The Efficacy of a Holding Community Program— Promoting Social Reflection at School

The authors developed a Holding Community Program to achieve the following objectives: (a) to increase the perspective-taking capacity of adolescents; (b) to promote interpersonal and intergroup harmony; (c) to empower school students to be more (pro)active in their communities and in public life. Apart from the intervention itself, the study comprised a pre-test and a post-test and involved a total of 240 Hungarian high school students (159 female, 66.3%). The students were aged 14–18 (Mage = 15.33; SDage = 0.88). They were recruited from four high schools. Control groups (N = 122) were chosen from the same institution and graded as experimental classes (N = 118, 7 classes). Both immediate and long-term effects of the intervention (4–6 months after the intervention) were explored. Quantitative analysis of the data indicated that the two-day intervention program had significantly increased the students’ perspective-taking capacity (short-term: F(1, 238) = 6.03, p < 0.05, long-term: n.s.) and efficacy beliefs (short-term: F(1, 238) = 3.83, p = 0.052, long-term: F(1, 238) = 3.38, p < 0.05). After the training, students were more willing to participate in collective actions (short-term: F(1, 238) = 7.32, p < 0.01, long-term: F(1, 238) = 3.83, p < 0.05). These results seem quite promising but the outcome was not significant regarding its effect on prejudice. Changing Societies & Personalities, 2021, Vol. 5, No. 3, pp. 422–437 423


Introduction
Institutionalized education is a constantly evolving social construct, which is changing alongside the understanding of knowledge construction, essential competencies, and the role of teachers and learners. Modernization of educational systems is a key issue in international policies, which in the last decade have highlighted the importance of learning in ensuring social mobility, equity, social cohesion and active citizenship (e.g., The Strategic Framework for European Cooperation in Education and Training: ET 2020). To achieve these objectives, it is necessary to enhance social and emotional skills because the development of these skills is crucial for promoting social inclusion and cohesion but also because of their importance for knowledge acquisition (Corcoran et al., 2018;Elias et al., 1997;Trilling & Fadel, 2009). Nonetheless, in general, national education systems still largely focus on fostering the students' academic performance rather than their social-emotional skills (Paksi, 2019;Sugimoto & Carter, 2015), the Hungarian public education system being no exception (Zsolnai et al., 2015). Furthermore, there is a perceived lack of preventive measures, e.g., strategies for prevention of abuse, violence and similar behaviour (Paksi, 2019). Hungarian society still has a long way to go in terms of social inclusion (Kende et al., 2018;Örkény & Váradi, 2010) and politically motivated intergroup violence is an existing problem (Faragó et al., 2019). From the psychological perspective, the implementation of widespread and evidence-based prevention programs remains an important educational goal for public schools.

Interpersonal Competencies: Essential, but Not Enough
Social and interpersonal competencies are seen as vital for academic progress of students (Webster-Stratton & Reid, 2004;Zsolnai, 2002) and also as essential for promoting diversity and inclusivity; for example, perspective-taking contributes to prosocial behaviour (Hodges et al., 2011), increases the propensity to help others and reduces outgroup bias (Batson et al., 2002;Vescio et al., 2003). Helping an individual to acknowledge that there is another viewpoint strengthens the self-other relationship by creating an overlap between the self and other cognitive representations: when people describe others, they tend to attribute a greater number of characteristics (especially positive) to the target of perspective-taking (Davis et al., 1996) and KEYWORDS perspective-taking, prejudice, active citizenship, program evaluation, education perceive themselves as being closer to those whose perspectives they are taking (Myers & Hodges, 2012).
Although an attempt to discover shared reality may motivate perspective-taking, such motivation in itself does not guarantee its successful outcome (Hodges et al., 2018). In some cases, perspective-taking is used to reinforce the differences between the self and the other with the other being perceived as a threat. In these cases, perspective-taking increases avoidance and social distancing (Bigazzi et al., 2019;Hodges et al., 2018;Vorauer & Sasaki, 2009). It creates a barrier to one's understanding of the mental state of the other and impedes cooperation.
To prevent this rebound effect, it is important to ensure that competence development should meet certain requirements. To reduce negative biases, educators have to put effort into creating and maintaining a safe space (Twemlow et al., 2001), strengthen students' positive self-views (Todd & Burgmer, 2013), promote inclusive, complex identification-the process of associating oneself closely with other ingroup or outgroup individuals and their characteristics or views- (Brewer, 2000) and help students find common values and goals to overcome the differences between them (Allport, 1958;Aronson & Patnoe, 1997).

