I find Sendec's points regarding a proper challange of their conclusions and / or findings well put.
I guess part of my problem is that I'm not seeing much of a conclusion or finding. The
jointogether article doesn't have any conclusions other than that the number of suicides with firearms involved surprise some people. Several of the more "interesting" remarks are sourced as "...said in an interview". Which is a nice way of just tossing out something that can't be reasonably vetted.
The link to FICAP has a pdf of a resource monograph
here. (Click "Firearm Injury in the US" to get to the resource book download.) The resource monograph appears to be well cited until it gets to pages 27 through 30 where we get interventions and conclusions.
Maybe it's me, but I can't see where all the gruesome stats are actually used as a basis for the proposed interventions, other than occuring in the same publication. In fact on page 28 of the monograph, we get
The effectiveness of these potential interventions is largely unknown, but they provide reasonable strategies that merit further consideration and study.
In other words, they're guessing. The GOA would likely guess otherwise.
I don't see a conclusion to refute - at least not in the article originally posted.
I doubt I'll ever be able to take the mixture of epidemiolgical disciplines and inorganic non-replicating tools, but that's just me. Being old enough to have grown weary with junk-science causality claims has jaded me, perhaps. Real scientific proof of causality via epidemiology is tough and rigorous - proving non-causality is even tougher (else there would not have been the flap over artificial sweeteners, breast implants, toxic mold, etc, etc).
I share a respect for the scientific method. If I were to start crunching numbers, I'd likely start
here. It admittedly addresses only the juvi population but that's 25% of the target stats.
So, to actually address Sendec's concerns:
1. Agreed - we are not refuting their conclusions - we're griping about a generally undisciplined conglomeration of observations that we "think" is "heading somewhere" - the article doesn't really say anything at all.
2. The disease model is imperfect. However, we're not the ones trying to assert a cross-discipline causality. Epidemiology is only marginaly better suited to gun violence than, say, climatology (my opinion).
3. I always tread carefully around the "you can't prove a negative" maxim. However, I believe it applies here. You'll never be able to prove guns aren't to blame. Think Aspartame: 90 countries, 200 peer reviewed studies showing no problems and a half-dozen hyperventilating web sites with anecdotes get intelliigent people frightened of the stuff.
4. Proving bias IS relevant. As the other posters have pointed out, there is an obvious agenda shared by the funding organizations. If a study came out along the lines of "NRA data driven study shows guns are good" it would never have any credibility whatsoever. The same should apply to the Joyce Foundation - just check the charter.
Sendec, you may not agree with anything I'm saying, but at least I read what you're saying - sorry I didn't do it earlier.