UPSC Prelims GS Paper I carries 200 marks across 80 questions, with a negative marking of 0.67 marks for every wrong answer. This means every decision you make about whether to attempt or skip a question has a direct, calculable effect on your final score — and making these decisions systematically, rather than on instinct, is one of the most practical improvements any aspirant can make to their Prelims performance.
This guide builds a clear, step-by-step decision framework for the skip-or-attempt question, grounded in the mathematics of UPSC’s marking scheme and the practical reality of how options can be eliminated under exam time pressure.
The Mathematics: Why Random Guessing Does Not Pay
Before building the decision framework, it helps to understand the expected value of a random guess on a four-option MCQ under UPSC’s marking scheme:
| Scenario | Marks Gained | Probability | Expected Value |
| Correct answer | +2 marks | 1 in 4 (25%) | 2 × 0.25 = +0.50 |
| Wrong answer | –0.67 marks | 3 in 4 (75%) | –0.67 × 0.75 = –0.50 |
| Net expected value from pure random guess | — | — | 0.50 – 0.50 = 0.00 |
A completely random guess produces zero expected value — it neither helps nor hurts in the long run. But you are almost never in a position of zero knowledge. Almost every question allows you to eliminate at least one option based on partial knowledge, which changes the calculation significantly.
Expected Value When You Can Eliminate Options
| Options Eliminated | Options Remaining | Probability of Correct Guess | Expected Value | Decision |
| 0 eliminated | 4 options | 25% | 0.00 (neutral) | SKIP unless you have other signal |
| 1 eliminated | 3 options | 33.3% | +0.22 (marginally positive) | Consider attempting — marginally worth it |
| 2 eliminated | 2 options | 50% | +0.83 (clearly positive) | ATTEMPT — expected value is strongly positive |
| 3 eliminated | 1 option — you know it | 100% | +2.00 | Always ATTEMPT — this is certain knowledge |
The critical threshold: eliminating two options makes the expected value of attempting clearly positive (+0.83). Eliminating only one option makes it marginally positive (+0.22) — enough to attempt if you have a slight lean toward one of the remaining options, but not if all three feel equally unfamiliar.
| Want to build your personal skip-or-attempt threshold through mock test analysis? → Explore Secure Prelims Program 2026 |
A Practical Three-Step Decision Process
Step 1: Immediate Recall Test (3 seconds)
Read the question stem. Do you immediately recognise the answer or know enough to eliminate two or more options? If yes — attempt. If no — move to Step 2 rather than spending time on it immediately.
Step 2: Elimination Scan (15–20 seconds maximum)
Scan all four options. Can you confidently rule out two or more based on any knowledge — even partial, even indirect? If yes — attempt from the remaining options. If you can only eliminate one, assess whether you have any directional lean toward one of the three remaining options. A genuine, specific lean (not a vague feeling) justifies attempting. Zero directional information beyond the one elimination — skip.
Step 3: Flag and Move (5 seconds)
Mark the question for review and move on immediately. Do not spend more than 30 seconds on a question you are unsure about on the first pass. Return to flagged questions if time permits. Making fresh decisions on returning often produces better results than agonising over the same question.
Building Your Personal Attempt Threshold
Your personal attempt threshold — the level of confidence at which attempting is better than skipping for your specific accuracy rate — should be calibrated from your mock test performance, not from generic advice.
After each mock test, separately score: questions you were confident about (accuracy rate A), questions where you attempted with one elimination and a lean (accuracy rate B), and questions where you guessed with no strong signal (accuracy rate C). If rate C is below 33%, your random guessing is costing you marks — tighten the threshold. If rate B is above 45%, your lean-based attempts are net positive — keep attempting when you have a lean.
This personalised calibration — available only from analysing your own performance data across multiple mock tests — is more useful than any general rule.
Common Mistakes in Skip-or-Attempt Decisions
- Over-attempting with false confidence: Feeling like you ‘kind of know’ a question is not the same as being able to eliminate two options with specific, grounded reasoning. False confidence is the most common cause of negative marking damage.
- Under-attempting from excessive caution: Aspirants who leave 25–30 questions unattempted out of excessive caution are leaving significant potential marks on the table — particularly if their accuracy on attempted questions is high.
- Spending too long deciding: The decision process above should take 30 seconds maximum. Spending 2–3 minutes agonising over a single uncertain question costs you the time you need to answer 2–3 questions you do know.
- Not calibrating from mock tests: Generic advice about attempt thresholds may not match your specific accuracy profile. Your mock test data tells you your actual expected value at different confidence levels — use it.
Frequently Asked Questions
What is the ideal number of questions to attempt in UPSC Prelims?
There is no universal ideal number — it depends entirely on your accuracy rate. An aspirant with 80% accuracy on attempted questions should attempt aggressively (65–70 questions). An aspirant with 55% accuracy should attempt conservatively (50–55 questions). The mock test calibration process above tells you your specific optimal range. Anyone who gives you a generic number without knowing your accuracy profile is not giving you useful advice.
Should I attempt all 80 questions to maximise score?
Only if your accuracy rate on uncertain questions is above 40%. If your accuracy on low-confidence attempts is below 33%, attempting all 80 questions will produce a lower score than attempting 60–65 confidently. The mathematics in the table above show exactly why — below 33% accuracy, the expected value of attempting is negative.
Is it worth returning to flagged questions at the end?
Yes — but with a specific discipline. On returning to a flagged question, do not revisit your original reasoning from scratch. Ask only: is there any new signal I noticed elsewhere in the paper that applies here? If yes, apply it. If no — and your original confusion remains — the default should be to skip rather than to guess.
Does the same skip-or-attempt logic apply to CSAT?
Yes — CSAT has the same marking scheme (2 marks correct, 0.67 marks deducted for wrong). However, CSAT is qualifying (33% of 200 = 66.67 marks) rather than merit-scoring, so the risk calculation is different. In CSAT, once you are confident of clearing the qualifying threshold, additional attempt risk decreases in value. Focus on securing the qualifying marks confidently rather than maximising score.
Conclusion
The skip-or-attempt decision in UPSC Prelims is not a matter of intuition — it is a matter of applied expected value thinking, personalised through mock test data. Eliminate two options and attempt. Eliminate one with a genuine directional lean and attempt. Zero eliminations or no lean — skip and move on. Calibrate this threshold to your own accuracy data from mock tests. The aspirant who follows this system consistently will outperform both the over-attempter and the over-cautious skipper.
| Riyasat IAS Mentorship — Personal Guidance for UPSC CSE Direct mentorship from Riyasat Ali Sir │ Hindi & English Medium │ 100% Online — Apply Now |
