HOW DO COGNITIVE ABILITY AND STUDY MOTIVATION PREDICT THE ACADEMIC PERFORMANCE OF IT STUDENTS? Margus Pedaste, Olev Must, Gerli Silm, Karin Täht, Külli Kori, Äli Leijen, Mari-Liis Mägi University of Tartu (ESTONIA) Abstract Information technology (IT) students’ dropout during the first year of their higher education studies is a big problem in Estonia. In general, students’ cognitive ability and motivation to study have been described as important factors in several models predicting students’ academic achievement, which is directly linked to dropout. However, the findings are sometimes contradictory and not specifically about IT studies. In the case of IT students, we might expect differences compared with other students because of the recent trends where more digitalized learning environments are used. Therefore, our aim was to discover how motivational aspects and cognitive ability of digitally competent students are combined in predicting academic performance and dropout and what the role of both of them is. In our study, we collected data from 44 first-year IT students. We asked them to fill in a study motivation questionnaire and do a cognitive ability test. In addition, data about their academic performance was collected from the higher education institutions’ study information systems. Our findings demonstrated that students’ cognitive ability correlated with their intrinsic study motivation, but not with their extrinsic motivation or amotivation. Gender plays an important role in predicting study motivation – girls are shown to be more motivated. A path analysis showed that gender had a similar effect on both intrinsic and extrinsic motivation, but no effect on students’ cognitive ability. However, whereas students’ grade point average is positively influenced by their cognitive ability and intrinsic motivation, extrinsic motivation seems to have a negative effect on academic performance. Keywords: IT education, study motivation, cognitive ability, academic performance, dropout. 1 INTRODUCTION About one third (32%) of Estonian information technology (IT) students drop out during the first year of higher education studies [1]. Several authors have shown that dropout is strongly linked to lower academic performance, which, in turn, is influenced by the characteristics of learners and the learning community (see [1], [2], [3], [4]). There are six groups of factors influencing students’ dropout from higher education studies according to the review by Kori et al. [1]: the student’s income, demographics, institutional characteristics, psychological factors, social integration, and academic performance in higher education studies. However, Chen [5] has found that college grade point average, one of the main characteristics of academic performance, is the biggest dropout predictor. Nowadays, the meaning of and engagement in a learning community is changing because of the expansion of digital communities. This is especially the case in IT studies, where students are more familiar with IT technologies for learning. Therefore, IT students are a promising group to study in order to discover how cognitive abilities and motivational aspects of digitally competent students are combined in predicting academic performance and what the role of both of them is. 1.1 Cognitive ability Many studies have demonstrated that academic performance correlates with students’ cognitive abilities. Combining the results of cognitive ability tests with high school average scores has led to the strongest correlations with academic performance in the first university year [6]. In the Estonian context, Niitsoo et al. [7] have also shown that mathematics state exam results are one of the predictors of students’ academic performance in IT studies. Cognitive abilities have been described as a set of mental abilities needed in carrying out a task. They are measured through different intelligence tests, and strong correlations between students’ academic performance and cognitive abilities have been revealed. For example, Lynn, Meisenberg, Mikk and Williams [8] found correlations from .83 to .92 between TIMSS and PISA test results and the average national IQ scores. The general effect of intelligence on educational achievement based on the meta- Proceedings of ICERI2015 Conference 16th-18th November 2015, Seville, Spain ISBN: 978-84-608-2657-6 7167