"Clayton E. Cramer" <[EMAIL PROTECTED]> writes:
> > > > > Not quite. They looked at raw death rates (all causes) and
> > > > > concluded that there hadn't been a population change in DC during
> > > > > that period. That's clearly wrong.
> > > >
> > > > What is wrong is your claim that they did not look at rates per
> > > > 100,000 population. They did. And I've even provided the figures for
> > > > you.
> >
> >
> > > No, you provided the rates from their article, which are raw counts
> > > per month.
> >
> > No I didn't. They were rates per 100k population. The DC ones were
> > age-adjusted as well.
> I don't have a copy of the article where I can find it. If it were
> just me, I would wonder if I completely misread their article. But
> I find that others read Loftin and McDowall's article and came to
> the exact same conclusion--they only used raw homicide and suicide
> deaths per month, and did not adjust for population changes.
Well, they're wrong too. I have the paper in front of me. Appendix A
contains the results from the raw data. Appendix B contains the
results using rates per 100k population. I posted them. You
insisted, apparently without even looking at them, they I had posted
raw counts. I am baffled as to why you continue to make such false claims.
> > > They even made a point of saying in the article that they recognized
> > > that raw counts might not be sufficient, and even claimed to have
> > > "conducted a similar analysis using annual mortality rates."
> > > Loftin, McDowall, Wiersema, and Cottey, 1616.
> >
> > That's right. And those were the figures I posted, from Appendix B of
> > their paper. Appendix B is in the four pages of supplementary
> > material referenced as NAPS document no 04909 mentioned in footnotes
> > on pages 1617 and 1618. I have the appendix right in front of me as I
> > type.
>
> Was this not in the published article?
The appendices were not published in the NEJM, but as NAPS document no
04909. In their NEJM article they wrote:
"The analysis of annual mortality rates gave results similar in
general pattern to those of the analysis of the monthly data. The
Box-Tiao estimates are available elsewhere. In the District of
Columbia the rates of both homicides and suicides by firearms
declined in the period after the law went into effect (p<0.001 and
p=0.085, respectively); at the same time, the rate of homicides
committed by other means increased (p = 0.082) and that of
suicides by other means did not change (p=0.653). In the
surrounding metropolitan area there were no significant changes in
the annual mortality rates."
How did you manage to miss that passage?
> Can you put the article up somewhere where I can see it?
No. You should be able to get it from the library. I already posted
the figures from appendix B.
> It almost sounds
> like they added something to their article after the NEJM published it.
No, they didn't.
> I don't much care if Loftin and McDowall have some numbers somewhere
> that claim that there was a real reduction in homicide and suicide
> rates. Unless they can either demonstrate that the population
> figures from the sources cited above are wrong (and the population
> of DC did NOT fall by 15% during this period), or repudiate their
> published raw count numbers, the homicide and suicide rates
> experienced only trivial reductions.
Your figures do not adjust for demographic change. I'll go with their
age adjusted rates, thank you.
--
Tim
