AI engineering glossary · Statistics inference
What is SRM (sample ratio mismatch)?
Sample ratio mismatch (also called SRM): A warning sign in an experiment where the observed split of users between groups differs from the intended split by more than chance would allow. It usually points to a bug in assignment or logging, and invalidates the results.
Sample ratio mismatch: an example
A planned 50/50 split that shows 50,900 versus 49,100 users out of 100,000 fails a chi-square check and needs investigation.
Related terms
More Statistics inference terms
- Sampling distribution
- Sequential testing
- Significance level
- Spearman rank correlation
- Standard error of the mean
- Statistic
- Statistical power
- Student's t-test
Where you learn it
Sample ratio mismatch is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.