Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Nov 30.
Published in final edited form as: Peace Confl. 2014 Feb;20(1):26–38. doi: 10.1037/a0035581

Youth in contexts of political violence: A developmental approach to the study of youth identity and emotional security in their communities

Christine E Merrilees 1, Laura K Taylor 2, Marcie C Goeke-Morey 3, Peter Shirlow 4, E Mark Cummings 5
PMCID: PMC4664468  NIHMSID: NIHMS725611  PMID: 26633936

Abstract

Going beyond the association between youth exposure to political violence and psychopathology, the current paper examines within-person change in youth strength of identity with their ethno-political group and youth reports of the insecurity in their communities. Conceptually related but growing out of different paradigms, both group identity and emotional insecurity have been examined as key variables impacting youth responses to threats from other group members. The goal of the current study is to review previous studies examining these two key variables and to contribute new analyses, modeling within-person change in both variables and examining co-variation in their growth The current paper uses data from 823 Belfast adolescents over 4 years. The results suggest youth are changing linearly over age in both constructs and that there are ethno-political group differences in how youth are changing. The results also indicate that change in insecurity is related to strength of identity at age 18, and strength of identity and emotional insecurity are related at age 18. Implications and directions for future work in the area of youth and political violence are discussed.

The study of youth in contexts of political violence has increased in number and quality since the call for such work by Cairns and Dawes in 1996. At the time of that writing, they noted that the study of youth in these contexts needed to go beyond the simple associations between exposure to political and ethnic violence and psychopathology. In particular they called for a more nuanced approach to understanding the complexities surrounding the ways in which growing up in a context of political violence affects youth development across domains, not just assuming it leads to psychopathology. In fact, Cairns and Dawes noted that the link between political violence and psychopathology is not as strong as one might expect given the destructive nature of such violence. At the conclusion of their commentary they identified three directions for future research: (1) use of ethnographic approaches that “enable us to examine the language and practices through which children and adults co-construct a violent conflict situation” (Goodnow, Miller, & Kessel, 1995); (2) longitudinal research in contexts where peace has been established and (3) research conducted within developmental theory.

This work called attention to the need for theoretical models to shape our examination of how and why, and for whom and when, political violence affects child development, with the goal of identifying the multiple psychological mechanisms that unfold over time for youth in contexts of political violence. Data for the current paper comes from a project designed to incorporate many components of this vision.1 The project included a qualitative investigation that grounded the work in the current perspectives of youth and mothers in segregated and economically-deprived neighborhoods in Belfast (Taylor et al., 2011). The goals of this phase were to embed ourselves as researchers in the words and experiences of families and to develop questionnaires of relevant experiences in their communities that were culturally informed. We then completed a two-wave pilot study (Goeke-Morey et al., 2009) to validate the measures we developed that served as cornerstone constructs for the main project. Finally, the main longitudinal project resulted in surveys with 999 mothers and their pre-adolescent and adolescent children over 6 years.

The longitudinal project was rooted in a social ecological framework, incorporating key variables at different levels of the social ecology (Cummings, Goeke-Morey, Schermerhorn, Merrilees & Cairns, 2009). Placing the study in this larger framework encouraged the inclusion of multiple perspectives to identify the many paths through which political and ethnic violence affect youth. Drawing from this developmental approach that emphasizes cultural context, two variables emerged as foci of the project: emotional security theory (EST: Cummings & Davies, 1996; Davies & Cummings, 1994) coming out of developmental psychopathology, and the social identity paradigm stemming from social psychology. In our collaborative longitudinal study of political violence and children in Belfast, we worked on advancing group identity processes and emotional security as explanatory processes for understanding children’s development. Although we have published numerous studies on the distinct contributions of each paradigm, we have not yet accomplished an integrative treatment that examined the mutual relations between these constructs. Thus, the goal of the current paper is to bring together these two major theoretical approaches that have been developed to explain the effects of political violence on youth.

Social Identity Paradigm

Drawing from the wealth of theoretical and empirical work provided through the lens of social psychological theories such as social identity theory (Tajfel & Turner, 1979) and social categorization theory (Turner, Hogg, Oakes, Reicher, & Wetherell, 1987), the study included both youth and mother reports of their strength of identity with one of two main ethno-political in-groups: Catholics and Protestants. Social identity theory, and the empirical and theoretical literature that stem from it, have sought to explain the intergroup processes and conditions that promote negative intergroup attitudes and behaviors such as prejudice and group-based aggression (Tajfel & Turner, 1979). The social identity paradigm (Tajfel & Turner, 1986) describes the motivation for group members to positively differentiate their own group from relevant out-groups in order to achieve a sense of positive identity. Research from this paradigm has supported the notion that individuals with a stronger sense of identity with their in-group are more likely to have negative attitudes toward out-group members, develop prejudiced attitudes toward out-group members, and even promote the use of aggression against out-group members (Bizman & Yinon, 2001; Branscombe & Wann, 1994; Fischer, Haslam, & Smith, 2010; Struch & Schwartz, 1989). At the same time, there are benefits derived from group memberships that promote individual and group well-being through processes such as social support and a motivation to work toward social change (Hammack, 2010). Group memberships also foster positive intergroup relations through processes such as empathy and helping (Tropp & Mallett, 2011). There are clear implications for understanding these social psychological processes for youth developing in contexts where group identity is salient (Cairns, 1996).

