The aim of this study is to examine the relationship between primary school teachers' 21st century skills self-efficacy levels and their computational thinking skills and STEM implementations self-efficacy levels. The sample of the study consists of 440 primary school teachers. While determining the sample of the research, convenience sampling technique, one of the purposeful sampling methods, was used. In order to collect data within the scope of the research, "Personal Information Form", "STEM Implementations Teacher Self-Efficacy Scale", "Computational Thinking Skill Scale" and "21st Century Skills Self-Efficacy Perception Scale" were applied. SPSS 22 package program was used to analyze the data obtained in the study. Since the data of the research showed normal distribution, Pearson correlation and multiple regression were analyzed. According to the results of Pearson correlation analysis conducted in this study, it was determined that there was a weak positive relationship between STEM implementations self-efficacy variable and 21st century skills self-efficacy computational thinking skills variables. In addition, it was determined that the relationship between the 21st century skills self-efficacy variable and the computational thinking skills variable was positive and moderate. According to the results of multiple regression, it was determined that primary school teachers' computational thinking and 21st century skills self-efficacy levels significantly predicted STEM implementations teacher self-efficacy level. As a result, it was determined that 21st century skills self-efficacy and computational thinking skills affect STEM implementation self-efficacy levels.
UĞRAŞ et al. (Sat,) studied this question.