Development and validation of the Computer Science Attitudes Scale for middle school students (MG-CS attitudes)

Primary and secondary students' exposure to computer science-related activities in school has been steadily increasing, heightening the need for valid measures regarding impact of these activities on students. This study reports on the development and validation process of an instrument to measure students' affective state as it relates to computer science in an academic setting. The self-report instrument, Computer Science Attitudes Scale for middle school students (MG-CS Attitudes), was developed based upon expectancy-value theory, which assumes two attitudinal constructs: self-efficacy and outcome expectancy. A set of ten initial items was administered to 663 middle-grade students from sixth to eighth grade (11-13 years of age). A combination of classical test theory and item response theory approaches were used to evaluate and validate the instrument using wellestablished construct validity frameworks to guide the process, leading to nine final items. The multi-stage validation process has resulted in a robust, well-functioning instrument, which can be used by researchers and evaluators to study CS-related educational interventions.