Learning Analytics to Determine Profile Dimensions of Students Associated with Their Academic Performance

With the recent advancements of learning analytics techniques, it is possible to build predictive models student academic performance at an early stage a course, using student’s self-regulation and affective strategies (SRLAS), their multiple intelligences (MI). This process can be conducted determine most important factors that lead good performance. A quasi-experimental study on 618 undergraduate students was performed profiles based these two constructs: MI SRLAS. After calibrating students’ profiles, techniques were used relationships among dimensions defined by constructs principal component analysis, clustering patterns, regression correlation analyses. The results indicate logical-mathematical intelligence, intrinsic motivation, have positive impact In contrast, anxiety dependence external motivation negative effect priori knowledge characteristics sample its likely behavior predicted may provide both teachers with early-awareness alert help in designing enhanced proactive strategic decisions aimed improve reduce dropout rates. From side, about main profile will sharpen metacognition, which