Knowledge Maxxing← Back to topic library
Responsible Tech MaxxingIntermediateFairness90 sec
Equalised odds
Equalised odds is a fairness criterion requiring both the true positive rate and the false positive rate to be equal across groups.
Prerequisite: Equal opportunity
What matters
By requiring both rates to match, this criterion is stricter than equal opportunity alone, which only requires the true positive rate to match across groups.
Example
A screening model satisfies equalised odds only if it correctly identifies qualified candidates at the same rate, and incorrectly flags unqualified ones at the same rate, across groups.
Common misconception
Equalised odds is not automatically satisfied just because a model achieves equal opportunity; equalised odds additionally requires false positive rates to match too.
Recall question
What additional requirement does equalised odds add beyond equal opportunity?
Keywords and related concepts
equalised oddstrue and false positive ratesfairness metricequal opportunity