StackedML
Practice
Labs
Questions
Models
Pricing
Sign in
Questions
/
Model Evaluation & Experimentation
/
Experimentation
/
Simpson's paradox
← Previous
Next →
200.
Definition of Simpson's Paradox
easy
What is Simpson's Paradox?
A
statistical artifact where correlation coefficients become negative when computed on standardized variables across different subsets of the data
B
A bias introduced when multiple hypothesis tests are conducted on the same aggregated dataset simultaneously
C
A trend that appears in aggregated data reverses or disappears when the data is disaggregated by a confounding variable
D
A phenomenon where a model's training accuracy consistently exceeds its test accuracy across all subgroups
Sign in to verify your answer
← Back to Questions