Design and Validation of Questionnaire to Assess the Validity of Designed Models in the Humanities Using a Mixed Method

The main objective of the present study was to design and validate a questionnaire for evaluating the validity of designed models in humanities using a mixed-method approach. This research is applied in terms of its goal and exploratory mixed-method (qualitative-quantitative) in terms of data type, consisting of qualitative meta-synthesis and quantitative descriptive-analytical (correlation) components. The qualitative study population included all theoretical foundations and relevant literature in domestic and international databases, which were selected using purposive non-random sampling and systematic elimination based on the PRISMA model flowchart. A total of 18 scientific works were chosen. The quantitative study population included all professors and faculty members of humanities departments at universities in Tehran. Based on sample size calculations for confirmatory factor analysis and multi-stage cluster random sampling, 370 respondents were selected as the sample size. Data collection tools included systematic literature review for the qualitative section and a researcher-developed questionnaire with 34 items derived from the qualitative section for the quantitative part. It is worth noting that validity in the qualitative section was calculated using the 27-item checklist based on the PRISMA model, and reliability was calculated using Cohen’s kappa coefficient. Additionally, in the quantitative section, validity was assessed through face validity, content validity, and construct validity, while reliability was calculated using Cronbach’s alpha and composite reliability. Validity and reliability were confirmed in both qualitative and quantitative sections. Data analysis methods included thematic analysis for the qualitative section and descriptive and inferential statistics (confirmatory factor analysis) using PLS software version 3 and SPSS software version 23 for the quantitative section. The findings indicated that the 34-item questionnaire could be used to assess the validity of models from the perspective of experts. The questionnaire evaluates external validity through components such as purpose (4 items), research method design (4 items), controlling confounding variables (8 items), and alignment (8 items), as well as internal validity through components such as logical review (3 items), expert feedback (4 items), and sensitivity analysis (3 items).