Adaptation of Artificial Intelligence Literacy Scale: Latent Profile Analysis

Abstract

Artificial intelligence literacy is vital for individuals' adaptation to the future workforce and societal changes by enabling them to understand and effectively use AI technologies and critically evaluate their impact on society. In this study, the validity and reliability of the artificial intelligence literacy scale in Turkish language were tested and the latent profiles of the students were determined. This methodological study was carried out with a total of 729 students between December 2023 and February 2024. Validity and reliability analyses were conducted with SPSS 27 and AMOS 24, and latent profile analysis was handled with R programming language. According to the results of the CFA analysis of the Artificial Intelligence Literacy Scale, the fit indices were found to be significant (X²/sd= 3.832, RMSEA=.062, CFI=.949, AGFI=.933, GFI=.960, NFI=.949, TLI=.928, IFI=.916). Considering the Cronbach Alpha value of the scale consisting of 4 sub-dimensions and 12 items, the internal consistency coefficientwas found to be 0.814. Since the lowest BIC value in the latent profile analysis was found in the VVV model, the VVV model was considered as the appropriate one in the study, and the class analyses were carried out through this model. With the LPA analysis, it was designated that the scale was divided into 3 classes. It was determined that the Artificial intelligence literacy scale is a valid and reliable measurement tool. After latent profile analysis, it was found out that the scale was divided into 3 classes.

Ähnliche Arbeiten