The lack of disclosure of marks of selected candidates in the Union Public Service Commission (UPSC) Civil Services (Main) examination in each General Studies paper, optional subject paper, and the final Personality Test seems to be impacting aspirants’ preparation process, as they are not able to track performance trends in different optional subjects and make informed choices for subsequent attempts.
(Sign up for THEdge, The Hindu’s weekly education newsletter.)
Experts claim that it also reflects a certain lack of transparency in the system, especially since the practice of disclosing the detailed distribution of marks obtained by all the candidates selected in the Civil Services (Main) examination was discontinued since 2018 without any official record of the decision being made.
More recently, the Central Information Commission (CIC) urged the Union Public Service Commission (UPSC) to restore the practice of displaying the detailed distribution of marks obtained by all the candidates selected in the Civil Services Main examination after it observed that the Department of Personnel and Training (DoPT) had no record of any formal decision made to discontinue the practice from 2018. The UPSC, in response, told the Commission that, post declaration of the final result, the detailed break-up of marks is forwarded to the DoPT and the Lal Bahadur Shastri National Academy of Administration (LBSNAA), while the DoPT maintained that the detailed marks are generated and maintained by the UPSC.
The issue came to the fore when a Right to Information (RTI) application was filed by a candidate seeking the comparative data of marks obtained by all selected candidates in each General Studies paper, optional subject paper, and the Personality Test over 10 years, up to 2023. The applicant was keen to find out how selected candidates had fared in different optional subjects. The petitioner stated that while paper-wise marks of recommended candidates were publicly disclosed back in 2017, only aggregate marks are now made available from 2018 onwards. The CIC noted that despite being provided a final opportunity to produce official records, the DoPT failed to submit any file noting, office memorandum, policy decision, or other record upholding the discontinuation of publication of the UPSC CSE (Main) marks across different subjects from 2018 onwards.
“In the absence of official paper-wise data for all candidates, aspirants depend upon marksheets voluntarily shared by only some of the successful candidates publicly, which may not be indicative of the overall picture. Such data is stray and incomplete and may present a distorted picture of what constitutes a good score in a particular paper. However, it is difficult to establish a direct causal relationship between the discontinuation and aspirants’ actual exam performance. The current practice has led to a less scientific and more luck/hunch-based exam preparation,” Pranay Aggarwal, Director, IAS Gurukul and Sociology optional faculty, told The Hindu.
He attributes the discontinuation of subject-wise marks of selected candidates to several possible factors. “The decision may have been motivated by a concern for the privacy of the individual candidates and the disclosure of their paper-wise scores. Another possibility is bureaucratic resistance within the UPSC to greater transparency. While a third possibility could be an attempt to prevent what the UPSC deems as avoidable litigation regarding the selection process,” he added.
With the breakup of marks not being available, it has reduced the amount of information that aspirants can access. “The Civil Services Examination is unusual because there is no official answer key or standardised benchmark for the descriptive Mains papers. Consequently, aspirants learn a great deal by studying the performance of successful candidates. Paper-wise marks allowed them to understand where marks were being made or lost—Essay, individual GS papers, the optional subject, or the Personality Test. Once only aggregate written marks are available, two candidates with an identical Mains total may have arrived there through completely different performance profiles. That information is valuable both for first-time aspirants and for candidates deciding how to improve in another attempt,” Former IRS Officer and UPSC mentor Ravi Kapoor, said.
Aspirants speak
Bihar native Amir Siddique, 27, who is aspiring for his fourth attempt and currently residing in Delhi, feels that candidates do not have a clear idea about what is the highest and lowest marks in the respective GS and optional papers. “This makes planning the next attempt extremely confusing,” he says.
“Earlier, when UPSC published the entire list with paper-wise marks, everyone had a clear benchmark of what was working in GS and optional papers. Now, students get to know some of the marksheets through Telegram/Youtube/Instagram which are posted by a few selected candidates. Most of the aspirants end up changing optionals or sticking with them based purely on coaching claims and topper talks, not real data,” Mr. Siddique adds.
