3 ms·

This is a pretty disappointing list. First, the items do not tend to be mutually exclusive - there's a lot of overlap with each other. I interpret that as a s

This is a pretty disappointing list. First, the items do not tend to be mutually exclusive - there's a lot of overlap with each other. I interpret that as a sign of poor organization. It makes the whole thing hard to follow. Second, most of this stuff isn't remotely evidence based! Plus, there's no way to understand the effect size even when some technique does have evidence supporting it. Overall, this is too much of a data dump to be helpful. The author should have ranked ordered these items based on the cumulative evidence supporting each one - that would have been extremely useful.

BenjaminTodd 9y ago
Hey, I clarified the situation in the intro: > In many cases, the evidence isn’t as strong as we’d like. Rather, it’s the best we’re aware of. We’ve tried to come to an all-considered view of what makes sense to try, given (i) the strength of the empirical evidence, (ii) whether it seems reasonable to us, (iii) the size of the potential upside, (iv) how widely applicable the advice is, and (v) the costs of trying. The details are given in the further reading we link to and the footnotes. > We’ve put the advice roughly in order. The first items are easier and more widely applicable, so start with them, then move on to the more difficult areas later. The order is also partly based on an upcoming analysis of which skills are most valuable. I agree some of the advice overlaps, but it's a little hard to avoid. Most of the areas are inter-related. I'd be interested in ideas on how to reorg.