@sandals said in #30:
> Factor analysis is mostly fake. In this particular case, your 4 factors depend heavily on (a) the arbitrary choice of deciding to fit 4 factors (rather than 3 or 5), and (b) the rotation strategy (arbitrarily chosen to be "promax"). You would get vastly different factors if you changed one of these arbitrary parameters. The same thing happens regularly in psychometrics; factor analysis should be viewed with extreme skepticism.
Do you have alternative approaches that you think would be more appropriate? These aren't *my* factors, remember - I'm talking about what this research group did. Certainly happy to hear your ideas for how to analyze this dataset differently.
You're right that FA (and related techniques) require decisions about dimensionality and such, but there are also ways to check for robustness and try to make those decisions in principled ways, or at least with transparency. For my part, I think "mostly fake" is too strong, esp. if we don't differentiate between the technique itself and how researchers apply it.
Thanks for reading!
> Factor analysis is mostly fake. In this particular case, your 4 factors depend heavily on (a) the arbitrary choice of deciding to fit 4 factors (rather than 3 or 5), and (b) the rotation strategy (arbitrarily chosen to be "promax"). You would get vastly different factors if you changed one of these arbitrary parameters. The same thing happens regularly in psychometrics; factor analysis should be viewed with extreme skepticism.
Do you have alternative approaches that you think would be more appropriate? These aren't *my* factors, remember - I'm talking about what this research group did. Certainly happy to hear your ideas for how to analyze this dataset differently.
You're right that FA (and related techniques) require decisions about dimensionality and such, but there are also ways to check for robustness and try to make those decisions in principled ways, or at least with transparency. For my part, I think "mostly fake" is too strong, esp. if we don't differentiate between the technique itself and how researchers apply it.
Thanks for reading!