Comparing effectiveness of additive, interactive and quadratic models in detecting combined effects of achievement goals on academic attainment

Abstract

This study compared effectiveness of additive, interactive, and quadratic statistical models in detecting the combined effects of achievement goals on academic achievement. In a prospective study that aimed to predict college students’ grades in an English course, we found that the quadratic model was more effective in detecting the combined effects of achievement goals on course grades than the additive and interactive models. In addition, a response surface analysis showed that the combined effects of achievement goals on course grades corresponded to a goal profile that involved tendencies to endorse mastery goals at high levels and performance goals at moderate levels. Findings suggest that the quadratic model is a viable data analytic technique that assists researchers in detecting combined effects of achievement goals on academic achievement.