Training vs. Gun Deaths

I'm a big fan of training for most things. Want to ride a motorcycle?  Not a bad idea to get some instruction first.  Want to skydive?  Might be smart to get instruction before jumping out first.  The funny thing is that firearms training for CWP permit holders does not seem to directly affect the rates of deaths or injuries.   Of course, Joan and the Brady Campaign think it does.  Joan has basically asserted, "More training = fewer gun deaths."  John Lott's research shows that "more training = fewer permits idssued = more crime."

I did a quick look at the stats provided by Joan to assess the validity of her hypothesis that "more training = less gun death."

This chart shows hours of training required to get a CWP on the X axis vs. rate of firearms deaths on the Y axis.

Data Sources:

• Hours of Training.   John R. Lott Jr.. More Guns, Less Crime: Understanding Crime and Gun Control Laws, Third Edition (Kindle Location 955). Kindle Edition. Lott did not include data for several states.  I thus surrogated as follows:  CA (4 hrs = minimum for renewal), IA (8 hrs = NRA basic pistol qualifies), KS (8 hrs = training req'ts unclear, I assume NRA basic pistol qualifies), NEBRASKA (8 hrs = NRA basic pistol qualifies).  Illinois and Wisconsin are excluded as they were "no issue" at time of data.  Delaware, Hawaii, Maryland, Massachusetts, New Jersey, New York, and Rhode Island are marked as "24 hours" as they are discretionary "may issue."  Who knows what training is necessary if you haven't donated enough to the local police ball fundraiser?
• Death Rate:  This is the "Death Rate" as calculated by VPC based on WISQARS data from the CDC 2008 based on a citation from Joan Peterson, Brady Campaign Board Member.  I'm sure that it includes suicides, LEOs and civilians committing justifiable homicides, murders, accidents, etc.

Visually looking at the data there is a very small negative correlation:  i.e. as training requirements for CWPs increase, the rate of firearms deaths appears to decrease.  Sadly, I have Excel Starter right now and it does not appear to be able to do a linear regression.  Still, looking at the correlation, it appears very weak.  This makes sense:  CWP holders are a small percentage of the population.  Why would increasing the training burden on a small fraction of individuals impact the leading cause of gun deaths, i.e. suicide?  How does decreasing the training requirement to carry a concealed handgun in Alaska to zero (Constitutional Carry) result in more people in remote native villages eating a shotgun?

I suspect that there is some sort of lurking variable here.  For example, it is rather plausible that the states with high training requirements (or straight up may issue or no issue) have low rates of firearms ownership which leads to lower rates of firearms death.  This may not correlate with lower rates of criminal activity or homicide or suicide, and indeed, probably does not.  It just means that in anti-gun Illinois suicidal people are more likely to hang themselves than to eat buckshot.

So while I fail to reject the Brady hypothesis based on this quick data analysis, I suspect the Brady hypothesis would fall apart under a bit more scrutiny.  This weak analysis shows, at best, a very weak correlation (which has no bearing on causation, of course).

UPDATE:  I'm not suggesting that this is an appropriate metric to use, either.  Imagine if you compared "Hours of training required to obtain a commercial driver's license for a tank truck" to "all fatalities involving automobiles."  I would expect the former to have only a small correlation, at best, to the latter.  I'm just using statistics provided by the Brady Campaign to show that even in the most favorable interpretation they don't really support the Brady position.