New Survey of Gun Ownership Shows Bias Before Publication

A large online survey of gun ownership has been done by researchers at Harvard and Northeastern Universities.  It is not expected to be published until 2017.  Some preliminary results have been released, but only to two publications.

The choice of those publications is highly instructive of the motives of the researchers.  The publications are The Trace, funded by Michael Bloomberg to promote citizen disarmament in the United States, and The Guardian, the British publication that is highly antagonistic to widespread gun ownership and the Second Amendment. From theguardian.com:

The unpublished Harvard/Northeastern survey result summary, obtained exclusively by the Guardian and the Trace, estimates that America’s gun stock has increased by 70m guns since 1994. At the same time, the percentage of Americans who own guns decreased slightly from 25% to 22%.

The lead author of the survey is Dr. Deborah Azrael.  In a Salon interview in 2015, she revealed a number of preconceptions that could contribute to selection bias in the survey, the questions asked, or interpretation of the results.  From salon.com:

“What we know is that if a woman is going to be killed by a firearm, she’s most likely to be killed by a current or former intimate partner. What we know is where there are more guns, more women die,” Azrael explained. “That’s just incontrovertibly true.”

That contention is hotly contested.  It is easy to find counter examples.  For example, gun ownership in Chicago is quite low, but the death rate for women and children from being shot, is high.  Gun ownership in Vermont is high, but the death rate for women and children from being shot, is low.  Gun ownership in Brazil, Venezuela, and Jamaica is low.  Homicides, including homicides with guns, are very high. It is a complicated question that does not lend itself to simple analysis, and it clearly is not "incontrovertibly true".

The following were responses from Dr. Azrael that were  "condensed and lightly edited for clarity" in the Salon Article.

That’s right. The two big sources of data on self-defense gun use are the National Crime Victimization Survey and non-governmental phone surveys, like the survey we conducted, in which people were asked, “Have you used a gun in self-defense to protect yourself against someone?” What we know from those surveys is you get much bigger estimates [on the use of guns in self-defense] from the type of survey we did for a complicated set of reasons.

It is possible to break down any system into two categories.  But to claim that the National Crime Victimization Survey is one of two "big sources" for self defense gun use is misleading.  The survey does not ask a direct question about defensive gun use, which is why the Center for Disease Control considered its numbers to be an outlier from most surveys.  From the CDC (pdf):

Defensive use of guns by crime victims is a common occurrence, although the exact number remains disputed (Cook and Ludwig, 1996; Kleck, 2001a). Almost all national survey estimates indicate that defensive gun uses by victims are at least as common as offensive uses by criminals, with estimates of annual uses ranging from about 500,000 to more than 3 million (Kleck, 2001a), in the context of about 300,000 violent crimes involving firearms in 2008 (BJS, 2010). On the other hand, some scholars point to a radically lower estimate of only 108,000 annual defensive uses based on the National Crime Victimization Survey (Cook et al., 1997). The variation in these numbers remains a controversy in the field. The estimate of 3 million defensive uses per year is based on an extrapolation from a small number of responses taken from more than 19 national surveys. The former estimate of 108,000 is difficult to interpret because respondents were not asked specifically about defensive gun use.

Dr. Azrael points to some anecdotal evidence.  Anyone who follows these stories knows that mistaken identity shootings are a rare subset of all shootings.  At Gun Watch, we have recorded thousands of news stories on defensive shootings since 2005.  From Dr. Azrael:

At this point there are countless horrible stories about people being mistaken for someone else, misapprehension by a gun possessor that somebody is a threat to them and people dying as a result.

Having actively looked for such stories, I cannot recall ever finding more than five such cases in any one year, out of thousands.


Dr. Azrael gives this indication of her fundamental assumption about access to firearms:

To me the fundamental problem is that everything we know suggests that access to firearms increases the likelihood of death and injury. Disproportionately to women, disproportionately to children, but to everyone. The notion that somehow increasing access to guns to women is going to be an exception to that rule is, to me, unfounded.

Her assumption is the fundamental assumption that arguments for public disarmament in the United States are based on.  It is an assumption.  There is considerable evidence that it is not true. Dr. Azrael indicates she has never found any contrary evidence.  Her reading list must be highly limited or self censored.  She does not seem to have read Dr. Gary Kleck, or Dr. John Lott,  both highly respected and published researchers in the field.

