Data
The data on this site are taken from undergraduate classified examination results published by the University of Cambridge.
"Other" results are excluded, since the category covers various circumstances that cannot be distinguished. The numbers in the "other" category are small so the choice to exclude them has minimal influence on the tables. The total numbers for each College are the total numbers of classified exam results reported. Half integral numbers arise where subjects award two separate classes to one student. Each class adds one half to the total to avoid double weighting the performance of a single student. Undivided seconds are also scored halfway between a 2.1 and a 2.2.
The published data are broken down by year group and gender, so this site allows the tables to be broken down by those categories. The publically available data do not break down results by subject, so it is not possible for this site to produce tables by subject.
The data for 2020 and 2021 are impacted by the COVID-19 pandemic. In 2020, there are no classified results for Years 1 and 2. Results for Years 3 and 4 are affected by a "safety net" policy, whereby students were guaranteed a result at least equal to their result in the non-COVID-affected Year 2. In 2021, classified results in Years 1 and 2 are available in only some subjects. The tables for 2020 are definitely not comparable to other years. The tables for 2021 should also be viewed cautiously.
Table Type
Baxter: the applied scoring scheme is: 1st = 5 points, 2.1 = 3 points, 2.2 = 2 points, 2(undivided) = 2.5 points, 3rd = 1 point. The same scheme is applied to calculate the Norrington Table at the University of Oxford.
Good Degree: a "good degree" is either a 1st or a 2.1. The applied scoring scheme is: 1st = 1 point, 2.1 = 1 point, 2.2 = 0 points, 2(undivided) = 0.5 points, 3rd = 0 points. The overall score is equivalent to the percentage of good degree results. These values may not match exactly those given on the primary publication site, because "other" results are excluded and undivided seconds also contribute to the totals on this site. The purpose of the Good Degree table is to present the information in a consistent manner to the Baxter tables, including a measure of statistical significance.
Table Contents
The tables list the total number of results (N) and the counts of each class. There is an option also to show the percentage of each class. The score for each College is calculated by applying the relevant scoring scheme to yield a sum, then dividing by N to yield the mean. The score is presented as 100 times the mean. In the Baxter table, the maximum score is 500.** In the Good Degree table, the score corresponds directly to the percentage of good degrees.
The Z score provides a measure of the statistical significance of the result. It is a measure of the extent to which the sample of results at a particular College differs from the results of the population as a whole. In the tables, the "population as a whole" is the collection of results that appear in the table. For example, if the table is restricted to five Colleges, the population as a whole comprises all results for those five Colleges. The sign of the Z score indicates whether the College score is greater or less than the mean of the population as a whole. The magnitude of the Z score indicates how significant that difference is. A Z score with magnitude greater than 1.645 is considered to be a significant difference at the 90% confidence level. In the tables, such values are coloured green (significantly good compared to the population as a whole) or pink (significantly poor compared to the population as a whole). The magnitude of the Z score is influenced both by the extent of the difference between the College score and the score of the population as a whole, and by the size of the College sample compared to the size of the population as a whole. For example, a smaller College could be ranked above a larger College on score, but it may have a smaller Z value because its sample size is smaller. Smaller sample sizes yield less confident conclusions.
** The Norrington Table published by the University of Oxford expresses the score as a percentage of the maximum (divide the Baxter score by 5).
Course Years
The course years refer to the categorisation reported on the primary publication site. Since different courses have different exam structures, some students may not take classified exams in some years of their course. For an overall picture of exam performance at a College in a given academic year, the default table merges all course years. Individual or multiple course years can be selected. Categorisation of "Final Course Year" formally requires results by subject, so this is not included on this site.
Genders
The genders refer to the categorisation reported on the primary publication site. The default table merges all genders. Results recorded as "other" gender are excluded on the primary publication site. The numbers are small so the choice to exclude them has minimal influence on the tables.
Colleges
The default selection comprises the undergraduate Colleges of the University of Cambridge. Colleges omitted from the default selection are Hughes Hall, St. Edmund's, Wolfson, and Lucy Cavendish (until 2025; see below). The Colleges to be included in the table can be selected through the Colleges link in the navigation menu. Setting to "Custom" opens a selection pane where any College can be selected or excluded. The selection made in this pane is remembered until it is changed again or the Colleges setting is changed back to "Default". For calculating Z values, the "population as a whole" comprises only the selected Colleges. In some circumstances, a College may have no results: in that case, the College will not appear in the table (even if selected).
Lucy Cavendish (LC): the first full set of undergraduate exam results is Year 1 in 2022. By 2025, LC will have a complete set of undergraduate year groups, and will be included in the default selection. In the meantime, the College has exam results only for some course years. The implementation on this site is to omit LC from the default selection until 2025. It is possible to add LC in the Colleges selection.
Data by Subject
Since the publically available data do not break down results by subject, it is not possible for this site to produce tables by subject, or by Arts/Sciences, etc.