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

Abstract

IntroductionEarlier research has suggested that that the international large-scale assessment, PISA (Programme for International Student Assessment), may be looked upon as a form of school test that is mostly explained by participating students’ socioeconomic status, non-cognitive factors, and various school factors, whereas another strand of research focuses on the similarities between PISA and cognitive ability assessments such as IQ tests. The latter position does also highlight the strong relationships between PISA scores and IQ test scores, typically aggregated to the country level. The current article adds to this scholarly debate by examining the latest PISA survey from 2018.MethodsCorrelation, regression, moderator, and mediation analyses were run for aggregated country-level data (N = 77) from PISA 2018.ResultsThe results indicate that PISA scores indeed are strongly associated with both cognitive ability test scores and socioeconomic status indicators such as average annual national income.DiscussionA more nuanced position suggests that PISA should be labelled a test that measures cognitive school skills rather than a proxy of national IQ, as the link between country-level abilities and school age abilities is neither causal, nor theoretically and empirically palpable, yet partly robust.

Interplay of socioeconomic status, cognition, and school performance in the ABCD sample

“…We conjecture that it is more likely that both cognitive ability and parental education exert substantial individual influence on children’s grades and that a potential interplay between the two is more complex than what can be captured by a simple regression model. This finding echoes earlier work at the country level in which the relationship between school performance and SES was not affected by cognitive ability 60 . The effect sizes we observed were modest and did not differ considerably between the 2-year and the 3-year follow-up assessment.…”

“…While effects of cognitive abilities on academic achievement, and, in turn, of SES on both the former and the latter are well documented, interactions between the three are less explored. Earlier work has investigated whether cognitive ability affects the relationship between SES and school performance on a country level and has not confirmed such effects 60 . By contrast, in this work, we seek to determine whether the relationship between socioeconomic background and school achievement will differ as a function of cognitive ability on an individual level.…”

Human capital indicators as influenced by SES, cognitive and non-cognitive skills: Evidence from Taiwan’s most recent PISA survey

“…As a robustness check at the cross-national level, three more PISA participants for 2018 (China, Kazakhstan, the Philippines) were included in the data set and tested in various OLS models. That is because outliers such as a highperforming country like China (i.e., Beijing-Jiangsu-Shanghai-Zhejiang) and low-performing countries like the Philippines and Kazakhstan might affect the results (Boman, 2022b(Boman, , 2023b. The relationships were roughly similar when these were included, although PIAAC scores became more predictive when replacement values were used for these three countries.…”