10 Comments

I think we do not prove causality. It just survives another disproof. That's what just happened.

Expand full comment

Agree on the latter point!

Expand full comment

One question I have here is why there were so many people in the sample who did not receive the CTC? The sample says "83% not child tax credit recipients". My understanding of the CTC is that it was (essentially) supposed to go to all parents, except for those with very high incomes. Any ideas?

Expand full comment

Thanks! I know there are many people who should be receiving it but aren't for a host of reasons. That does help answer why the sample split was 83/17 in this Data for Progress poll. https://money.yahoo.com/child-tax-credit-payments-are-missing-vulnerable-families-165515979.html

Expand full comment

It could be that the poll isn't representative of the share who got the CTC!

Expand full comment

Probability of contacting 14,000 individuals and 83% not receiving child tax credits; what if we assume 50% of 35.2 million families ~17 million individuals. .

Expand full comment

Well, the probability of being significantly off due to sampling error alone would be very small at 14k, but uniform bias is harder to predict. (Maybe parents with a lot of kids aren't sitting around taking online polls very regularly.)

Expand full comment

So, wait - ~17% of the electorate =35.2 million families? Number registered voters 2020 = 168.21 million, per https://www.statista.com/statistics/273743/number-of-registered-voters-in-the-united-states/. 17% of 168.2 M = 28.6 M. 28.6 million individual voters = 35.2 million families?

Expand full comment

Their poll says 17% of individual voters are CTC recipients... NOt sure how that translates to families though!

Expand full comment