BEGIN:VCALENDAR
VERSION:2.0
METHOD:PUBLISH
PRODID:-//Telerik Inc.//Sitefinity CMS 15.1//EN
BEGIN:VTIMEZONE
TZID:Eastern Standard Time
BEGIN:STANDARD
DTSTART:20251102T020000
RRULE:FREQ=YEARLY;BYDAY=1SU;BYHOUR=2;BYMINUTE=0;BYMONTH=11
TZNAME:Eastern Standard Time
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20250301T020000
RRULE:FREQ=YEARLY;BYDAY=2SU;BYHOUR=2;BYMINUTE=0;BYMONTH=3
TZNAME:Eastern Daylight Time
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
DESCRIPTION:CPR Methods Seminar: Sharad Goel\, Harvard Kennedy School\, wil
 l present "A Simple\, Statistically Robust Test of Discrimination."Abstrac
 t:In observational studies of discrimination\, the most common statistical
  approaches consider either the rate at which decisions are made (benchmar
 k tests) or the success rate of those decisions (outcome tests). Both test
 s\, however\, have well-known statistical limitations\, sometimes suggesti
 ng discrimination even when there is none. Despite the fallibility of the 
 benchmark and outcome tests individually\, I'll show a surprisingly strong
  statistical guarantee: under a common non-parametric assumption\, at leas
 t one of the two tests must be correct\; consequently\, when both tests ag
 ree\, they are guaranteed to yield correct conclusions. I'll present empir
 ical evidence that the underlying assumption holds approximately in severa
 l important domains\, including lending\, education\, and criminal justice
  -- and that our hybrid test is robust to the moderate violations of the a
 ssumption that we observe in practice. Applying this approach to 2.8 milli
 on police stops across California\, we find evidence of widespread racial 
 discrimination that would have been missed by standard methods.
DTEND:20241107T220000Z
DTSTAMP:20260513T103231Z
DTSTART:20241107T203000Z
LOCATION:
SEQUENCE:0
SUMMARY:CPR Methods Seminar: A Simple\, Statistically Robust Test of Discri
 mination
UID:RFCALITEM639142507512773168
X-ALT-DESC;FMTTYPE=text/html:<p>CPR Methods Seminar: Sharad Goel\, Harvard 
 Kennedy School\, will present "A Simple\, Statistically Robust Test of Dis
 crimination."</p><p><strong>Abstract:</strong></p><p>In observational stud
 ies of discrimination\, the most common statistical approaches consider ei
 ther the rate at which decisions are made (benchmark tests) or the success
  rate of those decisions (outcome tests). Both tests\, however\, have well
 -known statistical limitations\, sometimes suggesting discrimination even 
 when there is none. Despite the fallibility of the benchmark and outcome t
 ests individually\, I'll show a surprisingly strong statistical guarantee:
  under a common non-parametric assumption\, at least one of the two tests 
 must be correct\; consequently\, when both tests agree\, they are guarante
 ed to yield correct conclusions. I'll present empirical evidence that the 
 underlying assumption holds approximately in several important domains\, i
 ncluding lending\, education\, and criminal justice -- and that our hybrid
  test is robust to the moderate violations of the assumption that we obser
 ve in practice. Applying this approach to 2.8 million police stops across 
 California\, we find evidence of widespread racial discrimination that wou
 ld have been missed by standard methods.</p>
END:VEVENT
END:VCALENDAR