Trust, forgiveness, and peace: The influence of adolescent social identity in a setting of intergroup conflict

Following the signing of peace agreements, post-accord societies often remain deeply divided across group lines. There is a need to identify antecedents of youth’s support for peace and establish more constructive intergroup relations. This article explored the effect of out-group trust, intergroup forgiveness, and social identity on support for the peace process among youth from the historic majority and minority communities in Belfast, Northern Ireland. The sample comprised 667 adolescents (49% male; M = 15.74, SD = 1.99 years old) across two time points. The results from the structural equation model suggested that out-group trust was related to intergroup forgiveness over time, while forgiveness related to later support for the peace process. Strength of in-group social identity differentially moderated how out-group trust and intergroup forgiveness related to later support for peace among youth from the conflict-related groups (i.e., Protestants and Catholics). Implications for consolidating peace in Northern Ireland are discussed, which may be relevant to other settings affected by intergroup conflict.

The declaration of ceasefires and signing of peace agreements does not instantly rebuild trust or foster forgiveness for past wrongs (MacGinty et al., 2007). It may take considerable time to consolidate peace (Lederach, 1997). A majority of research in conflict settings has focused on understanding how the younger generation maintains intergroup hostilities by displaying negative out-group attitudes (e.g., prejudice, Turner et al., 2013) and behaviors (e.g., discrimination, Štambuk et al., 2020;aggression, Taylor et al., 2016). Less is known, however, about the factors that can promote youth's constructive contributions to rebuilding society after conflict (Taylor et al., 2020b). Past research with adults has found that intergroup trust and forgiveness are important for reconciliation (Tam et al., , 2009) and peace (Noor et al., 2015); understanding the influence of these factors on youth's future-oriented peacebuilding may hold promise. Building on this work, the Developmental Peacebuilding Model (DPM; Taylor, 2020) outlines how children and adolescents can foster peace across different levels of the social ecology. The DPM outlines that understanding the social identities of conflictrelated rivals may be important to build peace; therefore, this article integrates Social Identity Development Theory (SIDT; Nesdale, 2004) with the DPM. Given that durable peace necessitates finding workable solutions for all sides (Wagner & Druckman, 2017), the strength of in-group identity for both majority and minority groups is considered Merrilees et al., 2014). Therefore, the current article explores the moderating influence of strength of in-group social identity in the links between out-group trust and intergroup forgiveness on support for the peace process over time among youth growing up in post-accord Northern Ireland.

Developmental Peacebuilding Model
The DPM (Taylor, 2020) integrates a growing body of work that has begun to identify how children and adolescents can contribute to peace. For example, support for peace may involve endorsing symbolic, material, and relational actions intended to promote constructive patterns of engagement across group lines . The DPM outlines how peacebuilding is futureoriented, moving beyond interpersonal or intergroup prosocial relations  to focus on developing positive pathways for systemic peace at the macrosystem level Trust, forgiveness, and peace: The influence of adolescent social identity in a setting of intergroup conflict (Institute for Economics and Peace [IEP], 2016). This broader aspect of societal peace is relatively understudied in psychological science. Youth support for structural and cultural changes through a formal peace process, however, may be influenced by the strength of group identities as well as the levels of trust and forgiveness toward conflict rivals.