Identity in Play: Reducing Negative Interpersonal and Intergroup Attitude
In research literature, prejudice is usually defined as negative evaluations, beliefs, or feelings directed at people because of their perceived group membership. Several studies that compared populations of diverse background have found that people begin to develop prejudices at the age of 4 or 5, while adolescence is a critical period in which intergroup attitudes are formed and solidified (Aboud et al., 2012).
To explore, challenge, and overcome biases, various methods were implemented in recent decades. The Common Ingroup Identity Model (Gaertner & Dovidio, 2000) proposes that prejudice can be overcome with the help of recategorization. Members of different groups can share a common ingroup membership focusing on a more inclusive superordinate group representation. There is research evidence that shows that perceived shared identity leads to lower levels of intergroup threat (Riek et al., 2010) and to more positive evaluations of outgroup members (Guerra et al., 2013). The approach of Dual (Gaertner et al., 2016) and Multiple Social Identity (Brewer, 2000) takes into consideration the fact that apart from one common identity representation, separate/initial group identities can also be maintained within the context of a superordinate category. Some findings suggest that groups in a vulnerable position (e.g., minority groups) prefer a dual identity representation, where identification with the majority group does not require denial and devaluation of the minority position. Thus, people become more willing to address injustices without having to face a perceived identity threat. On the other hand, those in the position of power prefer a one-group representation because they are suspicious of any collective action that challenges the status quo or the boundaries of the majority group (Gaertner et al., 2016). The Intergroup Contact Theory states that under optimal conditions (equal status, common goals, no intergroup competition and authority sanction), increased contact with outgroup individuals could lead to better mutual understanding, more positive intergroup attitudes, and willingness to engage in contact Pettigrew, 1998, Pettigrew et al., 2011. Even an indirect, vicarious contact (e.g., by observing members of outgroups through the mass media or having a friend who knows someone from an outgroup) can be beneficial (Paolini et al., 2004). Thus, even indirect exposures can provide counterstereotypical information that may change the pre-existent intergroup representations.
Within educational contexts, the Intergroup Dialogue is a similar pedagogical approach that seeks to establish a common understanding among the people whose social identities and life experiences differ from one another (Ford, 2018). In line with our approach, the goal of the Intergroup Dialogue is to create a space that would promote the exchange of different perspectives and seeking mutual understanding. It also addresses the issues of collective identity and promotes discussions on how to overcome intergroup conflict, which keeps individuals alienated from others, through non-violent and collaborative negotiations. Among young adults, participation in the Intergroup Dialogue has proven effective in fostering cross-cultural communication and creating awareness of social justice issues (Ford, 2018;Hammack & Pilecki, 2015). One question remains open: How can be prejudice reduction effective when the target of prejudice is not part of the dialogue? It is important to reduce uncertainty by getting to know outgroup members and thus improve the intergroup relationship, but in educational settings sometimes a specific constellation of diversity is not available or is less accessible. Not to mention how can psychological intervention make a difference for empowering adolescents living in contexts where discrimination is socially acceptable or even supported?

