Adolescents’ Social Goal Orientations

Associations With Parenting Styles

To date, most studies examining potential predictors of adolescents’ social goal orientations have focused on individual factors (e. g., narcissism, Findley & Ojanen, 2013; temperament, Ojanen et al., 2012; callous-unemotional traits, Goagoses, Schipper, & Koglin, 2022). However, it is important to further examine how familial factors could mold social goal orientations. Parents play a vital role in the development of their children’s social competencies (Baumrind, 1991), including their social information processing (Rah & Parke, 2008). Baumrind (1991) proposed that different parenting styles differentially foster adolescents’ tendencies toward agency (status, power, dominance) and communion (affiliation, solidarity). For instance, adolescents perceive authoritative parents as more loving and influential and are more likely to emulate their parents’ prosocial behaviors and comply with their rules of conduct (Baumrind, 1991). Authoritarian parents have more difficulties maintaining interpersonal relations with their adolescent children, who in turn lack social consciousness, individuation, and autonomy (Baumrind, 1991).

Drawing upon attachment theory (Bowlby, 1969), coercion theory (Patterson, 2016), and social learning theory (Bandura, 1973), McDonald et al. (2013) linked parenting to social information processing and examined how parenting behaviors are related to highly aggressive preadolescent children’s social goals. The authors found that child-reported positive parenting (i. e., parents’ use of positive reinforcement for good behaviors) negatively predicted dominance goals and positively predicted affiliation goals 1 year later; in contrast, parent-reported positive parenting showed no such association. Furthermore, parent-reported use of corporal punishment positively predicted boys’ dominance goals but not that of girls. Lastly, parent-reported use of corporal punishment was not associated with affiliation goals. In an earlier study, Heidgerken et al. (2004) found that harsh discipline correlated positively with dominance/revenge goals and negatively with affiliation goals in middle childhood. Similarly, results presented by Sypher et al. (2019) indicated that positive parenting (i. e., involvement) is negatively associated with dominance goals, and that harsh parenting (i. e., conflict) is positively associated with dominance goals in middle childhood.

The parenting behaviors examined above can be embedded in core parenting style dimensions, specifically under parental warmth and parental coercion (Skinner et al., 2005). Parental warmth is characterized by the expression of love, affection, and support when children seek comfort, and within parent-child interactions focused on teaching and discipline (Skinner et al., 2005). Parental coercion, a feature of authoritarian parenting, is characterized by restrictive overcontrolling, obedience demands, and punitive discipline (Skinner et al., 2005). The findings reported above (Heidgerken et al., 2004; McDonald et al., 2013; Sypher et al., 2019) indicate that parental coercion is positively associated with dominance-oriented goals and negatively with affiliation-oriented goals – and vice versa for parental warmth. Skinner et al. (2005) also named parental autonomy support as an essential parenting style dimension characterized by encouraging freedom to express preferences and opinions. Parental autonomy support has been linked with the development of social competencies (Su-Russell & Russell, 2021) as well as fostering friendship quality (Xiang et al., 2023) and could thus also be associated with adolescents’ social goal orientations.

Associations With Externalizing and Prosocial Behaviors

Social goal orientations direct the social strategies selected for interactions, which then translate into social behaviors (Ojanen et al., 2005). Drawing on the work of Chung and Asher (1996), Ojanen et al. (2013) proposed that agentic goals are related to aggressive strategies, where aggression is used to gain control and status over peers. In turn, communal goals are linked with prosocial, accommodating, and conflict-resolution strategies, which aid in the development and maintenance of intimate relationships (Ojanen et al., 2013). Empirical studies corroborate that social goals are associated with specific behavioral patterns in peer interactions and conflicts. Ojanen et al. (2005; see also Salmivalli et al., 2005) found that communal goals are positively associated with prosocial behavior in early adolescence, and that agentic goals are negatively associated with prosocial behavior. Furthermore, agentic goals showed a positive association with bullying perpetration in early adolescence, while communal goals showed no direct association (Caravita & Cillessen, 2012; see also Pan et al., 2020). Similar, albeit somewhat inconsistent, findings were found for the association with aggressive behavior: Ojanen et al. (2005) and Salmivalli et al. (2005) reported that agentic goals were positively associated with aggressive behavior, while communal goals revealed no such association. Ojanen et al. (2013) found a negative association between communal goals and aggression in middle childhood but no association between agentic goals and aggression. Ojanen and Findley-Van Nostrand (2014) found that agentic goals could longitudinally predict relational aggression but not physical aggression in early adolescence, and vice versa for communal goals.