Out-group Trust and Intergroup Forgiveness
Trust is a social bond associated with feelings of security and confidence in the intentions and behaviors of others (Tropp, 2008). Trust can benefit youth by reducing loneliness and contributing to positive adjustment (Rotenberg et al., 2005(Rotenberg et al., , 2010. Furthermore, trust has the potential to reduce aggressive acts and promote prosocial behaviors among children (Song et al., 2018(Song et al., , 2020. Trust has also been shown to have important implications for the development of positive intergroup relations (Kramer & Carnevale, 2001). That is, trust may encourage greater cooperation and communication with out-group members ). Yet, out-group trust can be difficult to cultivate and easily broken (Cook et al., 2005). For example, individuals in opposing social groups (e.g., partisan political parties) may misinterpret out-group actions as efforts to undermine in-group interests and values, which can foster a lack of trust across group lines (Chambers & Melnyk, 2006). In conflict settings, trust involves taking a risk that the out-group will not engage in further acts of violence or exploit the in-group (Tam et al., 2009). Establishing mutual trust can be a prosocial facilitator for more peaceful intergroup relations by fostering positive outgroup attitudes (McKeown & Psaltis, 2017) and constructive engagement across group lines (Kenworthy et al., 2016), including intergroup forgiveness (Noor et al., 2015;Tam et al., 2007).
Children acquire the ability to forgive those that have wronged them from an early age, which is essential for repairing relationships (Oostenbroek & Vaish, 2019). More broadly, forgiveness can break the cycle of violence in conflict-affected societies by letting go of past injustices and the desire for revenge toward those who have harmed the individual or social group (Noor, 2016). Intergroup forgiveness is directed toward the out-group as a whole, rather than individual group members (Wohl & Branscombe, 2005), and can be unilateral (Roe, 2007). In conflict settings, youth are often socialized by narratives of collective victimhood (e.g., Reidy et al., 2015), while also negotiating intergroup tensions and political violence . Young people may, therefore, indirectly or directly feel like targets of out-group aggression, which may be foster a motivation for retribution and act as a barrier to forgiveness (O'Driscoll et al., 2021a). However, forgiveness can provide an opportunity to heal divisions (Myers et al., 2009). That is, forgiveness may be a key antecedent for reconciliation and the consolidation of peace Kosic et al., 2012). Both out-group trust and intergroup forgiveness, however, have been found to vary based on the social identities of the actors involved (e.g., Van Tongeren et al., 2014).

Social Identity Development Theory
The DPM outlines how children's social identities have implications for the antecedents of peacebuilding (Taylor, 2020). To better understanding how social ecologies shape those identities, along with implications for other intergroup relations, SIDT (Nesdale, 2004) integrates cognitive, affective, and social approaches (Nesdale & Flesser, 2001). Here we review the broader literature supporting SIDT, before proceeding to the context-specific application in this study.
An emerging body of research has begun to adopt SIDT as a theoretical framework in conflict settings (e.g., Merrilees et al., 2018;Shamoa-Nir et al., 2021), finding subtle nuance and differences between majority and minority groups in such contexts (e.g., Tomovska Misoska et al., 2020). SIDT proposes that the threat of violence heightens children and adolescents' strength of in-group identification (Nesdale et al., 2005a(Nesdale et al., , 2005b. That is, social identity may motivate youth to defend, maintain or improve the status of their group when facing intergroup conflict (Duffy & Nesdale, 2010). For example, children and adolescents with a greater strength of in-group identity are more likely to display prejudice (Merrilees et al., 2018), withhold resources (O'Driscoll et al., 2018), and engage in aggressive behaviors against the out-group (Merrilees et al., 2013). Thus, social identity may accentuate factors, such as negative outgroup attitudes and behaviors, which contribute to the maintenance of intergroup hostilities with long-term implications for supporting peace.
Alternatively, social identity may foster social change in settings of intergroup conflict (Haslam & Reicher, 2006). That is, if group members have little or no feasible options of separating from their in-group, social identity may bolster group cohesion and promote a collective motivation to enhance group status (Hammack, 2010). Social identification and group membership, may also protect youth against the development of negative psychosocial outcomes associated with exposure to political violence (e.g., for internalizing difficulties, see Merrilees et al., 2014; for externalizing difficulties, see Merrilees et al., 2013) and allow access to greater social support (Muldoon, 2013), positive intragroup behaviors (e.g., helping, Sabatier, 2008) and a framework to generate meaning from conflict-related adversity (Barber, 2008). Less is known, however, about how social identity relates to support for peace, or influences the effect of outgroup trust and intergroup forgiveness to this outcome, for youth growing up in contexts of intergroup conflict.
Moving beyond the dichotomous categorizations of social identity, previous research points to the importance of assessing the strength of in-group identification on intergroup attitudes and behaviors (e.g., out-group trust, intergroup forgiveness). Stronger in-group identity may lower out-group trust among adults. For example, those that identify more strongly with their in-group may be less likely to trust out-group members, and in turn, hold more negative out-group attitudes in situations of threat to group values, beliefs, or norms (Voci, 2006). Within societies emerging from protracted intergroup conflict, stronger in-group identification may heighten perceptions that trusting out-group members is a risk to the in-group, which in turn, has the potential to negatively impact intergroup relations and initiate a return to violence . However, previous research with young people voluntarily taking part in an intergroup peacebuilding program found that strength of social identity did not influence out-group trust (Leonard et al., 2015). Given potential selection bias (i.e., youth chose to participate in an intergroup program), there is a need to examine the effect, and potential interaction, of social identity and out-group trust on support for peace among a broader sample of youth.
A meta-analysis of research with adults found a negative relation between in-group identity and intergroup forgiveness in conflict settings (Van Tongeren et al., 2014). For example, perceptions of victimhood for past transgressions or atrocities carried out against the in-group may heighten group identification and reduce willingness to forgive the out-group (Cairns et al., 2005;Noor et al., 2008). To the best of our knowledge, however, no studies have explored the effects and possible moderating role of child and adolescent social identity on intergroup forgiveness and support for peace. Therefore, there is a need to understand the influence of conflict-related social identities on the potential link between intergroup forgiveness and peace for young people growing up in divided societies.

