Low math anxiety prevents math achievement decline for urban but not rural Chinese schoolchildren

Across a number of countries, rural schoolchildren systematically underperform in math in comparison with their urban counterparts. To plan targeted interventions, it is important to understand if there are any differences in factors influencing math performance in rural and urban areas. The purpose of the study was to compare the association between math performance, socio-demographic variables, nonverbal intelligence, and math anxiety in rural and urban China. The sample consists of 1,412 urban schoolchildren (51% females, M = 12.04) and 1,032 rural schoolchildren (50% females, M = 13.63). Raven's Progressive Matrices were used to measure nonverbal intelligence, and the Abbreviated Math Anxiety Scale served as a measurement tool for math anxiety. Overall, nonverbal intelligence is positively associated with math performance, whereas math anxiety is negatively associated. In contrast to the subtle decline in math performance with age in urban regions, in rural areas, math performance deteriorates sharply, widening the achievement gap over time. Notably, high nonverbal intelligence, low math anxiety, and high maternal education hinder the decline in math performance in urban areas, whereas in rural areas, only high nonverbal intelligence attenuates the decline. Furthermore, in urban areas, females show lower math achievement and higher math anxiety than males, while no gender differences are observed for rural areas. Potential explanations of the difference between rural and urban contexts and practical implementation are discussed.