Parenting and Courage: Exploring the Mediating Role of Self-esteem and Emotion Regulation among Adolescents

Abstract

With adolescence being a period for potential challenges but also positive growth, courage can aid adolescents in rising to some of these challenges. Yet few studies have investigated predictors and mechanisms of courage. The present study examined how three parenting constructs (attachment, autonomy support, and helicopter parenting) predict courage and if self-esteem or emotion regulation serve as mediators. The present study assessed predictors at Time 1 and courage at one-year follow-up. The sample included 203 American adolescents (51.7% girls, 14–15 year olds). Results suggested that relations between several parental variables (anxious attachment to parents, avoidant attachment to fathers, autonomy support of parents) and adolescent courage were mediated by self-esteem. In addition, cognitive reappraisal mediated associations between two mother-related variables (avoidant attachment, helicopter parenting) and courage. Findings contribute to the limited empirical research on adolescents’ development of courage.

Adolescence is a period for individuals to cultivate autonomy and independence (Zimmer-Gembeck & Collins, 2008). Although research has often focused on risks during adolescence (e.g., greater problem behavior), recent literature (e.g., Dahl et al., 2018) describes this period as a distinct critical phase that can potentially set stage for positive growth and positive risk-taking (defined as “socially acceptable and constructive risks” Duell & Steinberg, 2019, p.1). Courage helps teens to take positive risks and develop individuality and autonomy (Dahl et al., 2018; Gal & Rucker, 2021). Yet, there is little empirical evidence of how courage develops. One study found that aware parenting (i.e., supportive and receptive parenting) predicted more moral courage through self-esteem and peer social competence among twelfth grade girls but not boys (Bronstein et al., 2007). However, our study extends this work and contributes to the literature by testing whether multiple types of parenting factors (attachment, autonomy support, and helicopter parenting) and mediating factors (self-esteem and emotion regulation) predict a broader construct of general courage (which includes moral courage as well as other types of courage – e.g., physical, social) over the course of one year during mid-adolescence (14–16 year olds).

Courage

Courage is a form of positive risk-taking that manifests in everyday life (Lopez et al., 2003). As reviewed by Woodward and Pury (2007), courage includes different types of courageous behaviors, including physical courage (engaging in behavior despite the risk of encountering physical harm), moral courage (standing up for societal and ethical values despite the risk of challenging the opposing group), and social courage (intentional altruistic risky behavior that may harm one’s self-esteem, social relationship, or social image). Some prototypical examples of courage are facing one’s anxiety and performing on stage, standing up for people and one’s beliefs, coming out to family members, and riding a roller coaster despite the fear of heights. Although some scholars argue that courage involves a “noble good or worthy end” (Rate et al., 2007, p. 95) in line with a moral subtype of courage, others acknowledge that a broader conceptualization is needed as many behaviors can be considered subjectively courageous to individuals (Norton & Weiss, 2009; Woodward, 2004) and it is unclear how much some types of courage can be meaningfully distinguished.

For instance, Woodward and Pury (2007) question whether moral courage (e.g., standing up to a bully; Ossward et al., 2012) is distinct from social courage (e.g., giving an honest opinion despite the risk of damaging a relationship; Howard et al., 2016) given both involve social risks. Moreover, Szagun (1992) found that adolescents’ definitions of courage include both moral and physical risks. Finally, although courage is generally considered to be a desired virtue, Pury (2015) notes that bad courage is possible when the subjective value of the goal is only valuable to the individual but not for others (e.g., acts of terrorism or suicide). For these reasons, courage may be best considered broadly. Thus, we rely on a broad conceptualization of courage in the present study to capture the varied behaviors that adolescents may consider courageous.

Courage among Adolescents

While courage could be an asset at any age, it may be especially important for adolescents to develop courage and to be able to behave courageously. As teens gain independence, they have more opportunities for making impactful decisions. This increased autonomy coincides with their increased propensity to take risks and be influenced by peers (Steinberg, 2007), which could further lead to adaptive or maladaptive outcomes (Duell & Steinberg, 2019). As outlined by Dahl and colleagues (2018), adolescence is a pivotal time where neural, hormonal, and cognitive changes lead to greater salience of social information and greater motivational sensitivity to peers and rewards. They suggest that courage may be an especially important strength for adolescents to harness and use adaptively to make wise and healthy choices for their lives, which could expand self-knowledge, identity, and social prestige (Dahl et al., 2018). Research has supported this idea by findings that courage is positively associated with career adaptability and life satisfaction in Italian adolescents (Ginevra et al., 2018), less anxiety in Dutch children (Muris et al., 2009), and less internet addiction in Korean adolescents (Choi et al., 2015). Given the challenges but also potential for growth during adolescence, understanding which factors promote courage and through which mechanisms is important during this pivotal developmental period.

Theoretical Framework

The rationale for studying the development of courage stems from positive psychology, developmental research and theories, and the courageous mindset model (Hannah et al., 2010). Positive psychology highlights strengths and healthy behaviors that promote flourishing (Seligman & Csikszentmihalyi, 2000). Positive youth development (PYD; Lerner et al., 2009) builds on positive psychology by studying developmental processes that lead to healthy development and programs that can foster strengths of youth. PYD emphasizes that relational and developmental systems (e.g., parents, schools, counselors) can cultivate the development of strengths by nurturing favorable feelings of oneself or modeling emotion socialization in everyday life (Lerner et al., 2011). Therefore, from a prevention standpoint, PYD seeks to identify pathways that encourage the development of positive human characteristics, such as courage, as well as understanding which relational or developmental systems foster these qualities (Lomas et al., 2020).

Similarly, within developmental theories, including Bronfenbrenner’s ecological systems approach (1979; microsystem) and Sameroff and Chandler’s transactional model (1975), there is an emphasis on how family and social contexts contribute to child development, particularly early in development (Sameroff, 2010). Substantial research supports the proposition that parental factors such as parenting style, attachment, and parent-child interactions contribute to various child attributes, such as self-esteem (Pinquart, 2022), emotion understanding (Cooke et al., 2016), negative risky behavior (Dou et al., 2020), and prosocial behavior (Carlo et al., 2018). Thus, both developmental theories and the PYD support the proposition that broader systems, such as family contexts, will influence children’s development.