Post-accord Northern Ireland
Conflict between the Catholic and Protestant communities can be conceptualized as a struggle over issues of nationality and the constitutional status of Northern Ireland (MacGinty et al., 2007). The Catholic community, the historic minority in regard to population and perceived power, often identifies as Irish and wants to unify Northern Ireland with the Republic of Ireland. The Protestant community, the historic majority in relation to population and perceived power, tends to identify as British and wants to remain in the United Kingdom.
The most recent period of armed conflict, known as the Troubles, occurred between 1968 and 1998. The Troubles began following civil rights demonstrations by the Catholic community against economic and political inequalities. For example, in the 1970s, Catholics were disadvantaged in access to employment, education, and housing in comparison with Protestants (Ferguson et al., 2014). The civil rights movement contributed to an escalation of intergroup tensions and outbreak of sustained violence between state forces, paramilitary groups, and civilians (Cairns & Darby, 1998). A majority of conflict-related deaths comprised unarmed civilians, who tended to be Catholic (e.g., Mesev et al., 2009). The conflict officially ended with the signing of the Good Friday/Belfast Agreement and marked the beginning of the peace process, which established a power sharing government that needed majority approval from both communities to enact important decisions. Similarly, policies were implemented to promote intergroup equality and attempt to rectify the historic exclusion and discrimination of the Catholic community. That is, the peace process reflects the long-term path to social reconstruction and rebuilding relations between the two communities. The peace process is a commonly used and understood phrase among adults and young people in Northern Ireland (McKnight & Schubotz, 2017).
Within present-day Northern Ireland, both Catholic and Protestant communities have roughly equal populations (41% Catholic and 42% Protestant; Northern Ireland Statistics and Research Agency [NISRA], 2012). However, group-based differences in status and power remain. For example, Catholics often have lower socio-economic status in comparison with Protestants, and political parties representing the Catholic community are the minority faction within the power sharing government (Wright et al., 2017). The post-accord society continues to be divided along traditional group lines; day-to-day events are still viewed and understood within the context of group membership , even among children . Relations between the two communities also remain fraught. For example, young people may be directly exposed to outbursts of political violence (McEvoy-Levy, 2006). Similarly, youth may contribute to the maintenance of the cycle of violence by engaging in tit-fortat acts of intergroup aggression (Taylor & McKeown, 2019). Thus, the post-accord generation lives with the legacy of the conflict as well as on-going experiences with sectarianism (Noor et al., 2017). The post-accord generation, however, can also play an important role in peacebuilding . These children and adolescents may be more willing to reject long-standing conflict dynamics and engage in constructive intergroup behavior .

