates of: 89% at age 14, 88% at age 15, 88% at age 16, 90% at age 17, and 74% at age 18. Of the 184 teens who participated at age 13, 94% also participated at age 24. Individuals who completely dropped out of the study (participated at age 13 but did not participate at age 24) did not differ from teens who remained in the study on any baseline measures (friendship quality, emotional support behaviors, and emotional support competence), gender, or familial income. Nevertheless, to best address any possible biases due to attrition in longitudinal analyses or missing data within waves, we used full-information maximum-likelihood methods with the full original sample for all analyses. These methods have been found to yield the least-biased estimates when all available data are used for longitudinal analyses (vs. listwise deletion of missing data; Arbuckle, 1996).

Measures

Quality of Relationship with Friends (Age 13–18).

The Inventory of Parent and Peer Attachment (Armsden & Greenberg, 1987) was used annually to assess adolescents’ perceptions of the overall quality of their relationship with their friends in terms of the degree of trust, communication, and alienation in the relationship. All items asked teens to reflect on their relationship with their closest friend. A composite score was obtained from 25 five-point Likert scale items, such as “They respect my feelings” (1 = Never true, 2 = Seldom true, 3 = Sometimes true, 4 = Often true, 5 = Almost always true). This composite measure has been shown to have good test-retest reliability and has been related to measures of psychological well-being, self-satisfaction, decreased depression, higher likelihood of seeking social support, and less symptomatic responses to stressful life events (Armsden & Greenberg, 1987). Internal consistency in the present sample was excellent (Cronbach’s α’s = .92-.94).

Friend-Rated Emotional Support Competency (Age 13–18).

The Interpersonal Competency Questionnaire (Buhrmester, 1990) was annually administered to the target teen’s nominated closest friend, to assess that friend’s perception of the target teen’s interpersonal abilities. Stems were phrased such that friends were responding to items that directly described the interpersonal competencies of the target participant. Friends responded to questions such as “How good is [participant] at making someone feel better when they are unhappy or sad?” on a scale from 1 = Poor at this (defined as “they would be so uncomfortable and unable to handle this situation that it would be avoided if at all possible”) to 5 = Extremely good at this (defined as “they would feel very comfortable and could handle this situation well”). Eight items were summed together to obtain the Emotional Support subscale. Scores have previously been found to be related to sociability, self-esteem, and reduced anxiety and depression, but are discrete from perceived intimacy in the relationship (Buhrmester, 1990). Internal consistency was excellent (Cronbach’s α’s = .91-.95).

Emotional Support Received (Ages 13–18, 24).

Target teens and their nominated close friends participated in an observed social interaction task in private offices within a university building annually from age 13 to 18. At age 18 and 24, participants engaged in this same task with a romantic partner. In the six-minute task, participants asked their interaction partners for advice on a self-selected topic, to approximate natural social support processes. This interaction was then coded for the level of Emotional Support Received from the interaction partner using the Supportive Behavior Task Coding System for Adolescent Peer Dyads (Allen et al., 2001).

The Emotional Support Received code describes the level to which the partner attempts to understand and support the feelings raised by the participant, through processes such as sympathizing, naming the emotion, eliciting further emotion, or making a commitment to be emotionally available. Scores range from 0 = no attempt to emotionally support; to 4 = clear recognition of emotional distress (as defined by attempts to draw the speaker out, and clear expressions of warmth, concern, and sympathy throughout most of the interaction). The Intraclass Correlations, a measure of inter-rater reliability, ranged from .61-.78, which is considered in the good- to excellent-range for this statistic (Cicchetti & Sparrow, 1981).

Partner-Rated Support in Relationship (Age 18, 24).

The Network of Relationships Inventory (NRI; Furman & Buhrmester, 1985) was administered to participants’ romantic partners at ages 18 and 24 to assess partners’ perception of support in their relationship. Partners responded to questions, such as “When you are feeling down or upset, how often do you depend on this person to cheer you up?” on a scale from 1 = never/none to 5 = extremely much. Three items from the NRI were summed to obtain the Support subscale, which assesses the degree to which individuals perceive the relationship as a source of support. Internal consistency was good (Cronbach’s α = .81).

Romantic Attachment Security (Age 18, 24).

Romantic attachment style was assessed using participant self-report on the Experiences in Close Relationships questionnaire (Brennan et al., 1998). The Experiences in Close Relationships questionnaire is a 36-item measure, which assesses avoidant (18 items, including “I feel comfortable depending on romantic partners”) and anxious attachment styles (18 items, including “I worry that romantic partners won’t care about me as much as I care about them”) with romantic partners in general. Items are scored on a 7-point Likert scale where 1 = disagree strongly, 4 = neutral/mixed, and 7 = agree strongly, where several items are reverse coded so that higher scores indicate greater anxious or avoidant feelings and behaviors. For the purposes of this study, the total sum score of all items was reversed so that higher scores reflect more secure attachment to remain consistent with the other variables (for which higher scores reflect greater or more of that particular construct). This approach has been identified as a sufficient measure of secure attachment (Brenning et al., 2015). The Experiences in Close Relationships questionnaire has demonstrated strong validity with theoretical accounts of attachment dimensions (Bartholomew & Horowitz, 1991). The total sum score of the Experiences in Close Relationships questionnaire demonstrated excellent internal consistency in this sample (Cronbach’s α’s = .96).

Participant Demographic Information (Age 13).

Participant gender was collected at the start of the study, participants were asked to choose between “male” and “female” as their gender identity. Participants’ parents reported on annual household income on a scale where 1 = under $5,000; 2 = $5,000–9,999; 3 = $10,000–14,999; 4 = $15,000–19,999; 5 = $20,000–29,999; 6 = $30,000–39,999; 7 = $40,000–59,999; and 8 = $60,000 or more.

