Download Public OMMS Examiners Report 2021.pdf and more Study notes Statistics in PDF only on Docsity!
Examiners’ Report
Oxford Masters in Mathematical Sciences (OMMS)
Trinity Term 2021
Part I
A. STATISTICS
- Numbers and percentages in each class. See Table 1, page 1.
- Numbers of vivas and effects of vivas on classes of result. Not applicable.
- Marking of scripts. All Mathematics examination scripts were, as is the normal practice, single-marked according to carefully checked model solutions and a pre-defined marking scheme which is closely adhered to. The Mathematics dissertations and mini-projects were double-marked. A comprehensive independent checking procedure is also followed. (See the Part C Mathematics Examiner Report for details.)
Table 1: Numbers in each class
Number Percentages % 2021 2020 2019 2021 2020 2019 Distinction 21 27 16 40.38 65.85 50. Merit 9 7 6 17.31 17.07 18. Pass 21 5 9 40.38 12.2 28. DDM - 2 - - 4.88 - Fail 1 0 1 1.92 0 3. Total 52 41 32 100 100 100
B. New examining methods and procedure in the 2020 examinations
In light of the ongoing Covid 19 pandemic, the University changed the examinations to an open-book format and rolled out a new online examinations platform. An additional 30 minutes was added on to the exam duration to allow candidate the technical time to download and submit their examination papers via Inspera.
C. Changes in examining methods and procedures currently under discus-
sion or contemplated for the future
The department intends to hold in person exams in Trinity Term 2022.
D. Notice of examination conventions for candidates
The first notice to candidates was issued on 26th February 2021 and the second notice on 30th April 2021. These contain details of the examinations and assessments.
All notices and the examination conventions for 2021 are available online at:
https://www.maths.ox.ac.uk/members/students/undergraduate-courses/teaching-and-learning/part- c-students/examinations-and-assessments/part-c-and-omms
C. Detailed numbers on candidates’ performance in each part of the exam
See Table 3, page 4 for the number of candidates taking each Mathematics paper, together with statistics for the raw marks (average and standard deviation), and USMs (average and standard deviation) attained on each paper by this cohort. All papers listed are units except the Mathematics Dissertation, which is a double unit. The total maximum raw marks for a unit is 50 whilst the USMs are scaled to a maximum of 100. In accordance with University guidelines, statistics are not given for papers where the number of candidates was five or fewer.
Table 3: Statistics by paper (Mathematics papers)
Paper Number of Candidates AvgRaw StdevRaw AvgUSM StdevUSM C1.3 7 17.29 4.68 54.71 8. C2.4 7 26.14 7.71 64 8. C2.7 8 22.25 10.08 54.12 20. C3.1 8 26.38 14.38 67.62 19. C3.2 7 22.57 6.8 64.14 3. C3.3 15 18.27 7.59 56.4 15. C3.4 6 37 8.1 70.33 15. C3.5 7 23.29 8.18 58.14 13. C3.8 8 26.12 5.87 59.88 5. C3.11 6 23.5 3.83 67.67 5. C4.1 7 19.14 7.71 53.43 15. C4.3 6 24.67 8.87 59.83 10. C4.6 6 27.5 9.35 60 13. C6.1 11 26.18 13.2 64.73 18. C6.2 12 28.67 6.65 63.17 12. C6.3 7 31.57 8.92 68.71 12. C6.4 6 31.83 8.42 66 13. C8.1 8 36.38 8.18 66.5 14. C8.3 7 25 10.72 60.43 13. SC1 10 31.8 6.66 63.7 10. SC2 8 34.38 7.07 69.88 10. SC4 12 27.42 9.16 63.5 15. SC5 8 38.88 8.53 64.5 11. SC10 7 25.14 3.44 60.71 4. C5.4 19 - - 73.11 11. C6.5 18 - - 65.89 5. CCD 52 - - 73.83 11.
D. Recommendations for Next Year’s Examiners and OMMS
Supervisory Committee
Most of the recommendations are shared with the Part C Mathematics Examiner Board including
- The scaling algorithm which used Part B exam data should be reviewed, and even more preferably changed.
- Length of dissertation should be checked by other means instead of or in addition to the 7500 word limit, which was hard to verify.
- Penalties for late submission of online exam should be graduated penalties rather than a harsh cut-off.
- The rule for distinction/merit/pass should be more nuanced. In addition, it should be better communicated to assessors. For example, it was felt that not all assessors understand what the borderlines for these classifications are in their assessor reports.
- The effect of the new degree classifications on the scaling practice should be reviewed.
- Scaling practice should be done locally before the final board meeting.
- It should be better communicated to external examiners how their comments were taken into account.
For further details regarding the above recommendations, see Part C Mathematics Examiner Report.
We turn to the recommendations or comments which are more applicable to the OMMS examiner board and mostly concern discrepancies in the treatments for OMMS and Part C students. The board recognised that efforts have been made in previous years to ensure more transparency, agreements and fairness in the treatment of OMMS and Part C students. The board however felt that a number of differences still exists which should be addressed.
- Scaling algorithm. The current scaling algorithm use solely the data from the previous year Part B exam results. It is arguably unclear whether or how this practice is fair to OMMS students who had entirely different education before coming to Oxford.
- Exam prizes. The board felt that there is a strong case to match OMMS exam prizes with Part C exam prizes, in terms of both number of prizes and amount for each prize.
- Dissertation feedback. Currently, feedback on dissertations are given to OMMS stu- dents but not Part C students. If this is to stay in the future, assessors should be made aware of this fact prior to their assessing students’ work.
Finally