The courageous mindset model (Hannah et al., 2010) proposes a dynamic processing system that delineates why some people may exhibit courage in certain situations. Hannah and colleagues (2010) suggest that the perception of risk elicits an emotional and a physiological response, which could lead to the individual deciding to behave courageously in pursuit of a designated goal. The model further argues that courage can be activated through mediating or moderating factors such as cognitive affective units (e.g., values, emotions, and self-regulatory competencies), which are also impacted by internal (e.g., self-concept, positive emotions and beliefs, previous experience of courage) and external (e.g., social forces) resources. Importantly, by integrating the role of intrapersonal and interpersonal resources, this model provides a framework that allows empirical work to examine individual characteristics that could lead to courageous behaviors.

Consistent with these theoretical frameworks and models, in one of the only studies examining predictors of courage in youth, Bronstein and colleagues (2007) tested how parenting predicted intrapersonal competencies (i.e., self-esteem and social competence), and moral courage. Specifically, supportive and responsive parenting (measured at fifth grade) predicted moral courage in 12th grade adolescent girls, though no association was found for boys. In addition, self-esteem (measured at 12th grade) mediated the associations between parenting and moral courage for only girls. Such findings emphasize the important role of parents on adolescent behavioral outcomes. Although it is puzzling why parenting was not related to boys’ moral courage, given that women are found to engage in more interpersonally-oriented decision-making than men (Delaney et al., 2015), perhaps parenting only related to girls’ moral courage because it includes more interpersonal risks (than other types of courage like physical courage). As others acknowledge that a broader conceptualization is needed as many behaviors can be considered subjectively courageous to individuals (Norton & Weiss, 2009; Woodward, 2004) and because it is unclear how much some types of courage can be meaningfully distinguished, expanding courage to a broader scope in the present study can be useful. The present study also extends Bronstein et al. (2007)’s work by examining whether novel parenting factors (i.e., attachment, autonomy support, helicopter parenting) relate to fostering courage among teens through adolescent competencies (i.e., self-esteem and emotion regulation).

Interpersonal predictors

Attachment to parents.

The attachment system begins as a dyadic relationship between the infant and caregiver (Bowlby, 1982), but one’s attachment style becomes internalized as part of one’s own emotional and cognitive patterns that are stable over time. Research suggests that adult attachment is best captured in terms of anxiety and avoidance dimensions (Brennan et al., 1998). Individuals who are more anxious have greater fear of rejection and preoccupation with attachment needs and attachment figures, whereas more avoidant individuals have discomfort with closeness and tend not to use attachment figures for support (Brennan et al., 1998; Lee & Hankin, 2009). Furthermore, because various studies have demonstrated the differential impact of fathers’ and mothers’ attachment to their child in relation to depressive symptoms (Pan et al., 2016) and prosocial behavior (Gryczkowski et al., 2018), there is a clear need to examine the role of both mothers and fathers.

Anxious attachment is often associated with more negative outcomes. Individuals higher in attachment anxiety tend to develop more negative views of themselves because they have internalized perceived rejection from others (Mikulincer, 1995). Links between anxious attachment and lower self-worth have been found in youth (Gamble & Roberts, 2005) and adult samples (Mikulincer, 1995). Additionally, due to attempts to keep attachment figures (e.g., parents or romantic partners) nearby and attentive, more anxiously attached individuals tend to use emotion strategies that maintain or even increase negative affect (Mikulincer & Shaver, 2019). Specifically, research has shown adolescents high in attachment anxiety engage in less reappraisal (Karreman & Vingerhoets, 2012), which often is ineffective at decreasing negative affect (Gross, 1998). Although links between anxious attachment and suppression are often non-significant, research indicates that more anxiously attached people have difficulty disengaging from negative thoughts and emotions (Gentzler et al., 2010; Silva et al., 2012), which is counter to suppressing expressions of negative emotion. Therefore, it is expected that adolescents with anxious attachment will have less courage in part due to lower self-esteem and less use of reappraisal or suppression.

Avoidant attachment is linked to some outcomes that may promote courage. For example, by developing more self-reliance and self-esteem (Dan et al., 2014; Gamble & Roberts, 2005), avoidant individuals could be assertive and have greater courage. In terms of emotion regulation, avoidant individuals are theorized to minimize and suppress expressions of distress to downplay the need for support from close others (e.g., Cassidy, 1994; Mikulincer & Shaver, 2019). The tendency for individuals high in attachment avoidance to suppress and minimize emotions has been supported in infants (Cassidy, 1994; Martins et al., 2012), children and adolescents (Dubois-Comtois et al., 2011), and adults (e.g., Caldwell & Shaver, 2012). However, others suggest that attachment avoidance should relate to less courage because avoidance is related to less exploration due to not having a secure base (Aspelmeier & Kerns, 2003) and having less empathy (Spielmann et al., 2013). Therefore, in some contexts, individuals high in attachment avoidance may have less courage (e.g., not supporting a bullied friend). Thus, although prior evidence is somewhat equivocal, we hypothesize that more avoidant teens may report greater courage due to greater self-esteem and use of suppression.

Autonomy support and helicopter parenting.

Parental autonomy support refers to parenting behaviors that encourage the child’s self-initiating behaviors and decisions (Grolnick & Ryan, 1989). Various studies with adolescent samples have shown that perceived autonomy support is linked to adaptive developmental outcomes, such as higher self-esteem (Van Petegem et al., 2018). Further, research indicates that greater parental autonomy support relates to less frequent negative risk behaviors among adolescents (Vrolijk et al., 2020). Although associations between autonomy support and courage are not known, parental behaviors that support their child’s autonomy should help the child to engage in self-initiation, develop autonomous motivations, and achieve perceived competence (Marbell-Pierre et al., 2017). Therefore, autonomy support is likely to encourage forms of commitment to one’s values and further courageous behavior, which is also consistent with Burrow (2006) who argues that a relationship exists between autonomy and courage. Thus, in line with Bronstein et al. (2007) who measured supportive and responsive parenting, it is expected that autonomy supportive parenting will also be associated with higher reported general courage in youth. Because fathers’ and mothers’ autonomy support have different associations with adolescent problem behaviors (Vrolijk et al., 2020), we examined effects of mothers and fathers separately.

Related to lower autonomy support is helicopter parenting, wherein parents are excessively involved in their children’s or adolescents’ lives such as fixing difficult situations and making major decisions for teen (Cline & Fay, 1990). Research on helicopter parenting (or sometimes called overparenting; Bradley-Geist & Olson-Buchanon, 2014) has focused more on emerging adult populations. With a college student sample, one study on courage indicated that greater helicopter parenting (parental gender was not specified) related to less moral courage among college students (Evans & Karl, 2021).