Current Study
The current study explores the effects of out-group trust and intergroup forgiveness as antecedents of support for peace, and the influence of social identity on these relations for adolescents growing up as part of a post-accord generation. Adolescence is a critical time for exploring and establishing identities (McKeown et al., 2020;Merrilees et al., 2018). Adolescents tend to display an increased awareness of intra-and intergroup dynamics, develop more complex models of understanding the self and others, and place an emphasis on cultivating a perception of fitting in as part of social groups. The article also explores differences between the two conflicting groups (Schulz & Taylor, 2018). For example, social identity may function differently for both Catholics (historic minority) and Protestants (historic majority) in the post-accord era (Merrilees et al., 2011. Finally, in line with the DPM (Taylor, 2020), the article focuses on broader support for the peace process at the macrosystem level (Townsend et al., 2016).
We hypothesize that out-group trust (H1) and intergroup forgiveness (H2) will positively relate to support for the peace process. Based on research with adults, we expect a negative relation between strength of in-group identity and out-group trust (H3a) and intergroup forgiveness (H3b). Complementing these direct effects, we explore the moderating role of strength of in-group identity on these relations for the historic majority and minority groups over time (H4).

Procedure
Ethical approval for the study was secured from the University of Notre Dame [#12-11-568]. Families with a child in the target age range (10-17 years old) were recruited to participate in the larger study from working-class neighborhoods in socially deprived areas of Belfast. Study areas ranked in the lowest quarter of wards in terms of access to basic services and were relatively homogeneous in terms of Catholic and Protestant backgrounds. Finally, study areas were also selected to capture variability in terms of historic and ongoing political violence. That is, families and youth would have had different experiences with past and current tension between groups.

Participants
The current analyses included participants who contributed annually to waves 5 and 6 of a larger, longitudinal study of the psychosocial impact of political violence on children and families in Belfast, Northern Ireland (see . In this sample, 667 adolescents (49% male) were 15.74 (SD = 1.99) years old at Time 1; given the timing of data collection, all would have been born just prior to or following the peace agreement. Reflecting the racial majority in Northern Ireland, all participants were White, coming from both Catholic (39%) and Protestant (61%) backgrounds.
Sample Size Planning. A Monte Carlo simulation for structural equation models (SEMs), including regressive paths and confirmatory factor analyses, using full information maximum likelihood (FIML) and a statistical power of 80% or greater (with α = .05) indicated that a sample of approximately 450 would be sufficient to detect regression coefficients of β = .25 and latent factors with loadings of .50 (Wolf et al., 2013); therefore, the current sample size (N = 667) was sufficiently powered for the proposed model.

Measures
The following describes the available data for each construct; that is, out-group trust and intergroup forgiveness were only measured in the final two time points, while support for the peace process was only included in the final time point.
Out-group Trust. Adolescents' out-group trust was assessed using an adapted version of a scale designed for Northern Ireland . This measure was developed to capture the degree to which an individual believed that they would not be harmed by the out-group across a range of domains. In the current study, factor analyses revealed that a four-item version of this scale was the best fit for the data, which was then used at Time 1 and Time 2. This included the indicators of "I can trust them when they say they are sorry," "I can trust them when they say they want peace," "I can trust the other community not to take all the jobs if they have the chance," and "I cannot trust the other community because they want revenge for things we have done to them" (reversed). All items were responded to using a 5-point Likert-type scale ranging from 0 (strongly disagree) to 4 (strongly agree). Higher scores indicated a greater trust in the other community. The scale had excellent internal consistency (α = .91).
Intergroup Forgiveness. This adapted scale measured the extent to which individuals recognize the need for forgiveness between communities (Hewstone et al., 2006). The original measure was developed based on focus groups (McLernon et al., 2002) and refined through application in Northern Ireland (Hewstone et al., 2006). At both time points of the current study, two reversescored items were used based on factor analyses: "It is important that my community never forgets the wrongs done to us by the other community" and "it is important that my community never forgives the wrongs done to us by the other community." Participants responded on a scale from 0 (strongly agree) to 4 (strongly disagree). Higher scores indicated greater intergroup forgiveness with good internal consistency (α = .81).

