I conducted an EFA on four newly created measures: 1) awareness of inclusive teaching practices, 2) confidence in inclusive teaching practices, 3) reflection on inclusive teaching practices, and 4) likelihood to implement inclusive teaching practices. Together there are 64 items and I have a sample size of 210 (the sample size is 420, but split in half to conduct an EFA on a randomized half and a CFA on the other half).
Question: since these are new measures, should I have instead combined the four measures and conducted an EFA on all 64 items? (I am following recommendations of looking a factor loadings, variables per factor, and communality, all which are high - I recently started a new EFA with all 64 items). Or are there methodological reasons that conducting a separate EFA on each of these measures is acceptable? If so, could you provide references that I could cite?
Much of what I am finding is because these are newly created measures that I should have originally completed an EFA on all 64 items (i.e. the four measures).
Thank you, -Vanessa
Dear professor Vanessa, I hope this finds you well. If you are dealing with four new constructs in a second-order model (like in developing a scale, I suppose), you could have the items distributed along the dimensions you propose, theoretically, or you can let the EFA make the grouping of the 64 items. However, there has been a discussion on whether to conduct a CFA after the EFA is needed. This is because they are different models by nature. In the past, we usually conducted the EFA-CFA approach because the EFA software did not provide adequate indexes to evaluate the best solution. I suggest you use the FACTOR software for the EFA analysis with the 64 items. Factor provides a lot of indicators to analyze the factorial solution adequately. See the following manuscripts and other materials:
Lorenzo-Seva, U., & Ferrando, P. J. (2006). FACTOR: A computer program to fit the exploratory factor analysis model. Behavior research methods, 38(1), 88-91.
Ferrando, P. J., & Lorenzo-Seva U. (2018). Assessing the quality and appropriateness of factor solutions and factor score estimates in exploratory item factor analysis. Educational and Psychological Measurement, 78, 762-780. doi:10.1177/0013164417719308
Timmerman, M. E., & Lorenzo-Seva, U. (2011). Dimensionality Assessment of Ordered Polytomous Items with Parallel Analysis. Psychological Methods, 16, 209-220. doi:10.1037/a0023353
https://psico.fcep.urv.cat/utilitats/factor/Download.html
https://www.youtube.com/watch?v=bAjAEDYUJ5c
Dear Vanessa Johnson-Ojeda
As these are newly created measures, you should conduct an EFA on all 64 items.
According to their Eigen-values and factor loadings, you should decide the number of factors/ constructs.
I hope it helps :)
- Badges
- Science topic
- Similar topics
- Methodology
- Research Methodology
- Research Papers