A few studies have examined helicopter parenting during adolescence, but none have specifically examined courage. For example, helicopter parenting was linked with lower mental health and less self-efficacy for learning among 15–17 year old Malaysians (Ganaprakasam et al., 2018). Another study found that maternal helicopter parenting was associated with 17–19 year old Italian adolescents’ negative risky behavior (i.e., greater alcohol use), but that greater paternal helicopter parenting was related to less alcohol use (Pistella et al., 2022). In addition, US parents of 11–17 year olds reported that overparenting had positive relations with parents’ affect management (i.e., parental facilitation of encouraging child to feel positive emotions) and risk aversion (i.e., guarding child from engaging in perceived risks; Gagnon & Garst, 2019), potentially suggesting youth with helicopter parents might learn better emotion regulation but be risk averse and thus lower on courage. In sum, the limited evidence suggests that the role of helicopter parenting on adolescents’ psychosocial development is complex. Additionally, given that different adolescent outcomes are associated with maternal and paternal helicopter parenting, we investigated the mothers and fathers separately.

Intrapersonal mediators

Based on developmental theoretical models (e.g., Sameroff, 2010) and evidence demonstrating that parenting fosters self-esteem and emotion regulation competencies in youth (Morris et al., 2017) and that parenting can indirectly predict moral courage (Bronstein et al., 2007), the present study investigates intrapersonal mediators that may connect parenting to teens’ development of courage.

Self-esteem.

As self-esteem comprise individual’s belief of one’s value and self-worth (Coopersmith, 1967), we expect it to be an important predictor of courage. Adolescents with higher self-esteem could be better equipped to willingly stand up for their beliefs despite the presence of uncertainty (Sonnentag et al., 2018) due to greater self-respect and wanting to maintain an optimistic self-evaluation (Hannah et al., 2007). Further, theoretical views (e.g., Hannah et al., 2010) and empirical work indicate that higher self-esteem relates to having more optimistic thoughts and engaging in behaviors that strengthen one’s self-esteem (Judge & Bono, 2001). Therefore, individuals with greater self-esteem may be more confident and courageous, paralleling with other outcomes such as employment, life satisfaction, and flourishing (Marcionetti & Rossier, 2016; Parola & Marcionettil 2023; Salmela-Aro & Nurmi, 2007). Given that only one study has demonstrated positive links between supportive parenting and moral courage through greater self-esteem among US adolescent girls (but not for boys), more work is needed to replicate and extend Bronstein et al. (2007)’s findings to courage more broadly defined.

Emotion regulation.

Another potentially important predictor of courage may be emotion regulation (i.e., the ability to adjust and modify the experience and expression of emotions; Gross & Thompson, 2007). Given that adolescence is a period that is characterized by autonomy-seeking, experiencing mood fluctuations of both positive and negative affect, and having a heightened sensitivity to positive or rewarding experiences (Schweizer et al., 2020; Spear, 2011), it is critical for adolescents to learn how to regulate emotion regulation skills more independently (Steinberg, 2007). For example, research shows that poor emotion regulation relates to increased negative risk-taking behavior, such as greater hard drug use (Hessler & Katz, 2010). Thus, emotion regulation may help adolescents cope with challenges (Riediger & Klipker, 2014) and promote positive decision-making in emotional contexts. The current study focuses on two commonly assessed emotion regulation strategies, cognitive reappraisal and expressive suppression, because both can be fostered through parenting and attachment (Brenning et al., 2015; Mikulincer & Shaver, 2019) and are hypothesized to facilitate courage development.

Cognitive reappraisal is an antecedent-focused strategy that involves the reinterpretation of an emotion-triggering event to change its emotional impact that is linked to more positive outcomes (e.g., greater self-esteem, higher well-being; Gross & John, 2003). Expressive suppression is a response-focused strategy that involves actively inhibiting the observable expression of emotional experience and is often linked to negative outcomes (e.g., more depressive and anxiety symptoms, Gross & Cassidy, 2019). Despite the varied effectiveness, we expected both regulation strategies to aid in courageous behaviors. Heilman and colleagues (2010) demonstrated cognitive reappraisal reduces negative affect and risk aversion, leading to greater risk-taking propensity (measured with experimental risk-taking tasks). Thus, individuals who can successfully reappraise negative situations can reduce negative affect may have greater ability to exercise their intentions and be more courageous. Although suppression can be less effective in reducing negative affect (Heilman et al., 2010), suppression can be beneficial with high-intensity emotions (Gardner et al., 2020). Because courageous acts are done mainly in the presence of fear and uncertainty, it may be advantageous to be able to suppress negative or fearful emotions in order to deliberately act courageously.

Gender.

Researchers have speculated that courage is influenced by gender stereotypes (e.g., boys are encouraged to display more physical acts of courage whereas girls may be socialized in more relational ways to promote moral courage; Kinsella et al., 2017). However, few studies have empirically investigated the role of gender. One study empirically examined a lay theory (i.e., men are more courageous than women) and demonstrated no gender differences in social courage (Howard & Fox, 2020). As mentioned, teen gender served as a moderator in that parents’ supportive and responsive parenting predicted greater moral courage for 12^th grade adolescent girls but not for boys (Bronstein et al., 2007). The mixed findings suggest the need for further examining the role of gender on courage. As a result, this study considered adolescents’ gender as a covariate or moderator of other predictors of courage.

The Present Study

Research has shown that parenting influences many domains of a child’s development, such as emotional competencies (e.g., emotion regulation; Morris et al., 2017) and self-views (e.g., self-esteem; Gecas & Schwalbe; 1986). To our knowledge, no study has investigated the development of adolescents’ general courage. Guided by positive psychology and developmental models (e.g., Sameroff, 2010), the courageous mindset model (Hannah et al., 2010), and previous literature on moral courage (Bronstein et al., 2007), we examined how adolescents’ perceived parenting processes (attachment, autonomy support, helicopter parenting) are associated with their socio-emotional competencies (self-esteem, emotion regulation), and in turn, courage (see Figure 1). Because courage can help adolescents be more proactive and build personal and social resources (Dahl et al., 2018; Hannah et al.,2010), it is important to study antecedents of courage so that we can better understand why and how some adolescents are able to engage in courageous acts.

Figure 1.

Conceptual model

Note. The hypothesized model examines the longitudinal association between parenting behaviors and courage, with self-esteem and emotion regulation as mediators. Adolescents reported on both mother and father-figure for each interpersonal factor.