Strength of In-group Identity. At each time point, adolescents
were asked a series of five questions that assessed the strength of in-group identity with the Catholic or Protestant community (Brown et al., 1986). Participants could respond on a 5-point Likert-type scale to items assessing the extent to which they identified with the group and considered the group important. Responses could range from 1 (never) to 5 (very often), with higher scores indicating greater strength of in-group identification. This scale has been used previously in Northern Ireland with good reliability and validity. At Time 1, the measure displayed excellent internal consistency (α = .97).

Support for the Peace Process. At Time 2, adolescents' support
for the peace process was assessed using a measure of peacerelated attitudes used in settings of protracted intergroup conflict (e.g., Kurdish-Turkish, Uluğ & Cohrs, 2017; Israeli-Palestinian, Uluğ et al., 2021). In the current study, a single item was used: "I support the peace process in Northern Ireland." The single item about "the peace process" was chosen because the peace accord in Northern Ireland has existed for two decades; it is the only peace-related option that these young people have ever known. Participants responded on a scale from 0 (strongly disagree) to 4 (strongly agree). Higher scores reflected greater support for the peace process.

Data Analytic Plan
SEM using FIML will test the moderated cross-lagged effects in AMOS Graphics 18 (Arbuckle, 2009). Under the assumption that data are missing at random (Enders, 2010), FIML accurately estimates the parameters of interest. For each set of analyses, model fit will be evaluated using χ 2 /df index ⩽ 3, a Tucker Lewis Index (TLI) and comparative fit index (CFI) ⩾ .90, and a root-meansquare error of approximation (RMSEA) ⩽ .08 (Hu & Bentler, 1999).

Results
The means, standard deviations, and bivariate correlations for all controls, manifest variables, and composite scales for the constructs of interest are presented in Table 1.

Preliminary Analyses
Measurement Models. A confirmatory factor analysis using maximum likelihood estimation and promax rotation for a twofactor solution, supported by the scree plots and Eigen values, was conducted at each time point. At Time 1, the items for outgroup trust and intergroup forgiveness loaded on their respective factors, with no cross-loadings over .20; while each factor uniquely explained 39% and 34% of the variance (total 73%), the factors were also correlated (r = .57). A similar pattern was found at Time 2; that is, no cross-loadings over .20 and each factor uniquely explaining 38% and 32% of the variance (total 69%) with a correlation of r = .57 (indicating approximately 32% shared variance across the two constructs). In summary, given the good internal consistency of scales, low multicollinearity across constructs, and large sample size (Grewal et al., 2004), this empirical evidence suggests that the factors are related, yet conceptually distinct.
Therefore, latent variables were estimated for out-group trust and intergroup forgiveness. At Time 1 and 2, the four manifest indicators for out-group trust were fitted to latent variables; error variances were allowed to correlate overtime for each indicator. The fit for the measurement model of out-group trust over time was good, χ 2 (15) = 24.65, p > .05, N = 667; χ 2 /df = 1.64; TLI = .99; CFI = .99; RMSEA = .031 (CI = [0.001, 0.052]). For intergroup forgiveness, the two items were loaded onto a latent variable for forgiveness at each time point (with two manifest indicators, the model fit cannot be estimated).
Manifest Indicators. Adolescent age and gender were entered as manifest variables at Time 1, along with the composite score of strength of in-group identity given the high internal consistency (Perera & Maxwell, 2010). The composite scores of outgroup trust and intergroup forgiveness were centered and multiplied by in-group identity, respectively, resulting in two interactions terms (Cohen et al., 2018). As a single-item, support for the peace process was included at Time 2 as a manifest variable.

