Skip to main content

Omitted Variable Bias

Go Search
New Atlas
Atlas, A-Z
Atlas Maps
MPP/MPA Programs
Core Topics
Illustrative Courses
Topic Encyclopedia
Concept Dictionary
Career Tips
Best Practices Project

PPGPortal > Home > Concept Dictionary > N, O > Omitted Variable Bias

Omitted Variable Bias  

The average effect of X on Y is biased if other variables that determine Y and are correlated with X are left out of the regression model.

(Phil Oreopoulos, Toronto PPG 1004H)


Omitted variable bias is also referred to as the problem of excluding a relevant variable or underspecifying the model.

The conditions for omitted variable bias in multiple regression are similar to those for a single conditional variable: if an omitted variable affects Y and is correlated with at least one of the conditional variables, then the value of the coefficient of interest will be biased for causal interpretations.




Important Notices
© University of Toronto 2008
School of Public Policy and Governance