In examining the complexity of identification with an ethno-political group in socially deprived neighborhoods in Belfast, our project has generated support for stronger identification with the in-group functioning as both a protective and a risk factor. We found that stronger social identity helped to buffer Catholic mothers’ well-being related to the negative impact of the Troubles (Merrilees et al., 2011a). On the other hand, social identity acted as a risk factor for Protestant mothers in that it increased the likelihood of maternal adjustment problems. Greater social coping also helped to protect mothers’ mental health from current nonsectarian violence, but exacerbated depressive symptoms when used in response to daily sectarian violence (Taylor et al., in press). Mothers also reported that they pulled together in the face of sectarian conflict (Taylor et al., 2011). Greater social identity thus may increase social cohesion in these communities, helping mothers to work together for positive social change. For youth, identification with the in-group also had positive and negative effects. Youth with higher identification with their in-group reported less overall aggression in the face of sectarian threat, but engaged in more out-group aggression and delinquency (Merrilees et al., in press-a). At the same time, youth with stronger social identity were less likely to experience emotional problems over time, even when they experienced sectarian antisocial behavior in the community (Merrilees et al., in press-b). This protective role of social identity was particularly salient for Protestant youth.

Emotional Security Perspective

Emotional security theory (EST, Cummings & Davies, 2010; Davies & Cummings, 1994) was proposed as a developmental theory explaining the long-term effects of living in a context of destructive marital conflict for children and adolescents. Having its roots in attachment theory, emotional security theory describes the child’s need to maintain a set goal of felt security in the family system as a whole. This concept has since been expanded to include a child or adolescent’s need to maintain that same sense of felt security in their community (Cummings et al., 2010a). This focus on the critical role that non-familial group connections and identifications play clearly bridges with the social identity paradigm (Tajfel & Turner, 1979; Turner et al., 1987).

EST was originally developed to explain the negative effects of marital conflict on youth development. Extending notions of a child’s need for a secure base in the parent-child relationship (Bowlby, 1969), EST posits that children also maintain a sense of security with the family system. Reactions to family stress, such as conflict and other cues of family instability are motivated by a higher-order goal of maintaining a sense of security. When the security system is activated by family conflict and other indicators of family instability, children react affectively, behaviorally, cognitively, and physiologically in an effort to regain their felt security. Over prolonged exposure to negative and destructive marital and family conflict, children may become sensitized to cues of conflict and learn to over generalize their negative responses. Cummings and colleagues (Cummings & Davies, 2010; Cummings, Goeke-Morey, Schermerhorn, Merrilees, & Cairns, 2009) have recently extended this notion of emotional security to meaningful social systems outside of the home. Drawing from a social-ecological perspective, emotional security has been developed to understand children’s responses to conflict and instability in their communities. Specifically in communities where group-based identities (ethnic, racial, religious, and political, etc.) are salient, conflict between groups may trigger similar patterns of responses with increased exposure over time.

Using longitudinal mediation tests, multi-level modeling, and trajectories of individual change, there is strong support for the important role of child emotional security about community and family for child adjustment in Belfast. Emotional security about community has been shown in cross-sectional and three-wave longitudinal tests to mediate the impact of sectarian, but not nonsectarian, antisocial behavior on youth internalizing and externalizing problems (Cummings et al., 2011; Cummings et al., 2010a, 2010b). Emotional security in the family system has also been shown to be an important mediating process for child adjustment in the face of sectarian violence (Cummings et al., 2012; Cummings et al., 2010b). Looking at within-person change, emotional security about community may also explain the relation between experience of sectarian threat and youth adjustment (Cummings et al., in press a and b).

Integrating a Developmental Perspective

Central to the study of child development and conflict and peace studies is how the constructs of interest change over time. Researchers across contexts and conflicts have begun to tackle this challenge by examining longitudinal relations for youth in settings of protracted conflict. For example, political violence has been found to increase other forms of conflict (e.g., community and family conflict and violence). These numerous forms of stress have negative implications for child well-being (Boxer et al., 2013; Cummings et al., 2010a; Cummings et al., 2010b). Other community factors, such as stigma for child soldiers, are also associated with greater internalizing problems years after the end of war (Betancourt et al., 2013). However, individual, family and community-level factors such as self-esteem, positive parenting and family cohesion may also protect children from developing adjustment problems (Cummings et al., 2010a; Dubow et al., 2012) and child soldiers who experienced community acceptance over time also reported greater development of adaptive attitudes and behaviors (Betancourt et al., 2010).

As Cairns and Dawes noted in 1996, the field requires longitudinal models and analysis that allow for the assessment of change over time. Although longitudinal studies on youth in contexts of political violence are emerging, few have examined both within- and between-person change processes. These two sets of questions are equally important and relevant to identifying risk and protective processes for youth in contexts of political violence. Within-person change processes indicate a change relative to where a person was at a previous point in time, relative to herself. Between-person change, on the other hand, indicates an individual’s change in relation to the other individuals in the sample. The distinction seems subtle, but within- and between-person changes may indicate different conclusions regarding relations between two variables. For example, Rankin & Maggs (2006) found differences in the within- and in the between-person effects in the association between negative mood and drinking behaviors in college students. The within-person analysis suggested that when an individual person reported lower negative mood on a given day relative to the previous day, they were less likely to drink alcohol. On the other hand, looking at between-person differences, the data suggested that young adults with lower negative mood were more likely to drink than young adults reporting higher mood scores. The authors conclude that within-person change in negative mood, or individual mood swings, was an important factor in predicting problem drinking. This finding may not have been revealed through between-person analyses which often obscure individual patterns of change.

