How should I review flagged questions at the end of a certification exam?
Work through flagged questions in order, applying fresh eyes and the elimination technique. For each, spend no more than 60-90 seconds before committing to a final answer. Do not change answers from your first pass unless you identify a specific logical reason -- new information in a later question, or a clear elimination you missed. Resist changing answers based on anxiety alone.
The mark-for-review feature in certification exam software is one of the most powerful tools available to test-takers, and most candidates use it poorly. They either mark too many questions (creating an unmanageable review pile) or too few (missing the chance to revisit genuinely uncertain answers with fresh perspective). They then mismanage the review itself -- changing good answers based on anxiety, or spending too long on genuinely unknowable questions.
This guide establishes a complete system for marking, managing, and reviewing flagged questions on major certification exams, based on cognitive research on decision-making under uncertainty and the specific mechanics of testing platforms.
When to Mark a Question for Review
Apply a consistent marking standard: mark a question if and only if you are below a defined confidence threshold.
A practical three-tier system:
| Confidence Level | Action |
|---|---|
| High (80%+ certain) | Answer, do not mark, move on |
| Moderate (50-79% certain) | Answer with best guess, mark for review |
| Low (under 50% certain) | Answer with best guess, mark for review |
The critical discipline is: always select an answer before marking. Do not mark unanswered questions unless you are completely out of time. Most certification exams end the moment time expires -- if you have flagged questions with no answer selected and time runs out, they score as wrong. A best-guess answer selected before marking always outperforms leaving blank.
Managing the Mark Count
If you reach the end of the first pass with more than 20-25% of questions marked, you have a problem. A 90-question exam with 25+ flagged questions leaves inadequate time for meaningful review at 90-second average.
During the first pass, if you notice your mark count climbing rapidly, recalibrate. You may be marking questions you actually know well but are second-guessing. If a question took you fewer than 30 seconds to eliminate down to one answer, that is likely a correct answer -- unmarking it protects your review time for genuinely uncertain questions.
"The optimal strategy for adaptive and linear certification exams involves a two-pass approach, but the key constraint is that first-pass time must be rationed so the second pass has meaningful review time available. A common failure mode is spending too long on first-pass questions, exhausting time before the second pass can begin." -- Dr. Wim van der Linden, Center for Measurement in Education and Research
The Review Order Strategy
When you reach the review phase, do not simply start with the first flagged question. Use this order:
- Questions you marked because you needed to think longer (had two viable options, needed more consideration)
- Questions you marked because the scenario was complex (needed to re-read carefully)
- Questions you marked because you had very low confidence (genuine uncertainty)
Work the high-confidence flags first. These are the ones most likely to resolve quickly and bank easy points. Leave the genuinely unknowable questions for last -- they may not resolve regardless of time spent.
The Fresh-Eyes Technique
The primary value of reviewing flagged questions is temporal distance. You are no longer mid-stream in the exam; you are looking at the question with fresh perspective after completing other questions.
Fresh eyes sometimes reveal:
- A phrase in the question stem you initially misread
- An obvious eliminator you glossed over under time pressure
- A connection to a later question that contained relevant context
When reviewing, re-read the question stem completely before looking at the answer choices. Do not start from your marked answer and work backward -- this anchoring bias makes it harder to spot errors in your initial reading.
When to Change an Answer
The research literature on answer changing is unambiguous: first instincts are right more often than second-guessing under time pressure. A meta-analysis by Benjamin et al. (1984) analyzing 19 studies found that students change from a wrong answer to a right answer more often than from right to wrong, but this effect disappears when the reason for changing is anxiety rather than new information.
