BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:When do statistics provide evidence for discrimination by police? 
 A causal approach - Naftali Weinberger (Munich Center for Mathematical Phi
 losophy\, LMU Munich)
DTSTART:20220518T120000Z
DTEND:20220518T133000Z
UID:TALK174191@talks.cam.ac.uk
CONTACT:Matt Farr
DESCRIPTION:Benchmark tests are widely employed in testing for racial disc
 rimination by police. Neil and Winship (2019) correctly point out that the
  use of such tests is threatened by the phenomenon of Simpson's paradox. N
 evertheless\, their analysis of the paradox is inadequate\, in ways that p
 oint to a more general problem with how they relate statistical quantities
  to discrimination hypotheses. Simpson's paradox reveals that the statisti
 cs employed in benchmark tests will not\, in general\, be invariant to upd
 ating on new information. I argue that as a result of this\, benchmark sta
 tistics should not by themselves be taken to provide any evidence for or a
 gainst discrimination\, absent additional modeling assumptions. Although N
 eil and Winship highlight ways in which benchmark statistics appearing to 
 provide evidence for discrimination no longer appear to do so given additi
 onal assumptions\, they lack an account of which sets of assumptions would
  ensure invariance. Causal models provide such an account. This motivates 
 the use of causal models when using statistical methods as evidence for di
 scrimination.
LOCATION:Arts School Lecture Theatre A
END:VEVENT
END:VCALENDAR