Omitted Variable Bias Analysis
Updated 2026-09-11
INTRODUCTION
English translation pending.
CORE DEFINITION
Omitted variable bias arises in statistical estimation when a variable that influences both the treatment and the outcome is left out of the model, so part of its effect is wrongly attributed to the treatment. The standard analysis first determines the direction of the bias from the signs of the two relationships, then uses sensitivity analysis to ask how strong an omitted confounder would have to be to overturn the finding. The qualification is that the bias can rarely be eliminated, only bounded, so the honest output is a range and a statement of fragility.
SCAFFOLDING EFFECT
Reduce cognitive load
- Sign check: determine the direction of the bias from how the missing variable links to both sides. - Fragility test: ask how strong an unobserved factor must be to erase your result. - Confounder hunt: draw the causal graph and name every arrow you have not measured.
Anchor fast decisions
If an unmeasured factor raises both the chance of receiving the treatment and the level of the outcome, the estimated treatment effect absorbs that factor's contribution and is biased upward. The reverse holds when the two links point in opposite directions. Because the size of the distortion scales with both links, quantifying the plausible strength of each lets you bound how far the estimate could move.
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/Omitted-variable_biasverified
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