How to Conquer GS in UPSC Mains, Explained By Anudeep Durishetty AIR 1 CSE 2017 Writing a good Essay in UPSC Mains, Explained by Anudeep Durishetty Answer Writing Focus Group for Mains 2018 – Details DOWNLOAD: UPSC Prelims 2018 GS Paper I UPDATE : Answer Writing Group ( Offline | New Delhi ) Registrations Are Now Open!
27 JULY | 200+ Issues for Mains 2018 ForumIAS is conducting classes for Current Affairs Mains. Cover upto 200+ Issues important for Mains , with specific focus on holistic coverage and content differentiation. Attend the first session ( Open to All ) at 11AM, on Friday July 27 at ForumIAS Offline Center, New Delhi.
CLICK HERE to read more and register for the First Open Class
Current Affairs ( Pre + Mains ) 2019 | 4 Aug | Attend First 3 Classes Free ForumIAS Weekly Current Affairs Classes commence from Saturday, 4 August, 5AM, at ForumIAS Offline Center, New Delhi.We invite you to attend the first three classes free
CLICK HERE to register for attending the classes on August 4, 5 and 11.

Scaling in UPSC Civil Services Mains Examination - Explained

edited February 2013 in Miscellaneous
Thanks to @asg who discovered this text, I am reproducing it below:-

Scaling in UPSC Mains Explained



image
The head examiner is also the paper setter.
If the number of papers to be examined are less than 350, the head examiner checks all the copies but if the number exceeds this limit then for each additional 300 copies an additional examiner is appointed. The head examiner convenes a meeting of all the examiners in the UPSC premises within seven days of the conduction of an optional subject. All of them are asked to prepare model answers for all parts and sub parts of the question paper. These questions and answers are then thoroughly discussed. At the end of the discussion the head examiner with the consensus of the additional examiners issues final guidelines.



After this the examiners examine all the copies within a specific time frame.Once this is over, all the copies are arranged in vertical stacks (examiner-wise) with the head examiner. The head examiner then conducts a sample survey.



In the sample survey he chooses 20 copies from each stack (examiner). These 20 copies are - 5 which scored the highest in the stack; 5 which scored the lowest in the stack and 10 selected randomly. The head examiner then conducts upward or downward moderation of the marking of these copies, if the need be. If the discrepancy or deviation in any of the stack(s) is found erratic, he moderates the entire stack then.




Once the subject scrutiny is over the UPSC resorts to a statistical process known as 'normative linear regression' and brings all the subjects on a common plane. In the case of GS there is no scaling as such, although the rest of the process, before that, remains the same.
In the case of essay the process is entirely different.

There are no head examiners. There are a score of examiners appointed 2 per 300 copies. Each copy is evaluated by two examiners and the mean score of the two scores is taken into consideration. the marks awarded to a script by the first examiner is kept confidential from the second examiner.
*No good deed goes unpunished* | I am no knight, do not call me Sir | I write articles for Civil Services Aspirants, you can find them at http://blog.forumias.com/author/neyawn |

Comments

  • Further to this, scroll down to the bottom of Page 8 in the following:
    http://www.cic.gov.in/HC-Rulings/UPSC-Vs-CICandOrs.pdf
    Fate, it seems, is not without a sense of ïrony
  • Thanks @asg those really are some awesome findings
    and Drona asked,"what do you see?","the sparrow's eye!"......and Krishna replied,"Turning their discerning mind to That, directing their whole conscious being to That, making That their whole aim and the sole object of their devotion, they go whence there is no return"
  • @Neyawn @asg =D> ^:)^ ^:)^ ^:)^
    Man is sometimes extraordinarily, passionately, in love with suffering.
  • Can't take credit....just copy-pasted from somewhere
    Fate, it seems, is not without a sense of ïrony
  • @Anurag_Sinha Yahan ka asli Hero @robinolds hai. Joki aaj gayab hai :-?
    *No good deed goes unpunished* | I am no knight, do not call me Sir | I write articles for Civil Services Aspirants, you can find them at http://blog.forumias.com/author/neyawn |
  • edited February 2013
    @Neyawn robinolds rhymes with robinhood. hum garibon ka rakhwala.. \:D/
    Man is sometimes extraordinarily, passionately, in love with suffering.
  • This is certainly the best community. I was trying to find out scaling process by the UPSC for some time. Now I am going to study normative linear regression. :)