Procedure and Participants

The current study draws on data collected in 2020 as part of a larger research project, which the Ethics Committee and Regional School Authority Board approved. We recruited the adolescents from secondary schools in northern Germany (Grades 5 – 13). Participating schools sent an online parent questionnaire and adolescent questionnaire to parents (legal guardians), who could then voluntarily opt to participate in the project with their children. We obtained informed consent from both parents and adolescents. To address the current research question, we used the data of 354 adolescents (56.5 % female; M_age = 14.11 years, SD_age = 1.79) who were attending various school forms; 91.5 % indicated that they had been born in Germany. We used the data of 121 parents, with most questionnaires being answered by mothers who provided answers for both parents. Most parents had been born in Germany (61.2 %) and revealed varying levels of school completion. Table E1 (Electronic Supplementary Material, ESM 1) provides detailed demographic descriptions.

Instruments

We assessed the adolescents’ prosocial behavior with the self-report version of the Strengths and Difficulties Questionnaire (SDQ; Goodman, 1997). The Prosocial Behavior scale includes 5 items (e. g., kind to younger children; α = .70, ω = .70), with responses given on a scale of (1) not true to (3) certainly true. We assessed the adolescents’ externalizing behavior problems with the self-report version of the Child Behavior Checklist (YSR/11 – 18R; Achenbach, 1999; Döpfner et al., 2014;). The Aggressive Behavior scale includes 17 items (e. g., getting in fights, destroying belongings; α = .82, ω = .83), and the Rule-Breaking Behavior scale includes 15 items (e. g., lying, stealing, breaking rules; α = .78, ω = .78), with responses given on a scale of (1) not true to (3) very/often true.

We assessed the adolescents’ social goals with the Interpersonal Circumplex Inventory for Children (IGI-C; Ojanen et al., 2005). The IGI-C was backtranslated into German, and 2 items were rephrased to be more appropriate for adolescents (see Goagoses, Schipper, & Koglin, 2022, for confirmatory factor analysis based on the same data). The IGI-C contains 32 items that address the importance of adolescents’ social goals, i. e., what is desired when interacting with peers, with responses given on a scale ranging from (1) not at all important to (4) extremely important. The circumplex dispositions of social goals along the dimensions of agency and communion were initially assessed with eight subscales, namely, agentic (e. g., respect and admiration from peers), agentic-separate (e. g., control over peers), separate (e. g., keeping thoughts and feelings from peers), submissive-separate (e. g., avoiding humiliation in front of peers), submissive (e. g., avoiding conflict with peers), submissive-communal (e. g., pleasing peers), communal (e. g., mutual closeness with peers), and agentic-communal (e. g., expressing opinions to peers; see Table E2 (ESM 1) for reliability coefficients). These subscales are used to obtain vectors scores for the global agentic and communal goals, based on the calculation procedure reported by Ojanen et al. (2005; see also Locke, 2003):

• •
  Agentic Goals: Vector = agentic – submissive + [.707 x (agentic-communal + agentic-separate – submissive-communal – submissive-separate)]
• •
  Communal Goals: Vector = communal – separate + [.707 x (agentic-communal + submissive-communal – agentic-separate – submissive-separate)].

A high agentic goal vector score indicates a stronger desire for authority, power, and appearing confident before peers (i. e., agency), while a lower score indicates a desire to avoid confrontation and arguments with peers (i. e., submission). A high communal goal vector score indicates a desire for developing and maintaining closeness with peers (i. e., communion), while a lower score indicates the desire to withhold from peers (i. e., separation).

