uchicago-computation-workshop / ben_golub

0 stars 0 forks source link

Discrepancy between goals of Indian government and study's measured outcomes. #16

Open shugamoe opened 6 years ago

shugamoe commented 6 years ago

You didn't show a statistically significant difference (p=.104) between (Seed, CK) and (Broadcast, CK).

Parts of your whole paper, like the title seem to tout the relative effectiveness of Seed vs. Broadcasting. However, I don't see any government caring about "conversation volume" or "knowledge" versus simply making a "higher quality choice" (choosing the Rs. 500).

On that note, I think a reasonable goal for the Indian government to have in this case would simply be something like: "Make sure everyone with currency that we are going to make worthless redeems its value before the clock runs out." However, out of the outcomes measured in your study, the quality of the incentivized choice is the only thing that seems to come close to this. (The government doesn't care about conversations, or the "grades" of their "knowledge", as long as people make the right choice. At the very least, the right choice takes priority.)

TLDR; How would you convince a government to adopt a seeding vs. broadcast strategy given the discrepancies between the outcomes you measured and the governments actual goals, especially given the lack of a significant difference between (Seed, CK) and (Broadcast, CK) for the outcome that I think most closely aligns to a government's actual goals.

shugamoe commented 6 years ago

Don't be a lazy POS and upvote the earlier questions just because they are earlier. 287qwr

shugamoe commented 6 years ago

last-minute-enough-upvotes-approved