The reliability and validity study of the Turkish form of the Self-efficacy Perception Scale for Programming

The purpose of this study is to adapt "Computer Programming Self-Efficacy Scale", developed by Ramalingam and Wiedenbeck (1998), to Turkish and to examine individuals' programming self-efficacy scores across demographic variables. The study group is consisted of 152 undergraduate students from different academic departments who have taken at least one programming course earlier. The original scale was a 7-likert type with 32 items in 4 factors (independence and persistence, ability to perform simple programming tasks, ability to perform complex programming tasks, self-regulation). In the adapted version, the scale was finalized with 9 items in 2 factors (ability to perform simple programming tasks and ability to perform complex programming tasks) in Turkish form. Cronbach alpha coefficient was found to be .928 and McDonald's. (omega), which is proposed as being more appropriate for congeneric measurement, was found to be .956. As a result of explanatory factor analysis, two factors have explained 80,814 % of the variance together and this factor structure has been confirmed with confirmatory factor analysis, which showed a valid and reliable scale to measure programming self efficacy.