Predicting problem-solving skills of students from different socio-economic background using math- and self-related variables

The purpose of this study is to examine relationships between math- and self- related variables and problem-solving skills of students on PISA 2012 Turkish sample. Students’ socio-economic status was used as a control variable. Turkish sample was divided into 4 groups using PISA’s ESCS index to create socio-economically different subgroups. The data were analyzed with hierarchical multiple linear regression in three blocks. The finding illustrated, firstly, socio-economic status gives positive association with problem-solving in each block. One of the mathematics factors is that self-efficacy has highly positive relationship with problem-solving. The other factors give different results for different social groups.