The tested population and context represent the proposed: Which prior assumption no longer holds?
Changing from adult applicants to secondary-school students invalidates the prior assumption that tested population and context were representative.
The question
An education administration model was tested using adult continuing-education applicants and accepted for release. The organization now wants to use it for secondary-school students, whose records and decision impacts differ. Which prior assumption no longer holds?
Preparing for AIGP? Take the free 5-min readiness quiz →
- The tested population and context represent the proposed student use. ✓The prior acceptance depended on evidence from adults, so representativeness for secondary-school students and their context no longer holds.
- The original reviewers documented the model’s limitations.Good documentation remains useful, but it cannot make adult-population evidence representative of a newly proposed student population.
- The model’s software version remains unchanged.An unchanged version does not preserve validity when the population, data characteristics, and consequences of use have materially changed.
- The original acceptance metrics were computed correctly.Correct computation does not establish that metrics from adults transfer to secondary-school students or their distinct decision context.
The trap
Track the population and context behind acceptance criteria; a new affected group can invalidate representativeness without any code change. How to remember it
Changing from adult applicants to secondary-school students invalidates the prior assumption that tested population and context were representative.
How many of these would you get right?
One of 1581 AIGP questions on Certsqill. Take a free five-minute check and see your score per domain — not one number, but which section to open tonight.
Test your AIGP readiness — freeMore Understanding How to Govern AI Development questions
- Map dependencies before shutdown: What should it do first? →
- The intended-use assumption no longer holds: What limitation is most directly implicated? →
- Validate outcomes before attributing degradation: What prior assumption must be tested before concluding →
- All 426 Understanding How to Govern AI Development questions →
Part of the Certsqill AIGP question bank · Understanding How to Govern AI Development ·
Every answer, right and wrong, comes with its own explanation.