Current Study

In the current paper we take an initial step toward integrating these two theoretical models by examining within-person trajectories of youths’ reports of their strength of social identity with their Catholic and Protestant groups and their security in their communities. The goal of modeling within-person changes in this way provides an insight into how these processes may shift as youth develop through adolescence. Having grown out of different paradigms, strength of identity and emotional security have different theoretical bases, but both are expected to change in the face of intergroup threat as they share a motivational base of increasing a sense of individual well-being in the short-term. However, they also share the potential of leading to negative consequences over the long-term. As youth develop a greater awareness of group distinctions, naturally occurring changes in the sense of community and group identity may shift. To test this notion, latent growth curve modeling was used to model within-person change processes. Between-person changes were examined by evaluating ethno-political group differences in these processes, and by looking at co-variation in the two growth processes and their intercepts.

Method

Participants

The participants in the current analyses included 823 adolescents (n = 482; 48% male) who have participated in waves 3 through 6 of a longitudinal study of the impact of political violence on children and families in Belfast, Northern Ireland. Average age for youth across the four waves was 13.61 (SD = 1.99) at Time 1, 14.66 (SD = 1.96) at Time 2, 15.75 (SD = 1.97) at Time 3, and 16.83 (SD = 1.99) years old at Time 4. Age range across the sample spans for the current analyses ranges from 10 to 18 years old. Youth came from the two primary ethno-political groups in Belfast; 62% were from Protestant and 38% were from Catholic backgrounds. Consistent with the demographics in Belfast, all participants were White.

Examining the patterns of retention across the four time points, between any two time points, 80% to 87% of the families returned. Eighty percent of the families from Time 1 participated at Time 2, 81% from Time 2 to Time 3, and 87% from Time 3 to Time 4. These rates are at the higher end of the range of retention with high-risk samples (e.g., Betancourt et al., 2010; Browning, Burrington, Leventhal, & Brooks-Gunn, 2008; Kronenberg et al., 2010).

Procedures

A political demographer, expert in patterns of sectarian strife in Belfast, identified study areas that were homogenous, interfaced neighborhoods in socially-deprived wards that had variation in experience of the Troubles. Therefore, the majority of study areas were ethnically homogenous neighborhoods (greater than 90% Catholic or Protestant), and interfaced, or separated by “peace walls” or major roads from a neighborhood in which the other group is the majority. This physical basis for segregation was enhanced by social patterns which promoted isolation between the two groups along personal, educational, and professional lines (Shirlow & Murtagh, 2006). To minimize differences in socio-economic (SES) status among families, all study areas ranked in the lowest quartile of socially-deprived wards in Northern Ireland. There was a range in the severity of the number of conflict-related deaths during the Troubles across all study areas; this variability is useful for understanding current patterns of sectarian antisocial behavior. Given this profile for study areas, stratified random sampling was used to recruit approximately 35–40 families in each neighborhood.

Study design and procedures were approved by the Institutional Review Board at all participating universities. Professional interviewers from Belfast working with an established market research firm collected all of the annual data through face-to-face interviews in the participants’ homes. Youth provided assent, with parental consent, prior to participating at each time point. Each interview lasted approximately 45 minutes, and families were given £40 at time 1 and 2 and £50 at time 3 and 4 for their participation.

Measures

Strength of social identity (SSI)

The level of identification or attachment to a socially-relevant in-group was measured using Brown’s strength of social identity scale (Brown, Condor, Mathews, Wade, & Williams, 1986), which has been used previously in Northern Ireland with regard to belonging to the Catholic and Protestant communities respectively (Cairns, Kenworthy, Campbell, & Hewstone, 2006; Merrilees et al., 2011a; Merrilees et al., in press-a). Youth first state the group with which they identity, and then answer five follow up questions, including items such as I strongly identify with the (insert Catholic or Protestant) community and I am a person who feels strong ties with the (insert Catholic or Protestant) community. The response range is a 5-point Likert scale from 1= “never” to 5 = “very often,” with a range from 5 to 25 and higher scores indicated stronger social identity. Youth scores were added for a composite manifest variable of social identity. The internal consistency of social identity across the four time points was strong (Cronbach’s α = .94, .94, .97 and.98, respectively).

Emotional insecurity in the community (SIC)

Youth reported on emotional insecurity in the community on a measure that was developed specifically for this longitudinal study. Qualitative focus groups with mothers in Belfast generated the items to assess a child’s behavioral, affective, and cognitive responses to their sense of safety and threat in the community. A two-wave pilot study established the psychometric properties and predictive power of the mother report on this scale (Goeke-Morey et al., 2009). Sample items include I feel threatened by people approaching from the other community and Sometimes I feel that something very bad is going to happen in our community. Adolescents responded to 7 statements using a 5-point Likert-type scale ranging from 1 = “not at all like me” to 5 = “a whole lot like me.” Higher scores indicated greater insecurity about community with a range from 7 to 35. Cronbach’s α for this scale across the four time points was .86, .85, .82, and .74 respectively.

Analysis Plan

Latent growth curve modeling (LGCM) was used to assess within-person changes in strength of identity with the in-group and insecurity in the community. For the current analyses, the models were parameterized such that the intercept is the last time-point, age 18. Modeling the data with LGCM will tell us if and how youth are changing in levels of strength of identity and emotional insecurity as they age. Maximum likelihood estimation was used, which accurately estimates parameters with missing data under the assumption that the data is missing at random (Singer & Willett, 2003). One benefit of using LGCM with ML estimation is that cases with missing data are not excluded from the analysis. Between-person differences in the within-personal parameters were also examined. Specifically we looked at differences between Catholics and Protestants in their slopes, or rates of change, for both identity and insecurity in the community. We also looked at differences between Catholics and Protestants in their intercepts, or their estimated score on each variable at age 18. All analyses were conducted in MPLUS (Muthén & Muthén, 2006). Means and standard deviations for all study variables appear in Table 1.