Based on the reviewed literature, several mediation hypotheses were proposed. Specifically, it was expected that 1) anxious attachment predicts lower courage through lower self-esteem, less reappraisal, and less suppression; 2) avoidant attachment predicts greater courage due in part to greater self-esteem or suppression; 3) greater autonomy support promotes courage through higher self-esteem; and 4) helicopter parenting promotes less courage due to lower self-esteem or less reappraisal. Given that studies on negative risk-taking has found that socioeconomic status (SES) is related to less risky behavior (Huang et al., 2016; Wagner et al., 2019; Young et al., 2018) and more prosocial behavior (Lin & Shek, 2022), SES was expected to also play a role in positive risky behavior such as courage. Furthermore, recent findings indicated that higher SES (indexed by parental education) related to more positive risk-taking (Duell & Steinberg; 2020). Therefore, in addition to gender (Bronstein et al., 2007) SES was explored as moderators and covariates.

Method

Participants

The sample included 203 adolescents (51.6% female; Time 1 M_age = 15.08, SD = .48). Most identified as white (88.30%) and the remaining identified as multiracial (4.67%), Asian, 2.34%, Black, 2.33%, Hispanic, 1.87%, and Native American (0.47%). The majority of the participants perceived themselves to belong in the mid-range of SES ladder (Goodman et al., 2001) and reported that they had little difficulty paying bills. Based on an apriori power analysis (Faul et al., 2007), the study was sufficiently powered.

In the larger longitudinal study, 299 adolescents answered surveys up to three time points over the course of a year. However, only 228 participants completed the courage measure at the one-year follow-up, and 206 passed the validity checks at the Time 1 and follow-up. Three multivariate outliers also were identified and excluded.

Procedure

The adolescents were part of a larger study (taking place from 2018–2020) that assessed participants at three points across one year: an initial assessment (Time 1), a 6-month follow-up, and 1-year follow-up. After IRB approval through the institution’s ethic committee, researchers used a multi-faceted approach to recruit adolescents via flyers in the community, online advertisements, and in-person recruitment at high schools in an Appalachian area in the southeastern U.S. Prior to participating, parents gave active consent for their adolescents to participate in the study, and adolescents assented for their own participation. Time 1 data were collected in schools, during home visits, at a research lab, or at a public library, but in each scenario, researchers were there to address questions. For the one-year follow-up, adolescents completed their surveys in group settings at schools on paper or completed online or paper versions at their homes. We compensated participants $20 after the completion of each survey and provided a $20 bonus if they completed all three assessments.

For the current study, the predictors and mediators (parental attachment, parenting behaviors, self-esteem, and emotion regulation) were assessed at Time 1, and courage was assessed at the follow-up roughly a year after the initial assessment (M_days = 395.00, SD = 44.13). Although there were 3 data collection points in the larger study, all mediators were only measured at Time 1 but not at the 6-month follow-up. For parenting and attachment questions, most adolescents reported on both mothers and fathers (or mother-figure and father-figures), but a portion of adolescents only reported on one parent (7.64% did not report on father-related variables and 4.69% of adolescents did not report on mother-related variables).

Measures

Demographics

We assessed SES status using two questions. One question asked about the family’s difficulties paying bills on a 4-point Likert type scale (1 = A great deal to 4 = None/Not at all). The second question was the MacArthur Scale of Subjective Social Status (SES ladder; Goodman et al., 2001), where adolescents indicated on a 1–10 scale where their family ranked. Higher scores represented people who had the most money, best education, and respectable jobs.

Courage

Adolescents reported on their self-perceived general courage using the Courage Measure (CM, Norton & Weiss, 2009). The measure includes 12 items with a 7-point Likert type scale (1 = never to 7 = always). Example items are “I tend to face my fears.” “If there is an important reason to face something scares me, I will face it.” Higher averaged scores indicate greater courage. The scale had good reliability for the current study α = .87, which was similar to previous work with Dutch youth (α = .86; Muris et al., 2009).

Parental attachment

The Experiences in Close Relationships – Attachment Orientations (ECR-RS; Fraley et al., 2011) assessed adolescents’ perceived anxious and avoidant attachment with their father and mother figures. There are 6 items for avoidance (e.g., “I usually discuss my problems and concerns with this person,” reverse-scored) and 3 items for anxiety (e.g., “I often worry that this person doesn’t really care for me”) that are rated on a 7-point Likert scale (1 = strongly disagree to 7 = strongly agree). Fraley and colleagues (2011) had conducted an exploratory factor analysis, finding support for these 2 scales, which accounted for 69% of the variance for each parental figure (Fraley et al., 2011). Items were averaged with scale scores computed for each dimension and for each parent (α_{avoidance mother} = .91, α_{avoidance father} = .90, α_{anxious mother} = .86, α_{anxious father} = .92).

Parent autonomy granting

Adolescents completed the Parent Behavior Measure (Peterson et al., 1999), as a measure of their perceived autonomy support from parents. It includes 10 items for each parent regarding the extent that parents allow teens to make their own decisions regarding friendships, educational goals, career goals, and lifestyles using a 4-point Likert scale (1 = strongly disagree to 4 = strongly agree). A sample item is “This parent allows me to decide what is right and wrong without interfering too much.” Higher scores indicate higher levels of autonomy granting (α_mother = .88, α_father = .89).

Helicopter parenting

Teens reported on the degree to which each parent is excessively involved and controlling of their lives via the Helicopter Parenting measure (Padilla-Walker & Nelson, 2012). The scale has five items, answered on a 1 (not at all like me/him/her) to 5 (a lot like me/him/her) scale. An example item is “My parent solves any crisis or problem I might have.” Higher scores indicate more helicopter parenting (α_mother = .70; α_father = .78).

Self-esteem

Adolescents reported on their self-perceived self-esteem via the Rosenberg Self-Esteem scale (Rosenberg, 1965). The measure includes 10 items rated on a 4-point Likert type scale (1 = strongly disagree to 4 = strongly agree). An example item is “On the whole, I am satisfied with myself.” Higher scores indicate higher self-esteem (α = .91).

Emotion regulation

Adolescents completed the Emotion Regulation Questionnaire for Children and Adolescents (ERQ-CA; Gullone & Taffe, 2012). The ERQ-CA is a commonly used version of the ERQ (Gross & John, 2003) that Gullone and Taffe (2012) had slightly adapted for youth (e.g., with simplified wording). The measure includes 10 items, with six items assessing cognitive reappraisal (e.g., “When I want to feel less bad (e.g., sad, angry, or worried), I think about something different”), and four items assessing expressive suppression (e.g., “I control my feelings by not showing them”) on a 7-point Likert type scale (1 = strongly disagree to 7 = strongly agree). Gullone and Taffe (2012) had conducted a confirmatory factor analysis to provide evidence for these two scales with 13–15 year olds (CFI = .948, RMSEA = .071). We averaged corresponding items so that higher scores indicated greater use of each strategy (α = .80 for the reappraisal scale; α = .71 for the suppression scale).