Change an answer if:
- You find a specific logical error in your original reasoning
- A later question in the exam contained information that makes one answer clearly correct
- You misread the question stem on the first pass and your current reading supports a different answer
- You applied the wrong knowledge domain (e.g., answered for IPv4 when the question specified IPv6)
Do not change an answer if:
- You feel anxious about your original choice
- The alternative choice "looks better" without a specific reason
- You are second-guessing based on the difficulty of surrounding questions
"The advice 'go with your first instinct' is overly simplistic. The research shows first instincts are correct more often than changes -- but the key variable is whether the change is driven by new information or by anxiety. Anxiety-driven answer changes hurt performance; knowledge-driven changes improve it." -- Dr. Justin Kruger, NYU Stern School of Business
Question Review Mechanics by Platform
Different certification exam platforms handle flagging differently:
| Platform | Flag Icon Location | Review Access |
|---|---|---|
| Pearson VUE | Top of question screen | End of exam review screen, filterable |
| PSI Bridge | Side panel | Review list at end of section |
| Prometric | Status bar | Review mode navigates only flagged items |
| CompTIA CertMaster | Flag icon per question | Summary screen with filter |
Most platforms offer a review summary screen showing all questions, which are answered, which are flagged, and which are skipped. Use this screen to assess the scope of your review before diving in -- if you have 30 flagged questions with 15 minutes remaining, you have 30 seconds per question and need to work efficiently.
Time Management During Review
Set a ceiling for each flagged question during review:
- Type A flags (two viable options): 60-90 seconds maximum
- Type B flags (complex scenario): 90-120 seconds maximum
- Type C flags (low confidence, unclear): 45 seconds maximum -- your reasoning ability is not significantly improving after this point
When you hit your ceiling, commit to your current best answer and move to the next flagged question. Do not negotiate with yourself. The question is either answerable in 90 seconds or it is not -- and continued staring does not change that.
Handling Questions You Still Cannot Answer After Review
Some questions will remain genuinely uncertain even after review. For these:
- Apply the elimination filters one final time (absolute language, off-topic, constraint violations)
- Identify which of your remaining options is more aligned with domain best practice
- Select that option and commit
- Move on without looking back
A 50% confidence answer scores as well as a 100% confidence answer. Both count the same if correct. The only thing that matters is having an answer selected before time runs out.
Common Mistakes During Flagged Question Review
| Mistake | Why It Hurts | Fix |
|---|---|---|
| Changing answers based on anxiety | First instinct is usually right | Only change with specific new reasoning |
| Spending too long on hard flags | Exhausts time, may leave easy flags unreviewed | Set firm 90-second ceiling |
| Reviewing in sequential order | Hard questions at the top delay easy wins | Sort by confidence level first |
| Re-reading every answer choice | Time-wasting if reason for flag is known | Focus on your specific uncertainty |
| Ignoring the review screen summary | Misses the scope of the task | Always check summary before starting |
Frequently Asked Questions
Should I mark every question I am not 100% sure about? No. Mark questions where you are below 80% confidence. Marking every uncertain question fills your review queue with questions your instinct answered correctly, dilutes time for genuinely hard questions, and increases anxiety without improving outcomes.
What if I run out of time during the review phase? Check the review summary screen immediately and answer any completely unanswered questions first (these score as zero if left blank). Then work your low-confidence flags quickly. Do not spend time reconsidering high-confidence first-pass answers.
Can I see which questions I flagged after the exam? Most certification platforms do not preserve flag data after exam submission. Score reports typically show domain-level performance, not individual question-level data. Use the post-exam score report to identify which domains need reinforcement for potential retakes.
References
- Benjamin, L.T., Cavell, T.A., & Shallenberger, W.R. (1984). Staying with the initial answers on objective tests: Is it a myth? Teaching of Psychology, 11(3), 133-141.
- Kruger, J., Wirtz, D., & Miller, D.T. (2005). Counterfactual thinking and the first instinct fallacy. Journal of Personality and Social Psychology, 88(5), 725-735.
- van der Linden, W.J. (2011). Handbook of modern item response theory. Springer.
- Metcalfe, J., & Finn, B. (2011). People's hypercorrection of high-confidence errors: Did they know it all along? Journal of Experimental Psychology: Learning, Memory, and Cognition, 37(2), 437-448.
- Pearson VUE. (2024). Exam delivery platform candidate guide. Pearson VUE official documentation.
- Dempster, F.N. (1989). Spacing effects and their implications for theory and practice. Educational Psychology Review, 1(4), 309-330.