Self-efficacy is an important predictor of teaching behavior. Within several research traditions (TPACK framework, school achievement tests), different ICT self-efficacy scales have been developed. The empirical structure of existing questionnaires has frequently been researched and discussed within the TPACK framework. However, for teacher ICT self-efficacy, non-instructional aspects as well as alternative instruments have rarely been investigated and compared to standard TPACK self-report scales.
In this study, we administer two sets of non-subject-specific ICT scales to a group of pre-service teachers (N = 165). We investigate the empirical structure using structural equation modeling.
For both scales, the results show that instructional ICT self-efficacy forms a separate factor. For the remaining items, item difficulty and content drive further divisions. Further, more specific item formulations resulted in a higher range of scale means. Additionally, we find gender differences only on some non-instructional scales.
Results confirm that the distinction between instructional vs. non-instructional ICT self-efficacy is important when developing or using questionnaires for pre-service teachers. Results also indicate that the usage of more specific task-related items is a promising alternative to general TPACK items.