Measures of socioeconomic status (SES) are widely used in educational data analyses as control variables and to increase the statistical power of significance tests. The effectiveness of SES measures in these analyses depends on moderate correlations between measures and educational outcomes. However, meta-analyses of the relationship between SES and student achievement have produced average bivariate correlations in the.03–.32 range, implying these measures offer modest statistical control and power gains. Author provided an empirical example of using indirect effects in structural equation modeling to better capture the impact of SES analyzing data from the 2012 Program for International Student Assessment. The results illustrate the potential of indirect effects to fully capture the impact of SES in educational data analyses.
Bibliographical notePublisher Copyright:
© 2020 Elsevier Ltd
- Structural equation modelling
- Students’ achievement