Dr. Azrael's comments on women and children indicate a strange and unusual interpretation of uncontested facts.  All the literature shows that the vast majority of people murdered with guns, who commit suicide with guns, or who are victims of fatal gun accidents, are adult males. Murders and accidents are much more common among young adult males, with a substantial but minority subset of murder victims who are 14-17 year old male gang members.  Suicide with guns is overwhelmingly an older adult white male phenomena.

  Her surprising ignorance is reinforced with the following comment.

That’s an interesting framing of it and makes sense to me, the analogy to “stand your ground” and castle doctrine seem sound. I appreciate that you started out asking about community safety, because the question that doesn’t seem to get asked in any of these contexts is what happens to a community when more people start to carry guns?

 "What happens to a community when more people start to carry guns?" has been one of the most highly researched and written about subjects in the literature.  John Lott has devoted several books to the subject.

What is not disputed is that crime does not rise.  What is disputed is if the crime rate falls, and how much it falls.

This survey is unlikely to convince anyone.  It suffers from the major problem of all surveys about gun ownership.  A large segment of gun owners are unwilling to tell strangers whether they own a gun or not.  From prnewswire.com:

QUESTION: "If a national pollster asked you if you owned a firearm, would you determine to tell him or her the truth or would you feel it was none of their business?"

Gallup recently released a poll showing that gun ownership had declined from polls they had taken in an earlier time period. That number is inconsistent with the number of firearms that have been sold since President Obama took residency, but the difference can be answered by the Zogby Analytic question above. The poll indicates maintaining anonymity is a contributing factor

• 36% of Americans feel it is none of the pollster's business and that includes 35% of current gun owners 47% of Republicans and 42% of Independents  

If we take the lower number, 35%, and apply it to the lowest numbers for gun ownership from a national poll,  found in the General Social Survey (GSS), we see that it implies that gun ownership is up, not down.

The lowest number in the General Social Survey  is 32% of households.   For those of you who are challenged by algebra, bear with me.  If 35% of gun owners believe that gun ownership is something that is no business of national pollsters, is suggests that only 65% of gun owners would admit to gun ownership on national polls.  Thus, the true number of gun owners would be, using the lowest numbers from the GSS, 32% of households multiplied by 1/.65 or 1.54 X 32%, which comes to 49%.  49% happens to be about the number of households that reported gun ownership in the late 1970's.  That implies a higher level of per capita gun ownership, because the number of people per household has dropped by 19% from 1970.  In 1970, the number per household was 3.14.  In 2015, the number was 2.54. 

Trust in pollsters and the Government have both decreased dramatically in the last 50 years.  It is highly likely that the more restrictions on gun ownership are promoted, the less likely people would be to admit to ownership.

The new survey gives an estimate for privately owned guns in the United States.  Unsurprisingly, the estimate is quite low, at 265 million at the beginning of 2015.  If we apply the 1.54 correction for likely underreporting, the number becomes 408 million.  Latter in the Guardian article, is this nugget:

Manufacturing and import records suggest that more than 360m firearms entered the US market between 1899 and 2013, the new study’s authors noted. With gun sales spiking since 2013, some estimates would put the total number of American firearms today around 400m.

Another check is the number of guns added to the private stock in the U.S. since 1994.  The survey number shows 73 million more guns.  ATF figures show manufactured guns for the U.S. market, plus imports, minus exports, to be 137 million.  That does not include guns for the military.  The ATF figures are through the end of 2014.  The survey was conducted in early 2015, so they cover essentially the same time frame.  Yet the survey only catches 53% of what the ATF numbers show being added to the private stock.  This is consistent with significant underreporting to the survey.

I contacted Dr. John Lott.  He has not been sent a copy of the survey for peer review.

Dr. Lott published an article explaining the variability in surveys of gun ownership. He shows how the media has demonstrated bias in what surveys they promote. 

Update: His information is available in his new book War on Guns:

In fact, the media goes out of its way to find polls claiming that Americans are turning away from guns.  In their War on Guns they want to give the impression that gun owners are a small, fringe group.  Maybe they are hoping that this will have an impact on policy.  As General Social Survey director Tom Smith told me, a large drop in gun ownership would “make it easier for politicians to do the right thing on guns.”

©2016 by Dean Weingarten: Permission to share is granted when this notice and link are included.

Link to Gun Watch