Exploiting TIMSS and PIRLS combined data: Multivariate multilevel modelling of student achievement

Abstract

We exploit a multivariate multilevel model for the analysis of the Italian sample of the TIMSS&PIRLS 2011 Combined International Database on fourth grade students. The multivariate approach jointly considers educational achievement on Reading, Mathematics and Science, thus allowing us to test for differential associations of the covariates with the three outcomes, and to estimate the residual correlations between pairs of outcomes at student and class levels. Multilevel modelling allows us to disentangle student and contextual factors affecting achievement. We also account for territorial differences in wealth by means of an index from an external source. The model residuals point out classes with high or low performance. As educational achievement is measured by plausible values, the estimates are obtained through multiple imputation formulas. The results, while confirming the role of traditional student and contextual factors, reveal interesting patterns of achievement in Italian primary schools.

Italian Teachers' Well-Being Within the High School Context: Evidence From a Large Scale Survey

“…In order to explore the relationship between teachers' well-being and the teachers' working environment, considering the heterogeneity between disciplines and teachers' characteristics, a bivariate regression model has been adopted. Specifically, the use of a bivariate approach allows us (Leckie and Charlton, 2013; Sulis and Porcu, 2015; Grilli et al, 2016; Leckie, 2018) to consider the effect of teacher working environment and management practices and policies on both outcome measures selected to monitor teachers' well-being: namely Satisfaction with the current job environment () and Satisfaction with teaching profession (). Both variables are treated as response variables in a regression framework with the aim to shed light on which management practices and policies may have the greatest impact on improving teachers' well-being.…”

Do Textbooks Matter for Reading Comprehension? A Study in Flemish Primary Education

“…The parameter estimates in this model were not tangibly different from the univariate model parameter estimates. Generally, a multivariate model yields more efficient parameter estimates, but the efficiency gains of using a multivariate model versus a univariate model is a function of the correlation between the outcomes (e.g., Robins and Rotnitzky, 1995;Teixeira-Pinto and Normand, 2009;Grilli et al, 2016). Given that the correlation between reading comprehension and engagement was relatively low, we have no good statistical argument to use a multivariate autoregression model.…”

Studying the relationship between anxiety and school achievement: evidence from PISA data

“…In our sample, the average level of anxiety is equal to 0.48 (cf. Grilli et al, 2016;Bratti et al, 2007;Giambona and Porcu, 2015;Contini et al, 2017); in particular, boys tend to outperform girls in mathematics and science, whereas girls tend to outperform boys in reading (Invalsi, 2016). It is also interesting to note that we also tested whether the interaction term between gender and text anxiety was signicantly dierent from zero across quantiles.…”