Programming attitudes predict computational thinking: Analysis of differences in gender and programming experience

Programming attitude has received increasing attention in programming and computational thinking (CT) education. However, at present, analysis remains scant on the correlation between students' programming attitudes and their CT skills, and less attention is paid to the influencing effects of gender and programming experience. This study first validated a newly developed programming attitudes scale (χ2/df = 2.308, RMSEA = 0.073, CFI = 0.925, TLI = 0.912) among 1180 seventh-grade students (12–13 years old). It then analyzed the relationship between students' programming attitudes and CT skills considering gender differences. Students' programming attitudes significantly predicted their CT skills. Girls have higher CT skills than boys, but their more negative programming attitudes may affect their continued skill development. Third, the relationship among programming experience, programming attitudes, and CT skills was discussed among the sample (N = 175) with programming experience, revealing that programming attitudes played an intermediary role between programming experience and CT skills. Two specific factors of programming experience (first time learning programming and programming learning duration) had differing effects on students' programming attitudes and CT skills. For programming attitudes, the grade in which students got started with programming had no effect on their programming attitudes, but programming attitudes rose with increased programming learning time. For CT, students who learned programming starting in second grade possessed the highest CT level. In addition, there is an inverted U-shaped curve between students' programming learning duration and CT skills, peaked at the fourth year of programming learning career. Implications for promoting junior high school students’ CT skills and their positive attitudes toward programming are also discussed.