Advocating the need to bring back the pre-2018 format and release the full breakup of marks along with the final results, he says, “UPSC (Mains) evaluation has subjective elements, so keeping things transparent is the best way to maintain trust. The Commission gives an individual marksheet to every candidate who took the exam, so they know their own paper-wise numbers. But that individual marksheet is completely inadequate because the candidate has no official peer data to compare it against.”
Atharva Tamta, 22, a B.Tech. graduate in Computer Science Engineering and currently residing in Old Rajendra Nagar, Delhi’s UPSC coaching hub, holds similar views, noting that the biggest impact has been on the quality of information available to UPSC aspirants. “Data from UPSC annual reports show that first-attempt candidates have a selection rate of roughly 6–10%, which rises to around 12.65% for second-attempt candidates and reaches its highest level among third-attempt candidates, at approximately 22–24%. Success rates appear to peak or remain around 23% before showing diminishing returns. This reflects the importance of analysing a candidate’s previous attempts and providing holistic data on their individual strengths and weaknesses.”
The UPSC provides a wide spectrum of opportunities once a candidate is selected. “However, the lack of information provided by the institution to help candidates identify, understand, and correct their mistakes raises significant questions. As a result, students are compelled to rely heavily on anecdotal evidence and guidance from coaching centres, private tutors, and other external sources,” says Mr. Tamta, who has not personally struggled with this issue yet, as he will be appearing for his first attempt in 2027. However, within his study circle, he has seen his seniors and peers struggle to identify their strengths and weaknesses.
A subject-wise bifurcation of marks would allow students to identify the areas they need to improve and strengthen. “Without such information, preparation becomes somewhat of a blind game, where everyone assumes they know what they need to work on, while the reality may be quite different. Prior to 2018, the UPSC provided a much richer dataset that allowed students to analyse their performance and make more informed decisions about their preparation. Aspirants could identify whether they needed to work more on General Studies, their optional subject, or the essay paper,” he says.
Coming from a Computer Engineering background, Mr. Tamta finds the present system particularly challenging because the UPSC currently provides only three engineering-related optional subjects—Mechanical Engineering, Civil Engineering, and Electrical Engineering—while leaving out contemporary fields such as Electronics, Biotechnology, Aerospace, Robotics, and Computer Science.
“This can lead students from these backgrounds to shift towards humanities subjects such as PSIR, Sociology, or Anthropology, even when they may not have an academic background in those disciplines. The lack of subject-wise and candidate-wise data makes this decision even more difficult for aspirants whose graduation subjects are not represented in the UPSC optional list. If aspirants could examine the marks of successful candidates from similar academic backgrounds, they would be able to make a more evidence-based decision about their optional subject and preparation strategy,” he said.
Earlier system
Before 2018, the DoPT’s CSE Plus portal made available the subject-wise marks of recommended candidates in each of the Mains papers and the Personality Test. From 2018 onwards, the publicly available list shows only the aggregate written (Main) exam marks, Personality Test marks and the final total.
“The earlier system enabled aspirants to see that the same final score could result from very different combinations of performance. A candidate might compensate for moderate general studies marks through a strong Essay, optional, or interview score. This provided a more realistic understanding of the exam than the aggregate total alone,” Mr. Aggarwal said.
It also helped aspirants assess the maximum, average, and median marks scored in different optional subjects, thereby enabling a more data-backed optional subject selection in the initial stage of their preparation. “The previous system enabled a more systematic comparison of paper-wise and optional-subject performance. The subject-wise data allowed aspirants to compare the marks obtained by selected candidates in different optional subjects, observe their scoring ranges, and track year-to-year trends. A repeat aspirant could also compare their own optional marks with these benchmarks and decide whether to retain the optional, change it, or modify their preparation strategy,” he added.