Results

Descriptive Statistics

Raw means, standard deviations, maximum and minimum reported values, and sample sizes for all examined variables are presented in Table 1; variable intercorrelations are presented in Supplemental Table 3. Prior to latent change score modeling, variables were linearly transformed so that all scores were on similar scales (by dividing raw scores by their mean) and that the minimum scale value for all variables was greater than 0 (by adding 1 where necessary).

Table 1.

Variable Descriptive Statistics

Variable Friendship Quality Emotional Support Competence Emotional Support Received
M SD Min Max N M SD Min Max N M SD Min Max N
Age 13 102.21 13.93 58 125 183 27.56 6.48 13 40 173 1.07 1.04 0 3.75 179
Age 14 103.26 12.98 63 125 165 26.99 6.39 12 40 156 0.89 0.91 0 3.75 164
Age 15 101.91 14.50 74 125 162 27.61 6.52 8 40 142 1.05 0.98 0 4.00 149
Age 16 103.82 13.44 62 125 162 27.88 7.14 8 40 142 1.13 0.99 0 4.00 146
Age 17 106.76 14.24 67 125 166 29.44 6.08 14 40 145 0.67 0.87 0 3.50 141
Age 18 106.50 13.96 53 125 137 29.04 5.85 12 40 122 0.57 0.80 0 2.75 129
Variable Romantic Attachment Security Supportiveness in Relationship Emotional Support Received
M SD Min Max N M SD Min Max N M SD Min Max N
Age 18 165.89 30.92 89 215 97 11.56 3.08 5 15 86 0.64 0.83 0 3.2 64
Age 24 155.80 32.74 58 218 173 12.42 2.19 6 15 101 0.55 0.70 0 3.5 90
Open in a new tab

Note: To calculate romantic attachment, the Anxious and Avoidant subscales on the Experiences in Close Relationships questionnaire were combined and then reverse-scored so that higher values corresponded to greater attachment security.

Primary Analyses

R package lavaan version 0.6–8 (Rosseel, 2012) was used to perform nested model comparisons. A series of univariate dual change score model comparisons was performed for each of the three major constructs (friendship quality, support competence, and support received), in order to assess the extent to which presence of a quality at one point was predicted not just by overall trait-like levels of the quality or general patterns of growth, but by the degree to which that quality existed at the prior time point (McArdle & Hamagami, 2001). A dual change score model is a longitudinal extension of a latent change score model that decomposes the rate of growth into constant and proportional growth. Constant growth reflects an individual’s average rate of change over time akin to a slope parameter in a linear latent growth curve model. Proportional growth refers to deviation from the constant rate of growth over time. Positive proportional growth parameter estimates reflect an increasing rate of change over time whereas negative proportional growth parameters reflect a decreasing rate of change over time. Therefore, proportional growth parameters add nonlinearity to constant linear growth. All models include intercorrelations among constructs at each study time point, which were allowed to vary at baseline and then constrained to be equal across the five following time points. Adolescent gender and socioeconomic status were included as baseline covariates in all models by regressing the latent intercept and constant growth parameters on these demographic variables.

Univariate dual change score models can be extended to a multivariate framework through the inclusion of coupling paths (lagged associations between the two variables-of-interest) and correlations between the constant growth parameters (see Figure 2). For example, by including coupling paths, the rate of growth of friendship quality is informed not only by previous measurements of friendship quality, but also previous measurements of emotional support received and vice versa. Bivariate models were guided by best-fit univariate models. A series of nested model comparisons were used to determine whether including coupling parameters improved model fit.

Figure 2. Hypothesized Bivariate Dual Change Score Model with Cross-Lags.

Figure 2.

Open in a new tab

Includes latent slope and intercepts, constant change parameters (α), and proportional change parameters (β) for each variable. Coupling parameters are denoted as γ. Correlations between x and y at each time point were included in model calculations but are not pictured for clarity.

Follow-up predictions to supportive processes in adult romantic relationships were made by comparing predictions from adolescent peer relationship processes and adolescent romantic relationship processes in a hierarchical linear regression, controlling for adolescent gender and socioeconomic status, and using FIML to handle missing data.

Study Goal 1. To better understand univariate developmental trajectories of friendship quality, emotional support competence, and emotional support received from a peer. We hypothesized that friendship quality, emotional support competence, and emotional support received from a close friend would each increase across adolescence (indicated by a positive constant change term in latent change analyses). Additionally, we expected that after accounting for general increases across adolescence, individuals with relatively high amounts of friendship quality, emotional support competence, or emotional support received from a close friend at any time point would go on to demonstrate even greater gains going forward (as indicated by a positive proportional change term).

A linear dual change score model was the best fit model to describe the development of friendship quality (CFI = .817, AIC = 2163.2, BIC = 2192.0, RMSEA = .123 [90% CI = .100, .148], SRMR = .124; see Table 2 for model parameters, Supplemental Tables 4 and 5 for model comparisons). The positive constant change (β1 = 10.491, p < .001) and negative proportional change parameters (βx = −1.007, p < .001) describe the development of friendship quality in this sample as increasing across adolescence, yet with an asymptote for those who demonstrated relatively high friendship quality initially (i.e., individuals tended to increase from year to year, but when levels of friendship quality were higher, the rate of increase slowed).

Table 2.

