Abstract
Effect of Covarıate(S) on Model Data Fit in Confirmatory Factor Analysis
In this study, it was aimed to examine the effect of covariates on model data fit
in confirmatory factor analysis, which is frequently used in scale development and
adaptation studies. In terms of investigating the effect of covariate in confirmatory factor
analysis, this type of research is a basic research. Research is composed of students who
participated in PISA 2018-Turkey app. The data were obtained using various scales in
the student questionnaire of the PISA 2018 application. In data analysis; McDonald ω,
Cronbach α, confirmatory factor analysis and multiple indicator multiple questionnaire
of the PISA 2018 application. In data analysis; McDonald ω, Cronbach α, confirmatory
factor analysis and multiple indicator multiple cause modeling were used. According to
the findings obtained within the scope of the research, it was found that the model-data
fit obtained as a result of the confirmatory factor analysis was worse than the model-data
fit obtained as a result of multiple indicator multiple cause modeling. Accordingly, it is
thought that not only responses to scale items but also individual variables (covariates)
should be taken into account in scale development and adaptation studies. As a result of
the inclusion of the covariates in the analysis, the possibility of underrepresentation of the structure decreases and thus the validity of the measurements will be contributed.
Keywords
MIMIC Model, Confirmatory Factor Analysis, Covariate, Validity, Reliability.