Full Model
We report the findings for the full model before proceeding to the exploratory analyses on the moderating role of strength of ingroup identity. In the full model test, all endogenous variables were allowed to correlate; the error variances for the exogenous variables were also allowed to correlate. The cross-lagged effects of out-group trust and intergroup forgiveness, along with the lagged influence on support for the peace process among adolescents in Belfast for all participants, was a good fit to the data, χ 2 (91) = 131.32, p < .05, N = 667; χ 2 /df = 1.44; TLI = .98; CFI = .99; RMSEA = .026 (CI = [0.015, 0.035]).
First, we examine the control variables; age at Time 1 did not significantly relate to out-group trust, intergroup forgiveness, or support for peace at Time 2. Compared with boys, girls reported higher out-group trust at Time 2 (β = .10, p = .016) and support for the peace process (β = .21, p < .001); there were no gender differences in willingness to forgive the other community at Time 2. Considering community background, at Time 2, adolescents from the Catholic community reported higher out-group trust (β = .20, p < .001), intergroup forgiveness (β = .17, p = .002) and support for the peace process (β = .21, p < .001), compared with Protestant youth. Adolescents with stronger in-group identity at Time 1 reported lower trust (β = −.22, p < .001) and intergroup forgiveness a year later (β = −.25, p < .001), supporting H3a and H3b, but higher support for the peace process (β = .15, p = .002). Finally, we examined the auto-regressive stability parameters for out-group trust (β = .45, p < .001) and intergroup forgiveness (β = .36, p = .002), both of which were significant. That is, adolescents who were higher in out-group trust or intergroup forgiveness at Time 1 were also higher at Time 2.
Second, we examined the lagged effects. Adolescents who reported higher out-group trust at Time 1 also reported higher intergroup forgiveness at Time 2 (β = .28, p = .005). However, the reverse did not hold. That is, greater willingness to forgive the other community at Time 1 did not significantly relate to outgroup trust at Time 2. In terms of the main hypotheses, although out-group trust did not (H1), willingness to forgive the other community (H2) did significantly relate to support for the peace process 1 year later (β = .30, p = .003). In summary, although outgroup trust did not have a direct link, greater intergroup forgiveness was significantly related to higher support for the peace process among adolescents in Northern Ireland.

Multiple Group Comparison Model
The exploratory analysis then examined a multiple group model for Catholics and Protestants (H4). Following Bollen (1989), all paths were free to vary across groups in the full model. As each path was constrained to be equal across Catholics and Protestants, the resulting model fit was compared with the previous model using the χ 2 difference test. If the successive model was not significantly worse, the constraint was retained across groups. Following this procedure, only three paths differed across groups, and the final model was a good fit to the data, Figure 1; χ 2 (200) = 300.89, p < .05, N = 667; χ 2 /df = 1.50; TLI = .95; Out-group trust (Time 1) 3.36 1.46 0-6 .225** .490** −.059 -5 Intergroup forgiveness (Time 1) 1.43 0.99 0-3 .112* .308** −.188** .584** -6 Out-group trust (Time 2) 3.25 1.31 0-6 .258** .403** −.165** .629** .410** -7 Intergroup forgiveness (Time 2)  Note. *p < .05; **p < .01. The anchors for trust scales and peace item were strongly disagree to strongly agree, anchors for forgiveness were strongly agree to strongly disagree, and anchors for the strength of in-group identity were never to very often. CFI = .97; RMSEA = .028 (CI = [0.021, 0.034]). The stability parameter of intergroup forgiveness was significantly different, χ 2 diff (1) = 5.17, p < .05; Catholics: β = .11, p = .413; Protestants: β = .48, p < .001. In addition, both interaction terms were significantly different across groups. That is, for Protestants there was a significant negative effect of strength of in-group identity on the link between out-group trust and support for the peace process, χ 2 diff (1) = 5.19, p < .05; Catholics: β = .002, p = .985; Protestants: β = −.27, p = .007. Whereas for Catholics, strength of in-group identity was a significant and negative moderating effect on the relation between intergroup forgiveness and later endorsement of the peace process, χ 2 diff (1) = 11.23, p < .05; Catholics: β = −.43, p < .001; Protestants: β = .11, p = .297. In summary, significant differences emerged around the moderating effect of strength of in-group identity for Catholic and Protestant adolescents for the lagged effects of out-group trust and intergroup forgiveness on support for the peace process.