Foster Holding Communities
In their school years, children prioritize peer opinions and tend to identify with certain social groups. Peer group norms and attitudes become increasingly relevant and solidify in adolescence (Nesdale, 2007;Raabe & Beelmann, 2011;Váradi, 2014). Programs to prevent outgroup discrimination beyond any doubt should take into account the potential power of these increasingly important peer relationships. Promoting positive relationships and reciprocity between students is the key to cultivating an optimal learning environment in schools and promoting reflectivity, which can provide a fertile ground for social awareness.
The concept of a holding environment was first introduced in developmental and psychoanalytic research literature (Winnicott, 1965): according to this concept, optimal development depends on the mother's and attachment figure's ability to provide the child with a safe enough space to move gradually toward autonomy. Failures and acts of reparation provide opportunities for the child to experience the unknown and different perspective, giving them space to practice, change, and develop. In a good relationship, acceptance and mentalization (Fonagy et al., 1991)-the ability to reflect upon and to understand one's state of mind-leads to self-acceptance, self-reflection, and capacity for self-regulation. This dynamic over the last decades has been broadened from dyadic relationships to the group level: peer consultation groups (Minkle et al., 2008), work (Kahn, 2001), school (Hyman, 2012;Twemlow et al., 2001), and communities (Bigazzi et al., 2020;Jovchelovitch & Concha, 2013). In these settings, each member of the group shares the responsibility to support other group members, to create a safe space for conducting intrapersonal and interpersonal work, and to contribute to socially responsible efforts.
If children feel criticized or conditionally accepted in their school community, they may internalize the attitude that contributes to limited self-worth, feelings of isolation, while the feelings of acceptance and belonging provide the support necessary for people to thrive. An environment of acceptance minimizes the threat-induced defensiveness and enables children to work with self-other differences and create an inclusive space where new social perspectives and cognitive alternatives may emerge.
Youth empowerment and psychoeducational work should involve the entire school community (Twemlow et al., 2001). Children and adults are active agents whose reactions alter the dynamics of power by supporting or hindering others. If we enable students to be reflective, they will be able to manage social interactions on their own and actively participate in social life not through the absorption of readymade knowledge or the compliance with rules, but through dialogue and construction. On the other hand, teachers should also realize that they are part of the community and their (lack of) actions will have an impact.
Human interactions do not happen in a vacuum. From the socio-psychological perspective, youth empowerment should involve the whole community. A planned intervention should take into account not only its object, but also the social context that will be influenced indirectly by the intervention. Disadvantaged groups are often the target of interventions, while these interventions do not work with whole community that these groups are part of. When empowered members of the disadvantaged groups reposition themselves, they also reframe their relations with others, with the members of the whole community. If these others are not involved and strengthened to accept the change, the intervention loses some of its efficacy. The overlooked inequalities and power imbalance cause conflicts where neither the members of the majority nor the members of minority groups can engage in a safe dialogue. In this process, the indirectly involved actors should be aware of their superior power positions and make constant efforts to deconstruct them. Deconstruction is a prerequisite for changing the dominant forms of communication, creating space-an incubator-where new social realities and cognitive alternatives may emerge. In this space, minoritized or disadvantaged groups can elaborate their own perspectives, test the validity of possibilities, and later construct and disseminate their own versions of reality. (Bigazzi et al., 2020, p. 131) To enhance prosocial attitudes and active citizenship, it is important to strengthen social-emotional competencies and critical thinking skills of community members.

Intervention
The Holding Community Program (HCP) that we have designed is based on the methodology of conflict resolution and on the existing research evidence on complex social identity, intergroup dialogue, socio-emotional competence, prejudice reduction and human rights education.
The HCP aims to facilitate the dialogue between the students to promote critical thinking and to exploit the power of peer influence thereby stimulating the coconstruction of knowledge. The HCP encompasses an intensive two-day training and utilizes non-formal education methodology (Hamadache, 1991;Latchem, 2014) for experiential learning and knowledge co-construction. To ensure active learning, the ERR (evocation-realization-reflection) teaching framework was also used (Bárdossy et al., 2002). In line with the ERR framework, within each uninterrupted intervention period, three phases of the learning process were realized: the evocation stage (students are encouraged to think about what they already know); realization of the meaning (students are expected to come into contact with the new information); and reflection stage (students express their ideas and expand their own understanding). Specific techniques such as kinetic icebreakers, modelling, role playing, small group cooperation, sharing of personal narratives and reinforcement of positive interactions were used in order to stimulate communication.
The first day of the program is focused on the relationship between the self and the other. After the mutual trust between the participants is established, the training process concentrates on the following six objectives: first, to raise the awareness of the complex nature of (social) identity; second, to strengthen the feeling of community by paying attention to similarities and differences; third, to practice recognition of each other's emotions and mental states, to practice perspective-taking within the comfort zone; fourth, to practice mentalization and assertive communication in emotionally saturated, conflictual situations; fifth, to explore the psychological consequences of inclusion and exclusion and, finally, to discuss the values that are prevalent in society.
The second day takes further the results that have been achieved in the first day. The relationship between the self and other is expanded by adding the intergroup context and horizon of action to the dialogue. The objectives of the second day are as follows: first, to create awareness and critical understanding of the dynamics of power, socialization, and social inequalities; second, to experience the impact of stereotypes by taking the perspective of the privileged and disadvantaged; third, to explore the processes leading to discrimination and hate crimes; fourth, to strengthen the active bystander attitude and prosocial behaviour; fifth, to enhance self-efficacy in handling sensitive social issues and, finally, to discover cognitive alternatives for collective action.

Research Objective
This study aims to evaluate the results of the Holding Community Program. We expected that the adolescents who took part in this experiment would show better results in perspective-taking (primary outcomes). Furthermore, we assume that the improvements in social-emotional competencies would lead to the improvements in prosocial behaviour and help the participants overcome their prejudicial attitudes (secondary outcomes).

Procedure
The study used a non-randomized pre-test in combination with an immediate posttest and a delayed post-test. The program in the experimental classes was launched two to three weeks after the pre-test. The first post-test was administered immediately after the intervention, while the follow-up post-test was administered 4-6 months after the first post-test. The answers from the control classes were collected in the same periods as from the experimental classes. Classes of students were assigned through convenience sampling to the target group and control groups, and comparative analysis was conducted to ensure that there was no preliminary difference. The interventions were carried out with complete classes by two or three psychologists and consisted of two full day sessions per class. In total, five instructors participated in the research. The control group followed their regular curriculum. To ensure the ethical treatment of the participants, the experiment was approved by the local ethical board, the consent of both students and parents was obtained prior to the experiment.

