--- In [email protected], "authfriend" <[EMAIL PROTECTED]> 
wrote:
>
> --- In [email protected], "L B Shriver" 
> <[EMAIL PROTECTED]> wrote:
> <snip>
> > PANDITS HAD BEEN PART OF THE  ORIGINAL PROTOCOL.
> 
> They were not any part of the protocol that was made
> public, nor any part of the discussion and planning
> by the independent review board that was made public.
> I got on the mailing list for everything that was
> released about the study from its early stages, and
> there was no mention of pandits anywhere in it.

I was getting stuff from MIU faculty well before the study was 
publicly announced and it WAS part of theprotocol.

> 
> > They had been bought and paid 
> > for. Then they didn't show. So the group that participated was 
not 
> > as powerful as the group that had originally been anticipated.
> > 
> > After the scaling back of original reports claiming 25% reduction 
> > (might have been 20% come to think of it),
> 
> Yes, it was 20 percent.
> 
>  there was an ongoing effort of several months to make the data 
fit. 
> > My graduate student friend Mark ______ (last name still not 
> remembered) was a part of 
> > this. I had a standing joke with him about it: whenever I bumped 
> into him I would ask, 
> > "Seen any good statistics lately?" Then he would give me an 
> informal update. Let me be 
> > clear that this was not a conspiratorial relationship. Mark was 
> completely sold on the 
> > program and convinced that the correct interpretation of the data 
> would reveal the results. 
> > I was just an innocent bystander. Sort of.
> > 
> > Since I was not recording all the details for posterity at the 
> time, only the impressions 
> > remain. The impressions indicated that it took quite an effort 
> to "rectify" the findings 
> > based on their original model. I do not remember a single 
> alteration or adjustment, but 
> > something more like a scavenger hunt.
> > 
> > It is interesting to me how we are all quibbling about the 
details.
> > If anything is revealed here, it is that the "demonstration" 
> > demonstrated nothing. Except, perhaps, to the participants.
> 
> Even the *raw data*--the crime rate statistics--showed
> a very significant reduction from the rate the previous
> year for that period, considerably more than would have
> been expected from the overall crime trend.

I don't know that "very" significant was the appropriate term. At the 
least, the rawdata was in the predicted data.

> 
> What's more, that reduction occurred only during the
> demonstration period and for a few weeks afterward.
> Then it went right back up.
> 
> One of the problems the researchers encountered was
> obtaining the crime data in the way they had
> originally anticipated.  They had apparently been
> told by law enforcement (FBI or DC police, not sure
> which) that they would get it in a certain form, 
> broken down into certain categories, and they
> constructed their methodology around that understanding.
> 
> Whether they misunderstood or had been misinformed isn't
> clear, but a good deal of the fumfing around they had to
> do afterwards involved redoing the analysis to deal with
> the form in which they *did* get the data.  Plus which,
> there was a long delay in obtaining one major part of
> the data.
> 
> I don't remember the details, just the general outline.
> Some of this may be described in the study itself.
>






------------------------ Yahoo! Groups Sponsor --------------------~--> 
Get fast access to your favorite Yahoo! Groups. Make Yahoo! your home page
http://us.click.yahoo.com/dpRU5A/wUILAA/yQLSAA/JjtolB/TM
--------------------------------------------------------------------~-> 

To subscribe, send a message to:
[EMAIL PROTECTED]

Or go to: 
http://groups.yahoo.com/group/FairfieldLife/
and click 'Join This Group!' 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/FairfieldLife/

<*> To unsubscribe from this group, send an email to:
    [EMAIL PROTECTED]

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/
 



Reply via email to