Exploratory Path Analysis of Mathematics Ability Among At-Risk Students

Ace Amir Prescillas (St. Michael’s College of Iligan, Iligan City, 9200, Philippines)
Aniceto B. Naval (School of Graduate Studies, Northwestern Mindanao State College of Science and Technology, Misamis Occidental, Philippines)

Journal of Research in Mathematics, Science, and Technology Education
Vol. 3 No. 3 (2026): Journal of Research in Mathematics, Science, and Technology Education

Persistent mathematics underachievement among at-risk students remains a pressing educational concern because weak performance in mathematics can constrain later academic progression and learners’ ability to cope with more advanced school tasks. This study examined mathematics ability among at-risk students through an exploratory path analysis linking school-related, teacher-related, and student-related factors. Using a quantitative, cross-sectional, explanatory design, the study focused on students identified as at risk in mathematics through a TIMSS-like screening assessment. The respondents were 450 Grade 9 students from public and private junior high schools in Northwestern Mindanao, Philippines. There were 370 classified as at risk, and 368 cases were retained for the inferential analyses. Data were drawn from a mathematics ability test, a mathematics anxiety questionnaire, an attitude toward mathematics questionnaire, and a mathematics teaching strategy questionnaire. Analysis proceeded through descriptive statistics, regression-based screening, and exploratory path analysis using observed variables only. The final parsimonious model showed a good fit to the data. Appraisal and confidence emerged as the most immediate positive predictors of mathematics ability, while negative attitude showed a smaller negative role in the preliminary screening model. Teacher-related and school-related variables were linked to mathematics ability mainly through instructional and affective pathways. Reflective strategy showed positive direct and indirect associations through appraisal and confidence, whereas discussion showed negative indirect associations through the same mediators. The type of school also retained a small but significant direct contribution to mathematics ability, suggesting that broader school conditions remained relevant even after proximal affective and instructional factors were considered. The findings indicate that mathematics ability among at-risk students is best understood as a layered outcome shaped by school context, instructional processes, and students’ evaluative and competence-related responses to mathematics.