Apps

🧪 This platform is in early beta. Features may change and you might encounter bugs. We appreciate your patience!

Omitted Variable Bias

Also Known As: Confounding bias Unobserved heterogeneity
Statistical Error ID: omitted_variable_bias

Definition

Omitted variable bias occurs when a statistical model leaves out a relevant variable that is correlated with both the independent variable and the dependent variable. This causes the estimated effect of the included variable to absorb the influence of the missing one, leading to biased and inconsistent coefficient estimates. The direction and magnitude of the bias depend on the correlations between the omitted variable and the other variables in the model.

Examples

A study finds that ice cream sales are strongly correlated with drowning deaths and concludes ice cream causes drowning. The omitted variable is temperature: hot weather increases both ice cream consumption and swimming, which increases drowning risk.

A study finds that cities with more coffee shops per capita have higher rates of heart disease and concludes coffee consumption damages heart health. The omitted variable is urbanization: densely populated cities have both more coffee shops and more sedentary, high-stress lifestyles that independently increase cardiovascular risk.

An analysis of school test scores finds that schools with more computers per student perform significantly better academically and recommends buying more computers. The omitted variable is school funding: wealthier, better-funded schools both purchase more technology and attract more experienced teachers, which is the true driver of performance.

Verification Steps
Verification Steps
Binary yes/no questions that an AI must answer to detect a reasoning pattern in a text.
Each of the 452 aspects has verification steps — simple yes/no questions designed to systematically detect whether a pattern appears in a text. For ad hominem: "Does the argument attack a person rather than their claim?" For false dichotomy: "Are only two options presented when more exist?" This ensures consistent, reproducible analysis.

Binary (yes/no) questions an LLM must answer to identify this aspect:

  1. 1

    Is there a variable not included in the model that could plausibly affect the dependent variable?

    Type: binary
  2. 2

    Is the omitted variable likely correlated with one or more included independent variables?

    Type: binary
  3. 3

    Could including this variable substantially change the estimated effects of other variables?

    Type: binary
  4. 4

    Does the analysis claim causal effects without addressing potential omitted variables?

    Type: binary
Deep Dive
The expandable detail section on each aspect page with examples, psychology, and counter-strategies.
The Deep Dive section provides in-depth information about each aspect: a real-world example showing the pattern in action, an explanation of why it works psychologically, practical advice on how to counter it, alternative names, and links to related aspects.

Hierarchical Context