Regional differences in educational achievement: A replication study of municipality data

Abstract

The current study analyzed the relationships between explanatory variables such as socioeconomic status (SES), migration background (MB), and formal teacher competence, and aggregated grades in the Swedish lower-secondary school context by using aggregated municipality data from 2013, 2018, and 2019. SES indicators had larger effect sizes when data from different years were merged and when the outcome variable was changed to an alternative measure of educational achievement. In one model, the MB variable even became statistically insignificant. These results indicate that SES is an important variable which explains a substantial amount of variance in regard to school achievement indicators such as grade point average. Nonetheless, aggregated data may still suffer from omitted variable bias and biased effect size estimates.

Is the SES and academic achievement relationship mediated by cognitive ability? Evidence from PISA 2018 using data from 77 countries

“…As mentioned earlier, due to the cross-sectional design, it is not possible to control for participating students' previous academic achievement, effect sizes will be skewed upward, and causal relationships not possible to discern. Moreover, due to the aggregated data, it is not possible to make inferences at the individual or school level (Boman, 2022d). Some may still consider the Lynn and Becker (2019) data set to be unreliable and misguiding (Sear, 2022).…”