0
$\begingroup$

I have several measures on different memory tests. I consider these measures may actually measure different aspects of memory functioning and every measure contains a measurement error. I am thinking about two strategies. First one is using one measure as instrument variable. If errors are assumed independent, then IV gives a noise free measure. The second option is using factor analysis. I will derive the first factor and use it as an error-free measure.

Are these two methods valid? Thanks a lot.

$\endgroup$
1
$\begingroup$

Your question is somewhat unclear to me, but I believe that factor analysis is the way to go. In case, if there is no substantial prior theoretical knowledge (that is, your research is exploratory in nature), I would perform exploratory factor analysis (EFA) first. That would give your ideas about what constructs exactly are measured by your indicators (some might overlap or even measure the same thing). Then, if desired, you could perform a confirmatory factor analysis (CFA), which would validate your EFA results and hypothesized measurement model.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.