Artificial intelligence dependency among educators: a scale development and validation study

Abstract

This study aims to develop a valid and reliable measurement tool to assess the level of dependency on artificial intelligence (AI) among educators (teachers and academics). An exploratory sequential mixed methods design was employed. In the first phase, qualitative data were collected from 32 teachers and academics using a semi-structured interview form. Additionally, the literature review was conducted on AI dependency. Based on the findings, a 55-item pool was created. Content analysis was used for the qualitative data, while Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) were used to examine and validate the factor structure of the scale, respectively. The EFA results revealed a two-factor structure comprising 32 items: Dependency in Educational Processes and Dependency in Academic Processes, explaining 69.75% of the total variance. The Cronbach’s alpha reliability coefficient of the whole scale was found as 0.982. Additionally, to assess the scale’s reliability, a split-half method was used by dividing the items into two groups (odd and even). The Cronbach’s alpha coefficients for the first and second groups were 0.947 and 0.954, respectively, with a high and positive correlation between the two groups (r = .961). Furthermore, the Spearman-Brown coefficient was calculated as 0.980, and the Guttman split-half coefficient was also 0.980. Finally, CFA was applied on the 32-item version of the scale, and results confirmed the model with a chi-square/df ratio of 1.75 and an RMSEA value of 0.044. As a result, a valid and reliable tool was obtained to assess the dependency on AI among educators.

Ähnliche Arbeiten