Questions of educational equity and inclusive school development are increasingly framed through students’ subjective experiences in schooling. Against the backdrop of EU-level initiatives on whole-school well-being and mental health (e.g., UNESCO, 2025; OECD, 2023), school well-being (SWB) has become a key indicator for monitoring the quality of inclusive education (Kullmann et al., 2023). SWB, conceptualized as students’ cognitive and emotional evaluations of their school experiences, is widely recognized as vital for learning, socio-emotional development, and participation (Hascher et al., 2018). Schools are therefore not only institutional learning environments but also significant developmental contexts in which students’ emotional, social, and personal thriving must be actively supported (Schnell et al., 2025).

Within inclusive education, the question of which students experience lower or higher levels of SWB has gained particular relevance. Students with special educational needs (SEN) frequently face more challenging learning and developmental and are therefore assumed to be especially vulnerable. Yet empirical findings remain inconsistent: while some studies document lower SWB for students with SEN (e.g., Alnahdi & Schwab, 2021; Goldan et al., 2022), others report no meaningful differences (e.g., Kröske, 2020; Zdoupas et al., 2025). These inconsistencies partly stem from the fact that SEN is an administrative umbrella category encompassing highly heterogeneous support needs, limiting its usefulness as an explanatory indicator. In contrast, a more consistent body of research – both within and beyond Europe – identifies psychosocial problems as robust predictors of diminished SWB. Internalizing symptoms (e.g., anxiety, depression) and externalizing symptoms (e.g., aggression, conduct problems, hyperactivity) are systematically linked to lower emotional, social, and competence-related school experiences (Kim & Choe, 2022; Tobia et al., 2019). Epidemiological evidence across European countries suggests that psychosocial problems are widespread (pooled prevalence ≈ 15.5%), with notably higher rates for internalizing (~7–8%) compared to externalizing problems (~2–3%; Sacco et al., 2024). These difficulties often emerge early and frequently persist into adolescence, making them potent determinants of daily school functioning. At the same time, the research landscape on SWB remains conceptually heterogeneous. Existing studies employ global indicators, multidimensional frameworks, or domain-specific measures, complicating comparisons and theoretical integration (Hossain et al., 2023). The multidimensionality of SWB – spanning affective, social, and competence-related domains – has been repeatedly theorized but only unevenly tested empirically, especially within inclusive educational contexts.

The present study addresses these conceptual and empirical gaps by examining SWB in inclusive secondary schools through a multidimensional lens and by contrasting the predictive value of SEN with that of psychosocial symptoms. Specifically, the study investigates three core questions: 

1. What is the optimal dimensional structure of SWB in an inclusive secondary school context? We hypothesize that a multidimensional model will provide superior fit to the data compared to a unidimensional representation (H₁). 
2. Does the administrative category of SEN explain unique variance in SWB beyond that accounted for by students’ current psychosocial symptom profiles? We expect that SEN status will not contribute significant additional explanatory power (H₂). 
3. How do specific internalizing and externalizing symptoms differentially relate to distinct domains of SWB? Based on established vulnerability models, we hypothesize that internalizing symptoms will show stronger negative associations with affective and social SWB, while externalizing symptoms will primarily relate to reductions in social and competence-related SWB (H₃).

We surveyed N = 976 students (Grades 8-10) from seven inclusive secondary schools in North Rhine-Westphalia, Germany. The sample was distributed nearly evenly across grades 8 (39 %), 9 (31 %), and 10 (30 %). Ten percent of the students had diagnosed SEN, most commonly in the areas of learning disabilities (49 %), emotional and behavioral disorders (21 %), as well as speech and language impairments (17 %). SWB was assessed using the Questionnaire for Assessing School Well-Being in Inclusive Classes (Kullmann et al., 2023). The instrument comprises 24 items across six subscales: Attitudes Towards School (ATS; e.g., “I like going to school”), Affinity to Class (ATC; e.g., “I enjoy being with my classmates”), Academic Self-esteem (ASE; e.g., “I am satisfied with my academic achievements”), Concerns in School (CIS; e.g., “In the past weeks, I have been worried about my grades”), Social Problems in School (SPS; e.g., “In the past weeks, I had problems with specific classmates”), and Somatic Complaints in School (SCS; e.g., “In the past weeks, I had stomach aches because of school”). All items are rated on a six-point scale ranging from 0 (“does not apply at all”) to 5 (“fully applies”). Psychosocial problems were assessed using the self-report version of the Strengths and Difficulties Questionnaire (SDQ; Goodman, 2005). Four symptom domains were considered, each representing a distinct area of psychosocial problems. Emotional Symptoms capture core internalizing problems such as anxiety and depressive symptoms. Peer Problems reflect internalizing-related social withdrawal and interpersonal insecurity. Conduct Problems represent externalizing difficulties including aggressive and rule-breaking behavior. Hyperactivity/Inattention captures externalizing symptoms of impulsivity and concentration. All items are rated on a three-point scale ranging from 0 (“does not apply at all”) to 2 (“fully applies”). In addition, Prosocial Behavior was included as a control scale. Analyses proceeded in two stages. First, we used Confirmatory Factor Analysis (CFA) in R (lavaan package) to compare four competing measurement models of SWB, estimating model fit with robust maximum likelihood (MLR). Second, we conducted hierarchical multilevel regression analyses (lme4 package) with random intercepts for class membership. For each of the six SWB dimensions, we compared a null model, a model with only SEN status, a model with only the four psychosocial symptom scores, and a full combined model. Likelihood ratio tests and changes in explained variance (R²) were used to evaluate the incremental contributions of predictors.

