Study of adapting the Employee Advocacy Role Perception Scale into Turkish

AKSU M.

Ordu Üniversitesi Sosyal Bilimler Enstitüsü Sosyal Bilimler Araştırmaları Dergisi, cilt.15, sa.1, ss.340-369, 2025 (TRDizin)

Özet

This study aimed to verify the validity and reliability of the Employee Advocacy Role Perception Scale (EARPS), and to adapt it for Turkish organizations. In this study, the data obtained from 338 participants were divided into two parts: for exploratory factor analysis (EFA) and confirmatory factor analysis (CFA). Data analyses were performed with SPSS and AMOS software. The findings showed that the unidimensional scale was 64.779% of the total variance. Our analysis showed a KMO value of 0.947 and a Bartlett statistical test value of 𝜒2=3287.423, which was a significant difference (p<0.001), and the scale had a single-factor structure with item loads ranging 0.628–0.873. According to confirmatory factor analysis fit indices, the acceptable fit indices for Chi-squared test was 2.62 and a RMSEA value of 0.072 were established. The GFI (0.955), AGFI (0.920), and CFI values (0.980) showed good fit. The goodness-of-fit indices of the single-factor structure that emerged after the analysis revealed a good and acceptable fit. In addition, the standardized regression weights of the scale items varied between 0.548 and 0.885; t values from 9.626 to 11.695 that were statistically significant (p<0.001), and R2 values of 0.301–0.783. These results confirm the construct validity of the single-factor model with explanatory factor analysis. Regarding the internal consistency of the scale, Cronbach’s Alpha value of the relevant scale was 0.933, AVE was 0.647, CR was 0.959, and the correlation level between the two halves was 0.848 with a first half test reliability of 0.927 and a second half test showing a reliability value of 0.918. We found α=0.868 for odd-numbered scale items and α=0.876 for even-numbered scale items. We report an α=0.940 for the dataset that included 170 randomly selected items, and α=0.932 for the dataset of 168 items. When all these data were evaluated together, we showed that the reliability of the nine items constituting the relevant scale was high. When EFA and CFA results were evaluated together, the unidimensional structure of the EARPS, consisting of nine items, was valid and had a high reliability level. As a result, the compatibility of the relevant scale with the Turkish cultural structure was established, and it is expected that it will produce valid and reliable results for measurements regarding the employee advocacy role perception of human resources.