So this comment is sort of based on politics/ideology in research, but basically 50% of good empirical papers (Nature, AER) don’t generalize, and when examined, narrow specific actionable claims usually dissolve or were never there in the first place.
In this case, it’s not surprisingly that models don’t apply for spending that is an order of magnitude greater, for sort of outlier candidates.
So this comment is sort of based on politics/ideology in research, but basically 50% of good empirical papers (Nature, AER) don’t generalize, and when examined, narrow specific actionable claims usually dissolve or were never there in the first place.
In this case, it’s not surprisingly that models don’t apply for spending that is an order of magnitude greater, for sort of outlier candidates.