Analytic Plan

We conducted descriptive and correlation analyses in SPSS 28.0 and mediation models with the PROCESS macro (Model 4; Hayes, 2013). Although Baron and Kenny’s (1986) causal step approach was traditionally widely used in mediation research, other authors (see Hayes, 2009; MacKinnon et al, 2007) later criticized this approach because of its low power (large type 2 errors) and inability to test the indirect effect. In addition, research demonstrated that an indirect effect could exist despite the absence of a direct effect (Rucker et al., 2011). Therefore, the current study used Hayes’ (2013) process model, which utilizes a bootstrapping approach to determine the strength of indirect effects. In the current study, parental variables (anxious attachment, avoidant attachment, autonomy support, helicopter parenting) were independent variables, and self-esteem and emotion regulation (cognitive reappraisal and expressive suppression) were the mediator variables. The dependent variable was courage, assessed at the one-year follow-up. Gender was initially explored as a moderator due to prior research (Bronstein et al., 2007), but because it never was a significant moderator, gender (as well as SES) were included as covariates in all subsequent analyses.

Results

Preliminary analyses

We analyzed variables for missingness with Little’s MCAR test. Results indicated that data was not MCAR; χ² = 251.11, df = 172, p < .001. Only parent-related (i.e., attachment anxious mother and father-related variables) had more than 5% missingness. Thus, participants with missing data for father-related variables (n = 14–17) or mother-related variables (n = 7–13) were only excluded for those particular analyses. Scale scores were calculated if adolescents answered 80% of items that contribute to each scale. Due to skewness, square root transformations were used to normalize autonomy support scales for mother and fathers, and log transformations were conducted to normalize attachment anxious variables.

Attrition analyses were conducted with t-tests and chi-squares to examine whether there were associations between demographic factors (age, gender, race) and completion of the one-year follow-up surveys. Age and gender were not related to follow-up completion rates. However, race was related, x² = 13.71, p < .001, Cramer’s Phi = .22, p < .001, because 15.4% (standardized residual = −1.4) of white adolescents did not complete the one-year follow-up but 38% (standardized residual = 3.0) of other races/ethnicities did not complete it. Main variables (self-esteem, emotion regulation, attachment anxiety and avoidance, autonomy support) were unrelated to follow-up completion, except for helicopter parenting. Adolescents who did not complete the one-year follow-up reported higher helicopter parenting for mothers and fathers at Time 1 (M_mother = 3.27, SD = 0.96; M_father = 2.68, SD = 1.12) than did those who completed it (M_mother = 2.85, SD = 0.86; M_father = 2.34, SD = 0.94, t(282)_HP-mother = −3.06, p < .001; t(271)_HP-father = −2.26, p = .01).

Demographics and covariates

Variables’ means and standard deviations and bivariate correlations are presented in Table 1. Due to significant associations with courage, gender (M_boys = 4.73, SD = .98; M_girls = 4.35, SD = 1.01, t(201) = −.38, p = .01; males reported more courage) and SES-ladder (r = .19, p = .01; higher SES reported more courage) were included as covariates in all subsequent analyses.

Table 1.

Correlations between Time 1 intrapersonal and interpersonal predictors and Time 3 courage

	Mean (SD)	1.	2.	3.	4.	5.	6.	7.	8.	9.	10.	11.	12.	13.
1. SES ladder	6.27 (1.39)	---
2. Difficulty paying bills	3.35 (0.70)	.55***	---
3. Self-esteem	29.73 (6.39)	.35***	.19***	---
4. Reappraisal	4.14 (1.12)	.02	.08	.40***	---
5. Suppression	3.78 (1.25)	−.47***	−.13	−.30***	.01	---
6. Mother Autonomy	3.20 (0.55)	.11	−.05	.36***	.07	−.08	---
7. Father Autonomy	3.17 (0.59)	.15*	.02	.41***	−.03	−.14	.72***	---
8. Mother Anxiety	1.66 (0.76)	−.10*	.06	−.36***	−.14	.16*	−.46**	−.29***	---
9. Mother Avoidance	2.52 (1.14)	−.01	.02	−.29***	−.23***	.36***	−.31***	−.18***	.61***	---
10. Father Anxiety	1.85 (1.04)	−.18***	−.07	−.44***	−.11	.13	−.32***	−.41***	.62***	.21*	---
11. Father Avoidance	2.91(1.21)	−.19*	−.18*	−.42***	−.12	.24**	−.23**	−.46***	.30**	.35**	.62***	---
12. Mother HP	2.79 (0.82)	−.09	−.02	.07	.21**	−.12	−.19**	−.23**	−.08	−.31***	−.02	−.06	---
13. Father HP	2.34 (0.93)	.13	.17*	.17*	.09	−.17*	−.15*	−.03	.05	−.10	−.14	−.44***	.44***	---
14. Courage	4.54 (1.01)	.20**	.07	.39***	.23**	.04	.30***	.23**	−.09	−.08	−.11	−.14	−.16**	.03

Note. Parental attachment anxiety dimension abbreviated as Anxiety. Parental attachment avoidance dimension is abbreviated as Avoidance. Helicopter parenting is abbreviated as HP. P < .05.

^**p < .01.

^***p < .001.

Primary Analyses

Mediation models were conducted with self-esteem and reappraisal as potential mediators, but not with suppression because suppression and courage were not significantly correlated (see Table 1). We kept each parenting construct separate because the parenting variables were not highly correlated in this study (ranges from −.03 to .62, see Table 1) and we did not collapse across mothers and fathers due to previous studies showing differential effects of parenting styles of mothers and fathers on adolescent behavior (e.g., Vrolijk et al., 2020). Although several directional hypotheses were proposed, all directional paths were analyzed to be more comprehensive. As a result, we conducted sixteen models; of these, seven showed significant indirect paths (see Tables 2 and 3).

Table 2.