Best-Fit Univariate Model Parameters, Including Gender and Income as Baseline Covariates

Friendship Quality Emotional Support Competence Emotional Support Received
Intercept (β0) 1.022 (.010)*** 5.385 (.085)*** .743 (.183)***
Constant Change (β1) 10.491 (.103)*** .084 (.024)*** --
Proportional Change (βx) −1.007 (.007)*** -- --
Gender effect (βGender) .013 (.018) .798 (.114)*** .472 (.077)***
Income effect (βSES) .019 (.004)*** .044 (.029) .072 (.020)***
Covariance of β0 and β1 .082 (.013)*** −.060 (.034) --
Open in a new tab

Note: Constant change (β1) represents the slope, or general trend across the study. Proportional change (βx) represents the extent to which information about the previous time point is useful in predicting change in a construct, over and above constant change. Gender effect (βGender) and Income effect (βSES) represent predictions from participant-reported gender and baseline family SES to the latent variable-of-interest at time 1. Covariance of β0 and β1 represents the strength of covariance between intercept and slope terms in the model.

Standard error terms are described in parentheses.

***

p < .001

An intercept plus constant change model was the best fit for emotional support competence (CFI = .883, AIC = 2774.6, BIC = 2800.2, RMSEA = .058 [90% CI = .025, .087], SRMR = .098; see Table 2 for model parameters; see Supplemental Tables 4 and 5 for model comparisons). The positive constant change parameter (β1 = .084, p < .001) indicates that the sample was showing increases in emotional support competence, on average, across the study.

An intercept-only model was the most parsimonious descriptor for observed emotional support received from a best friend across adolescence, although model fit was poor (CFI = .439, AIC = 2393.615, BIC = 2409.6, RMSEA = .095 [90% CI = .071, .119], SRMR = .134; see Table 2 for model parameters; see Supplemental Tables 4 and 5 for model comparisons). This suggests that there is not a systematic, linear pattern of development identifiable in this sample.

Study Goal 2. To investigate potential inter-relations among friendship quality, support competence, and support received. We hypothesized that displaying relatively high friendship quality, emotional support competence, or support received at any given time would predict relative increases in the other constructs at the following time point (as indicated by positive coupling parameters in bivariate models).

Bivariate associations between friendship quality and emotional support competence were best described by a model that included coupling parameters from emotional support competence to friendship quality (CFI = .939, AIC = 4875.2, BIC = 4948.8, RMSEA = .048 [90% CI = .027, .066], SRMR = .090; see Table 3 for model parameters; see Supplemental Table 6 for model comparisons). Coupling parameter estimates were positive (γCompetence→Quality = 1.356, p < .001), indicating that higher emotional support competence at one time point was associated with an increase in the rate of friendship quality growth at the subsequent time point, even after accounting for baseline correlations and global patterns of change in these constructs. Covariate associations suggest that girls reported higher friendship quality (βGender = .077, p < .001) and were rated as more competent at providing emotional support at baseline (βGender = .655, p < .001), while teens from higher SES families reported higher friendship quality (βSES = .021, p < .001).

Table 3.

Bivariate Latent Change Score Model of Friendship Quality and Support Competence Across Adolescence

Parameter Estimate (Standard Error)
Friendship Quality
Intercept (β0) .771 (.044)***
Constant Change (β1) 3.103 (1.267)*
Proportional Change (βx) -1.024 (.008)***
Support Competence
Intercept (β0) 4.143 (.296)***
Constant Change (β1) -.063 (.241)
Proportional Change (βx) --
Coupling: Competence → Friendship Quality
Coupling γCompetence→Quality 1.356 (.230)***
Demographic Effect Estimates
βGender → Friendship Quality .077 (.019)***
βSES → Friendship Quality .021 (.005)***
βGender → Support Competence .655 (.130)***
βSES → Support Competence .032 (.023)
Covariances: Intercept & Constant Change Terms
β0Quality, β1Competence -.001 (.004)
β0Competence, β1Quality -.748 (.346)**
β0Quality, β1Competence .063 (.014)***
β0Competence, β1Competence -.040 (.024)
β0Competence, β0Quality .012 (.009)
β1Competence, β1Quality .038 (.027)
Open in a new tab

Note: Constant change (β1) represents the slope, or general trend across the study. Proportional change (βx) represents the extent to which information about the previous time point is useful in predicting change in a construct, over and above constant change. Coupling parameters describe the extent to which change in one variable from one time point to the next is predicted by the other variable at the previous time point, after accounting for covariances and univariate growth trajectories of each variable.

***

p < .001,

**

p < .01,

*

p < .05

Bivariate associations between emotional support competence and support received were best described by a model that included coupling parameters from emotional support competence to emotional support received (CFI = .722, AIC = 5242.8, BIC = 5303.9, RMSEA = .072 [95% CI = .054, .090], SRMR = .102; see Table 4 for model parameters; see Supplemental Table 6 and 7 for model comparisons). The intercepts of support competence and support received were significantly correlated, suggesting that these constructs are related at baseline (β0Competence, β0SuppRecv = .270, p < .001). Coupling parameter estimates were negative (γCompetence→SuppRecy = −.766, p < .01), indicating that individuals with greater emotional support competence at one time point experienced a relative decrease in the rate of growth emotional support received at the next time point. This model did not include gender or income as covariates, due to limitations to model convergence. This is likely due to the increased complexity of the model and reduced N for the emotional support received task, which was more time consuming than survey measures and required participants to visit the lab in-person.

Table 4.

