Reply #2: Shooting is Zen. FWIW, the NORC is one of the most respected social surveys in the USA

[benEzra]

and are affiliated with the University of Chicago, hardly a bastion of pro-gun sentiment.

http://www.norc.org/channels

From one site describing the survey, its funding, and its methodology:

THE GENERAL SOCIAL SURVEY

2005
Susan Carol Losh PhD
Department of Educational Psychology and Learning Systems
Florida State University
Tallahassee FL 32306-4453
[email protected]

What does America really think, whether it is about prejudice--or anything else? One invaluable source is the General Social Survey (GSS) which began in 1972. Once annual, it is now conducted every other year. The GSS is a representative public opinion survey of adults in the lower 48 United States and is part of what is called “social indicator research.” It is mostly funded by the federal government (the National Science Foundation).

The GSS is a state-of-the-art poll. It uses probability sampling and, unlike most other modern surveys, it utilizes in-person surveys rather than telephone interviews or Internet questionnaires. Its response rates are well over 70 percent and among the highest in the world. The original GSS questions were repeated from earlier work by Gallup, Roper, Harris, The Survey Research Center (Michigan) and other researchers to enhance over-time comparisons. Because many earlier questions had flawed wording, the GSS tests alternative item formats in an experimental, "split-ballot" method. This allows assessments of the difference question format can make and whether new question construction can substitute for the old with little loss of meaning.

Annual case bases range from 1300 (when the GSS was done annually) to over 3000. Currently, the GSS conducts about 3000 interviews every two years. Case bases vary on specific questions if split-ballot questions were used, but adjacent years may be combined to enlarge the case base. Interviews last about an hour. The most recent available is the 2002 survey. The 2004 survey will be released later in 2005 and plans for the 2006 General Social Survey began in 2004. I am proud to be a consultant on one of the 2006 GSS modules.

About one-third of the GSS questions are repeated every time. These are called "the core". Examples include detailed background characteristics: gender, age, ethnicity, religion and religiosity, income, occupation, presidential voting histories, and marital and family statistics. Several "social indicator" questions are also usually asked, for example: attitudes toward trust and society, or confidence in social institutions. Usually three topical modules comprise the remainder of the survey. Prior examples include attitudes toward gender and society, the military, and friendships. Modules in 1990, 1994, and 2000 (using some of the same questions) measure intergroup tolerance and prejudice. Items are designed to be intelligible to a national sample of respondents. While these may neglect specific local concerns, and samples of some groups (e.g., Southern Jews) are too small to generalize, the GSS allows us to "photograph" the mind of the general American public year after year. It is a national resource. Quite simply, there is no other comparable set of data.

I expect that some will dismiss the NORC survey results because the OP cited the WSJ instead of DailyKos or whatever. But the NORC survey itself is well respected by progressives, and it is actually one of the major indicators used by sociologists.

If anything, because NORC uses non-anonymous, in-person interviews, it may actually underestimate household gun ownership by a few percent (Hi, I'm from the government, do you own any guns?) but on the whole it's one of the best data sets out there on social factors.