Participants
In the pre-test and post-test, a total of 240 students participated (159 girls, 67%; Age: 14-18, M age = 15.33; SD age = 0.876). The four schools where the experiment was conducted were situated in Pécs, Hungary. Although the size of the classes sometimes varied, from each school an almost equal number of experimental and control classes took part and the age groups were identical. The effects of the intervention were further investigated with the help of a four to six-month follow-up test in which 125 students participated (92 girls, 74.6%; Age: 14-17, M age = 15.25; SD age = 0.922). Since the follow-up test coincided with the introduction of the distance learning mode due to the COVID-19 pandemic (the spring of 2020), many students declined to invest their time in the delayed post-test (the attrition rate was 48%). Comparative analysis was conducted to ensure that there was no difference between the dropouts and remaining participants in the relevant variables.

Instruments
First, the students were asked to provide their demographic data: age, gender, grade, and place of residence. After the demographic questions, all the other scales were measured on a 6-point Likert-type scale (1 = Strongly disagree/not at all, 6 = Strongly agree/Certainly). Since we planned to repeat the questionnaire survey, it was important to keep the questionnaire package short to ensure that students can pay sufficient attention.
To assess perspective-taking, we used the Interpersonal Reactivity Index (IRI, see Davis, 1983; Hungarian version in Kulcsár, 2002, pp. 411-427). The IRI is a self-report measure of individual differences in empathy with four subscales (personal distress, fantasy, empathic concern and perspective-taking). In our study, we administered the perspective-taking subscale (7 items), which measure the ability and tendency to look at the world from somebody else's point of view, e.g., When I'm upset at someone, I usually try to "put myself in his shoes" for a while. Cronbach's alphas at the pre-intervention, post-intervention and follow-up stages were .697, .765 and .747, respectively.
To assess Identity Threat, we used relevant scales of the Multiple Threat and Prejudice Questionnaire (MTPQ, see Bigazzi et al., 2019), which captures the subjective sense of danger and fear related to different identity aspects. Bio-National Identity in this questionnaire is based on the exclusive perception of belonging to a certain nation, rooted in one's biological heritage. In this case, the threat is seen as a genetic contamination of target groups (e.g., I think there is a real danger that white people become a minority in our homeland), while Gender Identity is worded around the traditional representation of gender roles (e.g., I would feel upset if I were considered a homosexual). We adjusted this instrument to make it more suitable for our age group, but the internal validity for these scales proved to be satisfying through time: .814 and .784;Gender .804,.846 and .854.
We adapted the Social Distance Scale of Bogardus (1933) to capture prejudice (this instrument was applied in other studies-see, for example, Faragó & Kende, 2017;Orosz et al., 2016). We measured the degree to which respondents would be willing to accept a member of each outgroup (e.g., Roma, migrant, homosexual) as a member of their class (1-1 item). Higher scores indicate higher levels of prejudice in this instrument. In selecting outgroups, we sought to cover the most relevant minorities in the Hungarian context.
We aimed to shed light on adolescents' motivation to participate in social actions, in other words, we were interested in how adolescents perceive their ability to participate in superordinate group matters and how willing they are to participate in collective action to improve intergroup relations and reduce social inequalities. Efficacy beliefs, which will be understood here as one's feeling that s/he is able to contribute to societal life, were measured with the help of the two items designed for this study: To what extent do you feel that you can take an active part in what happens to the Hungarians? and To what extent do you feel that you can take an active part in what's happening to people around the world? The alphas were .700, .729 and .784. Collective action was measured with the help of an item concerning the respondents' willingness to participate in collective action to increase social justice, therefore, it captures their intention to act rather than previous attendance. Students were asked to evaluate the following statements: I would like to organize programs to help the disadvantaged groups at school; I would like to go to a school community service for an organization that represents the interests of a disadvantaged group; and I would love to participate in a program aimed at reducing prejudice. The internal validity for this scale was acceptable through time: .799, .886 and .902.

Results
The statistical analyses were performed using IBM SPSS Statistics 25. Baseline equivalence was assessed to ensure that the experimental and control groups were not statistically different for the measured variables. For this purpose, the Independent Samples t-Tests were performed to compare the pre-test scores. The results showed the desirable correspondence (means and standard deviations are provided in Table 1).

