--- Martin Lewis <[EMAIL PROTECTED]> wrote: > > The Slate article you link to describes clustering > as a "time-honored > technique for many epidemiological studies". How > does this make it > extremely bad? There are difficulties with using > clustering but this > doesn't mean it is obviously wrong which is what Dan > asked.
It's extremely bad because warfare is not epidemiology. Casualties are not randomly distributed through Iraq - which is the assumption you would have to make to use clustering the way they used it. Apart from which, a moment's thought and a basic knowledge of military history would reveal that the report's estimate is absurd. The _nuclear bombing of Hiroshima_ killed 66,000 people. That involved the obliteration of an entire city. The report suggests that _100,000_ people were killed. If US forces were _trying_ to kill that many people they'd probably have real difficulty doing it. It would certainly have been reported by the literally thousands of journalists floating around Iraq. It would take an enormous, dedicated effort to kill that many civilians. It's not something that could happen by accident. It's _certainly_ not something that could happen without being reported on. Dismissals of the report's methodology are all over the web. It's a joke. It's only plausible if you really, really want to believe that American soldiers are genocidal mass murderers. ===== Gautam Mukunda [EMAIL PROTECTED] "Freedom is not free" http://www.mukunda.blogspot.com __________________________________ Do you Yahoo!? Check out the new Yahoo! Front Page. www.yahoo.com _______________________________________________ http://www.mccmedia.com/mailman/listinfo/brin-l
