The purpose of this study is to analyze how efficient online study groups can be formed among students based on their personality traits. A survey consisting of Ten Item Personality Inventory (TIPI) was conducted among the undergraduate students in a well-known university. Eighty-two students who did not know each other were assigned to 35 small online groups based on their personality characteristics. The group members were then asked to study collaboratively on a task by communicating via the university's learning management system (LMS) forums. It was found that other factors (such as gender) were more effective than personality traits on the group success, and groups with lower degrees of Emotional Stability scores obtained higher grades over the task. This study is one of the first examples that hierarchically show different factors affecting the success of online groups with data mining techniques. The findings of the study will contribute to the field of online collaborative learning that is one of the most prominent subject in distance education. (C) 2016 The Authors. Published by Elsevier Ltd.