Parents completed a Parenting Style Inventory (Eltern-Erziehungsstil-Inventar; EEI; Satow 2013) containing scales to assess Parental Warmth (e. g., providing security, trust, praise; α = .90, ω = .91), Autonomy Support (e. g., encouraging decision making, and responsibility; α = .82, ω = .82), and Parental Coercion (e. g., providing strict rules, and enforcing rules with punishments; α = .76, ω = .77) with 10 items each. Items were answered on a scale ranging from (1) not true to (4) very true.

Statistical Procedure

Preliminary Analyses

Table 1 displays the means and standard deviations per scale as well as their correlations. As expected (see Ojanen et al., 2005), agentic goals and communal goals were not correlated. Agentic goals were positively correlated with rule-breaking and aggressive behavior and negatively correlated with prosocial behavior. Communal goals were negatively related to aggressive behavior and positively correlated to prosocial behavior. Parental warmth was negatively correlated with agentic goals, and parental coercion was negatively correlated with communal goals. Furthermore, age was positively correlated with agentic goals, communal goals, rule-breaking behavior, and aggressive behavior. Males reported more rule-breaking behavior (t(352) = –2.11, p = .02, d = 0.25), and females reported more prosocial behavior (t(352) = 5.85, p < .001, d = 0.33); no other gender differences were significant (see Table E3, ESM 1).

Table 1 Means (M), standard deviations (SD), and correlations

	1.	2.	3.	4.	5.	6.	7.	8.	9.	M	SD
1. Agentic goals	−	-.01	.24**	.29**	-.14**	-.23**	-.01	-.12	.21**	-0.99	1.25
2. Communal goals		−	-.03	-.12*	.25**	.02	-.21*	.00	.12*	1.95	1.68
3. Rule-breaking behavior			−	.70**	-.24**	-.33**	-.15	-.16*	.31**	1.28	0.25
4. Aggressive behavior				−	-.27**	-.22**	-.11	-.05	.10*	1.38	0.27
5. Prosocial behavior					−	.21*	.21*	.19*	-.06	2.67	0.35
6. Parental warmth						−	.15*	.74**	-.10	3.68	0.42
7. Parental coercion							−	.35**	-.07	2.88	0.44
8. Parental autonomy support								−	-.04	3.37	0.41
9. Age									−	14.11	1.79

Note. *p < .05, **p < .01. Descriptive statistics and correlations with parenting styles are based on a sample of n = 121, while the others are based on a sample of n = 354.

Table 1 Means (M), standard deviations (SD), and correlations

View as image HTML

Path Analyses

Figure 1 displays the examined model, which generally demonstrated a good/acceptable model fit (χ²(21) = 1.80; p = .01; CFI = .97; TLI = .92; RMSEA = .05). There was a significant direct effect on agentic goals from parental warmth (β = -.31, p = .02) but not from parental coercion (β = .03, p = .71) and autonomy support (β = .09, p = .51). Together with age (β = .21, p < .001), the parenting styles accounted for a total of 10.3% of the variance in agentic goals. There was a significant direct effect on communal goals from parental coercion (β = -.26, p = .004) but not from parental warmth (β = .01, p = .97) and autonomy support (β = .10, p = .48). Together with age (β = .11, p = .03), the parenting styles accounted for a total of 7.3% of the variance in communal goals. Agentic goals had significant positive direct effects on rule-breaking behavior and aggressive behavior (β = .18, p < .001, and β = .28, p < .001, respectively) and a negative direct effect on prosocial behavior (β = -.13, p = .01). Communal goals had significant direct effects onto aggressive behavior (β = -.12, p = .02) and prosocial behavior (β = .22, p < .001) but not onto rule-breaking behavior (β = -.05, p = .32). Age had a direct effect on rule-breaking behavior (β = .26, p < .001) but not aggressive behavior (β = .04, p = .40). See Table E4 (ESM 1) for tabular representation of beta coefficients and standard errors. Gender had a direct effect on rule-breaking behavior (β = .15, p < .001) and prosocial behavior (β = -.29, p < .001). In total, 14.1% of the variance of rule-breaking behavior, 9.9% of the variance of aggressive behavior, and 15.0% of the variance of prosocial behavior were accounted for.

