I wrote about this in 2004:
Back in the 1950s, when the Gallup poll was almost the only game in town, it was rational to respond to the survey–you’d be one of about 1000 respondents and could have a reasonable chance of (indirectly) affecting policy. Now you’re just one of millions, and so answering a pollster is probably not worth the time (See here for related arguments).
The recent proliferation of polls—whether for marketing or to just to sell newspapers—exploits people’s civic-mindedness. Polling and polling and polling until all the potential respondents get tired—it’s like draining the aquifer to grow alfalfa in the desert.
I think the bigger issue is more that the FCC has failed at their duty to stop scam phonecalls.
The number of scam phonecalls is so overwhelming that I don’t answer my phone anymore for phone numbers that I don’t recognize. And I believe most Americans do the same.
Ethan:
Sure, but even non-spam polling is happening at an overwhelming volume. Too many polls—it’s just become too cheap and easy to call lots of people.
By some estimates, the average person receives an average of 28 robocalls per month. [1] On the other hand, look at polling. Say that pollsters collectively call a million people per year. Collectively that’s about than one call per American per year. I find it hard to believe that pollsters are more than a rounding error.
[1]: https://www.enterpriseappstoday.com/stats/robocalls-statistics.html
Was there a specific article/news item on the aquifer draining that sparked this post?
Ben:
It came from an email exchange with Art Owen at Stanford and Scott Keeter at Pew Research. Art is teaching a class on survey sampling and was having some discussion with Scott about different modes of surveying, and Art mentioned the idea of there being a “tragedy of the commons” with survey responses. Scott pointed to this 2019 paper by Thomas Leeper who raises the idea of survey response being a “a common pool resource” problem. This all reminded me of my blog post on the topic from 2004—but who ever cites old blog posts??—so I thought I’d repost, as people are still talking about the problem, nearly twenty years later.
But I guess the real-world physical problem of aquifer draining is much worse of a problem!
Makes sense. Thanks for article
“so answering a pollster is probably not worth the time.”
I wonder how you reconcile this with whether or not it is worth voting. You appear to support the idea that it is worth voting, and it is not clear to me that there is more at stake with a vote than responding to a poll (sad to say, in my opinion).
Dale:
See the first italicized paragraph above. I think that in the 1950s it was much more effective (in the sense of having a chance to influence policy) to answer a poll than to vote. Nowadays, though, there are so many surveys that I think voting is a more effective use of your time!
” I think that in the 1950s it was much more effective (in the sense of having a chance to influence policy) to answer a poll than to vote.” That seems very undemocratic to me – giving a number of randomly selected individuals a special influence. If it’s no longer like this, my first impulse is to think that that’s a good thing.
I agree with your last sentence – but I’m not convinced of the truth that it is no longer like this. There are myriad concurrent changes: due to gerrymandering and media influence the value of individual votes have declined; while the overabundance of polls have diminished their individual influence (as a number of people have noted) the attention to polls by politicians and the media can still make them influential (particularly as the half life of bad information keeps increasing); and the polarization of parties (in the US at least) seems to make tipping points in policy direction more pronounced (pure conjecture on my part). What I feel more confident about is that it is difficult to make any general statements about the relative value of voting vis a vis answering polls. I do the former but not the latter, out of a matter of principle not rationality.
This is the Prisoner’s Dilemma again. If the poll is worthwhile, it is rational to respond to it.
Too many polls.
Too many telemarketers.
Too much SUGging (Selling Under the Guise of taking a survey).
Too many spam calls.
Caller ID.
Too many surveys with poorly designed questions / poor design / obviously not pretested at all.
Surveys that are too long.
+1. I don’t think most people spend much time calculating the odds of affecting policy. Surveys have just become their own form of spam–a tax on every interaction with the outside world.
Jeff:
Regarding “I don’t think most people spend much time calculating the odds of affecting policy”: Agreed. Rationality is a theoretical concept. That said, I think it makes sense to consider rational odds, as I do think they exert some indirect pressure on behavior. Edlin, Kaplan, and I discuss this general point in Section 5.2 of our 2007 article.