    @Neyawn and @Anurag_Sinha kyo taang kheech rahe ho yaar.
    “So many books, so little time.” ― Frank Zappa | Indian Police Service 2013
  • I am going to study normative linear regression
    @robinolds
    Stats ka all time champ hona hai kya? Lol.
    I still cant believe you did stats fully!
    *No good deed goes unpunished* | I am no knight, do not call me Sir | I write articles for Civil Services Aspirants, you can find them at http://blog.forumias.com/author/neyawn |
  • hahaha
    I thought I should know how my marks are going to be up-scaled in Pub Ad and down-scaled in socio :)
    “So many books, so little time.” ― Frank Zappa | Indian Police Service 2013
  • @asg

    I wrote this in Indian Officer.
    When victory is certain, even the cowards can fight.
    Brave are those who fight even when their defeat is certain.
  • @Neyawn with reference to your first post, where did you get this data?
    any official source?
  • UPSC uses n-linear regression? Brilliant
    This institution never fails to surprise me.Too cool
    Pehle vey aap par dhyan nahi denge, Fir aap par hasenge, Fir aapse ladenge, Aur tab aapki jeet hogi.
  • Further to this, scroll down to the bottom of Page 8 in the following:
    http://www.cic.gov.in/HC-Rulings/UPSC-Vs-CICandOrs.pdf
    This is only for Pre and not for mains from a cursory look. Please correct me if am wrong.
    When the going gets tough...
    ... squeal like a pig.
  • Even i find it mentioning Prelims and not Mains. But there is no reason why the same cannot be followed in Mains.
    Management || The needs of the many outweigh the needs of the few - Star Trek II Wrath of Khan (1982)
  • edited January 2014
    Quoting from the pdf :
    The Normalised Equi-Percentile method used by the UPSC has been explained as under:-

    This method is based on the assumption of comparability among candidates taking various optional subjects. It is fair to assume that the mental ability (and consequent performance) of candidates in all optional subjects are about the same at very score range. We can assume that top 5% of say History candidates are comparable in ability to the top 5% of say Geography candidates. This assumption can be extended to other score range such as 10%, 15%, 20% etc. Thus, it is possible to statistically adjust the scores in various subjects.
    Further since the number of candidates for each subject is large (over 1000) it is reasonable to assume that the scaled marks should lie on a normal curve. For the normal distribution curve of each optional subject, mean of 150 and standard deviation of 30 (for a paper with maximum marks of 300) have been taken. The scaled marks are computed using the standard Statistical Tables-Areas under the standard normal curve.
    Management || The needs of the many outweigh the needs of the few - Star Trek II Wrath of Khan (1982)
  • if there is scaling than how come last year pub ad marks and maximum marks remain so low
    why wont our marks got scaled up.
    also i think if going by last year pub ad there is no separate cutoff in each paper too.
  • that pdf talks abt scaling for prelims-optionals-scaling only .. not about mains scaling ..
  • 1. I read the High Court and the Supreme Court order very carefully. The process described in the UPSC-Vs-CICandOrs.pdf is only for pre and not for mains and is not obsolete as both pre papers are common to all now.

    2. The process for mains is described very correctly in the very first post by Neyawn. However the paragraph 4 is misinterpretation on the lines of what @Noob mistakenly thought as well. The following is an excerpt from SC judgement which describes the process to bring parity among different subjects owing to varying levels of difficulty etc.
    "Apart from the differences in the scope and coverage of the syllabi; the difficulty level of the question-papers, and the standards of evaluation are therefore inevitably different and can vary from year to year across subjects/papers. Based on a holistic perspective, therefore, and with its decades of experience, the Commission applies upward or downward inter-subject moderation, wherever required. This is done to ensure a level playing field for all candidates. It is important to note that at this stage too, only statistics are taken into consideration with full anonymity as regards candidates' details."

    3. The judgement no where specifies the exact statistical methodology used like "regression" etc because as per the above quote no such method is employed. As can be inferred from the quote above inter-subject parity is accomplished only on the "holistic perspective" and "decades of experience" of the commission and not on any mathematical model. This reveals why certain optionals score low despite the belief of most candidates that scaling occurs (because there no "scaling" as such) and should answer @vinay88 's doubt.

