--- In [email protected], "authfriend" <[EMAIL PROTECTED]> wrote: > > --- In [email protected], "markmeredith2002" > <[EMAIL PROTECTED]> wrote: > > > > --- In [email protected], "authfriend" <[EMAIL PROTECTED]> > wrote: > > > > > > The crime rate reduction found by the D.C. study was a > > > sharp downward turn that correlated very closely with > > > the start of the eight-week project, not the kind of > > > gradual reduction over time you're talking about. > > > > > And as I recall, while crime remained lower (it did not > > > continue to drop) for several weeks following the end of > > > the project, it then climbed back up to "normal." > > > > > It should be fairly easy to look at the stats for a > > > comparable city to D.C. and see if the crime rate > > > pattern--a sharp drop at the beginning of the period, > > > followed by a rise to "normal" 12 or so weeks later-- > > > was also comparable. On the face of it, that seems > > > unlikely. > > > > The way I read the summary of the study is that they don't compare > > the crime rate during the course to that prior to the course, but > > to the same time period over the previous 5 yrs. > > Right, to control for seasonal variation. > > > If the DC crime rate was > > relatively flat during the 90s, maybe that's an ok methodology. But > > metro cities throughout the nation experienced a dramatic drop in > > crime rate starting around 1992-1993 and continuing for several yrs > > and therefore the study can't prove its point w/o controlling for > > this major factor. > > They did, by "predicting" what the crime rate *would* have > been for that period that year on the basis of the previous > five-year trend. It's true that there might have been > *somewhat* less of a reduction if the crime rate had started > going down in early 1993, but you would have no reason to > see the sharp, sudden drop they measured during the project > on the basis of the decline you're talking about (much less > the return to "normal" a few weeks after the study). > > > I've tried to search the uniform crime statistics but can't figure > > out how to isolate the variables needed. Also can anyone post the > > full study? > > I don't believe it's on the Web anywhere, nor do I think > you can generate the results from the statistics without > using time-series analysis. > > I still dont' trust how they massaged the raw data - > > the word here in fairfield after the course was that hagelin was > > disappointed in the data until they came up the adjusting weather > > variables. > > I already pointed out to you that the weather variable > was chosen *before* the study took place. The protocol > for the study was set and announced in advance precisely > so they couldn't be accused of after-the-fact tweaking > to make it come out the way they wanted. > > Hagelin may well have been disappointed, but if so it > was because they didn't get the 20 percent reduction > they had predicted--it was only 16-some percent.
The problem with this research design is that it is parallel to doing drug study on one subject. It might yield some nice antecdotal findings, but its far from conclusive. The intervention should be over multiple time periods and in multiple cities. Ideally with different size ME groups. For different intervention periods. In this example, each time period / intervention in each city would be one "subject" / observation. Testing for 10 interventions, across various seasons, in 10 cities, 100 observations, would provide enough data to begin clearly control for seasonal effects, weather, longer term crime trends (from abortion, etc), police on the street, policing practices, regional and education income levels, size of the group, length of the intervention, etc. On the other hand, one intervention in one city leaves so many posible other effects, including researcher bias, errors, etc, that caused the change in crime, that the study by itself is not conclusive, or even credible. Though it may provide some nice exploratory findings to justify larger studies. Just as a one person clinical drug study would not be credible or conclusive by itself. At best it would be indicative if possible results from more comprehensive studies. ------------------------ 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/
