Psychometric Properties and Factorial Structure of the Lithuanian Version of the "Student Engagement in School"

Abstract

Students’ engagement in school is an important phenomenon for their learning and overall well-being. The aim of this paper is to present the adaptation and validation of the Student Engagement in School Questionnaire (Lam et al., 2014) to the Lithuanian school context. The psychometric indicators and factorial structure of the questionnaire were assessed in two studies. The first study was carried out in June 2019. 476 students took part in this step. The second study was undertaken in February 2020. The sample was 445 pupils. To examine the reliability of the Lithuanian version of the Student Engagement in School Questionnaire, we calculated the internal consistency coefficients (Cronbach’s α and McDonald’s ω). The results of both studies showed good internal consistency of the questionnaire: the values of Cronbach’s α and McDonald’s ω were greater than 0.70 for all scales. To determine the factor structure of the measurement tool, first, exploratory factor analysis (EFA) was performed. The results of EFA only partially confirmed the authors’ proposed questionnaire structure. After applying the Varimax rotation, a six-factor structure was identified, which differed from the original three-factor model. To test the accuracy of the three-factor model and the higher-order factor model structure, confirmatory factor analysis (CFA) was performed. As the original three-factor model was not supported, the CFA results demonstrated that alternative hierarchical structures provided a good fit to the data. A second-order model with six first-order factors (SRMR = 0.056; RMSEA = 0.048; CFI = 0.905) as well as a more theoretically nuanced third-order model (SRMR = 0.056; RMSEA = 0.048; CFI = 0.908) were both found to be suitable. However, it is necessary to verify the stability of the obtained hierarchical model structures by replicating the study.