Immediate Effect
Paired t-tests were first conducted to find possible differences between the pre-test and post-test scores in the intervention and control groups. The results we obtained seemed quite promising as we found significant differences between the groups (Table 1). To test our hypothesis that the intervention would change the outcome variables, we then performed repeated measures ANOVA with INTERVENTION (the person participated in the Holding Community Program or not) as a between-subjects factor, and TIME (pre-intervention and post-intervention) as a within-subjects factor. The interaction effects are presented in Table 1.
The HCP was an effective tool to develop perspective-taking: after the training, students were more willing to participate in collective actions. Compared to those in the control group, participants of the intervention group tended to accept members of the Roma community and believed themselves to be able to take a more active part in social life. Members of the intervention group had lower scores on the explicit measure of threat and prejudice toward migrants and therefore showed a desirable change in their attitudes although it reached the conventional levels of significance only in the t-test. The repeated measures ANOVA revealed no significant interaction effect (see Table 1). The study's findings confirm the short-term effectiveness of the HCP, which appears to increase the perspective-taking capacity of adolescents and motivate them to be an active part of their environment.

Follow-up
Due to the high attrition rates, the Independent-Samples t-tests were carried out to compare the dropouts and remaining participants. The results show that there was no significant difference in terms of age (t(238) = -1.579, P = 0.116) and most of the observed variables at the baseline (see Table 2). However, more boys than girls tended to drop out (χ 2 = 6.30, P = 0.12) and those who did not complete the follow-up questionnaire showed a perceived greater threat to their gender identity (M remaining = 3.85, SD remaining = 1.62, M dropouts = 4.35, SD dropouts = 1.51) in the pre-test. The combination of these results is consistent with the previous findings of the studies using the MTPQ, where men were found to be more susceptible to the gender identity threat than women (Géczy & Varga, 2019).

Table 2
Means and Standard Deviations in Pre-, Post-and Follow-up-test  To test whether these effects would persist over time, repeated measures ANOVAs were conducted again, with three time points. While the change in collective action remained significant, the effect of the improvements in the perspective-taking capacity faded over months. Interestingly, the interaction effect on perceived efficacy became greater (see Table 2).

Discussion
The goal of this study was to measure the effectiveness of the HCP aimed at the development of perspective-taking, reducing prejudice and promoting active participation. Our findings suggest that this intervention can enhance socialemotional skills and civic engagement. Moreover, it encouraged student reflection about matters of social diversity and inclusivity. The intervention was less effective in dealing with prejudice. This result can be explained by the fact that the prejudice against minority groups has been strongly institutionalized in Hungary in the last decade. Declaring a positive attitude towards institutionally discriminated minority groups means not only reframing one's relationship with the minority members but also taking a stand against the authorities and political institutions normalizing such prejudice and discrimination. We might suggest that more targeted programs are needed to deal with the prejudice against specific minority groups.
The result shows that a change in perspective-taking was detectable but it has not persisted in the long-term, underlining the greater importance of systematic prevention and long-term intervention planning. Likewise, it deserves our attention because most research involving psychological field experiments merely focuses on the immediate effect (Paluck & Green, 2009) rather than long-term effectiveness.
When we asked the students how they benefitted from the program, they reported that the program helped them acquire new knowledge about the member of different groups and bond with them. They also said that they learned "new things about how to deal with different challenges in life" and realized their role in eliminating external stereotypes.
During the training sessions, students often demonstrated a fairly limited experience of social life. Research shows that engaging in collective action beyond educational settings often remains suspended in the absence of cognitive alternatives (Tajfel, 1978). We suggest that, in order to stimulate critical social awareness, educators should create more opportunities for exploring together with students' collective actions and their possible consequences.
Although we think that these results contribute to the psycho-educational field, some limitations of the study should be noted. Although the instruments we used have adequate psychometric properties, self-report measures can be subject to bias. In future studies, it would also make sense to enrich the self-administered questionnaires with more qualitative data and behavioural observations, which may provide us with some deeper insights. Further research may also explore the effects of longer interventions for the achievement better results sustainable over time.
Moreover, further research should also investigate if the effects of this intervention differ for those in the majority or minority positions. Another question worth analysing is how the impact changes for those who belong to minority groups in a broader social context but belong to the majority in their class. As empowerment cannot happen in a vacuum (Bigazzi et al., 2020), these conditions presumably alter the identification and development processes.
In conclusion, we found that the HCP is suitable for fostering positive interpersonal competencies. This tool can be used to encourage young people to reflect more on the world around them, in particular on the issues of social inequality. Educators should invest more energy into creating an inclusive space where students will be able to explore perspectives without the fear of being questioned.