Effects of demographic variables on mathematics self-esteem, test anxiety, and achievement in adolescents: a Bayesian beta regression analysis

Background: Mathematics achievement is shaped not only by cognitive skills but also by psychological factors such as self-esteem and test anxiety. Demographic characteristics, including gender, age, and school location, may further moderate these relationships, yet limited research has explored these dynamics in diverse educational contexts. Situated in Nigeria, this study examined how mathematics self-esteem and test anxiety predict mathematics achievement among adolescents, while considering the moderating role of demographic variables. Methods: A cross-sectional design was employed with data collected from 329 secondary school students. Validated instruments measured mathematics self-esteem, mathematics test anxiety, and mathematics achievement. Bayesian beta regression was used to analyse the relationships among psychological and demographic variables, offering a nuanced understanding beyond traditional statistical approaches. Results: Findings revealed that mathematics self-esteem positively predicted achievement, while test anxiety was negatively associated with performance. Older male students reported lower self-esteem, rural students reported higher test anxiety, and rural students demonstrated higher achievement despite elevated anxiety. Two-way interaction effects involving gender, age, and location accounted for additional variance in mathematics self-esteem and achievement. Conclusion: The results highlight the complex interplay between psychological and demographic variables in shaping mathematics outcomes. Strengthening students' self-esteem and addressing test anxiety are essential for promoting equitable learning environments. The findings underscore the importance of demographic context, pointing to the need for targeted interventions and policies that support resilience, reduce anxiety, and enhance adolescent mathematics achievement.