Bivariate Latent Change Score Model of Support Received and Support Competence Across Adolescence

Parameter Estimate (Standard Error)
Emotional Support Received
Intercept (β0) 1.986 (.066)***
Constant Change (β1) --
Proportional Change (βx) --
Support Competence
Intercept (β0) 5.390 (.085)***
Constant Change (β1) .214 (1.408)
Proportional Change (βx) --
Coupling: Competence → Support Received
Coupling γCompetence→SuppRecy -.766 (.262)**
Demographic Effect Estimates
βGender → Support Received --
βSES → Support Received --
βGender → Support Competence --
βSES → Support Competence --
Covariances: Intercept & Constant Change Terms
β0SuppRecv, β1Competence −.035 (.050)
β0Competence, β1SuppRecv .288 (.152)
β0SuppRecv, β1Competence .051 (.091)
β0Competence, β1Competence −.041 (.121)
β0Competence, β0SuppRecv .270 (.074)***
β1Competence, β1SuppRecv .014 (.086)
Open in a new tab

Note: Constant change (β1) represents the slope, or general trend across the study. Proportional change (βx) represents the extent to which information about the previous time point is useful in predicting change in a construct, over and above constant change. Coupling parameters describe the extent to which change in one variable from one time point to the next is predicted by the other variable at the previous time point, after accounting for covariances and univariate growth trajectories of each variable.

***

p < .001,

**

p < .01

Study Goal 3. To predict from adolescent friendship support processes to functioning in adult romantic relationships. We hypothesized that age 18 friendship quality, emotional support competence, and support received would predict age 24 romantic attachment, supportiveness in romantic relationships, and support received from a partner even after accounting for age 18 romantic partner variables.

Friend-reported emotional support competence at age 18 predicted emotional support reported by romantic partners at age 24 over and above emotional support reported by romantic partners at age 18 (β = .248, p < .05; see Table 5). No associations were identified from friendship quality to romantic attachment or from emotional support received from a peer at age 18 to support received from a romantic partner at age 18.

Table 5.

Prediction of Support in Romantic Relationships at Age 24 from Peer Support Competence at Age 18

Support in Romantic Relationship at Age 24 (RP Report of NRI)
β ΔR2 R 2
Gender −.403*** .123***
Family Income −.054 .001 .124
Partner-Reported Support in Romantic Relationship (Age 18) .119 .016 .142
Best Friend-Reported Emotional Support Competence (Age 18) .247** .047** .189**
Open in a new tab

Note:

***

p < .001.

**

p < .01.

β weights are from final model. RP = romantic partner; NRI = Non-Relational Interview.

Post-Hoc Analyses

The Role of Friendship Stability

We considered the possibility that continuity (or lack of continuity) of the closest friend over time might alter the relation between the development of friendship quality, supportive behaviors, and support skills within peer dyads. Codes were created at each wave indicating whether the same friend did vs. did not participate at the prior wave. These friendship stability codes were then regressed onto the latent change scores in the best-fit models, to determine whether the presence of the same or different peer at subsequent waves altered predictions across adolescence. In univariate models, no associations were identified between change scores and the presence or absence of the same peer, suggesting that development proceeded similarly for participants who repeatedly participated with the same best friend and for those whose best friends changed during the study.

The Role of Gender

We considered the possibility that gender may influence not only baseline functioning, but trajectories across time. Models were generated that analyzed predictions from gender identity to latent change factors in each of the best-fit models. Univariate model predictions indicated that girls’ reports of friendship quality increased more quickly than boys’ (βGender = .116, p < .001) and that girls’ reports of emotional support competence increased more quickly than boys’ (βGender = .068, p < .001). In both cases, model fit was good and all substantive estimates remained significant, though smaller (see Supplemental Table 8).

Mean Adolescent Peer Support Processes From Age 13 to 18 and Later Romantic Relationship Support

We also considered whether social experiences within peer relationships across adolescence (rather than just at age 18) predicted support in future romantic relationships. Summary variables of friendship quality, emotional support competence, and emotional support received across adolescence were created by taking the mean of each construct across all participating time points from age 13 to 18. Then, regressions using full information maximum likelihood and controlling for gender and income were used to evaluate whether friendship experiences predicted functioning in romantic relationships over and above romantic relationship support variables at age 18. Support received in the observational task from a romantic partner at age 24 was predicted by the mean amount of support a participant received from their peer in that task across adolescence, over and above the amount of support they received from their romantic partner in the same task at age 18 (β = .40, p < .001; see Table 6). No predictions were identified from mean level of friendship quality in adolescence to changes in attachment in adult relationships and from mean emotional support competence across adolescence to supportiveness in adulthood.

Table 6.

Prediction of Support Received from Partner at Age 24 from Mean Support Received from Peers in Adolescence

Support Received from Romantic Partner at Age 24 (Observed)
β ΔR2 R 2
Gender −.26* .
Family Income −.08
Support Received from Partner (Age 18) .24
Mean Support Received from Peer (Age 13–18) .40***
Open in a new tab

Note:

***

p < .001.

**

p < .01.

*

p < .05.

β weights are from final model.

Discussion

This study characterized the development of adolescent friendship quality, an adolescent’s friend-rated competence in providing emotional support, and the amount of support that teen receives from their friend from age 13 to 18, as well as predictions to related supportive processes within romantic relationships at age 24. Friendship quality increased across adolescence, with an asymptote effect for those who demonstrated higher friendship quality early on, emotional support competence increased and support received demonstrated no reliable pattern of change. Higher emotional support competence at one year predicted increases in friendship quality and decreases in emotional support received at future time points. Friend-rated emotional support competence at age 18 predicted romantic partners’ report of support in the relationship at age 24, over and above romantic partners’ report of support at age 18. Each of these findings is discussed in detail below, followed by consideration of their limitations.

Emotional Support Competence as a Driver of Relationship Development

In this study, emotional support competence arose as a key component in the development of high-quality, intimate friendships. Providing effective support to close others has been identified as a skill that promotes intimacy, engagement, and connectedness (Furman, 2018; Stotsky et al., 2020; Stern et al., 2021). On the other hand, higher emotional support competence at one time point also predicted relative decreases in support received from a friend at the following time point. It may be that emotional support competence also serves as a marker of adolescents’ broader emotional functioning, such that competent adolescents need less support going forward. Ongoing work has taken up this question, suggesting that adolescents’ burgeoning ability to care for close others and respond to emotional need is a major marker of long-term functioning (Allen et al., under review). This suggests that—while efforts to foster adolescent friendships may be difficult—targeting adolescents’ underlying competence in managing emotions may support their engagement in high-quality friendships (Allen et al., 2002; Costello et al., 2022; Wingspread, 2004).