Table 1.

Means and Standard Deviations for SIC and ISA Variables

Variable M SD
1. Age 10 SIC 9.26 2.38
2. Age 11 SIC 11.05 5.39
3. Age 12 SIC 11.00 6.06
4. Age 13 SIC 10.41 5.12
5. Age 14 SIC 9.95 4.38
6. Age 15 SIC 9.46 4.18
7. Age 16 SIC 9.21 3.62
8. Age 17 SIC 8.53 2.75
9. Age 18 SIC 8.55 2.92
10. Age 10 SSI 16.49 6.38
11. Age 11 SSI 16.89 6.12
12. Age 12 SSI 16.23 6.02
13. Age 13 SSI 17.44 5.85
14. Age 14 SSI 17.71 5.66
15. Age 15 SSI 17.91 5.36
16. Age 16 SSI 17.73 5.92
17. Age 17 SSI 18.17 5.60
18. Age 18 SSI 18.16 5.56
Open in a new tab

Note. SIC = Insecurity in the community; SSI = Strength of Social identity

Results

Models of Within-person Change

As a first step, two separate sets of models were run for the two main constructs of interest to assess the best model for within-person change. The best fitting model was determined by the chi-square difference test and the root mean square error of approximation (RMSEA; Wu, West, & Taylor, 2009). Acceptable model fit is indicated by an RMSEA≥.08 (Hu & Bentler, 1999). For both strength of in-group identity (χ2diff (3) = 16.03, p < .01) and emotional security in the community (χ2diff (3) = 259.3, p < .001), the linear model fit better than the intercept only model, and the quadratic models failed to converge. For both variables, emotional security in the community and strength of identity with in-group members, linear trajectories were used in subsequent models.

Strength of in-group identity

The results suggested that a linear model was the best fitting model (Linear model RMSEA = .040; Intercept only model RMSEA = .043). Examining the means of the intercept and slope parameters suggest that at age 18, the mean level of strength of identity with the in-group was M = 17.95 (p <.001) and that the average change in identity was not significantly different from zero. Looking at the variance estimates of these parameters suggests that there is individual variability in ending points (intercepts) and rates of change.

Emotional insecurity in the community

Similar to the models for identity, the model comparisons for SIC suggested a linear model best captures within-person change (RMSEA = .08). The means and intercepts for SIC indicate that youth are decreasing in insecurity as they age (M = −.44, p < .001) with a mean level of SIC of M = 8.32 (p < .001) at age 18. The variance components also suggest individual variability in ending points and rates of change.

Parallel process model

To assess changes in both variables over time, the next model included latent growth models of both variables and correlations between the parameters for estimates of the variables. This parallel process model allows for the time-varying nature of both processes simultaneously (See Figure 1). Estimating the model resulted in a negative variance estimate for the linear slope of the SIC variable. When this variance was fixed at 2, the estimated value of the variance for this parameter from the LGCM, the model terminated normally with no additional problems (RMSEA = 0.6). The results of this model show that on average youth are decreasing in insecurity over age (M = −.43, p < .001), and youth on average are not changing in strength of identity. The intercepts are significantly related indicating that youth with higher insecurity at 18 reported higher strength of identity at age 18 (r = .52, p < .001). The slope of insecurity and the intercept of strength of identity are also related indicating that change in insecurity is related to strength of identity at age 18 (r = .23, p = .048). This result suggests that youth who are decreasing in insecurity less slowly report higher strength of identity at age 18.

Figure 1.

Figure 1

Open in a new tab

Conceptual Model for the Interrelations Between Change in Strength of Identity and Change in Emotional Insecurity.

Parallel Process Model for Catholics and Protestants

Given the significant variability in model parameters, the models were next tested separately for Catholic and Protestant groups. One significant difference in the pattern of the models emerged. Results for Protestants suggest they are increasing in strength of identity over age (M = .26, p = .003) and Catholics show a trend toward decreasing strength of identity over age (M = −.22, p = .079). These trajectories are plotted with the full sample trajectory of emotional insecurity in Figure 2.

Figure 2.

Figure 2

Open in a new tab

Change in Strength of Identity and Emotional Insecurity by Age

Discussion

Cairns and Dawes (1996) challenged us to develop theory-driven, longitudinal research to understand the complex and nuanced experiences for youth in contexts of political violence. Responding to this call, the current paper examines links between two key variables thought to impact youth adjustment in contexts of political violence. More specifically, we examined the interrelations between youth strength of identity with their ethno-political group and their sense of emotional security in their communities. Coming out of social psychology, the impact of strength of identity with in-group members on intergroup attitudes and behaviors has been documented across contexts of intergroup tension and war. Emotional security theory, on the other hand, grew out of developmental approaches that explain the regulatory processes that organize youth responses to socially meaningful conflict (e.g., marital conflict, family conflict). The current study examines relations between these two variables using a longitudinal dataset, with measures designed through a grounded, qualitative approach. We hope the results of this study honor the vision outlined by Cairns and Dawes (1996) by demonstrating the pertinence of a more nuanced approach to understanding the complexity of factors that affect youth in divided societies.