Discussion
The current study explored the effects of out-group trust, intergroup forgiveness, and strength of in-group identity on support for peace among youth growing up in post-accord Northern Ireland. Framed by the DPM (Taylor, 2020) and SIDT (Nesdale, 2004), the study adds to an emerging body of work identifying factors that contribute to peacebuilding among children and adolescents growing up in conflict-affected societies. The full model revealed important cross-lagged and lagged effects. Higher out-group trust related to greater willingness to forgive the other community 1 year later, whereas earlier forgiveness did not relate to later out-group trust. This study expands previous work on trust with children (e.g., Rotenberg et al., 2005), focusing specifically on out-group trust. Consistent with previous research with adults (Noor et al., 2015;Song et al., 2020;Tam et al., 2007), out-group trust may provide a foundation for intergroup forgiveness for youth growing up as part of a postaccord generation. There was mixed support for H1 and H2 related to support for peace. Although out-group trust did not significantly relate to later support for peace, intergroup forgiveness did. Intergroup forgiveness may enable children and adolescents, even those born after the peace agreement, to give up desires for revenge and retribution; thus, forgiveness may help to interrupt the cycle of violence, facilitate social reconstruction, and promote cooperative intergroup relations in the future (Noor et al., 2015).
There were also significant direct effects of social identity on later out-group trust, intergroup forgiveness, and support for the peace process in Northern Ireland. Consistent with H3a and H3b, In-group Identity X Intergroup Forgiveness adolescents with stronger in-group identity reported lower trust and forgiveness a year later. It is possible that adolescents who identify more strongly as Catholic or Protestant may view outgroup trust as a threat to in-group security ; stronger identification with a conflict-related group may also be linked with reluctance to forgive the other community for past suffering (Cairns et al., 2005;Noor et al., 2008). Perhaps surprisingly, in the full model, young people who identified strongly with their community background reported higher support for peace a year later. Social identity may foster collective efforts to further achieve goals and enhance group status (Hammack, 2010). Integrating these findings: adolescents born after the agreement may recognize that both groups are locked into the peace accord. Thus, despite lower trust or forgiveness, high identifiers may simultaneously recognize that the best way to defend, maintain, and improve the status of their in-group (Nesdale et al., 2005a(Nesdale et al., , 2005b; is through the existing peace framework. Exploratory multiple group comparisons revealed divergent findings regarding the moderating effect of social identity. For both groups, stronger in-group identity moderated support for peace, albeit for distinct constructs. For Catholics (historic minority), strength of in-group identity increased the positive impact of earlier intergroup forgiveness on support for the peace process. For Protestants (historic majority), stronger in-group identity magnified the link between out-group trust and support for peace. These results suggest that for understanding support for peace among youth who strongly identify with their in-group, intergroup forgiveness for past wrongs (e.g., discrimination and collective suffering) is stronger for Catholics, while trusting outgroup members not to exploit their rising status (e.g., taking jobs and desiring revenge) is stronger for Protestants. The importance of trust in this point is underscored by its importance in the functioning of labor markets more broadly (Kirchler et al., 1996).
In line with the DPM (Taylor, 2020), the findings highlight the peacebuilding potential of young people, which have important implications for promoting meaningful change at the macrosystem level. Beyond improving interpersonal relations, this study examines factors that relate to youth support for the peace process, which outlines structural and cultural change in Northern Ireland. The nuanced findings suggest a "one-size-fits-all" approach will not work among the post-accord generation, as both groups negotiate both the legacy of the conflict and lasting asymmetric social status and power relations. Symbolic and materials steps toward peace may also foster civic and political engagement among youth in Northern Ireland . These constructive patterns of engagement may contribute to the unfreezing of traditional conflict dynamics, and in turn, consolidate peace.