Looking into the early twenties, emotional support competence in late adolescence predicted age 24 romantic partner reports of support in their relationship over and above age 18 romantic partner reports of support in the relationship. Previous work has questioned whether the skills developed in same-gender friendships can directly translate to cross-gender romantic relationships, or if teens must adapt those skills as the focal relationship changes (Connolly et al., 2015). Indeed, most work that addresses patterns from peer relationships to romantic relationships focuses on the pathway from deviant peer relationships to intimate partner violence (Dishion et al., 2012; Miller et al., 2009), but little is known about the development of supportive peer processes across these two central relationships. Furthermore, other work has suggested that the presence of romantic partners in adolescence may alter the way that teens interact with their peer network, and more work is necessary to disentangle this nuance (Zimmer-Gembeck, 2002). However, findings from the current study suggest that developing emotional support competence in late adolescence may lend itself to more evident support within a future romantic relationship. In particular, emotional support competence at age 18, when relationship focus begins to shift toward establishing long-term romantic relationships, was predictive of the amount of support perceived by future romantic partners; reinforcing the idea that skills developed in adolescent friendships are generalizable (Allen et al., 2020).

Generalizability of Emotional Support Competence and Skills

In addition to finding predictions from adolescent emotional support competence to adult romantic relationship supportiveness, we also identified several indicators suggesting that the skills developed within teen best friendships generalize to other relationships. Friend-rated competence in providing support also increased across adolescence for the sample as a whole. Given that participating best friends differed across years more often than not, these findings suggest that teens are learning about offering emotional support to close others in general, and not just to a specific friend. Indeed, while we are unable to interpret a null finding, results did not differ when accounting for stability vs. change in the peer that participated in the study at a given time point. Previous work has also observed that, even when interaction partners differ, the interaction dynamics may be comparable, maintaining teens’ developmental trajectories across friendships by similarly shaping their expectations of peers (Dishion et al., 1997; Loeb et al., 2018). The accumulation of experiences with different partners may shape expectations of future relationships, consistent with attachment theory (Mikulincer & Shaver, 2020); this interpretation is supported by the finding that average support received from friends across adolescence predicted support received from romantic partners in the twenties. As expectations accumulate and skills are honed, the adolescent best friendship serves as a template for future close relationships, including romantic partnerships.

The Promising Idea of General Growth and Catch-Up

Friendship quality and emotional support competence both demonstrated growth across the adolescent time frame. This lends itself to a promising conclusion that teens are, generally, learning the skills that are important for relationship maintenance. Additionally, although friendship quality was increasing generally over time, teens with relatively lower quality friendships at a given year tended to make relatively greater gains in friendship quality over the following year—evidence of a potential ‘catch up’ effect. Said differently: teens in low-quality friendships are not doomed to remain in those low-quality friendships forever. Over time, for teens across the spectrum of friendship quality, accumulating positive relationship experiences may consolidate the lesson that close relationships are rewarding; a conclusion that could have long-term implications for relationship functioning.

Support Received in Adolescence

Despite its intuitive and established importance (Priem & Solomon, 2014), observed levels of emotional support received from a best friend across adolescence demonstrated limited developmental change. Developmentally, this may reflect changes in support sought from different sources, and, although same-age peers are often a source of emotional support, they are also a key source of instrumental or informational support (Hombrados-Mendieta et al., 2012). Perhaps, as stressors vary across adolescence, teens adjust their priorities when seeking support. Different topics may better lend themselves to emotional support; for instance, managing a fraught parental relationship vs. discussing the pros and cons of two different universities in order to make a decision about where to apply (topics which may also elicit quite unique emotional responses for two different teens). Despite the fact that emotional support received across adolescence did not follow a particular developmental pattern, teens still demonstrated gains in emotional support competence. Ultimately, the amount of support received across adolescence predicted the amount of support received in adult romantic relationships, suggesting that supportive experiences in adolescence may shape expectations that influence partner selection and romantic behaviors that elicit support (Furman, 2018).

The Role of Context

Each of the support development processes was also linked to adolescents’ social context, although it is important to note that inferences from this study are limited in their scope and generalizability. Self-identified female gender was related to greater emotional support competence and emotional support received at age 13, and girls demonstrated greater gains in each than did boys. Adolescents from families with higher baseline income and girls were advantaged in terms of baseline and development of emotional support. It is possible that these advantages are rooted in societal expectations, socialization, and opportunities for practicing supportive experiences (Cohen & Garcia, 2008; Connolly et al., 2015; Gillespie et al., 2015). For instance, perhaps girls are surrounded by adults who reinforce behaviors that are related to emotional support, while boys are provided with less or different scaffolding (or, in concerning cases, are punished for displays of emotion and empathy). Nonetheless, other work has found that social support correlates with boys’ self-esteem and relates to the psychological health of economically disadvantaged teens, suggesting that it is not simply that boys and adolescents from families with lower incomes do not want (or need) emotional support from peers (Demaray & Malecki, 2002; Moran & Eckenrode, 1991). More work is clearly needed to understand the nuances of contextual expectation, intersectionality, and social development within and across dyads that are comprised of many different identities.

