Investigate missingness and add targeted labels: Which approach | AIGP
7-day money-back guarantee — full refund within 7 days of purchase if you've completed under 20% of the questions. See pricing →
Certifications Tools Flashcards Career Paths Exam Guides Blog Pricing For Teams About

Language

✓ EnglishDeutschEspañolFrançaisPortuguês
Check readiness — free →

Investigate missingness and add targeted labels: Which approach fits?

AIGP Understanding How to Govern AI Development Medium

Investigate missingness and obtain targeted labels while reporting remaining uncertainty.

The question

A public-benefit agency is evaluating an eligibility-support model. Many records lack outcome labels, and missingness is concentrated among applicants with language-assistance needs. Before selecting between feasible validation methods, the agency requires evidence covering that underrepresented group and uncertainty from missing labels. Which approach fits?

Preparing for AIGP? Take the free 5-min readiness quiz →

  1. Investigate missingness and add targeted labels.
    Investigating the missingness pattern and obtaining targeted labels improves evidence for the affected group while leaving residual uncertainty visible.
  2. Restrict validation to complete-label records and compare subgroup estimates.
    Complete-case validation excludes the group with concentrated missing labels, so subgroup estimates remain incomplete and potentially biased.
  3. Report aggregate performance from available labels and describe the missingness limitation.
    Reporting the limitation is useful, but aggregate available-case results do not provide evidence for the affected underrepresented group.
  4. Impute missing outcomes using the majority observed outcome for validation.
    Majority imputation can mask concentrated missingness and create unjustified confidence about applicants needing language assistance.
The trap
Ask who is missing from labeled data and whether the missingness pattern could change performance conclusions.

How to remember it

Investigate missingness and obtain targeted labels while reporting remaining uncertainty.

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 — free

More 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.