However, the marks of selected candidates alone cannot conclusively establish which optional is preferable for a given aspirant. It should be considered alongside a number of other factors, such as one’s academic background and one’s own inclination towards the subject.
On the whole, the aggregate marks reveal a candidate’s overall performance but not how those marks were distributed across papers. “ The issue is the absence of adequate information and a publicly explained rationale, rather than an unfair evaluation process. The disclosure of paper wise marks would enhance transparency and public confidence,” Mr. Agarwal said.
Reflecting upon the previous system, Mr. Kapoor highlighted thatsubject-wise marks of recommended candidates that were publicly available included marks across Essay, the four General Studies papers, the two optional papers, and the Personality Test, all of which created a much richer dataset.
“Instead of merely knowing that somebody scored, for example, a particular total in the written examination, aspirants could see the anatomy of that score. That matters a lot in UPSC preparation. A candidate could discover that apparently exceptional ranks were not necessarily the result of extraordinary scores everywhere. One candidate might have performed particularly well in essay, another in the optional, and another might have compensated for relatively modest written marks through a strong interview. It gave aspirants a more realistic understanding of how the examination actually rewards performance,” he said.
However, one should not conclude simply that an optional subject is “good” or “bad” based on the marks of successful candidates. “The candidates choosing different optionals are different populations, the number of candidates varies enormously, and difficulty and evaluation may vary from year to year,” Mr. Kapoor said.
Current practice
Presently, individual candidates can access their own detailed marksheets after completion of the entire examination process. For comparison with others, however, they largely depend upon marksheets voluntarily shared by successful candidates, or informed guesswork based on coaching-institute databases and crowdsourced spreadsheets. “These sources suffer from selection bias because unusually high marks are much more likely to be publicised than average or low marks. Aggregate written and interview marks are adequate for establishing the final merit position, but they are inadequate for serious comparative analysis,” Mr. Agarwal said.
If more granular information already exists, was historically disclosed, and can improve public understanding of a highly consequential examination, there should be a strong and clearly articulated reason for withholding it. “For an institution such as UPSC, transparency is valuable not merely because candidates want more information, but because transparency strengthens trust in the examination itself,” Mr. Kapoor added.
Viable solution
According to experts, the detailed distribution of marks should be released after the declaration of the final result, when the entire exam cycle has concluded (as was done before 2018). Releasing it before the interviews and final result may affect an ongoing selection process. “Once the final result has been declared, UPSC could publish a standardised dataset containing Essay, GS-I to GS-IV, Optional Paper I and II, written total, Personality Test marks and final total for recommended candidates. It should ideally be published every year in a consistent and machine-readable format so that serious longitudinal analysis becomes possible,” Mr. Kapoor said.
UPSC can also publish anonymised and aggregated statistics for the larger pool of Mains-appearing candidates– such as their subject-wise mean and median marks, score distribution, and number recommended. All this data is already readily available with the UPSC. “This would serve the genuine informational needs of aspirants and prevent misleading conclusions based only on a few publicly available toppers’ marks. Additionally, any future change in disclosure policy should be supported by a formally recorded and publicly explained decision,” Mr. Agarwal said.
In the current scenario, candidates can analyse their own marks, but the ecosystem-level comparative picture has become weaker. “Today, an aspirant largely relies upon his or her own marks, voluntarily shared mark sheets of other candidates, topper interviews, coaching institutes, and informal datasets compiled through Telegram, online forums, and social media. That is an inferior substitute for authoritative data because it suffers from selection bias. People who voluntarily publish their marks are not necessarily representative of the entire cohort. If the purpose is simply to know who qualified, the aggregate is sufficient. If the purpose is to understand how candidates performed across an examination consisting of seven merit-counting written papers, it clearly is not,” Mr. Kapoor said. He added that for good measure, removing data does not eliminate speculation; it often increases it.
(Rajlakshmi Ghosh is a journalist with over three decades of experience in media houses and educational institutions)
