The aim of this study is to determine the extent at which students' computational thinking (CT) skills are explained by certain variables. To this end, the existence of explanatory and predictive relationships between students' CT skills and the variables in question has been examined and Structural Equation Model was employed for building a model in order to establish whether the variables predict students' CT skills or not. A correlational survey method was used in this research and the data collection tools of "Personal Information Form", "Computational Thinking Skills Scale" and "Thinking Ways Scale" were administered to 405 students who were studying in 5th-12th grades in the 2018-2019 academic year in various schools in the Province of Edirne, Turkey. According to the research results, it was determined that CT skill was predicted by different ways of thinking; as well as other variables such as academic achievement in math, science and Information Technology (IT) courses, IT usage experience and level of education.