Explaining Social Disparities in Mathematical Achievement: The Role of Motivation

Applying Machine Learning and SHAP Method to Identify Key Influences on Middle-School Students' Mathematics Literacy Performance

The PISA 2022 literacy assessment highlights a significant decline in math performance among most OECD countries, with the magnitude of this decline being approximately three times that of the previous round. Remarkably, Hong Kong, Macao, Taipei, Singapore, Japan, and Korea ranked in the top six among all participating countries or economies, with Taipei, Singapore, Japan, and Korea also demonstrating improved performance. Given the widespread concern about the factors influencing secondary-school students' mathematical literacy, this paper adopts machine learning and the SHapley Additive exPlanations (SHAP) method to analyze 34,968 samples and 151 features from six East Asian education systems within the PISA 2022 dataset, aiming to pinpoint the crucial factors that affect middle-school students' mathematical literacy. First, the XGBoost model has the highest prediction accuracy for math literacy performance. Second, 15 variables were identified as significant predictors of mathematical literacy across the student population, particularly variables such as mathematics self-efficacy (MATHEFF) and expected occupational status (BSMJ). Third, mathematics self-efficacy was determined to be the most influential factor. Fourth, the factors influencing mathematical literacy vary among individual students, including the key influencing factors, the direction (positive or negative) of their impact, and the extent of this influence. Finally, based on our findings, four recommendations are proffered to enhance the mathematical literacy performance of secondary-school students.

Self-perceptions as mechanisms of achievement inequality: evidence across 70 countries

Children from lower socioeconomic status (SES) backgrounds tend to have more negative self-perceptions. More negative self-perceptions are often related to lower academic achievement. Linking these findings, we asked: Do children's self-perceptions help explain socioeconomic disparities in academic achievement around the world? We addressed this question using data from the 2018 Programme for International Student Assessment (PISA) survey, including n = 520,729 records of 15-year-old students from 70 countries. We studied five self-perceptions (self-perceived competency, self-efficacy, growth mindset, sense of belonging, and fear of failure) and assessed academic achievement in terms of reading achievement. As predicted, across countries, children's self-perceptions jointly and separately partially mediated the association between socioeconomic status and reading achievement, explaining additional 11% (ΔR2 = 0.105) of the variance in reading achievement. The positive mediation effect of self-perceived competency was more pronounced in countries with higher social mobility, indicating the importance of environments that "afford" the use of beneficial self-perceptions. While the results tentatively suggest self-perceptions, in general, to be an important lever to address inequality, interventions targeting self-perceived competency might be particularly effective in counteracting educational inequalities in countries with higher social mobility.

Mathematics anxiety and mathematical calculation in deaf children: A moderated mediation model of mathematics self-efficacy and intelligence

BACKGROUND

Previous studies have shown the relationship between mathematics anxiety and math performance in deaf students, but their inner influencing mechanism remains unclear.

AIM

To examine a moderated mediation model between mathematics anxiety and mathematical calculation, with intelligence as a moderator, and mathematics self-efficacy as a mediator.

METHODS

A sample of 247 deaf children from 2 special education schools and 247 hearing children (matched in intelligence) from one mainstream school in China completed computerized tests of intelligence and mathematical calculation and self-report questionnaires of mathematics anxiety and mathematics self-efficacy. Simple mediation analyses and moderated mediation analyses were conducted using PROCESS, and a simple slopes method was employed to plot the conditional indirect effects.

RESULTS

There was a significant negative correlation between mathematics anxiety and mathematical calculation, and between mathematics anxiety and mathematics self-efficacy in deaf children and hearing children. However, mathematics self-efficacy was positively associated with mathematical calculation in deaf children but not in hearing children, and the significantly negative relationship between mathematics anxiety and intelligence was observed only in deaf children but not in hearing children. Mathematics self-efficacy partially mediated the association between mathematics anxiety and mathematical calculation in deaf children; and the indirect effect between mathematics anxiety and mathematical calculation via mathematics self-efficacy was moderated by intelligence in deaf children but not in hearing children.

CONCLUSIONS

The results were discussed to illuminate the mechanism in relation to the practical implication for the intervention and early development of mathematics performance in deaf children.