Parenting Styles and Social Goal Orientations

The first aim of this study was to examine the association between parenting styles and adolescents’ social goal orientations. Only partially supporting our first hypothesis (Hypothesis 1a), we found that parental warmth was negatively associated with agentic goals but showed no association with communal goals. Thus, adolescents who receive ample parental support, affection, and kindness are less likely to desire authority and dominance among their peers. This aligns with previous research, showing that positive parenting is negatively associated with the expression of dominance goals in ambiguous peer provocations (McDonald et al., 2013; Sypher et al., 2019). During interactions with their parents, which are in part governed by parenting styles, children learn about relationships and interaction skills (Ladd & Pettit, 2002); parental warmth fosters children’s conflict resolution skills and socially-appropriate strategies for improving interpersonal relationships (Eisenberg & Valiente, 2002) and increases the extent to which children pursue emotional ties with their peers (Putallaz & Heflin, 1990, as cited by Ladd & Pettit, 2002). Through such mechanisms, parental warmth could hinder the development of agentic goals, as these are contrary to learned relationship standards. Arguably, these mechanisms should also drive the development of communal goals, indicating the need for more research to explain the lacking association. McDonald et al. (2013) found that parent-reported positive parenting showed no association with social goals (compared to child-reported), stating that differences emerge based on individualized reflective perspectives.

Our second hypothesis (Hypothesis 1b) was also only partially supported, as parental coercion was negatively associated with communal goals but showed no association with agentic goals. Hence, adolescents subjected to overcontrolling and restrictive parenting are less likely to desire affiliation and closeness with peers their own age. This aligns with attachment theory (Bowlby, 1969) and social learning theory (Bandura, 1973), which in turn also raises questions concerning the lacking association with agentic goals. It would seem plausible that adolescents subjected to parental coercion – characterized by authority and obedience – would internalize these social interaction norms and more likely endorse complimentary agentic goals. However, adolescents of directive parents (i.e., restrictive, obedience- and status-oriented) often lack individuation and autonomy and have a high external locus of control (Baumrind, 1991), which could suppress the development of agentic desires.

Lastly, the current findings do not support our third hypothesis (Hypothesis 1c), as autonomy support was associated neither with agentic goals nor with communal goals. Although parental autonomy support has been linked with adolescents’ academic achievement, well-being, and self-esteem (Vasquez et al., 2016), associations with social aspects are less clear. Drawing upon studies that examined peer interaction and relationship constructs reveals that parental autonomy support may not be a direct predictor. For instance, Gao et al. (2021) found that maternal autonomy support did not have a main effect on children’s sociability (i.e., having many friends, preferring to spend time with others instead of being alone) and assertiveness (i.e., willingness to express opinions, being a good leader). Similarly, Clark and Ladd (2000) reported that parental autonomy support did not account for variation in young children’s friendship quality, peer acceptance, or socioemotional orientations (i.e., inclinations for prosocial-empathic peer interactions). Taken together, our findings provide first evidence about the significance of parental factors for adolescents’ social goal orientations.

