[lbo-talk] gun play

Wojtek Sokolowski sokol at jhu.edu
Wed Apr 30 13:34:29 PDT 2003



> Chronicle of Higher Education - web daily - April 30, 2003
>
> Scholarly Debate Over Guns and Crime Rekindles as States
> Debate Legalization By DAVID GLENN

That statistical correlation does no prove causation is Research Methods 101. In order to claim any effect on crime rate reductions, one need to show a causal connection between the two, not just correlations between statistcal aggregates. Such connection would exist if crime victims (or potential crime victims) were using weapon, or more specifically guns, to protect themselves when attacked. However, National Crime Victimization Survey which is the most comprehensive data source on all (reported and unreported) vicitmization contradicts that. For example, in 2001, only 1.1% of all victims of attempted or completed violent crimes used a weapon (not necessarily a gun!) as a protective measure, and 2.% threatened to use a weapon (you donot need to have a weapon to make a such threat). In othere words, weapon (again, not necessarily a gun) is very seldom used inself-defence, and therefe it is unlikely to be a couse in ant reduction in crime rates.

Wojtek



More information about the lbo-talk mailing list