From the Physics Admissions Coordinator

Report on the Physics Admissions Exercise 2023

In 2023, Oxford Physics received a total of 1672 applicants for places in Physics or Physics and
Philosophy, an increase of 39 (2%) on the 2022 figures. Of these, 1652 applicants were
contesting the 187 places available for 2024 admission, or approximately 8.9 applicants per
place, with 20 applicants seeking deferred places. Compared to 2022, there was a slight
decrease in the number of available places, reflecting post-pandemic capacity pressures in
colleges.
Of the 1672 applicants, 1118 (66.9%) were classified as ‘UK’ applicants (63.3% in 2022), 96
(5.7%) were classified as EU but not UK (7.3% in 2022), and 458 (27.4%) were classified as
non-EU (29.4% in 2022).
Across the collegiate university, Physics usually aims to interview around 2.5 applicants per
place; this number was increased to about 2.9 this year for the reasons described below. For
short-listing, we use the results of the Physics Aptitude Test (PAT) as well as all other
contextual information described at
http://www.ox.ac.uk/admissions/undergraduate/applying-to-oxford/decisions/contextual-
data.
The contextualised GCSE 1 (cGCSE) score produced by the University was used as part of the
shortlisting information. This is a return to previous practice, and follows the reinstatement of
externally-assesed GCSEs rather than the teacher-assessed grades used during the pandemic.
The cGCSE score is based on the ability to compare GCSE results between applicants from
schools that are contextually ‘similar’.
The PAT has been run for many years, and it is a good predictor of future performance at
Oxford. The test is set to a defined syllabus and both the content and draft questions are
checked by school teachers to ensure that the level is appropriate. Maths and physics
elements are mixed together into a single two-hour paper. Each question is separately
double-blind-marked, with markers focusing on individual questions to ensure consistency of
approach. Further details, including the admissions criteria and sample papers, can be found
on the Oxford Physics Admissions website at:
https://www.physics.ox.ac.uk/study/undergraduates/how-apply/admissions-procedures-
physics-courses .
In 2023 it had been planned, in line with other Oxford admissions tests, to change the format
of delivery of the PAT such that questions were delivered online, with an online calculator,
and with responses written on paper. Due to significant technical issues that were
experienced during the delivery of other Oxford online admissions tests, the PAT was instead

1 The cGCSE score is expressed as the number of standard deviations the applicant is away from their ‘expected’ number of
A*/9/8 grades and was typically be in the range −3 to +3, expressed to 2 decimal places. Overseas applicants, or others lacking
GCSE information, were assigned a neutral cGCSE score of zero.

Page 1 of 5
From the Physics Admissions Coordinator

delivered in a paper format for most candidates. This change of format significantly reduced
the disruption that would have otherwise been caused to candidates. Despite these
mitigating actions, the late change of format, environmental factors, and other resulting
differences such as the change from an online to a physical calculator, caused disruption to
candidates.
Notices of special considerations resulting from the change of PAT format were received and
were considered at all stages of the subsequent process. Where candidates’ tests were
disrupted, the form of their disruption was considered in decision-making at the time of
shortlisting and of offering places. To accommodate candidates whose performance in the
PAT had been affected, the number of interviews was increased from 2.5 to about 2.9
applicants per place. A small number of candidates whose circumstances meant they were
unable to take the PAT, or who were the most severely disrupted by the change in delivery
format were invited to take an alternative PAT, during the same period as the interviews, the
results of which were also considered when making decisions about places.

Number of candidates (y-axis) against PAT score (x-axis) by

candidate outcome (Dec 2023)
50
45
40
35
30
25
20
15
10
5
0
0
3
6
9
12
15
18
21
24
27
30
33
36
39
42
45
48
51
54
57
60
63
66
69
72
75
78
81
84
87
90
93
96
99

All applicants Shortlisted Placed

We are always extremely grateful to all schools and test centres for hosting candidates for the
PAT. This is particularly true during this year in which the late change to the format caused
significant late changes both to students and to those responsible for delivering the tests. We
are also grateful for the yearly advice we receive from schools and teachers on adapting the
PAT to changes in school syllabi, and on the effect of the test delivery format. We expect to
continue to make further changes reflecting such advice in subsequent years.