Data for the current paper were pulled from the last four waves of our six-wave longitudinal study of youth and families in Belfast that began in 2006. The larger study has resulted in a growing and emerging body of published papers assessing multiple pathways of influence within the social ecology of political violence. Our previous findings support the notion that youth strength of identity with in-group members acts both as a risk and protective process. More specifically our studies have found that strength of identity buffers individuals from the mental health consequences of experiencing political violence, but that the power of identity to buffer the effects of exposure to political violence may differ between ethno-political groups (Merrilees et al., 2011; Merrilees et al., in press-b). At the same time, strength of group identity may also increase the likelihood of using aggression against out-group members (Merrilees et al., in press-a), consistent with previous research within a social identity paradigm. Another body of research has supported the notion that emotional insecurity in the community is an important regulatory process for youth exposed to repeated incidents of intergroup conflict (Cummings et al., 2010; 2011; in press-b). Developed from an attachment perceptive, these studies support the notion that children and adolescents develop regulatory processes that help them maintain felt security not just in the family system, but in other salient social groups such as their community (Cummings & Davies, 2010).

Utilizing the longitudinal approach called for almost two decades ago by Cairns & Dawes (1996), the current study explored within-person changes in youths’ sense of in-group identity and their sense of emotional security in the community, two constructs that have contributed to our understanding of risk and protective processes for youth in contexts of political violence. The results examine the individual trajectories of change in these two key psychological variables. Identifying within-person changes in these constructs may help identify naturally occurring developmental changes that should be considered in studies of youth development. Moreover, the analyses incorporate relevant between-group differences (i.e., Catholic/Protestant youth). This set of findings highlights that the individual developmental process for a given adolescent may be affected by the social group he/she belongs to.

The results show that youth are changing in their strength of identity with in-group members and in their sense of emotional insecurity. The overall model suggests that youth are increasing in their strength of identity with in-group members as they age. In other words, it is not just the case that older youth report higher strength of identity, but these results support the age-related development of that increase with repeated measures data. Looking at differences in the trajectories for Catholic and Protestant youth suggests that youth identifying with these two main ethno-political groups are changing differently; Protestant youth are increasing in their strength of identity as they age, while Catholic youth are reporting no change or possibly a slight decrease with age. One potential explanation for the difference in how youth are changing in the post-accord period could be related to the increased efforts made by cultural groups on the Protestant side to increase the participation of youth and families in cultural and fraternal organizations (McAuley, Mycock, & Tonge, 2012). The results also suggest that although change processes are different for Catholic and Protestant youth, there is no mean difference between Catholics and Protestants by age 18 in this sample. That is, over the course of adolescence, youth from the two communities experience differences in how strongly they identify with their in-group, but as they emerge into adulthood, strength of identity is comparable across Catholics and Protestants.

At the same time, youth in both groups are decreasing in their insecurity in the community as they age. The within-person change in emotional insecurity in the community suggests that, on average, insecurity decreases with age; in other words, youth are increasing in their sense of security in their communities through adolescence. One explanation for this pattern is an adaptive stress and coping response. That is, as youth experience threat in the community, they react through behavioral, emotional, and cognitive changes that satisfy their need for security. As youth transition from middle childhood through adolescence they may acquire greater peer and neighborhood resources outside of the family system that would be particularly relevant to providing a sense of security in the community. The between-person analyses found a lack of group differences in the trajectories of insecurity, which suggests that the process may be similar for Catholic and Protestant youth. Nonetheless, despite an average decline in insecurity in community, youths’ emotional insecurity about community remains pertinent to their adjustment and well-being and that within-person change in insecurity has implications for youth adjustment (Cummings et al., in press-b).

Examining the two processes together suggests that strength of identity with the in-group and insecurity in the community are positively related at age 18 even though the slopes of the two variables are not correlated. The overall model for both Catholics and Protestants suggests that youth with higher strength of identity report more insecurity in their community. The results also indicate that change in security in the community is related to youth reports of strength of identity at age 18. That is, the less quickly youth are decreasing in insecurity, the higher their reported strength of identity with their in-group. In other words, the youth who report persistent insecurity in the community through adolescence also reported higher strength of identity at age 18. It is possible that youth who continue to feel insecure in the community, respond by seeking more support from their family and in-group peers, strengthening their connection to own-group members.

These results further support the notion that youth are not just passive recipients of the effects of political violence (Barber, 2009). Results from previous studies and the current analyses suggest that youth are engaging in coping processes that serve to decrease their sense of insecurity (Cummings et al., in press-b), potentially through group-based processes and social support (Merrilees et al., 2011; Taylor et al., 2011; Taylor et al., in press). Previous research tested within the social identity paradigm suggest that regulating one’s sense of threat by pulling together with group members can have both positive and negative effects, with the former coming through increased sense of social support and the latter through coming together to derogate out-group members (Hammack, 2010; Merrilees et al., 2013; Merrilees et al., in press-b).

Assessing the short and long-term effects of these coping strategies will help untangle for whom and under what conditions group affiliations stimulate peacebuilding or destructive responses to out-group members. One limitation of the current approach is the inability to assess how variables are dynamically affecting each other over time. Accurate assessment of the interrelations between these variables requires data be collected on the time-scale in which theory or previous empirical research suggests change happens. Many longitudinal studies, including the current one, collect data on an annual basis, limiting the conceptualization of how and when change happens. For example, understanding how emotional responses unfold in a discrete time period (e.g., within days, or within weeks) may provide a window to understanding how self- and group-based regulation processes develop over time as group-based threats are experienced. It is possible that the short-term benefits of pulling together in the face of threat are maintained by derogating out-group members, psychologically and physically reproducing conflicts (Bar-Tal, 2007). On the other hand, if group-based processes take constructive forms that serve to increase within-group feelings without devaluing out-groups, the emotional benefits of such processes may not have intergroup costs.

