How can I test CFA between formative and reflective measure?

[Eung Il Kim]

Hello

I tried to run cfa between formative measure and reflective measure.
In the attached file, I like to test CFA between a and b variables. I tried to link these variables but could not find the result with "can't link exogenous to the endogenous variable".
Like to hear some comments about this problem.

Thank you

[Phil Bromiley]

You didn't get a quick answer. You will increase your chances of useful answer by following the FAQ on asking questions – provide Stata code in code delimiters, readable Stata output, and sample data using dataex. We don't even know exactly what code you ran. It is certainly possible that your models are not identified.

I don't know what you mean to run the analysis between formative and reflective measures – your diagram has different exogenous variables for the two underlying constructs. So, I'm not sure what it means to compare them. Note also that it is quite possible that models with arrows pointing different directions can actually have precisely the same mathematical representation in likelihood function and so cannot be differentiated based on the data.

[daniel klein]

Edwards (2011) discusses some of the (severe) problems with so-called formative models. Rönkkö et al. (2015) express similar skepticism. It is not my area of expertise but I have not come across any strong counter-arguments, yet.



Edwards J., R. 2011. The Fallacy of Formative Measurement. Organizational Research Methods, 14(2): 370--388.

Rönkkö, M., McIntosh, C. N., and Antonakis, J. 2015. On the adoption of partial least squares in psychological research: Caveat emptor. Personality and Individual Differences, 87: 76--84.