Sunday, March 19, 2017

Why Public Opinion Surveys Require Large Bags of Salt

Pew’s 2016 survey asked about bans on high capacity magazines and “assault-style weapons.”  In spite of several decades of mass murders often inaccurately reported as committed with such weapons, only 54% supported such bans.[1]  Yet Gallup in 2016 found only 34% supported such bans.[2]  Regardless of which is correct (and the enormous disparity on this question raises serious questions about the validity of the polling methods), the gun culture includes a substantial part of American society.

Other surveys seem to indicate declining gun ownership.  The National Opinion Research Center’s General Social Survey reports a decline in gun ownership from 47.0% to 31.0% from 1973 to 2014.  While the percentage refusing to answer the question rose from 1.0% to 3.2% during that period, this is not enough to explain the 39% in the 2016 Gallup survey and 31.0% in the NORC poll.[1] 

[1] NORC, “Trends in Gun Ownership in the United States, 1972-2014,”, last accessed March 7, 2017.

[1] Pew Research Center, “Opinions on Gun Policy and the 2016 Campaign,”, last accessed March 7, 2017.
[2] Art Swift, “In U.S., Support for Assault Weapons Ban at Record Low,”, last accessed March 7, 2017.


  1. The GSS has a confusing question: It asks "Do you happen to have in your home (IF HOUSE: or garage) any guns or revolvers?".

    That implies handgun ownership to people who are familiar with guns. After all, if someone asked you if you had a "motor vehicle or a convertible", would you necessarily answer yes if you had a motorcycle or pickup truck? It's a poorly worded question. Gallup's question is much better because it merely asks about a gun, not a "gun or revolver". And Gallup's question has remained steady at about the 40% rate since around 1996, bouncing up and down.

    It's also interesting to note that Gallup's results dropped from 51% in October 1993 down to 38% in July 1996. What happened during that time? Oh, the Brady Law, the Assault Weapons Ban, Gun Free Schools Act, talk about Brady II, etc. Prior to 1993, the number averaged about 50% of households.

    It's my contention that there is a built-in bias against admitting that you own a firearm in about 20% of gun owners that dates back to the actions of the Bill Clinton's first term. That's the only way to account for that drop. If you assume that the average gun owning household had just 2 guns, and that the US had about 96 million households, the number of gun owning households dropping from (96 * .51) = ~49 million to (96 * .83) = ~36.5 million, that's about 25 million guns flooding the used market, at a time when gun manufacturers were running flat out producing new guns.

    There's a book in there somewhere.....

  2. I aim to please. It's just something I noticed a while back while discussing this online. We used to talk about this kind of stuff back on talk.politics.guns