Is being a "small fish in a big pond" bad for students´ psychosomatic health? A multilevel study on the role of class-level school performance.

ABSTRACT

BACKGROUND: Features of schools and classes are closely related to students´ health and wellbeing. However, class composition (e.g. in terms of school performance) has rarely been examined in relation to students´ health and wellbeing. This study focuses on the so called Big-Fish-Little-Pond-Effect (BFLPE), by investigating whether the level of high-performing students in classroom is negatively associated with psychosomatic complaints of students who perceive themselves as poor performers. METHODS: Data were derived from the German sample of the WHO-Collaborative "Health Behaviour in School-aged Children (HBSC)" study 2013/2014. The sample included 5226 11-, 13- and 15-year-old students. Individual perceived school performance (PSP) was included (very good/good vs. average/below average PSP) at the individual student-level. At the class-level, school performance in class was generated by aggregating the share (in percentage) of students who report a very good/good PSP to the class-level, indicating the percentage of students with good/very good PSP in classroom. Using multilevel regression models, the association between class-level school performance (in percentage of students with very good/good PSP) and individual psychosomatic complaints were analyzed, stratified by students´ individual PSP. RESULTS: Students who report average/below average PSP showed higher likelihoods of psychosomatic complaints (Odds Ratio 1.75; 95% Confidence Interval 1.52-2.03) compared to counterparts with very good/good PSP. The aggregated class-level PSP was not significantly associated with psychosomatic complaints. However, in line with the BFLPE, results further revealed that students with average/below average PSP, who attend classes with a higher percentage of students who report very good/good PSP, had higher likelihoods of psychosomatic complaints (Odds Ratio 1.91; 95% Confidence Interval 1.01-4.01) compared to classmates with very good/good PSP. CONCLUSIONS: This study revealed that class composition in terms of PSP was differentially associated with students´ psychosomatic complaints, depending on their individual PSP. Findings highlight the vulnerability of students with poor PSP placed in classes with a higher percentage of students with good PSP. Results of this study therefore indicate a need for initiatives for low performing students from teachers and school staff in class.