Welcome to European Tribune. It's gone a bit quiet around here these days, but it's still going.
Display:
A Statistical Analysis of the ICTY's bias. | 220 comments (213 topical, 7 editorial, 0 hidden)
Re: A Statistical Analysis of the ICTY's bias. (none / 1)
JakeS:
he arbitrarily classifies 7 of the Serbian suspects as being related to Kosova
The classification needn't be arbitrary. In wikipedia's List of indictees of the International Criminal Tribunal for the former Yugoslavia each indictee is assigned an ethnic group and an incident.

One thing that Vladimir does wrong is compare Indictees to Convictions, implying that those not convicted have been acquitted. This is not the case. The proper comparison would be the ratio of convictions to acquittals/case-dismissals as there are a number of cases which are ongoing, have been transferred to national jurisdiction, were never concluded because the indictee died, or in which the indictee is at large. Vladimir posted the full stats
here (these do not distinguish between the different wars or regions, only ethnic groups):

Stats from ICTY (none / 0) Here they are



Most economists teach a theoretical framework that has been shown to be fundamentally useless. -- James K. Galbraith
by Carrie (migeru at eurotrib dot com) on Mon Mar 16th, 2009 at 05:50:00 AM EST
[ Parent ]
Re: A Statistical Analysis of the ICTY's bias. (none / 1)
The 7 didn't have a classification... it would have required hours of sifting through documents to find them.
by vladimir on Mon Mar 16th, 2009 at 05:55:26 AM EST
[ Parent ]
Re: A Statistical Analysis of the ICTY's bias. (none / 1)
Then you can't ascribe them to Kosova, especially given that Kosova is the conflict with the smallest number of indictees and arbitrarily assigning 7 Serbs to it biases the data more than, say, assigning the 7 to Krajina or Republika Srpska.

Most economists teach a theoretical framework that has been shown to be fundamentally useless. -- James K. Galbraith
by Carrie (migeru at eurotrib dot com) on Mon Mar 16th, 2009 at 05:58:04 AM EST
[ Parent ]
Re: A Statistical Analysis of the ICTY's bias. (none / 1)
Another usegul comparison would be between indictees and population. Serbs were the largest ethnic group in the former Yugoslavia by a factor of nearly 2. See the table on Ethnic composition of Yugoslavia in 1981 in Wikipedia's Demographics of SFR Yugoslavia. A log-log regression of indictees vs. population might be appropriate in this case, too.

Most economists teach a theoretical framework that has been shown to be fundamentally useless. -- James K. Galbraith
by Carrie (migeru at eurotrib dot com) on Mon Mar 16th, 2009 at 05:56:38 AM EST
[ Parent ]
Re: A Statistical Analysis of the ICTY's bias. (none / 0)
I've delved into their documents. I agree, it's an unorganized mess. A horror-show for scholars.

"It Can't Be Just About Us"
--Frank Schnittger, ETian Extraordinaire
by papicek (papi_cek_at_hotmail_dot_com) on Mon Mar 16th, 2009 at 05:59:33 AM EST
[ Parent ]
Re: A Statistical Analysis of the ICTY's bias. (none / 1)
OK, then that's kosher.

The vanilla averages instead of weighted averages is still nonsense, though.

- Jake

Friends come and go. Enemies accumulate.

by JakeS (JangoSierra 'at' gmail 'dot' com) on Mon Mar 16th, 2009 at 05:58:38 AM EST
[ Parent ]

A Statistical Analysis of the ICTY's bias. | 220 comments (213 topical, 7 editorial, 0 hidden)
Display: