Similarities and differences in social and emotional profiles among students in Canada, USA, China, and Singapore: PISA 2015

ABSTRACT

Although previous research showed that discrete social-emotional skills such as empathy, motivation, and social relationships in school significantly predict achievement, students tend to use various social-emotional skills in combination. As such previous investigations cannot comment on how different combinations or profiles of students’ social-emotional skills predict achievement relative to discrete skills. Likewise, little is known about cross-national comparisons of social-emotional skill profiles (SESP), and the extent to which SESP differ on their academic achievement. The purposes of this study were three-folded: 1) to determine whether a four-factor social-emotional skills model could be used for cross-national comparisons; 2) to identify social-emotional profiles in 15-year-old students from four different countries – Canada, the United States, China, and Singapore; and 3) to evaluate how different profiles predict students’ reading, maths, and collaborative problem-solving (CPS) test scores. Our results showed multigroup measurement invariant in the structure, loadings, and thresholds of the four-factor social-emotional skills model. We identified three profiles labelled Sociable, Reserved and Withdrawn in Canada, Singapore, and the United States; whereas, we found three profiles labelled Solitary, Team-oriented, and Reserved in students in China. Finally, the way each profile associated with reading, maths and CPS in each country appeared to align with the cultural expectations of learning.

KEYWORDS:

• Social-emotional skills
• latent profile analysis
• cross-cultural comparison
• standardised international student achievement

Disclosure statement

No potential conflict of interest was reported by the authors.

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Notes

1. It should be noted that in China, participants represented adolescents living in Beijing, Shanghai, Jiangsu and Guangdong.

2. PISA 2015 student background questionnaire was distributed to all student participating in the assessment (OECD, 2017). This means that all participants filled out the questionnaire. Missing data were handled using MPlus’ default estimation—Full Information Maximum Likelihood—a widely employed approach that is used in many statistical packages and has been found to be more superior than other techniques (e.g., listwise deletion; Enders, 2001).

3. CFI and RMSEA were chosen to evaluate the model fit, because they were the two most commonly (78.4% and 64.9%) reported fit indices in confirmatory factor analysis (Jackson, Gillaspy, Jr., & Purc-Stephenson, 2009).

4. We also conducted LPA with BCH separately for 10 plausible values in each subject area for each country. Given that there was no significant practical difference and the averaged values offered a practical avenue for interpretation on how students belong to different profiles did on these subject areas, we decided to use the results from averaged values for interpretation.

Additional information

Funding

This work was supported by the University of Manitoba [48452].

Notes on contributors

Virginia M. C. Tze

Virginia M. C. Tze is an Assistant Professor at the University of Manitoba. Her studies focus on emotions, social-emotional learning, emotional management, self-regulation, motivation, and cultural diversity.

Johnson C.-H. Li

Johnson C.-H. Li is an Associate Professor in the Department of Psychology at the University of Manitoba. His research interests include quantitative methods, effect size measures, bootstrapping and correction for range restrictions.

Lia M. Daniels

Lia M. Daniels is a Professor of educational psychology at the University of Alberta. She studies students’ and teachers’ motivation and emotions to create adaptive learning environments.