Limitations & Future Directions

Significant limitations of this study warrant consideration. Most importantly, we are unable to make causal inferences based on the interrelations observed here. Unobserved variables or processes may underlie the predictions identified by these models; variables such as parent socialization, teens’ relationships with siblings, empathy, social skills, or personality traits known to influence social behavior such as neuroticism and agreeableness. Relatedly, it is likely that participants’ ratings of friendship quality were influenced by that friend’s emotional support competence. While we attempt to handle this challenge by gathering information from multiple reporters, it is difficult to fully disentangle constructs that appear quite as inherently linked as friendship quality and emotional support. Power was also modest to detect relations between variables and limited our ability to thoroughly investigate moderators and covariates in support development. Further, the measures used provide only a small window into adolescents’ rich social lives, which undoubtedly vary across time, context, and interaction partners outside of the observed support task. Future work might consider the role that friendship stability may play in the development of friendship quality and emotional support capacities over time; perhaps the maintenance of consistent friendships or many different close friendships facilitates different growth trajectories over time.

There are several limitations related to the structure of this study. Repeated participation by the same close friend with two participants created some statistical interdependence, a limitation of our study and any study conducted within a school environment where participants regularly interact with one another. In addition, measures administered to close peers and romantic partners differed, and though chosen to be as similar as possible, this mismatch may have obscured some continuities. In terms of the sample, our findings are not generalizable to all dyads: for instance, we did not examine any cross-gender adolescent friendships, any romantic relationships that do not identify as heterosexual, and any romantic relationships that were shorter than three months in duration or non-monogamous, despite the significant developmental role those relationships may play.

This study is also limited in its consideration of gender identity only along a binary, and future work should consider gender identities outside of this binary, as well as friendships and romantic relationships that are composed of different gender and sexual identities. Additionally, although these findings suggest that the close peer dyad serves an important role in support development, the study did not consider the emotional content discussed during supportive interactions. Emotional support capacities may develop differently depending on the types of stressors adolescents face and the extent of alternative sources of support and stress in their lives. Future work is also needed to better understand temporal nuances of emotional support development.

Overall, however, these findings suggest that emotionally supportive conversations within friendships as early as age 13 set the stage for longer-term social development. Adolescents who function relatively well in terms of emotional support competence at baseline develop increasingly strong relationships and competence into late adolescence, which may lend themselves to the development of supportive future romantic relationships. Future work targeting adolescents’ emotional support skills may have promise in laying a foundation for the development of intimate, supportive relationships into adulthood.

Supplementary Material

SUPINFO

Figure 1. Hypothesized Univariate Dual Change Score Model Path Diagram.

Figure 1.

Open in a new tab

Dual change score model, where ΔK=constant (1), μ0=mean of latent intercept of a given variable, Δx[t]=latent changes at time t, μs=mean of latent slope, α=constant change score (or slope loadings), σ0s=correlation between latent intercept and slope, β=proportional change parameter.

Acknowledgements:

This study was supported by grants from the National Institute of Child Health and Human Development and the National Institute of Mental Health (5R37HD058305-23, R01HD058305-16A1, R01-MH58066).

