What you'll be able to do- ✓Work out who left from which groups by hand, and sort every driver into a lever, an attribute that must not be used, or an open question
- ✓Read a recruitment funnel on small numbers, and write a recommendation as strong as the evidence and no stronger
- ✓Compare a gender pay gap three ways (raw, within grade, controlled), read its interval, and say what cannot be concluded
- ✓Count a skills gap two ways, test how far self-ratings can be trusted, and spot a benchmark that is not in the data
- ✓Write a one-page CHRO brief with a claim ledger behind it and a data-handling and fairness note that you wrote yourself
Turn employee, application and skills files and a question from the CHRO into a one-page brief you can defend — leaving rates you rebuilt, a hiring recommendation sized to its evidence, a pay comparison with its interval, skills gaps as ranges, and an honest note on how people's data was handled.
⌁ How HR analysts use AI assistants on attrition, hiring, pay and skills — sort drivers into levers and attributes, read small numbers carefully, ask for the interval and the method, keep a ledger of where every figure came from, and state how personal data was handled.