Limitations and Future Directions
The current study has a number of strengths. For example, the longitudinal design allows for inferences about the direction of some effects. The study was conducted within a real-world context of intergroup conflict, which may enhance the overall ecological validity of results. Inclusion of both the historic minority and majority groups also provides a more complete picture of the possible steps on the path to peace. Using FIML along with the Time 1 variables accounts for potential missingness on those variables assuming that the data are missing at random (Enders, 2010). Yet, there are also a number of limitations that could be addressed in future research.
First, the longitudinal design was correlational with only a single reporter. The measures used in these analyses were added during the course of the study, as the youth aged, and are only available at the time points reported. Moreover, a range of intraand interpersonal factors, such as empathy, may shape peacebuilding (Noor et al., 2008) and more constructive intergroup relations (O'Driscoll et al., 2021b;Taylor et al., 2020b). Future research should extend the current model to consider the antecedents of out-group trust and intergroup forgiveness, as well as incorporate multiple reporters (e.g., parents), to provide a broader understanding of factors relating to youth support for the peace process.
Second, support for the peace process, a commonly used term to refer to the 1998 Belfast Agreement (McKnight & Schubotz, 2017), was measured using a single item previously used in conflict settings. Future research should consider adopting more comprehensive multi-item measures that assess both endorsement of specific peace-related policies (e.g., integrated education) or controversial issues (e.g., symbolic reparations) to gain a more nuanced understanding of support for peace among young people growing up in Northern Ireland .
Third, qualitative follow-up might provide more nuanced understanding of youth's conceptualizations of out-group trust and intergroup forgiveness. For example, cognitive interviewing might unpack the target more precisely. When adolescents born after the peace accord refer to the "other community" for forgiveness, do they have in mind those responsible for the harm caused during the Troubles? Or, are their responses informed by exposure to annual spikes of violence and intergroup tension? Teasing apart the contemporaneous along with the vicarious or indirect feelings around trust and forgiveness (e.g., Noor et al., 2017), may be a promising area for future research with the post-accord generation.
Fourth, social identity is complex and changes across time for children and adolescents in conflict settings (Merrilees et al., 2018). For example, key distinctions may exist between religious and national identities for young people living in post-accord Northern Ireland . SIDT outlines that social identity may strengthen the relation between conflict and youth intergroup attitudes and behaviors (Nesdale et al., 2005a). As such, future research could explore the effects of exposure to political violence on changes in identity, which in turn, may influence the antecedents for peacebuilding.

Implications
The findings of the current study have implications for promoting more peaceful intergroup relations in Northern Ireland. There is a need for peacebuilding initiatives to cultivate out-group trust as a foundation for the development of intergroup forgiveness and support for peace among youth (e.g., Noor et al., 2015). Given the benefit and burden of social identity for children and adolescents in conflict-affected societies (Hammack, 2010), results indicate the need for interventions tailored to the specific perspectives and needs of minority and majority group conflict rivals in the post-accord generation. On one hand, interventions for Catholics could emphasize the importance of forgiveness of historic injustices and victimization to build a shared future. On the other hand, interventions for Protestants could involve generating a sense of trust that members of the other community will not take advantage of their emerging status, including seizing all jobs or desiring revenge for past harms. Targeted interventions of this kind may be particularly beneficial for those that identify strongly with their in-group.
The current study makes a unique contribution by highlighting the effect of out-group trust, intergroup forgiveness, and social identity on support for the peace process among both majority and minority youth growing up in the post-accord generation. The findings have important implications for practitioners and policy makers within Northern Ireland and other conflict-affected settings.