Page 2 of 5
From the Physics Admissions Coordinator

There were a significant number of declared special circumstances, medical certificates or

letters drawing attention to adversities in applicants’ personal lives that may have affected
performance or ability to participate in the test. All of these were taken into account both in
making shortlisting and offer decisions.
The 2023 PAT was designed to be a little less time-constrained than the 2022 one, to smooth
the transition to the intended online format. The marks achieved by applicants who sat the
main PAT ranged from 8% to 96%, with a mean mark of 55.6% (previously 51.2% in 2022,
43.1% in 2021) and a standard deviation of 18.6% (16% in 2022, 17.7% in 2021). More details
are shown in the graph above (this only includes marks on the main PAT). The spike at ‘0’
includes applicants who withdrew, applicants who did not register for the PAT and/or
applicants who were unable to sit the main PAT. This year continued a pattern of recent years
whereby of the top 100 applicants by PAT mark, only around a quarter of these had all their
secondary schooling within the UK educational system (either state or independent sectors).
In a return to pre-pandemic practice, the principal determinant for shortlisting this year was
the R-score. Pre-interview this score is given by:
R-score pre-interview = PAT mark + 10 x cGCSE.
Our aim in the admissions process is to take the applicants whom we judge to have the most
ability and potential to benefit from our course and teaching. Likewise, in the admissions
process we are aiming to project applicants forward to their performance on-course, not just
comparing the levels attained at age 17 or 18.
The 346 applicants with R-scores equal or higher than 75% were automatically shortlisted for
interview, with a further 54 applicants with slightly lower scores also automatically shortlisted
after the inclusion of contextual data, giving a total of 400 automatically shortlisted applicants
(307 in 2022 and 298 in 2021). A further 190 applicants (compared to 164 in 2022, and 202 in
2021) who were below the automatic thresholds but whose application forms showed other
evidence of excellence and/or mitigating circumstances, including considerations of various
levels of technical disruptions during the PAT, were also shortlisted.
A total of 536 applicants were therefore invited for (remote) interview this year. A key goal of
the Oxford admissions process is that the probability of admission should not depend on the
applicant's choice of college. Short-listing is therefore followed by a reallocation process, in
which applicants are transferred from first-choice colleges with a large ratio of applicants per
place, to colleges with a smaller ratio of applicants per place. This aims to ensure that, for
each college, the ratio of interviewed first-choice applicants to places is as close as possible
regardless of college. This year 105 applicants were reallocated to a different first choice
college at the time of shortlisting. Reallocation has been practised by the University for many
years, ensuring that all strong applicants have the same chance of obtaining places at Oxford,
although possibly not at their first-choice college. Reallocation is not an indicator of the
strength or weakness of an applicant; applicants with very high PAT scores can be reallocated.
Every short-listed candidate has two interviews given by a first-choice college and one given

Page 3 of 5
From the Physics Admissions Coordinator

by a randomly allocated second-choice college. Each interview is marked out of 10 based on

the academic judgement of the interviewing tutors. The scale is such that a mark of 6 broadly
corresponds to ‘acceptable’; 7 corresponds to ‘good’; and an average interview mark of 8 or
higher will almost certainly result in an offer. Approximately 1% of interviews are scored as
‘10’.
Applicants are assessed based on the totality of information about the applicant with no one
interview, by itself, decisive. While most of the accepted applicants have three good
interviews (at least as viewed by the interviewers), 68 accepted applicants had one interview
which scored less than a 7, while 14 accepted applicants had one interview which scored
below 6. It is very hard for applicants to assess their own interview performance and we know
from conversations with accepted students that it is extremely common for applicants to
think that interviews which have gone well – even very well – have gone badly. For applicants
offered a place, the average interview mark this year was 7.95 (7.98 in 2022).
We would like to express our gratitude for the hard work of both applicants’ parents (for
applicants interviewing from home) and teachers and IT staff of applicants’ schools (for those
interviewing from school) for their work in facilitating the interviews and making appropriate
spaces available.
After the interviews, the three interview marks are combined into a single score (out of 100).
To guide admitting tutors, an overall ranking was produced based on the post-interview R-
score:
Post-Interview R-score = (PAT mark out of 100) + 10 x cGCSE + 2 x (Interviews out of 100)
This ranking is for guidance only; all applicants are assessed individually based on their R-
scores, PAT scores, interview results, and all information on the UCAS form, including
contextual information, and then compared centrally against all applicants applying to Oxford
Physics.
To ensure that the strongest applicants obtain places, all colleges have access to information
on all applicants through a central database, and colleges are actively encouraged to flag up
strong applicants they will be unable to offer a place to themselves. As a result, 15 applicants
were offered a place at a college that had not interviewed them at all, either as first or second
college.
Ultimately, 199 offers were made for entry in 2024. These include 9 open offers, in which a
college is not specified at the time of the offer. These are designed to cover the anticipated
withdrawal rate of applicants who are made an offer and subsequently either decline the
offer or fail to make the offer conditions. The offers include 16 offers made for Physics and
Philosophy. A further 2 deferred offers were made for entry into Physics in 2025.
Every year, there are applicants who are excellent physicists who we do not get a chance to
see at interview. There are also applicants who we interview and we would have liked to have
offered places to, but are unable to do so because of the finite capacity of the course. We
know that every year we turn down applicants who, in the end, turn out to be stronger

Page 4 of 5
From the Physics Admissions Coordinator

physicists than some of the applicants we do offer places to. We wish all applicants
enjoyment and understanding in their future pursuits of physics.

Page 5 of 5