BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CAIL//Events//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:CAIL Events
BEGIN:VEVENT
UID:2026-10-22-hannah-li@cail.columbia.edu
DTSTAMP:20261021T213738Z
DTSTART:20261022T190000Z
DTEND:20261022T200000Z
SUMMARY:Other: Hannah Li - Operational Dosage: Implications of Capacity Co
 nstraints for Experimentation
LOCATION:CSB 453
DESCRIPTION:Randomized controlled trials are the gold standard for evaluat
 ing interventions — yet a broad class of trials systematically violates 
 a core assumption of standard experimental design. When interventions are 
 delivered by capacity-constrained resources — teachers\, healthcare prov
 iders\, social workers\, and countless others — participants compete for
  finite service capacity\, creating hidden variation in the treatment itse
 lf. We call this operational dosage. Using queueing-theoretic methods\, we
  show that in such settings\, treatment effects are both capacity- and sam
 ple-size-dependent\, and that statistical power is non-monotone in sample 
 size: adding participants beyond a critical threshold reduces power\, dire
 ctly contradicting conventional power calculations. We validate these insi
 ghts through simulations calibrated to a real-world tuberculosis intervent
 ion trial in Kenya. Our results reveal a fundamental design lever — capa
 city selection — that has been largely ignored in experimental practice\
 , and offer a unifying explanation for why promising interventions so ofte
 n fail to replicate or scale.\n\nJoint work with Justin Boutilier\, Jonas 
 Jonasson\, and Erez Yoeli.\n\nhttps://cail.columbia.edu/events/2026-10-22-
 hannah-li
URL:https://cail.columbia.edu/events/2026-10-22-hannah-li
END:VEVENT
END:VCALENDAR