by Metsfanmax on Thu Dec 27, 2012 12:55 am
saxitoxin wrote:I reject the results of this study for 3 reasons: 1- the study was limited to 677 people
- sample set is too small
What would be an appropriate sample size for such a study? I am not sure, but at most this means there are significant error bars on that ratio of 4.5. I really doubt that the error bars approach the 100% level.
2- the study was limited to residents of Philadelphia
- sample distribution is too small
- Philadelphia has an atypical crime rate (http://blogs.phillymag.com/the_philly_p ... te-matter/)
- Philadelphia is an urban area; sample distribution limits can't account for different gun ownership and use trends in rural and suburban areas
- Philadelphia is a Northeastern city; sample distribution limits can't account for different firearms use cultures in western states
These aren't reasons to reject the study, they're just reasons to not generalize the results to outside of Philadelphia.
3 - most importantly - the study did not attempt to explain a difference in injury rates between persons lawfully possessing a firearm and those unlawfully possessing one (most likely a person unlawfully possessing a firearm will inherently be at greater risk of violent assault than a person lawfully possessing one or a person possessing no firearm at all; persons who have access to black market sources for firearms, logically, are moving in a different and more dangerous segment of society than the average person)
The 4.5 times more likely result was given after already having taken into account general characteristics that made a person at higher risk to be involved in gun violence. It does not seem to have taken into account the difference between legal and illegally owned guns, but I would argue that this is subsumed in the general "high-risk" category they describe in the paper.
Last edited by
Metsfanmax on Thu Dec 27, 2012 1:00 am, edited 1 time in total.