Bryan Caplan  

Smoking, Social Desirability Bias, and Dark Matter

PRINT
Civil Disobedience: King versu... Try Harder or Do Something Eas...
At the IEA blog, Kristian Niemietz points out that expenditure surveys fail to detect most of the tobacco sales visible in national product accounts
For most goods, the two show broadly the same pattern: with small errors, what people profess to buy grosses up to what is really being sold in the country. But tobacco is a big exception. Less then half of the recorded cigarette purchases shows up in the Living Cost and Food Survey. In the US equivalent, the ratio is not even 40%.
Why is tobacco dark matterSocial Desirability Bias!

The mismatch between what smokers say in surveys and what they do in practice is a classic example of the difference between "stated preferences" and "revealed preferences". Social engineers love stated preferences. Opponents of big supermarkets, too, always have a survey at hand, indicating that the vast majority of residents in their areas would never set foot in a discounter. But once it is there, it flourishes.

There is nothing schizophrenic about this behaviour. When asked whether you would shop in a big supermarket in your area, of course you respond something like "No! Small, local shops are much more charming and personal" - because that is the socially acceptable thing to say. When you smoke, saying that you want to quit makes you at least a repentant sinner.

Now ask yourself: Is voting more like a national product account - or a consumer expenditure survey?



COMMENTS (5 to date)
Andrew_FL writes:
Now ask yourself: Is voting more like a national product account - or a consumer expenditure survey?

An insightful question. I would say for example the answer is obvious in the case of a poll, and it fits with my feeling, for example, that the President's "personal likeability" ratings in polls are inflated relative to how people really feel, and to a lesser extent probably his job approval ratings as well. Saying you don't like Obama at least personally is highly socially unacceptable. Why it would make you a racist!

However, I'm less confident this would be true of a national election. Or at least that the effect is as strong in a national election. The outcome of an election has consequences, an opinion poll does not. It determines whether they get what they'd actually prefer or not.

The counterargument is probably that for the individual voter, their decision has little effect on the outcome, but has probably a significant and lasting social effect....as long as people know how they voted. Which is probably why exit polls are not reliable.

J.D. writes:

Or maybe they survey respondent misheard the surveyor and thought he or she asked about Tomacco.

KLO writes:

Back before it was possible to measure metabolic rate with a high degree of accuracy, researchers relied heavily on self-reported food intake and activity data. The consistent result of the self-reported data was that the fatter the person the less they reported eating. This baffled the researchers, many of whom assumed that there was not any systematic bias in their data. From the data, researchers were forced to conclude that food intake and weight were inversely correlated.

Then researchers came up with a way to measure metabolic rate over long stretches of time. These data showed that fat people had higher metabolic rates than thin people, which meant either that fat people were getting their energy from something other than food (the sun perhaps) of they were underreporting what they ate much more than thin people. Alas, the latter appears to be true for the same reason that people underreport the amount they smoke.

EclectEcon writes:

One reason for the difference in the numbers is the apparently widespread practice of smuggling, at least in some areas:
http://www.eclectecon.net/2014/03/taxes-incentives-and-cigarettes-people-respond-to-incentives.html

ThomasH writes:

Voting (as opposed to polling) is more like the production accounts as it's more likely what you really think?

Comments for this entry have been closed
Return to top