Disciplinary climate, opportunity to learn, and mathematics achievement: an analysis using doubly latent multilevel structural equation modeling

ABSTRACT

Disciplinary climate and opportunity to learn (OTL) are considered as effectiveness-enhancing factors that can improve mathematics achievement. In this study, we investigated whether the school-level aggregation of student-reported OTL could yield reliable and valid measures, and then explored the relationships among disciplinary climate, OTL, and mathematics achievement at both school and student levels. Doubly latent multilevel structural equation modeling was adopted to analyze data from 63 countries/economies measured in the Programme for International Student Assessment (PISA) 2012. Three key findings emerged: (1) both disciplinary climate and OTL were reliable constructs when used at the school level, (2) disciplinary climate and OTL had positive effects on achievement at the school level, and OTL mediated the influence of disciplinary climate on achievement, and (3) OTL was positively associated with student achievement at the student level. Methodological and practical implications were discussed.

KEYWORDS:

• Disciplinary climate
• opportunity to learn
• mathematics achievement
• doubly latent multilevel structural equation modeling
• PISA 2012

Disclosure statement

No potential conflict of interest was reported by the authors.

Availability of data

The data that support the findings of this study are openly available in OECD Education Statistics (database): https://www.oecd.org/pisa/data/pisa2012database-downloadabledata.htm

Notes

1 According to the guidance of theoretical (Asparouhov & Muthén, 2010; Stapleton, 2008) and substantive studies (Dettmers et al., 2009), replicate weights were not appropriate in a multilevel framework. These studies indicated that there are two methods to account for the complex design of PISA in Mplus. The first is to use replicate weights with “type = complex” to account for the clustering and stratification. The second is to use sampling weights with “type = twolevel”. The results of the two methods are very similar (Dettmers et al., 2009). In this study, we used the second method for three reasons. First, with a multilevel framework, we can directly decompose the variance of targeted variables (i.e., OTL and disciplinary climate) at the student and school levels, which is consistent with our research questions. Second, the multilevel framework enables us to explore the relationship between OTL, disciplinary climate, and academic achievement at both school and student levels. Third, our research aims to extend previous research by disentangling the variance of OTL and disciplinary climate into school and student level, which can be realized by using the multilevel framework.

Additional information

Notes on contributors

Faming Wang

Faming Wang is a PhD candidate at the Faculty of Education, University of Macau. His research interests include mathematics education, big data and international large-scale assessments in education, and motivation and engagement. His recent publications appear in Journal of Psychoeducational Assessment, Learning and Individual Differences, and Frontiers in Psychology.

Yaping Liu

Yaping Liu is a PhD candidate at the Faculty of Education, The University of Hong Kong. Her research interests include educational statistics and measurement, educational evaluation, and literacy assessment. Her recent publication appears in Frontiers in Psychology.

Shing On Leung

Shing On Leung is an associate professor at the Faculty of Education, University of Macau. His research interests include educational measurement and applications of statistics in education and social sciences. His recent publications appear in The Journal of Experimental Education, Multivariate Behavioral Research, System, and Journal of Applied Measurement.