Sorry, I didn't realize that length was an issue -- have trimmed the message and re-sending.
Responses interlaced and underlined. ________________________________ From: GunCite <[email protected]> To: Phil Lee <[email protected]>; "Olson, Joseph E." <[email protected]> Cc: "Firearms Reg, List" <[email protected]> Sent: Monday, March 25, 2013 11:30 PM Subject: Re: High, very high, rates of homicide and suicide in Black communities. 1. In addition to the mortrate10 link the CDC has similar data covering an earlier period at http://webappa.cdc.gov/sasweb/ncipc/mortrate9.html There is a difference in classification rules (which I don't remember) between the two pages that requires care. The difference is explained in the second link, but that data wasn't used by the study in question. Comment was for info, but duration of data can create a statistical significance issue. 2. The CDC data may be worthless -- I discovered significant differences between their counts of total firearm homicides in Maryland and the counts from the Maryland state police -- see Table 1 in my posting http://www.mcrkba.org/JHU_Study.pdf Is it worthless with respect to total homicide rates? I've used FBI data as well and get similar results for total homicides. Don't know for certain about CDC total homicide data (that's why I used "may"), but believe Maryland Police data and not CDC. To check would require state by state comparisons which are beyond me now. You should know that the FBI data has many problems -- non-reporting areas and resulting estimate errors, initial classification errors and more. 3. However, this abstract makes two assertions that I view as suspect. The first is that they used "state-level survey-based estimates of household firearm ownership". FYI they used the BRFSS Survey. Results are here: http://www.schs.state.nc.us/schs/brfss/2001/us/firearm3.html Now anyone who has studied this problem knows that surveys of gun ownership are known to be inaccurate for a variety of reasons that I'm not going into here. Agreed on both counts. Previous to my response, you wrote, "Both sides use data selectively to make arguments and being aware of these differences can speed the finding of the data to support an argument, but I think the side opposing gun control has the better data --if only pro-freedom groups would make the investment in good statistical analyses."As I tried to allude to earlier, there simply isn't more specific and reliable information available. I don't know how a "pro-freedom" group is going to obtain or generate reliable estimates of firearms prevalence by state. Why should a pro-freedom guy be interested in firearm prevalence by state? You are looking at data relevant to proving firearm ownership is bad. But the concealed carry results published by John Lott makes clear the value of firearms to the good guys. We'll never get good data if we leave the seeking of data to the anti-freedom crowd. One problem is that the NRA opposes data collection -- they seem to prefer being ignorant than taking a risk that the data is the usual garbage or worse. Or, maybe they think most people are too dumb to understand. 4. If you have a scientific justification for your claim that "the racial/ethnic categories serve as a proxy for socio-economic factors and nothing more", I would like to see a cite for it. It would be a useful shield against a bigotry claim. While this claim is an article of faith among some, I've never seen a study on the subject. If you're looking for a bigotry shield, so to speak, here are a couple of references. Kubrin and Wadsworth, "Identifying the Structural Correlates of African American Killings," Homicide Studies, Feb 2003. http://www.gwu.edu/~soc/docs/Kubrin_identifying_structural.pdf "The present study extends the understanding of the structural determinants of African American killings by analyzing the impact of key socioeconomic and demographic factors on disaggregated Black homicide rates in St. Louis neighborhoods.... The findings reinforce the necessity of dis-aggregating homicide rates to understand the race-violence relationship." There are further citations in the article. Another source: http://hsx.sagepub.com/content/10/1/3.abstract The link contains the abstract only, but page two of the paper provides a gender and race blanket, "Neglecting racial differences in women's use of deadly violence is important because mounting evidence indicates wide racial disparities in homicide rates, with homicides committed disproportionately by African Americans The omission of these intertwined status (i.e. race and gender) when studying the ecological correlates of high rates of homicide suggest that criminologists place little importance on racial dynamics except as they apply to male patters of offending." From the same page, "Although researchers now acknowledge the significance of examining race-specific homicide rates (e.g., Blakwell, 1990; Hareer & Steffensmeier, 1992; Krivo & Peterson, 1996, 2000; Messner & Golden, 1992: Peterson & Krivo, 1993; Shihadeh & Flynn, 1996)..." If you want those specific citations, please contact me off-list. Obviously the above is not intended to be a complete response. If I correctly understood your request, it merely shows that studies examining homicide by race are not taboo in homicide research publications. The research may not be taboo, but the results may. Thanks for the references. ________________________________ From: Phil Lee <[email protected]> To: GunCite <[email protected]>; "Olson, Joseph E." <[email protected]> Cc: "Firearms Reg, List" <[email protected]> Sent: Monday, March 25, 2013 5:41 PM Subject: Re: High, very high, rates of homicide and suicide in Black communities. To some of GunCite's points: 1. In addition to the mortrate10 link the CDC has similar data covering an earlier period at http://webappa.cdc.gov/sasweb/ncipc/mortrate9.html There is a difference in classification rules (which I don't remember) between the two pages that requires care. 2. The CDC data may be worthless -- I discovered significant differences between their counts of total firearm homicides in Maryland and the counts from the Maryland state police -- see Table 1 in my posting http://www.mcrkba.org/JHU_Study.pdf 3. The cited paper by Miller is behind a pay-wall so I've not seen any of it but the abstract. However, this abstract makes two assertions that I view as suspect. The first is that they used "state-level survey-based estimates of household firearm ownership". Now anyone who has studied this problem knows that surveys of gun ownership are known to be inaccurate for a variety of reasons that I'm not going into here. The second is the statement that "Although causal inference is not warranted on the basis of the present study alone, our findings suggest that the household may be an important source of firearms used to kill men, women and children in the United States." As a scientific claim the use of the word "may" renders this statement worthless (e.g. that "the sun MAY rise in the west tomorrow" is equally valid). If they really intended to use "is" instead of "may", the statement would almost certainly wrong since the experience of Baltimore Police is that on the order of 95% of murders are drug related most of which by criminals who can't own firearms. To fail to distinguish between these killings and those accidental or passion related killings by non-disqualified people make the intent of this paper political, not science. The political intent is clear when you see one of the authors is David Hemenway. 4. If you have a scientific justification for your claim that "the racial/ethnic categories serve as a proxy for socio-economic factors and nothing more", I would like to see a cite for it. It would be a useful shield against a bigotry claim. While this claim is an article of faith among some, I've never seen a study on the subject. For the most part, the science and statistics for most reports on violence are not worth the paper they are printed on. Phil
_______________________________________________ To post, send message to [email protected] To subscribe, unsubscribe, change options, or get password, see http://lists.ucla.edu/cgi-bin/mailman/listinfo/firearmsregprof Please note that messages sent to this large list cannot be viewed as private. Anyone can subscribe to the list and read messages that are posted; people can read the Web archives; and list members can (rightly or wrongly) forward the messages to others.
