Dear researchers,
Please help me with the following question. Can we conduct a factor analysis (EFA or CFA) for a scale that measures problem solving, decision making kind of constructs. where the items are developed as vignette questions with closed ended options with one correct answer (i.e. the scoring will be 0 and 1). Also please share any previous materials like research papers or books that can explain me this question as I could find only likert scales in factor analysis.
Thanking you in advance
Hello Anitha,
the common factor model assumes that several behaviors (e.g., respones, tasks) are due to one underlying cause. If this fits your vignettes, then of course you can do it.
All the best,
Holger
Factor analysis can be applied to dichotomous items. You typically want to use tetrachoric correlations (rather than Pearson correlations / covariances) as input and use specific estimation methods for categorical data (e.g., WLSMV in Mplus). An alternative are methods/models of item response theory that make more complete use of the information in dichotomous item data.
Thank you Dr. Holger Steinmetz and Dr. Christian Geiser for your inputs. these are very helpful.
- Badges
- Science topic
- Similar topics
- Biological Science
- Bioecology
- Systems Ecology
- Landscape Ecology
- Connectivity