Research paper on Multiple Victim Public Shooting
UPDATE; A debate on that work can be found here. Thanks to John Fund and Cal Blalik for sending me this link.
Two comments and responses:
1) "Prof. Lott wrote in an email that he counted less-severe incidents to get enough data for statistically significant results. He justifies his exclusion of gang murders because gun usage by chronic criminals "would not be directly affected by the passage of right-to-carry laws."
That seems to be precisely the reason to include them for a full picture of the effect of these laws. Of course, the complete picture frequently goes missing in this debate."
-- The point is that we tried and reported it many different ways (including the measure used by the New York Times, two or more murders, three or more murders, four or more murders, various combinations of that with injuries as well as injuries separately, and the number of attacks). If someone had a reasonable suggestion for a measure, we would have used it. Out of all those ways, one way was not statistically significant and for what I think is a very obvious reason: the basic model had almost as many control variables as events. My question is: why is it better to try only one measure and find that is statistically insignificant as the other paper you cite does? Besides saying it is too much work, what is there reason for only trying one measure? They also include crimes that we provide a theoretical reason for why they will move the coefficient towards.
2) "Grant Duwe, a researcher on the later study, said the news-archive approach was likely incomplete, because the media don't always give publicity to multiple shootings."
-- Duwe and co-authors had a choice between doing the costly and time consuming media approach that we did or they could use the government data that includes cases that would include cases that shouldn't have been included. An important question to answer in evaluating Duwe's answer above is: Does one really believe that there are really many multiple victim public shootings involving non-gang members and where four or more people are killed that don't get any news coverage? These authors were sent our data before they did their paper and did not provide any evidence that our method missed the types of cases that we were concerned about. From having checked our data against the FBI reports (which we had done years before their work), I would argue that the only cases they picked up were gang shootings of other gangs. Something that can be interesting, but would not be particularly relevant for our estimates of the impact of right-to-carry laws.