Revisiting the factor structure of the mathematics teaching efficacy through a bifactor approach in a Greek sample of special and typical preschool education teachers

Abstract Early childhood teachers’ self-perception of teaching mathematics is closely related to their responsibility for students’ achievement. Through this study, we intend to evaluate the psychometric properties, the factorial structure, and the measurement invariance of the Mathematics Teaching Efficacy Scale (MTES). A total number of n = 621 early childhood teachers, employed in typical or special education, participated in the study. They were administered two self-reporting scales, the MTES and the Responsibility for Student Achievement Questionnaire. Oblique and bifactor EFA in a random sub-sample of teachers ( n = 207, comprising $$\frac{1}{3}$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mfrac> <mml:mn>1</mml:mn> <mml:mn>3</mml:mn> </mml:mfrac> </mml:math> of the entire sample), supported the two-factor structure. The two competing models obtained from the previous analyses were tested in the remaining sample ( n = 414). Both models had an acceptable fit according to the CFI and the TLI, however, the bifactor model demonstrated a better fit. Afterward, we evaluated the measurement invariance across special and typical education teachers. All nested models were evaluated across the teachers’ profession and demonstrated a good fit. Statistical analysis supported that the bifactor model can explain the existing variance rather than the two-factor model. Moreover, no ceiling or floor effects were observed. The structural equation models suggested that the “Responsibility for students’ achievement” predicts “Mathematics Teaching Outcome Expectancy” better than “Personal Mathematics Teaching Efficacy Belief” and “Mathematics Teaching Efficacy”.