The purpose of this study is to develop a psychometrically sound Teacher Collaboration Scale based on Item Response Theory. The data obtained from 461 teachers were analyzed, and items were calibrated using Samejima’s Graded Response Model. Exploratory Factor Analysis confirmed a unidimensional structure explaining 53.51% of the variance, while local independence and item-model fit were rigorously tested. The final 18-item scale demonstrated high discrimination parameters ranging from 1.815 to 4.777. The analysis of the test information function indicated that the scale provided the highest precision and reliability for teachers with collaboration levels between -3 and +1 theta. Additionally, Differential Item Functioning analysis revealed no gender bias, confirming the scale was a valid and reliable instrument.
Hıdıroğlu et al. (Fri,) studied this question.