Topics: In Flanders, significant gaps in the sense of school belonging and student performance exist between majority monolingual Dutch-speaking and minority multilingual pupils speaking Dutch as well as a non-Dutch heritage language (HL)^{[1, 2]}. In this study, we investigate whether these differences in school outcomes may be a result of differences in children’s emotional competence. Children’s school outcomes critically depend on their emotional competence, i.e., the capacity to experience, express, regulate emotions, and understand their own emotions as well as those of others^[3]. Children need emotional competencies to enable learning in an inherently social context such as school. For example, children who are better at identifying emotions and, hence, regulating these emotions are more likely to establish positive and supportive relationships with teachers and peers^[3] which in turn may affect their sense of school belonging. Moreover, emotionally competent children generally feel more at ease at school, even in situations of stress. Consequently, they have more cognitive capacity to focus on learning than peers who are emotionally insecure, which in turn positively impacts their performance ^[4].

                   Prior empirical research has repeatedly shown how among monolingual children, language competence is associated with emotional competence^[5-7]. For example, the more emotion words children know, the better they recognize others’ facial emotions ^[8]. Similarly, children’s verbal skills are positively associated with emotional awareness of self and others ^[9]. However, to date, it remains unclear whether being able to speak and comprehend multiplelanguages goes hand in hand with increased emotional competence and subsequent school outcomes. Moreover, if the latter is the case, it remains unclear why multilingual minority pupils often have less positive school outcomes.

                   Although multilingual children are likely to have more emotion concepts due to their exposure to emotions in at least two languages (and corresponding emotion cultures), whether or not this multilingual advantage materializes may critically depend on children’s language proficiency profile and the language policy of the school. Multilinguals rarely have equivalent proficiencies in their languages due to differences in language use across social contexts (cfr. complementarity principle^[10]). If emotions are rarely discussed in Dutch (e.g. because they are less discussed in school as compared to the family), children’s language skills in Dutch may not be substantive enough to induce a multilingual advantage in emotional competence. In other words, children’s language proficiency profile may be a critical factor to take into account when trying to explain differences in school outcomes between monolinguals and multilingual children. Moreover, if schools adopt a language assimilation policy, multilingual children may not be provided with the opportunity to connect their HL skills with their Dutch language skills which may hamper multilingual children’s emotional development. Hence, the relationship between language proficiency profile and emotional competence may be moderated by school’s language policy.

Research questions: How are school outcomes related to children’s language proficiency profiles (RQ1)? Does emotional competence mediate the relationship between different types of language proficiency profiles and school outcomes (RQ2?) And is the relationship between children’s language proficiency profiles and emotional competence conditional upon the role of the school’s language policy (RQ3)?

We use survey data from primary school children (aged 10 to 12) from the Ethnic-Cultural Diversity in Schools (ECDIS) project. 3073 pupils were surveyed via a paper-and-pencil questionnaire during school hours across 59 Flemish schools. 1894 (63%) of the pupils in the study are multilingual. Emotional competence was measured by the “Differentiating Emotions” subscale of the Emotion Awareness Questionnaire, consisting of items such as ‘I am often confused or puzzled about what I am feeling’ and ‘I never know exactly what kind of feeling I am having’[11]. SSB was assessed by the Psychological Sense of School Membership Scale[12]. A standardized math test[13] was used as an indicator of student performance. We focused on math achievement because math tests are less linguistically biased than tests for other subjects. We constructed a variable “Language Proficiency Profile” with four categories. Monolingual children formed a first language proficiency profile. Then, we performed a Two-Step cluster analysis procedure in SPSS to classify multilingual children into language proficiency groups based on four language proficiency measures (rating on 1 to 5 on speaking and understanding Dutch and the HL). This analysis resulted into three multilingual profiles: (1) fluent multilingual in both languages, (2) fluent in Dutch but low proficiency in the heritage language, and (3) low Dutch proficiency but moderate proficient in the heritage language. School’s Language Policy was measured by a set of items referring to three different diversity models: (1) assimilation, (2) colorblindness and (3) pluralism[14]. SAS was used to conduct multilevel linear regressions as a first step. In a second step, MPlus was used to perform a multilevel structural equation model. Control variables are gender, grade, migration generation and SES.At this moment in time, we ran the analyses for Sense of School of Belonging. In the coming months, we will run the same analyses for Math Performance.