Indirect, direct, and total effects of mother-related models predicting adolescent courage

Predictor	Mediator	Indirect effects		Direct effect (c’)		Total effect (c)
		B	95% CI	B	95% CI	B	95% CI	R 2	F
Anxious attachment	Self-esteem Reappraisal	−.08	[−.16, −.01]	.09	[−.51, 1.61]	.02	[−.10, 1.21]	.07	4.19 (4, 176) *
		.00	[−.04, .03]	−.01	[−1.13, .96]	−.01	[−1.14, .99]	.06	2.82 (4, 179) *
Avoidant Attachment	Self-esteem Reappraisal	−.06	[−.15, .01]	−.02	[−.18, .14]	−.08	[−.24, .10]	.07	3.19 (4, 176) *
		−.04	[−.09. −.00]	−.01	[−.17, .15]	−.05	[−.21, .12]	.06	2.82 (4, 179) *
Autonomy support	Self-esteem Reappraisal	.09	[.04, .17]	.16	[1.56, .16] *	.25	[.60, 2.04] ***	.13	9.28 (3, 181) ***
		.01	[−.02, .04]	.24	[.54, 1.95] ***	.25	[.58, 2.01] ***	.12	8.54 (3, 184) ***
Helicopter parenting	Self-esteem Reappraisal	.04	[−.02, .09]	−.17	[−.38, −.05] *	−.13	[−.34, .01]	.20	11.06(4,182) ***
		.05	[.01, .10]	−.19	[−.41, −.07] *	−.15	[−.35., −.01] *	.08	5.39 (3,186) **

Note. Standardized coefficients are reported. Significant effects are bolded,

^*p < .05.

^**p < .01.

^***p < .001. For mother-teen models, Ns are between 181–190.

Table 3.

Indirect, Direct, and Total Effects of Father-related Models Predicting Adolescent Courage

Predictor	Mediator	Indirect effects		Direct effect (c’)		Total effect (c)
		B	95% CI	B	95% CI	B	95% CI	R 2	F
Anxious attachment	Self-esteem Reappraisal	−.08	[−.16, −.02]	.11	[−.32.1.43]	.03	[−.76, 1.05]	.07	3.39 (4, 173) *
		−.01	[−.06, .02]	.02	[−.81, .95]	.00	[−.87, .91]	.06	3.02 (4, 175) *
Avoidant Attachment	Self-esteem Reappraisal	−.07	[−.17, −.00]	−.01	[−.16, .14]	−.09	[−.23, .08]	.07	3.39 (4, 173) *
		−.02	[−.06. .02]	−.06	[−.20, .10]	−.07	[−.21, .09]	.06	3.02 (4, 176) *
Autonomy support	Self-esteem Reappraisal	.11	[.04, .19]	.06	[−.47, 1.07]	.17	[.09, 1.60] *	.11	7.00 (3, 173) ***
		−.01	[−.06, .02]	.17	[.10, 1.56] *	.16	[.02, 1.52] *	.09	6.14 (3, 176) ***
Helicopter parenting	Self-esteem Reappraisal	.05	[−.00, .11]	−.02	[−.17, .13]	.03	[−.13, .19]	.07	4.70 (3, 176) **
		.01	[−.01, .05]	.00	[−.15, .16]	.02	[−.14. .18]	.06	4.09 (3, 179) *

^*Note. Standardized coefficient values are reported. Significant effects are bolded, p < .05.

^**p < .01.

^***p < .001. For father-teen models, Ns are between 177–183.

Attachment predicting courage

For the models with attachment-anxiety with mothers and fathers, self-esteem, and covariates (gender, SES, attachment-avoidant), a significant negative indirect effect was found, indicating that adolescents who scored higher on anxious attachment had lower self-esteem, which predicted less courage (see Tables 2 and 3). The direct effects between anxious attachment and courage were not significant in either mother or father models (see Tables 2 and 3). The two models where anxious attachment predicted teen courage through teen reappraisal were not significant for mothers or fathers (see Tables 2 and 3).

For avoidant attachment with mothers, the mediating model with self-esteem was not significant, but the mediating model with reappraisal was significant (see Table 2). Adolescents with higher avoidant attachment with mothers used less appraisal, which was associated with less courage. On the other hand, for avoidant attachment with fathers, the model with self-esteem as the mediator revealed a significant indirect effect: avoidant attachment with fathers was significantly related to lower self-esteem, and self-esteem was positively related to courage. The direct effect of avoidant attachment with fathers on courage was not significant (see Table 3). The model for reappraisal as the mediator between avoidant attachment with fathers and courage was not significant (see Table 3).

Autonomy support predicting courage

The model with autonomy support, self-esteem, and covariates had a significant indirect effect for both mothers and fathers. Greater autonomy support was related to greater self-esteem, which in turn predicted greater courage (see Tables 2 and 3). The direct effect from autonomy support to courage was also significant for mothers (see Table 2), but not for fathers (see Table 3). The two models in which autonomy support with mothers and fathers predicted courage through reappraisal had nonsignificant indirect effects, but significant direct associations between autonomy support of mothers and fathers and courage (see Tables 2 and 3).

Helicopter parenting predicting courage

The two models for helicopter parenting predicting courage through self-esteem were not significant for either mothers or fathers. However, the direct effect for helicopter parenting on courage was significantly negative for mothers (but not for fathers), suggesting more maternal helicopter parenting predicted less courage (see Tables 2 and 3). The model with mothers’ helicopter parenting and cognitive reappraisal had a significant indirect effect (see Table 2). The positive indirect effect suggested that maternal helicopter parenting was significantly related to more usage of cognitive reappraisal, and cognitive reappraisal was significantly related to greater courage. The last model between fathers’ helicopter parenting and courage through cognitive reappraisal was not significant (see Table 3).

Discussion

This study provides new evidence of how parenting is associated with courage among adolescents through intrapersonal attributes of self-esteem and cognitive reappraisal. Investigating the antecedents and explanatory mechanisms of courage is important, as the majority of the literature on courage is not empirical. The present study contributes to the literature as it is the first to empirically assess predictors and mechanisms of general courage in adolescents. We integrated various developmental perspectives (e.g., Bronfenbrenner, 1979; Lerner et al., 2015; Sameroff, 2010), the PYD (Lerner et al, 2011), and the courageous mindset model (Hannah et al., 2010), which underscore how familial contexts influence adolescent development. Aligning with these theoretical perspectives, our findings provide preliminary evidence suggesting that to promote courage in teens, interventions could directly target teens’ self-esteem and tendency to reappraise situations that result in negative emotion or could focus on the parents to indirectly promote teens’ development of courage.

