Does UPSC Normalize Marks? Scaling vs Moderation

No. The Civil Services Preliminary Examination is a screening test whose marks are not counted for the final merit at all. The Prelims cut-off is fixed on your General Studies Paper-I marks (out of 200), adjusted only by the one-third negative-marking rule. The words "normalization" and "scaling" do not appear anywhere in the notification, and no percentile or normalization mechanism is applied to Prelims marks — everyone sits the same two papers in a single common session, so cross-shift normalization is structurally inapplicable.

No. General Studies Paper-II (CSAT) is a qualifying paper only — you must score at least 33% (66 out of 200) — and its marks do not count towards the Prelims cut-off or your rank. There is no normalization or percentile conversion of CSAT marks; you either clear the 33% gate or you do not.

No. Multi-shift exams like SSC CGL or NTA tests use percentile normalization because different candidates sit different question papers across shifts. UPSC CSE Prelims is held in a single common session with the same papers for everyone, so there is nothing to normalize across shifts. The framing that UPSC applies shift-based or percentile normalization is simply incorrect.

Moderation is an evaluation technique for descriptive answer scripts that corrects for individual examiners being strict or lenient within the same paper. The Supreme Court has described it as "inherent in the evaluation of answer scripts in any large scale examination, where there are more than one examiner" (Sanjay Singh v. UP PSC, 2007), and in UPSC v. Angesh Kumar (2018) it "approved the method of moderation adopted by the UPSC". So moderation of Mains descriptive papers is real and judicially endorsed — but UPSC does not publish the formula, and it is not applied to the objective Prelims paper.

They are distinct techniques. Moderation corrects examiner strictness or liberality within the same paper. Scaling, as the Supreme Court defined it, is "the exercise of putting the marks which are the results of different scales adopted in different subjects by different examiners into a common scale so as to permit comparison of inter se merit" — i.e. adjusting across different subjects. The Court held the scaling system "unsuited" for a common-subject examination. UPSC publishes no cross-subject scaling formula.

This is a popular belief, not a confirmed fact. There is no official source for any inter-subject scaling or normalization formula for optional subjects. The notification prescribes none, and the courts (Angesh Kumar, 2018) declined to compel UPSC to disclose any such methodology. What is confirmed is only that moderation — examiner-variability correction — applies to descriptive papers generally; that is not the same as a published formula that "equalises" easy versus hard optionals. Treat the optional-scaling formula as an unverified aspirant belief.

Only Mains Papers I–VII plus the interview. Essay (250), GS-I to GS-IV (250 each) and the two Optional papers (250 each) count for merit; the interview/Personality Test carries 275. Paper-A (an Indian language, 300) and Paper-B (English, 300) are qualifying only, needing 25% each. Prelims (both papers) and the two qualifying language papers do not add to your merit total.

In UPSC v. Angesh Kumar (2018) the Supreme Court held that UPSC cannot be compelled to disclose raw marks, the moderation method, cut-off methodology or evaluator details under RTI mechanically, because revealing the intermediate stages of evaluation would harm the integrity of a large-scale examination system. So the existence of moderation is confirmed, but the exact arithmetic is deliberately not public — do not trust any site claiming to publish UPSC's precise formula.