    References:
    Hon. High Court Judgement : WP(C) No.17583/2006
    Hon. Supreme Court Judgement: Special Leave Petitions (C) Nos. 11977/2012, 11978/2012, 11979/2012 11980/2012, 9333/2012 (this is available on UPSC website itself easily)
    When the going gets tough...
    ... squeal like a pig.
  • Please correct me and tag me if anyone finds me wrong. It will only enhance our understanding of this opaque process.
    When the going gets tough...
    ... squeal like a pig.
  • i guess they are having a scaling panel - 'discretionary'
    hey bhagwan unhe sadbuddhi dena [-O<
  • I think @Drjanitor is spot on in his interpretation. If the scaling worked scientifically, pub ad's topper should have got the same marks as another paper's last year.
  • edited January 2014
    I know shit about this normative linear regression or any other statistical methods but this is my "on the ground" understanding of the scaling method.
    Assume 100 guys have written a paper which was deadly difficult, now out of this according to me, the top 15 guys get their marks up-scaled, the next 35 remains constant and the rest 50 gets their marks down-scaled! I have no idea why and how this is followed, but this is the only way to explain the scaling process of last year's pub ad paper. The paper was very tough, yet the rank holders have got good marks while some have even got 8/9/ single dight marks in paper 1. There's no way someone can get a single digit mark unless it is down-scaled! So double whammy, first u get the paper tough and u get your marks down-scaled ! :-w
    |If you tremble with indignation at every injustice, then you are a comrade of mine|| ||Upsc wars veteran|
  • I think @Drjanitor is spot on in his interpretation. If the scaling worked scientifically, pub ad's topper should have got the same marks as another paper's last year.
    that's correct. the topper of pub ad last year has a rank in the 300s
    |If you tremble with indignation at every injustice, then you are a comrade of mine|| ||Upsc wars veteran|
  • I know shit about this normative linear regression or any other statistical methods but this is my "on the ground" understanding of the scaling method.
    Assume 100 guys have written a paper which was deadly difficult, now out of this according to me, the top 15 guys get their marks up-scaled, the next 35 remains constant and the rest 50 gets their marks down-scaled! I have no idea why and how this is followed, but this is the only way to explain the scaling process of last year's pub ad paper. The paper was very tough, yet the rank holders have got good marks while some have even got 8/9/ single dight marks in paper 1. There's no way someone can get a single digit mark unless it is down-scaled! So double whammy, first u get the paper tough and u get your marks down-scaled ! :-w
    That is precisely the misunderstanding. There is no such method of scaling like you imagine there to be. It is purely based on the past experience of commission officials and historical scoring of the optional as far as I can tell (And even this "moderation" is done only on discretion of the commission if it deems it necessary. It is not obligatory). People who have scored single digit marks are most likely victims of the so called "butchering" rather than the non-existent "scaling" process.
    When the going gets tough...
    ... squeal like a pig.
  • Thanks to @asg who discovered this text, I am reproducing it below:-

    Scaling in UPSC Mains Explained

    image
    The head examiner is also the paper setter.
    If the number of papers to be examined are less than 350, the head examiner checks all the copies but if the number exceeds this limit then for each additional 300 copies an additional examiner is appointed. The head examiner convenes a meeting of all the examiners in the UPSC premises within seven days of the conduction of an optional subject. All of them are asked to prepare model answers for all parts and sub parts of the question paper. These questions and answers are then thoroughly discussed. At the end of the discussion the head examiner with the consensus of the additional examiners issues final guidelines.

    Can you please indicate the source of this information?(where this text was discovered)
    In a RTI filed by mrunal, UPSC has held that it does not give any model answers.
    http://mrunal.org/2012/07/rti-upsc-cse2011-mru.html
  • @drjanitor you are correct about it.

    Scrubs fan?

    Nothing in this world worth having comes easy :-)
    *No good deed goes unpunished* | I am no knight, do not call me Sir | I write articles for Civil Services Aspirants, you can find them at http://blog.forumias.com/author/neyawn |
  • @ramesh,

    1. The source is http://www.cic.gov.in/HC-Rulings/UPSC-Vs-CICandOrs.pdf
    2. This was for prelims examination where there was an optional prelims paper unlike now when it has been replaced by the CSAT paper. Hence it is obsolete.
    *No good deed goes unpunished* | I am no knight, do not call me Sir | I write articles for Civil Services Aspirants, you can find them at http://blog.forumias.com/author/neyawn |
  • @drjanitor you are correct about it.

    Scrubs fan?

    Nothing in this world worth having comes easy :-)
    @Neyawn , More precisely Janitor fan ;)
    When the going gets tough...
    ... squeal like a pig.
Sign In or Join to comment.

Courses by ForumIAS for CSE
ForumIAS is trusted by over 10,000+ students for their Prelims, Mains and Interview Preparation and we currently run several assistance programs to help students from Civil Services prelims preparation to rank upgradation to IAS. You can enroll for these programs by visiting http://academy.forumias.com

Welcome!

We are a secret self-moderated community for Civil Services preparation. Feel free to join, start a discussion, answer a question or just to say Thank you.

Just dont spread the word ;)

Sign in or join with Facebook or Google

Subscribe to ForumIAS Blog