Some of the Errors in John Donohue’s Research on Guns

There are many errors of different types in John Donohue’s research.

There are some errors in the data set that ADZ produced. For example,

• The observations for county 2060 in Alaska are repeated 73 times for 1996. .

• The first full year of the shall-issue law for Kansas is coded as 1996 when in fact the law

was not passed until 2006.

• The first full year of the shall-issue law for Florida is coded as 1989 when in fact the law

was passed in 1987.

• The first full year of the shall-issue law for South Dakota is coded as 1987, however the

law was passed in 1985.

As can be seen from the last three columns in Table 1, the errors in ADZ’s data set biased the

results towards what they wanted to show.

Moody, Lott, Marvell, and Zimmerman, “Trust But Verify: Lessons for the Empirical Evaluation of Law and Policy,” Social Science Research Network, January 25, 2012.

Other papers point to biases in the reported standard errors and omitted variable bias.