Mediated Paths to Courage

Attachment

Findings regarding attachment dimensions partially supported our hypotheses. First and contrary to expectations, avoidant and anxious attachment were not directly associated with courage (either in bivariate correlations or mediational models). However, there were multiple ways that attachment indirectly predicted courage. Specifically, adolescents higher on anxious attachment with both mothers and fathers reported lower self-esteem, which was related to less courage. These findings are consistent with prior work that suggests that anxiously attached adolescents are likely to develop relatively negative views of themselves (Dan et al., 2014; Gamble & Roberts, 2005; Lee & Hankin, 2009), and advances the literature by illustrating that becoming less courageous is another potential outcome for these teens.

A significant indirect effect with avoidance with fathers (but not mothers) was found in that adolescents who reported more avoidant attachment with fathers had lower self-esteem, which in turn predicted less courage. These findings are inconsistent with previous work that suggests that avoidantly attached adolescents have positive views of themselves (Gamble & Roberts, 2005), and further suggest that those who feel like they cannot be vulnerable to their fathers or rely on their father for support may have more negative views of themselves and may lack confidence to risk being courageous. Although the indirect effect was not significant for mothers, some research suggests that mothers are more often used as a safe haven (i.e., emotional support when the child is upset) whereas fathers may more often serve as a secure base (i.e., promote the child taking initiative and exploring world; Kerns et al., 2015). Thus, teens’ reports of their avoidant attachment with fathers may be especially relevant to their self-views and likelihood of initiating courageous behaviors.

In contrast, avoidance with mothers (but not fathers) was indirectly related to less courage through lower levels of reappraisal. The link between avoidant attachment and less reappraisal was consistent with a study with college students suggesting avoidant attachment associated with less cognitive reappraisal (Liu et al., 2020). Yet our study’s finding with youth in mid-adolescence and attachment to mothers specifically (but not fathers) advances the literature. As mentioned, the tendency for children to rely on mothers as a safe haven when upset (Kerns et al., 2015) may suggest that children are especially likely to learn how to regulate negative emotions from mothers (Chang et al., 2003) whereas father-child attachment may be more linked to the pursuit or regulation of positive emotions (Gentzler et al., 2013; Gerhardt et al., 2019). Thus, teens who report more avoidance with mothers may be learning less effective emotion regulation from them (e.g., if mothers are not modeling positive reappraisal), which could inhibit teens’ development of courage.

Autonomy Support

The findings for autonomy support were mostly consistent with hypotheses. In line with other literature (e.g., Van Petegem et al., 2018), bivariate associations and direct paths between autonomy support indicated that teen-reported autonomy support of mothers and fathers were linked to greater courage. These findings are noteworthy as they extend prior research on parenting and moral courage (Bronstein et al., 2007) to the broader construct of general courage. Furthermore, the current study found that self-esteem mediated the positive association between mothers’ and fathers’ autonomy support and courage. The results are consistent with research suggesting that parental autonomy support is related to greater self-esteem across one year (Van Petegem et al., 2018). Additionally, the indirect effect is consistent with Bronstein et al. (2007)’s finding of more aware parenting predicting girls’ moral courage, but inconsistent in that autonomy support also significantly predicted boys’ courage in the present study. As previously noted, while some (Woodward & Pury, 2007) argue for different subtypes of courage (e.g., moral and physical courage), others (Norton & Weiss, 2009; Woodward, 2004) argue for a more broader operationalization due to the lack of agreement of how meaningfully different types of courage could be distinguished. Different types of courage may be encouraged in boys compared to girls, but the current research on the role of gender is inconclusive as some report that girls report more physical courage than boys and that boys report greater psychological courage than girls (Santilli et al., 2020). Overall, if gender plays a role in different types of courage, the utilization of a broader measure of courage may explain why the association was found for both boys and girls in the present study if girls and boys can think of what “courage” means to them. Further, Pury and colleagues (2007) speculate that different types of courage have distinct correlates and that relational factors may play a more salient role in females’ moral courage than it does in males’. Thus, parental autonomy support may have predicted general courage for both boys and girls in the present study because asking about general courage may evoke thoughts about a broader range of examples and motives than moral courage. Overall, these findings extend the literature by suggesting positive self-concepts fostered through autonomy support of parents contribute to greater courage, regardless of gender.

Helicopter Parenting

As expected, more helicopter parenting of mothers was associated with less courage. These results are somewhat consistent with findings indicating that parental reports of greater overparenting related to higher risk aversion (Gagnon & Garst, 2019) and that greater perceived helicopter parenting related to less moral courage among college students (Evans & Karl, 2021). However, both studies were limited because the measures did not specify the parents’ gender. Our study only found a negative relationship between maternal helicopter parenting and courage, but it advances this limited research as others have noted the importance of examining unique influences of both maternal and paternal helicopter parenting on child outcomes (e.g., Love et al., 2019). Maternal helicopter parenting may be related to less courage due to overparenting impeding adolescent autonomy and competency, which are important developmental goals during adolescence. Our research therefore provides new evidence that high school students’ perceptions of mothers’ helicopter parenting could interfere with their developing feelings of courage. Fathers’ helicopter parenting was unrelated to adolescent courage. Because the majority of the work on helicopter parenting has been done with emerging adult samples and sometimes does not include fathers, there is a need to further explore the role of parental gender in relation to helicopter parenting and psychosocial outcomes in mid-adolescence. Nevertheless, this study’s finding of maternal helicopter parenting relating with less courage is novel and can be applied to help design parenting programs to facilitate adolescents’ positive development.

Although helicopter parenting by mothers directly predicted less courage, there was an unexpected indirect effect through cognitive reappraisal in the opposite direction. Specifically, maternal helicopter parenting was linked with greater cognitive reappraisal, which then predicted greater courage. Counter to prior research that found helicopter parenting and reappraisal to be unrelated (Wenze et al., 2019), our findings suggest a positive relation. The positive relation does parallel some work conducted with parental reports on overparenting and work with emerging adult samples. Specifically, Gagnon and Garst (2019) found that parents who engaged in greater overparenting reported that they engaged in more affective management for their teens, which may reflect the positive associations between maternal helicopter parenting and cognitive reappraisal. Kwon and colleagues’ (2017) qualitive research found that some Korean American college students considered helicopter parenting to be beneficial towards emotional (e.g., feeling supported and loved) and academic outcomes. As proposed by Wenze et al. (2019) and based on Gagnon and Garst (2019), mothers who engage in helicopter parenting may model positive reappraisal of challenges to help their adolescents to reinterpret adversity, which in turn may lead to greater courage. This suppression effect is novel and contributes to literature by showing that even though helicopter parenting is sometimes related to negative outcomes (e.g., less autonomous children; Schiffrin et al., 2013), it could also be related to positive outcomes (i.e., reappraisal) because it involves providing support or encouragement to teens to achieve their goals (Padilla-Walker & Nelson, 2012). Overall, the findings highlight that helicopter parenting can be beneficial in some contexts but problematic in other contexts, as others have found (e.g., less alcohol use; Pistella et al., 2022; better positive affect management; Gagnon & Garst, 2019).