Social Goal Orientations and Behavioral Outcomes

The second aim of the study was to examine the association between adolescents’ social goal orientations and their externalizing and prosocial behaviors. Aligning with previous literature (Ojanen et al., 2005; Salmivalli et al., 2005), our first hypothesis (Hypothesis 2a), that agentic goals would be positively associated with rule-breaking and aggressive behavior and negatively associated with prosocial behavior, was fully supported by our findings. Hence, adolescents with a higher endorsement of agentic goals (i. e., desiring authority, status, and dominance above peers) reported more rule-breaking and aggressive behavior and less prosocial behaviors toward others. Previous studies indicated that aggression is associated with social status, popularity, and leadership during adolescence (Waasdorp et al., 2013), so adolescents with high agentic goals may use aggressive strategies to attain their desired status and dominance above peers (Ojanen et al., 2013). Research embedded within social identity theory similarly demonstrates that adolescents with a social dominance orientation are more likely to engage in physical and relational aggression (Gumpel & Gotdiner, 2023). Expanding on previous studies, we found that agentic goals were also positively associated with rule-breaking behavior, which during adolescence includes substance use, stealing, and truancy and has also been associated with increased social status, i. e., popularity (Franken et al., 2017). As such, it is likely to be used strategically to attain agentic goals.

Our second hypothesis (Hypothesis 2b), that communal goals would be negatively associated with rule-breaking and aggressive behavior and positively associated with prosocial behavior, was only partially supported by our findings. Adolescents with a higher endorsement of communal goals, i. e., a stronger desire for affiliations and closeness with peers, reported less aggressive behaviors and more prosocial behaviors. Again, this aligns with previous studies on social goal orientations (Ojanen et al., 2005; Salmivalli et al., 2005) as well as the findings by Wentzel and Erdley (1993), who determined that adolescents’ knowledge of appropriate strategies for making friends predicted social behavior, which in turn predicted peer acceptance. Similarly, aggressive behavior results in reduced likeability amongst peers (Franken et al., 2017; Waasdorp et al., 2013) and would thus be a behavior not enacted by adolescents aiming to develop or maintain relationships.

In summary, we were able to conceptually replicate and expand on previous findings, demonstrating that social goal orientations play a significant role in externalizing and prosocial behavior among adolescents in Germany. Hence, future research on externalizing behavior and prevention thereof should focus on social goal orientations in addition to social-emotional skills. This notion aligns with various models of social competence, from early childhood to adolescence (e. g., social competence prism, Rose-Krasnor, 1997; model of social-emotional learning, Denham & Brown, 2010; SIP, Crick & Dodge, 1994), which revealed the centrality of balancing personal and interpersonal social goals. Furthermore, Erdley and Asher (1999) noted that maladaptive social goals may be an important risk factor in the development of emotional and behavioral disorders, and that merely teaching social-emotional skills ultimately does not result in behavior change if the maladaptive social goals remain. Similarly, Trach et al. (2018) maintained that educators implementing social-emotional learning programs that address emotional and behavioral problems at school need to additionally target group processes and social dynamics.

Limitations and Future Research

Although we drew on relevant developmental theories, the current study utilizes cross-sectional data, so one cannot make inferences about the causality of the constructs without future longitudinal studies. For instance, parenting styles have been shown to predict change in externalizing behaviors over time, yet bidirectional associations in childhood and adolescence have also been observed (Pinquart, 2017). Future studies with bigger sample sizes should examine gender moderation (Mayeux & Kraft, 2018) and latent profile analysis to explore variations in the (simultaneous) endorsement of both agentic and communal goals. Future studies could examine whether parental social coaching (i. e., behavioral advice and cognitive framing about peer challenges, such as developing friendships, gaining acceptance, dealing with rejection; see Gregson et al., 2016) and parental pressure to be popular or socially preferred (see Wright, 2020) are related to social goal orientations. Furthermore, studies could explore the association between social goal orientations and emotion processes, such as emotion regulation, as these are vital for social information processing (Lemerise & Arsenio, 2000) and show associations with parenting styles (Goagoses, Bolz, et al., 2022) and externalizing behaviors (Compas et al., 2017). Given that most peer interactions occur within school contexts, future studies could further explore contextual moderators (Pan et al., 2020) or associations with academic adjustment (Goagoses & Koglin, 2022). Lastly, qualitative studies could provide insights into how social goal orientations influence social behaviors (e. g., selection and enactment of social strategies to attain agentic and communal goals).