That's right. I recall suggesting something for that but I don't recall what it was.

> I thought you did look at that and found no supporting evidence.  (Re: effect of bitcoin-usd price)

Eh, I remember it being tricky because it gets absorbed into the time effects.  So while I don't find any strong evidence for price mattering, I didn't find evidence against either.  I'd probably have to change the specification if I wanted to investigate that, because $$\log P_t Z_{it} = \log P_t + \log Z_{it}$$ and $$\log P_t$$ gets absorbed into the time fixed effects.

> Sort of a weird critique, considering that's orders of magnitude larger than most econ experiments.

I think the critique was more about generalizability than statistical power.  I am absolutely sure this is gonna be a critique a referee is gonna make.

SimpleStacker

what counts for quality on Stacker News might not be considered quality in the general population

Obviously this is correct, but it also implies that what got zapped on a more general platform would be different. So, it doesn't have anything to do with whether zaps encourage quality, as judged by the consumers.

Does anyone have any thoughts, or a better response to this?

We've had versions of this conversation several times and I think there is a real concern that a Keynesian Beauty Contest + sufficient ass-milkers will lead to an echo chamber. Right now we're protected by a fairly iconoclastic set of top zappers.

Another person pointed out that 200,000 posts and 11,000 users is tiny

Sort of a weird critique, considering that's orders of magnitude larger than most econ experiments.

I thought you did look at that and found no supporting evidence.

econ

Research in Public #13: Internal Department Feedback

> What would y'all have responded?

https://media4.giphy.com/media/v1.Y2lkPTc5MGI3NjExNmgxM3pqN2ZwZmg5M3A3ZmV4ZHU0eTRpbGMxaDV3c2UxYmIxZDVsdiZlcD12MV9pbnRlcm5hbF9naWZfYnlfaWQmY3Q9Zw/iMCdARp9vhOmF1EpUx/giphy.gif

> what counts for quality on Stacker News might not be considered quality in the general population

Obviously this is correct, but it also implies that what got zapped on a more general platform would be different. So, it doesn't have anything to do with whether zaps encourage quality, as judged by the consumers.

> Does anyone have any thoughts, or a better response to this?

We've had versions of this conversation several times and I think there is a real concern that a Keynesian Beauty Contest + sufficient ass-milkers will lead to an echo chamber. Right now we're protected by a fairly iconoclastic set of top zappers.

> Another person pointed out that 200,000 posts and 11,000 users is tiny

Sort of a weird critique, considering that's orders of magnitude larger than most econ experiments.

>  What say you all?

I thought you did look at that and found no supporting evidence.