References

  1. Allen JP, Costello MA, Pettit C, Stern JA, & Uchino BN Adolescent Caregiving Success as a Predictor of Social Relationship Qualities from Age 13 to 33. [Paper Under Review]. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Allen JP, Grande L, Tan J, & Loeb E. (2016). Parent and peer predictors of change in attachment security from adolescence to adulthood. Child Development, 89(4): 1120–1132. 10.1111/cdev.12840 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Allen JP, Hall FD, Insabella GM, Land DJ, Marsh PA, & Porter MR (2001). Supportive Behavior Task Coding Manual. Charlottesville, VA: University of Virginia. [Google Scholar]
  4. Allen JP, & Land D. (1999). Attachment in adolescence. In Cassidy J. & Shaver PR (Eds.), Handbook of attachment: Theory, research, and clinical applications (p. 319–335). Guilford Press. [Google Scholar]
  5. Allen JP, Marsh P, McFarland C, McElhaney KB, Land DJ, Jodl KM, & Peck S. (2002). Attachment and autonomy as predictors of the development of social skills and delinquency during midadolescence. Journal of Consulting and Clinical Psychology, 70(1), 56–66. 10.1037/0022-006X.70.1.56 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Allen JP, Narr RK, Kansky J, & Szwedo DE (2020). Adolescent peer relationship qualities as predictors of long‐term romantic life satisfaction. Child Development, 91(1), 327–340. 10.1111/cdev.13193 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Arbuckle JL (1996). Full information estimation in the presence of incomplete data. In Marcoulides GA, & Schumacker RE (Eds.), Advanced structural equation modeling: Issues and techniques (pp. 243–277). Lawrence Erlbaum Associates. [Google Scholar]
  8. Armsden GC & Greenberg MT (1987). The inventory of parent and peer attachment: Individual differences and their relationship to psychological well-being in adolescence. Journal of Youth and Adolescence, 16, 427–454. 10.1007/BF02202939 [DOI] [PubMed] [Google Scholar]
  9. Bauminger N, Finzi-Dottan R, Chason S, & Har-Even D. (2008). Intimacy in adolescent friendship: The roles of attachment, coherence, and self-disclosure. Journal of Social and Personal Relationships, 25(3), 409–428. DOI: 10.1177/0265407508090866 [DOI] [Google Scholar]
  10. Buhrmester D. (1990). Intimacy of friendship, interpersonal competence, and adjustment during preadolescence and adolescence. Child Development, 61(4): 1101–1111. 10.1111/j.1467-8624.1990.tb02844.x [DOI] [PubMed] [Google Scholar]
  11. Capaldi DM, Knoble NB, Shortt JW, & Kim HK (2012). A systematic review of risk factors for intimate partner violence. Partner abuse, 3(2), 231–280. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Cicchetti DV & Sparrow SA (1981). Developing criteria for establishing interrater reliability of specific items: Applications to assessment of adaptive behavior. American Journal of Mental Deficiency, 86(2), 127–37. [PubMed] [Google Scholar]
  13. Cohen GL, & Garcia J. (2008). Identity, belonging, and achievement: A model, interventions, implications. Current Directions in Psychological Science, 17(6), 365–369. 10.1111/j.1467-8721.2008.00607.x [DOI] [Google Scholar]
  14. Colarossi LG & Eccles JS (2000). A prospective study of adolescents’ peer support: Gender differences and the influence of parental relationships. Journal of Youth and Adolescence, 29, 661–678. 10.1023/A:1026403922442 [DOI] [Google Scholar]
  15. Collins A, & van Dulmen M. (2006). Friendships and romance in emerging adulthood: Assessing distinctiveness in close relationships. In Arnett JJ & Tanner JL (Eds.), Emerging adults in America: Coming of age in the 21st century (pp. 219–234). American Psychological Association. 10.1037/11381-009 [DOI] [Google Scholar]
  16. Collins WA, Welsh DP, & Furman W. (2009). Adolescent romantic relationships. Annual Review of Psychology, 60, 631–652. doi: 10.1146/annurev.psych.60.110707.163459. [DOI] [PubMed] [Google Scholar]
  17. Conger RD, Conger KJ, & Martin MJ (2010). Socioeconomic status, family processes, and individual development. Journal of Marriage and Family, 72(3), 685–704. 10.1111/j.1741-3737.2010.00725.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Consedine NS, Sabag-Cohen S, & Krivoshekova YS (2007). Ethnic, gender, and socioeconomic differences in young adults’ self-disclosure: Who discloses what and to whom? Cultural Diversity and Ethnic Minority Psychology, 13(3), 254. 10.1037/1099-9809.13.3.254 [DOI] [PubMed] [Google Scholar]
  19. Connolly J, Baird K, Bravo V, Lovald B, Pepler D, & Craig W. (2015). Adolescents’ use of affiliative and aggressive strategies during conflict with romantic partners and best-friends. Euroapean Journal of Developmental Psychology, 12(5), 549–564. 10.1080/17405629.2015.1066244 [DOI] [Google Scholar]
  20. Cramer D. (2010). Emotional support, conflict, depression, and relationship satisfaction in a romantic partner. The Journal of Psychology, 138(6), 532–542. 10.3200/JRLP.138.6.532-542 [DOI] [PubMed] [Google Scholar]
  21. Demaray MK, & Malecki CK (2002). The relationship between perceived social support and maladjustment for students at risk. Psychology in the Schools, 39(3), 305–316. 10.1002/pits.10018 [DOI] [Google Scholar]
  22. Dishion TJ, Eddy JM, Haas E, Li F, & Spracklen K. (1997). Friendships and violent behavior during adolescence. Social Development, 6(2), 207–223. [Google Scholar]
  23. Dishion TJ, Ha T, & Véronneau MH (2012). An ecological analysis of the effects of deviant peer clustering on sexual promiscuity, problem behavior, and childbearing from early adolescence to adulthood: an enhancement of the life history framework. Developmental psychology, 48(3), 703. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Evans GW, & Kim P. (2013). Childhood poverty, chronic stress, self‐regulation, and coping. Child Development Perspectives, 7(1), 43–48. 10.1111/cdep.12013 [DOI] [Google Scholar]
  25. Furman W. (2018). The romantic relationships of youth. In Eds. Bukowsky WM, Laursen B, & Rubin KH (Eds.), Handbook of Peer Interactions, Relationships, and Groups, Second Edition. Guilford Press. [Google Scholar]
  26. Furman W. & Buhrmester D. (1985). Children’s perceptions of the personal relationships in their social networks. Developmental Psychology, 21(6), 1016–1024. 10.1037/0012-1649.21.6.1016 [DOI] [Google Scholar]
  27. Gillespie BJ, Lever J, Frederick D, & Royce T. (2014). Close adult friendships, gender, and the life cycle. Journal of Social and Personal Relationships, 32(6), 709–736. 10.1177/0265407514546977 [DOI] [Google Scholar]
  28. Grimm KJ, Ram N, Estabrook R. (2017). Growth modeling: Structural equation and multilevel modeling approaches. Guilford Press. [Google Scholar]
  29. Hamaker EL, Kuiper RM, & Grasman RP (2015). A critique of the cross-lagged panel model. Psychological Methods, 20(1), 102. [DOI] [PubMed] [Google Scholar]
