Regression Discontinuity Design, RDD
Updated 2026-08-13
INTRODUCTION
English translation pending.
CORE DEFINITION
A quasi-experimental method used when a treatment is assigned by a score threshold, such as a 60-point pass mark. It compares the outcomes of units that fall just barely on either side of the cutoff. Near the threshold the two groups are essentially similar, separated only by a single point of fate, which makes the comparison nearly random and isolates a local causal effect without any random assignment.
SCAFFOLDING EFFECT
Reduce cognitive load
- Substitute for randomization: when groups cannot be assigned at random, the threshold yields a near-random comparison at the cutoff - Local causal estimate: it recovers the treatment effect specifically for units sitting close to the threshold - Testable assumptions: the no-manipulation condition around the cutoff can be checked directly
Anchor fast decisions
Treatment is assigned by a threshold on a continuous running variable. Units just above and below it are nearly identical in every expectation except the one-point difference in that variable, so the jump in their observed outcomes at the cutoff approximates the local causal effect of the policy, replacing randomization with a natural boundary.
MINIMUM ACTION
In progress 0/1Practice this model in one real situation:
account_treeGenealogyexpand_more
menu_bookReferencesexpand_more
Source support: Explicit
- en.wikipedia.orghttps://en.wikipedia.org/wiki/Regression_discontinuity_designverified
PRIVATE NOTES · Only visible to you
SAVED Q&A
ENTRY Q&A · Private saving available
Ask with a clear boundary
thinkingmodels answers from published entry context only.
Your question is sent to thinkingmodels. The answer uses public entry context only.
RELATED MODELS