The purpose of this research was to investigate the effect of certain variables on educational achievement in the subject of math by applying the path analysis method. The variables were self-efficacy, competency goals, functional-approach goals, functional-avoidance goals, learning strategies, and persistence. Tehran secondary schools – that were public – were stratified and 389 boy and girl students from among the third grade students of the secondary schools were randomly selected. A questionnaire consisting of 6 subscales as mentioned above (the variables) were used, which were selected from valid sources. These sources included Miller et. al. (1996), and Middleton and Midgley (1997). Furthermore, the students’ final math exam grades were used to assess their achievement.

Path analysis method was used for data analysis. The results revealed that there was a relatively good fit between the hypothesized model and the observed data. The direct effects of functional-approach goals, avoidance-approach goals, self-efficacy, learning strategies and persistence on mathematics achievement were confirmed. Self-efficacy had the most direct effect on mathematics achievement, compared to other variables of the proposed model. Competency goals had a significant indirect effect on mathematics achievement through self-efficacy, learning strategies and persistence. The mediating role of persistence between avoidance-approach goals and mathematics achievement was significant but small in magnitude.