confirmation bias
A cognitive bias where people give more weight to information that supports their existing beliefs or hypotheses and downplay or ignore conflicting evidence.
Key Points
- Leads people to seek, interpret, and remember data that aligns with their prior views.
- Skews estimates, risk assessments, and prioritization by filtering out disconfirming facts.
- Countermeasures include actively searching for disconfirming evidence, inviting diverse perspectives, and using decision checklists.
- In agile, rely on empirical validation (experiments, MVPs, A/B tests), sprint reviews, and retrospectives to challenge assumptions.
Example
A product owner believes a new feature will boost adoption and highlights a few enthusiastic stakeholder comments while ignoring analytics that show low usage of similar features. The team runs an A/B test and discovers the feature does not improve activation, prompting a pivot.
PMP Example Question
During backlog refinement, the sponsor insists a feature will increase revenue and cites only supportive anecdotes. What should the agile team do to best reduce confirmation bias?
- Invite only subject matter experts who agree with the sponsor to keep the discussion focused.
- Hold a pre-mortem and require evidence that could disprove the assumption before prioritizing the feature.
- Extend the next sprint to collect more opinions from the same stakeholders.
- Skip the sprint review metrics and rely on the sponsor's experience for the decision.
Correct Answer: B — Use a pre-mortem and seek disconfirming evidence
Explanation: Proactively looking for data that could invalidate the assumption is a direct way to counter confirmation bias and supports evidence-based prioritization.
HKSM