CFA results strongly supported a multidimensional structure. The six-correlated-factor model showed excellent fit (χ^2 (215)=720.42,CFI=.962,TLI=.955,RMSEA=.056,SRMR=.051) and was significantly superior to a unidimensional model (∆χ^2 (15)=1984,p<.001), confirming H₁. Hierarchical regressions revealed that psychosocial symptoms alone explained substantial variance across all SWB dimensions (∆R^2 range=.18 to .40,all p<.001). Crucially, adding SEN status to models already containing psychosocial symptoms yielded no significant improvement in model fit for any SWB outcome (χ^2 (1)=.37-2.02,all p>.05), supporting H₂. The association between SEN and SWB was fully subsumed by students' current psychosocial symptom profiles. Regarding symptom-specific associations (H₃), internalizing symptoms (Emotional Symptoms, Peer Problems) demonstrated the strongest and broadest negative effects, with standardized coefficients ranging from β=-.19 to β=-.56, all p<.001, particularly impacting affective and social SWB. Externalizing symptoms (Conduct Problems, Hyperactivity/Inattention) showed smaller, more domain-specific effects (β=-.07,p<.05 to β=-.24,p<.001), concentrated on social and competence-related SWB. Together, these findings challenge assumptions that SEN status represents a meaningful explanatory category for students’ well-being. The hierarchical modelling approach provides stronger evidence than previous simultaneous-entry models by demonstrating that SEN status did not significantly improve model fit. This supports empirical models that foreground individual symptom profiles as proximal determinants of SWB. By showing that specific psychosocial domains map onto specific SWB components, the study strengthens multidimensional frameworks of SWB and underscores the importance of moving beyond categorical classification systems toward symptom-sensitive approaches within inclusive education research. This underscores the need for school-based well-being frameworks that are symptom-sensitive and multidimensional. Practically, we recommend that inclusive schools: (1) prioritize the early detection of internalizing symptoms, which showed the strongest and most pervasive negative associations with SWB; (2) employ differentiated psychosocial assessments rather than relying on SEN labels as homogeneous risk markers; and (3) adopt monitoring systems that capture the distinct affective, social, and competence-related components of SWB to guide targeted interventions.

Alnahdi, G. H., and Schwab, S. (2021). Inclusive education in Saudi Arabia and Germany: students’ perception of school well-being, social inclusion, and academic self-concept. Eur. J. Spec. Needs Educ. 36(5), 773-786. Goldan, J., Nusser, L., and Gebel, M. (2022). School-related subjective well-being of children with and without special educational needs in inclusive classrooms. Child Indic. Ress 15, 1313-1337. Goodman, R. (2005). Strengths and Difficulties Questionnaire. Hascher, T., Morinaj, J., and Waber, J. (2018). “Schulisches Wohlbefinden“, in Leistung und Wohlbefinden in der Schule: Herausforderung Inklusion, eds K. Rathmann and K. Hurrelmann (Weinheim: Belz), 66-82. Hossain, S., O'Neill, S., and Strnadová, I. (2023). What constitutes student well-being: A scoping review of students' perspectives. Child Indic. Res. 16(2), 447-483. Kim, E. K., and Choe, D. (2022). Universal social, emotional and behavioral strength and risk screening: relative predictive validity for students’ subjective well-being in schools. Sch Psychol. Rev 51(1), 40-54. Kröske, B. (2020). Schulisches Wohlbefinden, Zugehörigkeit und Unterstützung bei Schülerinnen und Schülern im gemeinsamen Unterricht der Sekundarstufe I. Unterrichtswissenschaft 48, 243-272. Kullmann, H., Zentarra, D., Lütje-Klose, and Geist, S. (2023). Wohlbefinden und Inklusion an der Laborschule Bielefeld (WILS): Ausgewählte Ergebnisse der Fragebogenerhebungen 2013 – 2018 in den Jahrgangstufen 6 – 10. Schule – Forschen – Entwickeln 2, 77-110. OECD (2023). PISA 2022 Results (Volume II): Learning During – and From – Disruption, PISA. Paris: OECD Publishing. Sacco, R., Camilleri, N., Eberhardt, J., Umlauf, R., and Cedro, A. (2024). A systematic review and meta-analysis on the prevalence of mental disorders among children and adolescents in Europe. Eur. Child & Adolesc. Psychiatry 33, 2877-2894. Schnell, J., Saxer, K., Mori, J., Buchmann, M., and Ragni, T. (2025). Feeling well and doing well. The mediating role of school engagement in the relationship between student well-being and academic achievement. Eur. J. Psychol. Educ 40, 48. Tobia, V., Greco, A., Steca, P., and Marzocchi, G. M. (2019). Children’s wellbeing at school: a multi-dimensional and multi-informant approach. J. Happiness Stud. 20, 841-861. UNESCO (2025). Integrating health and well-being into education policy and planning: a handbook. Paris: UNESCO, Paris. Zdoupas, P., Vösgen-Nordloh, M., and Kulawiak, R. (2025). Subjektives schulisches Wohlbefinden in inklusiven Schulen im Spannungsfeld zwischen sonderpädagogischem Unterstützungsbedarf und psychosozialen Problemen. ESE 7, 124-139.