Measuring Student Socioeconomic Status in Large Cross-National Studies: A Review and Critique of the Literature

• In Event: Measuring Student Home Background in Large Cross-National Studies: Conceptual and Methodological Issues

Abstract

Despite the importance of constructing valid and reliable SES measures in research using cross-national datasets, there is little consistency in the variables and methods that researchers employ when using these data to explore important educational questions. In this paper, we report the results of a review of recent cross-national research designed to assess how researchers label, construct, and employ measures of student SES in cross-national research. The inclusion criteria for our search included: relevance, an empirical nature, and quality, as evidenced by publication in a peer-reviewed journal or similar outlet. Relevance of the articles was determined by including only articles that used concepts of socioeconomic status in their models. We further limited the search to articles published post 2000 using data from 2000 and after. This second restriction was made in order to provide an overview of recent trends and to limit the number of reviewed articles. The search was further restricted to three of the most commonly used cross-national datasets: TIMSS, PISA, and PIRLS. We used the following electronic search engines: JSTOR, ERIC, Social science abstract, SSCI, Education full text, EconLit, Wilson web, and WorldCat.
After thorough review and coding, 55 articles matched the inclusion criteria and used measures of SES in their analysis. Among these articles, we found tremendous variation in nomenclature, variable inclusion, scaling procedures, treatment of missing data, and operationalization of student SES. Further, while some authors provided comprehensive descriptions of their approaches to measure SES, others provided little or no mention of their SES variable(s). In this paper, we present these results, discuss differences across datasets (PIRLS, PISA, and TIMSS), and offer “lessons learned” for conducting research using large cross-national datasets.

Authors

• Thomas F. Luschei, Claremont Graduate University

• Loris Fagioli, Claremont Graduate University

• Amita Chudgar, Michigan State University