  30. Haydon KC, Collins WA, Salvatore JE, Simpson JA, & Roisman GI (2012). Shared and distinctive origins and correlates of adult attachment representations: The developmental organization of romantic functioning. Child Development, 83, 1689–1702. doi: 10.1111/j.1467-8624.2012.01801.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Hombrados‐Mendieta MI, Gomez‐Jacinto L, Dominguez‐Fuentes JM, Garcia‐Leiva P, & Castro‐Travé M. (2012). Types of social support provided by parents, teachers, and classmates during adolescence. Journal of Community Psychology, 40(6), 645–664. [Google Scholar]
  32. Loeb EL, Stern JA, Costello MA, & Allen JP (2020). With (out) a little help from my friends: insecure attachment in adolescence, support-seeking, and adult negativity and hostility. Attachment & Human Development 23(5), 624–642. 10.1080/14616734.2020.1821722 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Loeb EL, Tan JS, Hessel ET, & Allen JP (2018). Getting what you expect: Negative social expectations in early adolescence predict hostile romantic partnerships and friendships into adulthood. The Journal of early adolescence, 38(4), 475–496. doi: 10.1177/0272431616675971. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. McArdle JJ, & Hamagami F. (2001). Latent difference score structural models for linear dynamic analyses with incomplete longitudinal data. In Collins LM & Sayer AG (Eds.), New methods for the analysis of change (pp. 139–175). American Psychological Association. 10.1037/10409-005 [DOI] [Google Scholar]
  35. Meuwly N, Bodenmann G, Germann J, Bradbury TN, Ditzen B, & Heinrichs M. (2012). Dyadic coping, insecure attachment, and cortisol stress recovery following experimentally induced stress. Journal of Family Psychology, 26(6), 937–947. 10.1037/a0030356 [DOI] [PubMed] [Google Scholar]
  36. Mikulincer M. & Shaver PR (2016). Attachment in Adulthood: Structure, Dynamics, and Change, 2nd ed. Guilford Press. [Google Scholar]
  37. Mikulincer M. & Shaver PR (2020). Enhancing the “broaden and build” cycle of attachment security in adulthood: From the laboratory to relational contexts and societal systems. International Journal of Environmental Research and Public Health, 17, 2054: doi: 10.3390/ijerph17062054 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Miller S, Gorman-Smith D, Sullivan T, Orpinas P, & Simon TR (2009). Parent and peer predictors of physical dating violence perpetration in early adolescence: Tests of moderation and gender differences. Journal of Clinical Child & Adolescent Psychology, 38(4), 538–550. 10.1080/15374410902976270 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Moran PB & Eckenrode J. (1991). Gender differences in the costs and benefits of peer relationships during adolescence. Journal of Adolescent Research, 6(4), 396–409. 10.1177/074355489164002 [DOI] [Google Scholar]
  40. Narr RK, Allen JP, Tan JS, & Loeb EL (2017). Close friendship strength and broader peer desirability as differential predictors of adult mental health. Child Development, 90(1), 298–313. 10.1111/cdev.12905 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Oudekerk BA, Allen JP, Hessel ET, & Molloy LE (2015). The cascading development of autonomy and relatedness from adolescence to adulthood, Child Development, 86(2), 472–485. 10.1111/cdev.12313 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Priem JS & Solomon DH (2014). Emotional support and physiological stress recovery: The role of support matching, adequacy, and invisibility. Communication Monographs, 82(1), 88–112. 10.1080/03637751.2014.971416 [DOI] [Google Scholar]
  43. Reis HT (2014). Responsiveness: Affective interdependence in close relationships. In Mikulincer M. & Shaver PR (Eds.), The Herzliya series on personality and social psychology. Mechanisms of social connection: From brain to group (p. 255–271). American Psychological Association. 10.1037/14250-015 [DOI] [Google Scholar]
  44. Roisman GI, Masten AS, Coatsworth JD, & Tellegen A. (2004). Salient and emerging developmental tasks in the transition to adulthood. Child Development, 75, 123–133. 10.1111/j.1467-8624.2004.00658.x [DOI] [PubMed] [Google Scholar]
  45. Rubin KH, Bowker J, & Gazelle H. (2010). Social withdrawal in childhood and adolescence: Peer relationships and social competence. The development of shyness and social withdrawal, 131–156. Guilford Press. [Google Scholar]
  46. Seiffge-Krenke I, Persike M, & Shulman S. (2015). Gendered pathways to romantic attachment in emerging adults: The role of body image and parental support. European Journal of Developmental Psychology, 12(5), 533–548. 10.1080/17405629.2015.1044963 [DOI] [Google Scholar]
  47. Shulman S, Laursen B, Kalman Z, & Karpovsky S. (1997). Adolescent intimacy revisited. Journal of Youth and Adolescence, 26(5), 597–617. 10.1023/A:1024586006966 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Simpson JA, Rholes WS, & Nelligan JS (1992). Support seeking and support giving within couples in an anxiety-provoking situation: The role of attachment styles. Journal of Personality and Social Psychology, 62(3), 434–446. 10.1037/0022-3514.62.3.434 [DOI] [Google Scholar]
  49. Stern JA, Costello MA, Kansky J, Fowler C, Loeb EL, & Allen JP (2021). Here for you: Attachment and the growth of empathic support for friends in adolescence. Child Development. 10.1111/cdev.13630 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Stotsky MT, Bowker JC, & Etkin RG (2019). Receiving prosocial behavior: Examining the reciprocal associations between positive peer treatment and psychosocial and behavioral outcomes. Journal of Research on Adolescence, 30(2), 458–470. 10.1111/jora.12537 [DOI] [PubMed] [Google Scholar]
  51. Sullivan HS (1953/2013). The interpersonal theory of psychiatry. Routledge. [Google Scholar]
  52. US Census Bureau (1999). Money income in the United States: Current population reports. Consumer Income, 0–206. [Google Scholar]
  53. Waters E, & Sroufe LA (1983). Social competence as a developmental construct. Developmental Review, 3, 79–97. 10.1016/0273-2297(83)90010-2 [DOI] [Google Scholar]
  54. Wingspread D. (2004). Wingspread declaration on school connections. Journal of School Health, 74, 233–234. 10.1111/j.1746-1561.2004.tb08279.x [DOI] [PubMed] [Google Scholar]
  55. White AM, & Gager CT (2007). Idle hands and empty pockets? Youth involvement in extracurricular activities, social capital, and economic status. Youth & Society, 39(1), 75–111. 10.1177/0044118X06296906 [DOI] [Google Scholar]
  56. Zimmer-Gembeck MJ (2002). The development of romantic relationships and adaptations in the system of peer relationships. Journal of Adolescent Health, 31(6), 216–225. doi: 10.1016/s1054-139x(02)00504-9 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

SUPINFO
NIHMS1843469-supplement-SUPINFO.docx (44.1KB, docx)

RESOURCES