In sum, current findings related to the parenting variables align with the theoretical frameworks (e.g., Lerner, 2009) by suggesting that mothers and fathers may have an essential role in fostering courage through facilitating self-esteem and emotion regulation.

Suppression and Courage

Although we had posited that emotional suppression may relate to more courage, this hypothesis was not supported. As hiding emotional displays may be taxing (Butler et al., 2003), this suppression process may limit the resources to behave courageously. In future studies, it may be worth examining whether suppression predicts some types of courage (e.g., suppression helps with physical but not psychological courage), but not others.

Gender and SES

Due to Bronstein et al.’s (2007) findings, gender was explored as a moderator between parenting and courage. However, all interaction terms were not significant, suggesting that, counter to Bronstein et al. (2007), parental processes in relation to courage operated similarly for girls and boys. More work is needed to better understand the relationship between gender and courage.

Additionally, the current study found that higher SES was associated with greater courage. Although this finding is similar to a recent study indicating that greater parental education related to greater positive risk-taking among adolescents (Duell & Steinberg, 2020), there may be more nuances of this association (e.g., it may depend on the type of courage or other moderating factors) that could be investigated in future research.

Limitations and Future Directions

Despite the contributions of the current study, this study was not without limitations. First, the sample consisted primarily of white adolescents in the Appalachian area of the US, where courage may be encouraged through various sources (e.g., cultural values of grit; environmental context; Patrick et al., 2023). Although courage appears to be valued across cultures (Lopez et al., 2010), different cultures and contexts may not consider identical behavior to be courageous, which can limit the generalizability of the findings. Interestingly, a study with Chinese adults found an unexpected greater emphasis of individually-oriented courage rather than socially oriented courage through qualitative analyses, but they also reported that responsibility was a unique cultural characteristic of courage among Chinese adults, whereas components of persistence and breakthrough (i.e., surpassing the limits of a current situation) were universal aspects of Eastern and Western courage (Cheng & Huang, 2016). Second, social desirability effects may have occurred (Krumpal, 2013). Future studies should incorporate multiple informants of courage (i.e., peers, parents, teachers) or observational data (e.g., creating an adolescent version of a task similar to Sasse et al. (2020) where participants intervened to restore justice in a staged embezzlement situation). Third, because the courage measure used in this study assessed general courage, adolescents may have varied in what types of behaviors they thought about and may not have considered relatively abstract risks related to psychological (e.g., addressing phobias; Putman, 1997) or existential courage (e.g., searching for meaning in life; Maddi, 2004). Fourth, because predictors were measured at Time 1 and courage was only measured at the one-year follow-up, without a Time 1 courage assessment to covary, the time scale of the prediction of courage (and if courage levels changed over time) is unclear. Fifth, although the courageous mindset model (Hannah et al., 2010) is the only to-date model on courage, its emphasis on the in-the-moment process of courage underscores the variability of courage across contexts (e.g., peers vs. family, home-setting vs. school, being with others vs. being alone). However, our study was unable to address the role of context or how various forms of courage may differ across situations. Finally, although the current study explored multiple predictors in a single study, additional factors may also contribute to adolescent courage development. Given that peers are extremely important to adolescents (Sanders, 2013), future work should assess relationships with friends and friends’ courageous behaviors to explore its link with teens’ developing courage.

Theoretical and Practical Implications

Overall, the findings of this study have several theoretical and practical implications. Theoretically, the links between some novel parenting (e.g., helicopter parenting and attachment) and intrapersonal factors (e.g., emotion regulation) are in line with developmental and Positive Youth Development perspectives (Lerner et al., 2015) and with the model of courage that highlight how children’s parenting and relationships may influence their developing self-views and regulatory capacities (Hannah et al, 2010). The findings also add to developmental research by showing how paternal and maternal parenting relates to courage, self-esteem, and emotion regulation in various ways.

Our study also demonstrated practical implications such that parenting, self-esteem, and cognitive reappraisal could be targeted in interventions to promote courage among adolescents. Importantly, findings also suggest that courage can be nurtured through parenting and intrapersonal factors. Based on the theoretical framework (Lerner et al., 2011, Sameroff, 2010) that suggests home contexts are important settings for adolescents, findings suggest that familial context can provide healthy resources for positive development. Thus, to boost courage in adolescents, targeted interventions for both parents and teens can be specifically designed to teach the importance of attachment patterns, ways to support teens’ autonomy, the strengths and weakness of helicopter parenting during adolescence, the importance of self-esteem, and healthy emotion regulation skills.

Conclusion

As courage relates to greater well-being (Ginerva et al., 2018), it is worthwhile to understand factors and the processes that may foster or stifle adolescent courage, as Gillham and Seligman (1999) suggest. This study was the first to empirically explore the antecedents of adolescent general courage. The current findings advance the literature by showing that parental autonomy support, attachment, helicopter parenting may encourage adolescent courage through self-esteem and cognitive reappraisal. In sum, this study’s focus on identifying antecedents and processes that lead to adolescent courage is an important step in empirically examining what fosters courageous behavior among teens. Interventions designed to help parents form secure attachment relations and give healthy levels of autonomy support to teens may also be promoting higher self-esteem and as a result, courage in their children. However, parents and clinicians should also be cautious and evaluate a child’s values and goals before engaging in courageous behavior because bad courage does exist (Pury et al., 2015). Nonetheless, as courage is more often considered a desirable character strength (Seligman & Csikszentmihalyi, 2000), parents would likely be interested in how their relationship and parenting style could affect their child. Interventions directly with youth could encourage them to use their courage for adaptive decisions and living according to their ideals, in line with Dahl and colleagues’ (2018) emphasis to intervene in adolescence to channel teens’ natural proclivity for risks and peer reinforcement into healthy activities.

Figure 2.

Significant models between interpersonal variables and courage through intrapersonal variables

Note. The arrows were added only for the significant models.

• Parenting behaviors (autonomy support, avoidant attachment, maternal helicopter parenting) predict adolescent courage through self-esteem and cognitive reappraisal.

• Helicopter parenting can relate to both positive and negative outcomes.

• Greater self-esteem and reappraising emotions relate to greater courage.

• Both mothers and fathers have a role in adolescent courage.