Using exploratory factor analysis on data with pro max variation the KMO value is indicated as not positively defined. I have 56 variable and pilot study is for 137 respondents. Can anyone explain as to why it is not getting defined.
https://link.springer.com/content/pdf/10.3758/BF03192753.pdf
https://revistas.um.es/analesps/article/download/analesps.33.2.270211/210801/0
Hello Anoop,
The most likely explanation is that your set of variables (that you are trying to factor the scores from) has linear dependencies. That is, at least one variable is redundant with others in the data set.
You might try looking at pairwise correlations (an easy inspection). If that doesn't highlight obvious cases, then run collinearity diagnostics on the set of variables. That will help. In SPSS, collinearity statistics may be had from the regression subprogram. You will have to enter some other variable as the "dependent variable" in order to elicit the results, but just plug in all your variables of interest in as "independent variables."
Good luck with your work.
- Badges
- Science topic
- Similar topics
- Mathematics
- Algebra
- Matrix