Much of the research being conducted across contexts of political and ethnic conflict focus on maladjustment, such as internalizing problems and post-traumatic stress. Thus, still more work needs to be done to understand the ways in which youth are constructing their experiences of their context and in particular, their conceptions of well-being (Barber, McNeely, & Spellings, 2012). Recognizing and understanding youth agency and the potential constructive contributions to rebuilding their societies should also be pursued, calling attention to the notion that children are not merely passive victims, but agents of constructive change (Cairns, 1996; McEvoy-Levy, 2006). Youth visions of the future have implications for reconstructing physical infrastructure and reweaving the torn social fabric of communities devastated by war (Lavi & Solomon, 2005; Solomon & Lavi, 2005). Many researchers have responded to this vision, showing how youth respond in adaptive and pro-social ways when faced with political violence (see Barber, 2009; Betancourt et al., 2010; Cummings et al., 2010b; Keresteš, 2006). Future research should continue to build our understanding of how young people can be peacemakers and peacebuilders amid situations of ethnic strife and intergroup divide.

Acknowledgments

This research was support by NICHD Grant 046933-05 to the last author. We would like to thank the many families in Northern Ireland who have participated in the project. We would also like to express our appreciation to project staff, graduate students, and undergraduate students at the University of Notre Dame and the University of Ulster.

Footnotes

1

Ed was a pioneer in the psychological study of children who are growing up in contexts of political violence. In our collaborative longitudinal study of political violence, his vision inspired many aspects of the project. This proposed integrative statement on the intersection between group identity and emotional insecurity was an important goal for Ed and many of us over our nearly decade-long partnership. In Ed’s honor, we grapple with this issue in print for the first time.

Contributor Information

Christine E. Merrilees, University of Notre Dame

Laura K. Taylor, University of Notre Dame

Marcie C. Goeke-Morey, The Catholic University of America

Peter Shirlow, Queen’s University, Belfast.

