| GunCite Home|

John Lott responds to criticism from Stephen Teret JD MPH of the Johns Hopkins Violence Research Group

(This information is excerpted from this URL. Edgar A. Suter, MD, is Chair of, Doctors for Integrity in Research and Public Policy.)

1. Lott and Mustard (L&M) do not take into account rates of arrest or convictions in their samples. (i.e., if arrest rates and conviction rates were going up before the scope of the study, then through the theory that the best deterrence is certainly of punishment, a theory I think we all agree on, violent crime rates may have gone down anyway.)

This is a strange comment. We do control for arrest rates in all the regressions (something no other gun control study has done), and when data is available on the conviction rates we also control for that.

The general changes in crime rates is not a problem for our paper since we control for individual year dummies which take out any year-to-year changes that are occurring in crime rates.

2. In addition to arrest and conviction rates, L&M don't take into account other factors that might account for rising or falling crime rates. Were more cops put on the streets? A new anti-drug initiative that may have worked? Community policing? Their samples are so widespread that it would be impossible to draw any conclusions without ensuring that each area shares the same characteristics, something they did not do in much detail.

We also control for police employment and payroll. We control for changes in several different types of gun laws. If you can think of some other law that just happened to change in all these different states in the different years that they adopted the "shall issue" law, I would be most interested in hearing what it is. We tried circulating the paper to gun control advocates before we released it, but out of all the people that we sent it to only one gave us more than the most perfunctory comments. Anything that was brought up to us we controlled for.

3. They admit outright that demographic trends have changed, but refuse to admit that it might affect their study. They acknowledge that the population percentage of black males aged 10-19 has gone down significantly. If this segment of the population was responsible for a disproportionate amount of murders, then the proportion of murders would change significantly...

We control for the % of the population that is black males between 10 and 19, the % of the population that are black males between 20 and 29, the % of the population that are black males between 30 and 39, the % of the population that are black males between 40 and 49, the % of the population that are black males between 50 and 64, the % of the population that are black males over 65, and the same break downs for white males, black females, white females, other males, and other females. No study of crime has had as detailed a breakdown of demographic changes as this study.

4. (Taken right from their study): "Using county level data has some drawbacks. Frequently, because of the low crime rates in many low population counties, it is quite common to find huge variations in the arrest and conviction rates between years."

Their solution is "to limit the sample to only counties with large populations. For counties with a large numbers of crimes, these waves have a significantly smoother flow of arrests and convictions relative to offense," but this again ignores the 3 points I raised above.

They say that quot;an alternative solution is to take a moving average of the arrest or conviction rates over several years," but then go on to say that this "reduces the length of the usable sample period, depending upon how many years are used to compute this average. Furthermore, the moving average solution does nothing to alleviate the effect of multiple suspects being arrested for a single crime."

Apparently you think that everyone who works in this area is an advocate for a particular position. We were not. We attempted to openly point out both the strengths and weaknesses of the data that we used. I would also point out that we used both state level and county level data and obtained fairly similar results with respect to the "shall issue" variable for violent crime.

5. (Also straight from their study): "The only real effect from making concealed handguns legal could arise from people being more willing to use handguns to defend themselves, though this might also imply that they more likely to make mistakes using these handguns."

Thus, they admit there is a possibility that accidental deaths will increase under right-to-carry laws.

When one does hypothesis testing, assuming one is objective, one should, to the best of their ability, clearly state the strongest hypotheses that exist on either side of a debate. The point is then to test these hypotheses. We did that and the evidence rejects the hypothesis advanced in point 5.

| GunCite Home|