Preliminary Findings for Sense of School Belonging: Only multilingual children with low proficiency in Dutch and moderate proficiency in the heritage language score significantly lower on sense of school belonging than monolingual children. Other multilingual groups did not significantly differ from the monolingual reference group. Also, the ability to differentiate emotions is significantly positively related to sense of school belonging but emotion differentiation only partially mediates the relationship between language proficiency profiles and sense of school belonging scores. Multilevel analyses will be conducted in the coming months to investigate the role of language policy in explaining why a multilingual advantage does not materialize. Analyses at the individual level, without controlling for data clustering in schools, do not provide evidence of moderated mediations.

1.Van Der Wildt, A., P. Van Avermaet, and M. Van Houtte, Multilingual school population: ensuring school belonging by tolerating multilingualism. International Journal of Bilingual Education and Bilingualism, 2015. 20(7): p. 868-882. 2.Celeste, L., et al., Can School Diversity Policies Reduce Belonging and Achievement Gaps Between Minority and Majority Youth? Multiculturalism, Colorblindness, and Assimilationism Assessed. Pers Soc Psychol Bull, 2019. 45(11): p. 1603-1618. 3.Denham, S.A., Emotional Competence During Childhood and Adolescence, in Handbook of Emotional Development, V. LoBue, K. Pérez-Edgar, and K.A. Buss, Editors. 2019, Springer: Cham. p. 493-541. 4.Oberle, E. and K.A. Schonert-Reichl, Social and Emotional Learning: Recent Research and Practical Strategies for Promoting Children’s Social and Emotional Competence in Schools, in Handbook of Social Behavior and Skills in Children, J.L. Matson, Editor. 2017, Springer: Cham. p. 175-197. 5.Salmon, K., et al., The Role of Language Skill in Child Psychopathology: Implications for Intervention in the Early Years. Clin Child Fam Psychol Rev, 2016. 19(4): p. 352-367. 6.Beck, L., et al., Relationship between language competence and emotional competence in middle childhood. Emotion 2012. 12(3): p. 503-514. 7.Cole, P.M., L.M. Armstrong, and C.K. Pemberton, The role of language in the development of emotion regulation, in Child development at the intersection of emotion and cognition, S.D. Calkins and M.A. Bell, Editors. 2010, American Psychological Association: Washington. p. 59-77. 8.Streubel, B., et al., Emotion-specific vocabulary and its contribution to emotion understanding in 4- to 9-year-old children. J Exp Child Psychol, 2020. 193: p. 104790. 9.Mancini, G., et al., Predictors of emotional awareness during childhood. Health, 2013. 05(03): p. 375-380. 10.Grosjean, F., Bilingual. Life and Reality. 2010, Cambridge: Harvard University Press. 11.Rieffe, C., et al., Psychometric properties of the Emotion Awareness Questionnaire for children. Personality and Individual Differences, 2007. 43(1): p. 95-105. 12.Goodenow, C., The psychological sense of school membership among adolescents: Scale development and educational correlates. Psychology in the Schools, 1993. 30(1): p. 79-90. 13.Dudal, P. and G. Deloof, Vrij centrum voor leerlingenbegeleiding. Leerlingenvolgsysteem. Wiskunde: Toetsen 5 – Basisboek. 2004, Antwerpen: Garant. 14.Konings, R., O. Agirdag, and J. De Leersnyder, Development and Validation of Domain Scpecific Diversity Model Scales among Pupils and Teachers: A Multilevel Approach. Social Psychology of Education, accepted.