E. Mark Cummings, University of Notre Dame.

References

  1. Barber BK, editor. Adolescents and war. New York, NY: Oxford University Press; 2009. [Google Scholar]
  2. Barber BK, McNeely C, Spellings C. Role of political factors in wellbeing and quality of life during long-term constraints and conflict: An initial study. Lancet. 2012 published online Oct 8. http://www.thelancet.com/health-in-the-occupiedpalestinian-territory-2012.
  3. Bar-Tal D. Sociopsychological foundations of intractable conflicts. American Behavioral Scientist. 2007;50:1430–1453. http://dx.doi.org/10.1177/0002764207302462. [Google Scholar]
  4. Betancourt TS, Agnew-Blais J, Gilman SE, Williams DR, Ellis BH. Past horrors, present struggles: The role of stigma in the association between war experiences and psychosocial adjustment among former child soldiers in Sierra Leone. Social Science & Medicine. 2010;70:17–26. doi: 10.1016/j.socscimed.2009.09.038. http://dx.doi.org/10.1016/j.socscimed.2009.09.038. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Betancourt TS, McBain R, Newnham EA, Brennan RT. Trajectories of internalizing problems in war-affected Sierra Leonean youth: Examining conflict and post-conflict factors. Child Development. doi: 10.1111/j.1467-8624.2012.01861.x. in press. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bizman A, Yinon Y. Intergroup and interpersonal threats as determinants of prejudice: The moderating role of in-group identification. Basic and Applied Social Psychology. 2001;23:191–196. http://dx.doi.org/10.1207/153248301750433669. [Google Scholar]
  7. Bowlby J. Attachment and loss: Volume 1. Attachment. New York: Basic Books; 1969. [Google Scholar]
  8. Boxer P, Huesmann RL, Dubow EF, Landau SF, Gvirsman SD, Ginges J. Exposure to violence across the social ecosystem and the development of aggression: A test of ecological theory in the Israel-Palestinian conflict. Child Development. 2013;84:163–177. doi: 10.1111/j.1467-8624.2012.01848.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Branscombe NR, Wann DL. Collective self-esteem consequences of outgroup derogation when a valued social identity is on trial. European Journal of Social Psychology. 1994;24:641–657. http://dx.doi.org/10.1002/ejsp.2420240603. [Google Scholar]
  10. Brown R, Condor S, Mathews A, Wade G, Williams J. Explaining intergroup differentiation in an industrial organization. Journal of Occupational Psychology. 1986;59:279–304. http://dx.doi.org/10.1111/j.2044-8325.1986.tb00230.x. [Google Scholar]
  11. Browning CR, Burrington LA, Leventhal T, Brooks-Gunn J. Neighborhood structural inequality, collective efficacy, and sexual risk behavior among urban youth. Journal of Health and Social Behavior. 2008;49:269–285. doi: 10.1177/002214650804900303. http://dx.doi.org/10.1177/002214650804900303. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Cairns E. Children and political violence. Oxford, UK: Blackwell; 1996. [Google Scholar]
  13. Cairns E, Dawes A. Children: Ethnic and political violence—A commentary. Child Development. 1996;67:129–139. doi: 10.1111/j.1467-8624.1996.tb01715.x. http://dx.doi.org/10.2307/1131691. [DOI] [PubMed] [Google Scholar]
  14. Cairns E, Kenworthy J, Campbell A, Hewstone M. The role of in-group identification, religious group membership and intergroup conflict in moderating in-group and out-group affect. British Journal of Social Psychology. 2006;45:701–716. doi: 10.1348/014466605x69850. http://dx.doi.org/10.1348/014466605X69850. [DOI] [PubMed] [Google Scholar]
  15. Cummings EM, Davies PT. Emotional security as a regulatory process in normal development and the development of psychopathology. Development and Psychopathology. 1996;8:123–139. [Google Scholar]
  16. Cummings EM, Davies PT. Marital conflict and children: An emotional security perspective. New York and London: The Guilford Press; 2010. [Google Scholar]
  17. Cummings EM, Goeke-Morey MC, Schermerhorn AC, Merrilees CE, Cairns E. Children and political violence from a social ecological perspective: Implications for research on children and families in Northern Ireland. Clinical Child and Family Psychology Review. 2009;12:16–38. doi: 10.1007/s10567-009-0041-8. http://dx.doi.org/10.1007/s10567-009-0041-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Cummings EM, Merrilees CM, Schermerhorn AC, Goeke-Morey MC, Shirlow P, Cairns E. Testing a social ecological model for relations between political violence and child adjustment in Northern Ireland. Development and Psychopathology. 2010a;22:405–418. doi: 10.1017/S0954579410000143. http://dx.doi.org/10.1017/S0954579410000143. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Cummings EM, Merrilees CE, Schermerhorn AC, Goeke-Morey MC, Shirlow P, Cairns E. Longitudinal pathways between political violence and child adjustment: The role of emotional security about the community in Northern Ireland. Journal of Abnormal Child Psychology. 2011;39:213–224. doi: 10.1007/s10802-010-9457-3. http://dx.doi.org/10.1007/s10802-010-9457-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Cummings EM, Merrilees CE, Schermerhorn AC, Goeke-Morey MC, Shirlow P, Cairns E. Political violence and child adjustment: Longitudinal tests of sectarian antisocial behavior, family conflict and insecurity as explanatory pathways. Child Development. 2012;83:461–468. doi: 10.1111/j.1467-8624.2011.01720.x. http://dx.doi.org/10.1111/j.1467-8624.2011.01720.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Cummings EM, Merrilees CE, Taylor LK, Shirlow P, Goeke-Morey M, Cairns E. Longitudinal relations between sectarian and non-sectarian community violence and child adjustment in Northern Ireland. Development and Psychopathology. doi: 10.1017/S0954579413000059. in press-a. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Cummings EM, Schermerhorn AC, Merrilees CM, Goeke-Morey MC, Cairns E. Political violence and child adjustment in Northern Ireland: Testing pathways in a social ecological model including single- and two-parent families. Developmental Psychology. 2010b;46:827–841. doi: 10.1037/a0019668. http://dx.doi.org/10.1037/a0019668. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Cummings EM, Taylor LK, Merrilees CE, Goeke-Morey MC, Shirlow P, Cairns E. Relations between political violence and child adjustment: A four-wave test of the role of emotional insecurity about community. Developmental Psychology. doi: 10.1037/a0032309. in press-b. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Davies PT, Cummings EM. Marital conflict and child adjustment: An emotional security hypothesis. Psychological Bulletin. 1994;116:387–411. doi: 10.1037/0033-2909.116.3.387. Extracted for TESP Project (The Tampere Corpus of English for Specific Purposes). doi: http://dx.doi.org/10.1037//0033-2909.116.3.387. [DOI] [PubMed] [Google Scholar]
