Re: [SLUG-POL] Racial Profiling: was open source projects...

From: Robert Haeckl (rhaeckl@tampabay.rr.com)
Date: Sun Oct 14 2001 - 19:28:18 EDT


Paul M Foster wrote:
>
> On Sun, Oct 14, 2001 at 04:18:53PM -0400, Robert Haeckl wrote:
>
> > Paul M Foster wrote:
> > >
> > > On Sun, Oct 14, 2001 at 11:53:33AM -0400, Robert Haeckl wrote:
> > >
> > > > Paul M Foster wrote:
> > > > >
> > > > > On Sat, Oct 13, 2001 at 10:35:35PM -0400, Robert Haeckl wrote:
> > >snip
> > > I was thinking about this last night. It occurs to me that what we're
> > > calling "profiling" is really inductive or deductive reasoning
> > > (depending on the circumstances). Sherlock Holmes was the master of
> > > this. Scientific inquiry is pulled along by this as well. In the absence
> > > of plain and clear facts, one must rely on probabilities and tendencies.
> > >
> >
> > No, inductive and deductive reasoning have nothing to do with
> > statistics. If profiling is a screening mechanism, it must have known
> > qualitative and quantitative error rates. If you can find me even the
> > semblance of a scientific study that shows racial profiling as a
> > screening tool with qualitative and quantitative accuracy rates
> > approaching 90%, I'll give you the benefit of the doubt. But I'm sure
> > it won't be there.
> >
>
> Well, I don't have time to do a full web search and read everything
> there is on the subject to find the nugget you want. I did check the
> Academy of Behavioral Profiling, though (http://www.profiling.org). They
> seem to have quite a few reference materials on the subject. Are you
> saying that all profiling done by law enforcement agencies is based on
> false premises and is invalid? If so, they've made a considerable
> investment in something that doesn't work. Wouldn't be the first time,
> but.... You're singling out racial profiling. Is other profiling okay,
> just not racial profiling?
>
> I think Jim Wildman was correct-- we all profile other people and
> circumstances for any number of reasons. We don't normally call it
> "profiling", but that's effectively what it is. I would tentatively
> define it as, "The use of statistics or past experience to predict
> current or future events."
>
> Paul

No, I don't really expect you to do much research on it, because it's
not there. We don't hold our law enforcement to the same standards as
say, a laboratory. Cost and cooperation I guess would make it
prohibitive. Yes, there is "statistical accounting" that provides
composites. Composites are descriptive, but hardly stand up to any
measure of a good screening procedure, one that is both specific and
sensitive. If you get a diagnostic lab test done, you expect it to rule
out a cancer or at least be correct if it rules it in. If you're told
that the test misses half the occurences or may report that you have
cancer when in fact you don't, you may very well not want that test.
Profiling would be akin to a very poor screening test, except selected
individuals will have no choice to submit to it or not. If you want
profiling, than I say everybody should be profiled. At least then it
would be random. I'd submit, but I would no longer consider this a free
country but an evolving fascist state.

Maintaining a relatively intact free culture will require a continuing
worldwide campaign against terrorism, in my opinion. It was started 12
years ago, but there was no follow-through over the last 9 years, and
about 6000 people and innumerable families and friends just paid the
price. I don't want them to have lost their lives for a fascist state.
We have a responsibility to try to maintain this country as it is and
improve on it. Condoning any form of ethnic isolation or targeting is a
step backwards.

I am not against profiling, per se. Jim is obviously correct, we do it
all the time on an individual basis. But there is a very fine line
between profiling and socialized bigotry and we must be sensitive to
that. Heck, we're fighting against it.

-Robert



This archive was generated by hypermail 2.1.3 : Fri Aug 01 2014 - 19:43:54 EDT