I use an example to illustrate my question.
I have a model that explains choice of low fat vs full fat milk, that was actually bought in a store. We model it with a binary logistic regression.
The model parameters mostly stem from a questionnaire, some are established scales, like the New Environmental Paradigm (NEP) Scale, published in The Journal of Environmental Education by R. E. Dunlap and K. D. Van Liere (1978). We build one factor NEPValue, the larger, the higher the environmental concern.
For simplicity of this question assume we only look at the following model:
High Fat Milk Purchase (Yes/No) = b0 + b1*NEPValue + b2*OtherCovariate + error
One of the reviewer strongly argues us to debate, why we use these scales. We argue, this is because the literature and our hypothesis states "Higher environmental concern leads people to buy the more "natural" thing, high fat milk c.p. (i.e., when controlling for e.g. weight loss concern/fitness/lactose intolerance and a bunch more)".
But I wonder, what is the general rule: When would a scale introduce endogeneity? Is there anyone who could help me, by pointing to the relevant literature?