  25. Dubow EF, Huesmann LR, Boxer P, Landau S, Dvir S, Shikaki K, Ginges J. Exposure to political conflict and violence and posttraumatic stress in Middle East youth: Protective factors. Journal of Clinical Child and Adolescent Psychology. 2012;41:402–416. doi: 10.1080/15374416.2012.684274. http://dx.doi.org/10.1080/15374416.2012.684274. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Fischer P, Haslam SA, Smith L. “If you wrong us, shall we not revenge?” Social identity salience moderates support for retaliation in response to collective threat. Group Dynamics: Theory, Research, and Practice. 2010;14:143–150. http://dx.doi.org/10.1037/a0017970. [Google Scholar]
  27. Goeke-Morey M, Cummings E, Ellis K, Merrilees C, Schermerhorn A, Shirlow P, Cairns E. The differential impact on children of inter- and intra-community violence in Northern Ireland. Journal of Peace Psychology. 2009;15:367–383. doi: 10.1080/10781910903088932. http://dx.doi.org/10.1080/10781910903088932. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Goodnow JJ, Miller PJ, Kessel F, editors. New Directions for Child Development. 67. San Francisco: Jossey-Bass; 1995. Cultural practices as contexts for development. [Google Scholar]
  29. Hammack PL. Identity as burden or benefit? Youth, historical narrative, and the legacy of political conflict. Human Development. 2010;53:173–201. http://dx.doi.org/10.1159/000320045. [Google Scholar]
  30. Hu L, Bentler PM. Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling. 1999;6:1–55. doi: 10.1080/10705519909540118. [DOI] [Google Scholar]
  31. Keresteš G. Children’s aggressive and prosocial behavior in relation to war exposure: Testing the role of perceived parenting and child’s gender. International Journal of Behavioral Development. 2006;30:227–239. http://dx.doi.org/10.1177/0165025406066756. [Google Scholar]
  32. Kronenberg ME, Hansel TC, Brennan AM, Osofsky HJ, Osofsky JD, Lawrason B. Children of Katrina: Lessons learned about postdisaster symptoms and recovery patterns. Child Development. 2010;81:1241–1259. doi: 10.1111/j.1467-8624.2010.01465.x. http://dx.doi.org/10.1111/j.1467-8624.2010.01465.x. [DOI] [PubMed] [Google Scholar]
  33. Lavi T, Solomon Z. Palestinian youth of the intifada: PTSD and future orientation. Journal of the American Academy of Child & Adolescent Psychiatry. 2005;44(11):1176–1183. doi: 10.1097/01.chi.0000177325.47629.4c. http://dx.doi.org/10.1097/01.chi.0000177325.47629.4c. [DOI] [PubMed] [Google Scholar]
  34. McAuleyJWMycock A, Tonge J. Political socialization and cultural learning in the Orange Order in Northern Ireland. Paper presented at the 35th Annual Scientific Meeting of the International Society of Political Psychology; Chicago, Il. 2012. Jul, [Google Scholar]
  35. McEvoy-Levy S. Troublemakers or peacemakers?: Youth and post-accord peace building. Notre Dame Press; Notre Dame, IN: 2006. [Google Scholar]
  36. Merrilees CE, Cairns E, Goeke-Morey MC, Schermerhorn AC, Shirlow P, Cummings EM. Associations between mothers’ experience with the Troubles in Northern Ireland and mothers’ and children’s psychological functioning: The moderating role of social identity. Journal of Community Psychology. 2011;39:60–75. doi: 10.1002/jcop.20417. http://dx.doi.org/10.1002/jcop.20417. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Merrilees CE, Cairns E, Taylor LK, Goeke-Morey MC, Shirlow P, Cummings EM. Social identity and youth aggressive and delinquent behaviors in a context of political violence. Political Psychology. doi: 10.1111/pops.12030. in press-a. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Merrilees CE, Taylor LK, Goeke-Morey M, Shirlow C, Cummings P, Cairns E. The protective role of group identity: Sectarian antisocial behavior and adolescent emotion problems. Child Development. doi: 10.1111/cdev.12125. in press-b. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Muthén LK, Muthén B. Mplus user’s guide (Version 4) Los Angeles, CA: Muthén & Muthén; 2006. [Google Scholar]
  40. Rankin LA, Maggs JL. First-year college student affect and alcohol use: Paradoxical within- and between-person associations. Journal of Youth and Adolescence. 2006;35:925–937. http://dx.doi.org/10.1007/s10964-006-9073-2 http://dx.doi.org/10.1007/s10964-006-9073-2. [Google Scholar]
  41. Shirlow P, Murtagh B. Belfast: Segregation, violence and the city. London, England: Pluto Press; 2006. [Google Scholar]
  42. Singer JD, Willett JB. Survival Analysis. In: Schinka JA, Velicer WF, editors. Handbook of Psychology, Volume 2: Research Methods in Psychology. New York, NY: John Wiley & Sons; 2003. pp. 555–580. [Google Scholar]
  43. Solomon Z, Lavi T. Israeli youth in the second intifada: PTSD and future orientation. Journal of the American Academy of Child & Adolescent Psychiatry. 2005;44:1167–1175. doi: 10.1097/01.chi.0000161650.97643.e1. http://dx.doi.org/10.1097/01.chi.0000161650.97643.e1. [DOI] [PubMed] [Google Scholar]
  44. Struch N, Schwartz SH. Intergroup aggression: Its predictors and distinctness from in-group bias. Journal of Personality and Social Psychology. 1989;56:364–373. doi: 10.1037//0022-3514.56.3.364. http://dx.doi.org/10.1037/0022-3514.56.3.364. [DOI] [PubMed] [Google Scholar]
  45. Tajfel H, Turner J. An integrative theory of intergroup conflict. In: Austin WG, Worchel S, editors. The social psychology of intergroup relations. Monterey, CA: Brooks-Cole; 1979. pp. 94–109. [Google Scholar]
  46. Tajfel H, Turner JC. The social identity theory of intergroup behavior. In: Worchel S, Austin WG, editors. The psychology of intergroup relations. Chicago: Nelson Hall; 1986. pp. 7–24. [Google Scholar]
  47. Taylor LK, Merrilees CE, Campbell A, Shirlow P, Cairns E, Goeke-Morey MC, Schermerhorn AC, Cummings EM. Sectarian and nonsectarian violence: Mothers’ appraisals of political conflict in Northern Ireland. Peace and Conflict: Journal of Peace Psychology. 2011;17:343–366. doi: 10.1080/10781919.2011.610199. http://dx.doi.org/10.1080/10781919.2011.610199. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Taylor LK, Merrilees CE, Cairns E, Goeke-Morey MC, Shirlow P, Cummings EM. Risk and resilience: The moderating role of social coping for maternal mental health in a setting of political conflict. International Journal of Psychology. doi: 10.1080/00207594.2012.658055. in press. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Tropp LR, Mallett RK. Charting new pathways to positive intergroup relations. In: Tropp LR, Mallett RK, editors. Moving beyond prejudice reduction: Pathways to positive intergroup relations. Washington DC: American Psychological Association; 2011. pp. 3–17. [Google Scholar]
  50. Turner JC, Hogg MA, Oakes PJ, Reicher SD, Wetherell MS. Rediscovering the social group: A self-categorization theory. Cambridge, MA, US: Basil Blackwell, Cambridge, MA; 1987. [Google Scholar]
  51. Wu W, West SG, Taylor AB. Evaluating model fit for growth curve models: Integration of fit indices from SEM and MLM frameworks. Psychological Methods. 2009;14:183–201. doi: 10.1037/a0015858. http://dx.doi.org/10.1037/a0015858. [DOI] [PubMed] [Google Scholar]

RESOURCES