Figures released after a two month investigation into the Cambridge supervision system have revealed major disparities in supervision provision across the university. While anecdotes about some colleges working their students harder than others have long been circulated, the investigation uncovers the detail behind the rumours.

The news calls into question a fundamental feature of the collegiate system: across a number of courses, the average student at some colleges can receive over 100 per cent more time in supervisions and college classes than their peers studying elsewhere - despite paying the same tuition fee. The disparities appear to be particularly pronounced for first year students beginning their Cambridge education.

The data covers supervisions, college classes and college seminars from the last academic year. The figures, released under the Freedom of Information Act, also show which colleges rely more on graduate student supervisors than experienced academics to supervise their undergraduates.

Small group supervisions represent the defining feature of an undergraduate education at Cambridge and Oxford and while the University of Cambridge prospectus tells applicants and students that “it’s not the case that some are better for particular courses”, these new figures suggest that the college you choose can have a serious impact on the supervisions you receive.

Economists face one of the starkest inequalities in supervision provision. For first year economics students the supervision gap can be as much as 71 hours, with the average Newnham student receiving 115 hours over the year, but only 43 hours if they attended Sidney Sussex last year. First years at Gonville & Caius saw supervisors for 89 hours respectively, while those at King’s and Trinity had 66.

As part of the investigation, a Varsity survey of Cambridge students revealed conflicting views on the strength of graduate student supervisors compared with more experienced academics, and showed that the supervision gap is being felt by students across the university.

Historians also face wide variation in college contact time hours. History prelim students at Girton, Selwyn and Homerton received around 25 hours of supervision on average last year, less than half the college contact time had by King’s, Peterhouse and Newnham first years with 53, 55 and 59 hours respectively.

One King’s history student told Varsity that they knew “other colleges have more classes organised internally by directors of studies”, giving one explanation for this variation.

The gap between the college providing the most and the least hours of supervision for English students studying for Part I is 39 hours; the average undergraduate received 108 hours with academics last year at Magdalene and 104 at Trinity, but only 69 if they were at Peterhouse or Robinson.

For computer scientists, the supervision gap can be as much as 51.5 hours. First year students at Magdalene were in supervisions for an average of 111 hours last year, while those at Girton received 60, 61.75 at Newnham and 68 hours at Selwyn and Caius on average.

Archeology & anthropology students in Part I of their degree at Sidney Sussex received an average of 57 hours of supervision time, while their Trinity counterparts received 43. Those at Homerton had 38 and at Robinson only 33.

The supervision gap for second year mathematicians was up to 33 hours, with Lucy Cavendish offering 62 hours of college contact time compared to just 28 hours at Homerton and 30 at Magdalene.

For engineers the supervision system is also a lottery: first years received an average of 67 hours at Homerton, but only 37 hours at Robinson, while second years could receive 71 hours at Homerton, but 47 if they attended Lucy Cavendish.

One English second year at Emmanuel College said they “receive very little supervision” for certain papers, even though they know that students at “some colleges receive weekly supervisions and essays” for them.

The student, who wished to remain anonymous, also told Varsity that despite the fact “Dissertation and Portfolio supervisions for Part I are limited” in number by the Faculty of English, “it is well known that certain supervisors will give students extra supervisions.”

While many subjects saw great variations in contact time across the university, the investigation also reveals the courses in which students have a more consistent supervision experience. The supervision figures for geography, history of art, natural sciences, PPS, music, land economy and law are relatively similar whichever college undergraduates attend, reflecting the fact that supervisions in several of these courses are centrally organised.

In Varsity’s survey, one Trinity law student highlighted that the central organisation of supervisions “makes the system feel quite fair” in comparison to other courses. Similarly, a second year geographer pointed out that for first years there are “vast differences in the numbers of supervisions the different colleges receive” when contact time is not organised by the department.

The figures also underline the significant variations in contact time between different subjects. While third year English students in Cambridge can expect to receive an average of 51 hours of college contact time, finalists studying philosophy only see supervisors for an average of 19 hours across the whole year. It is little better for third year geography and PPS students, who tend to see supervisors for an average of 21 and 22 hours respectively, while those reading MML and law receive around 42 hours of sessions with academics.

According to data, in a number of subjects at several small and medium-sized colleges, such as Magdalene and Clare, graduate student supervisors outnumber academic supervisors employed by the college to supervise for a range of courses. Across the university, the figures show there seems to be a shortage of academic supervisors for philosophy, linguistics, computer science and engineering.

One second year geographer at Sidney Sussex told Varsity they thought they were given too little contact time, and complained that graduate students were “not as good” as more experienced academics “as they get mixed up and often don't specialise in what they're supervising us on. One had to email us after the supervision because he'd told us completely the wrong thing.”

A Magdalene natural scientist said that “not only do graduate students make sometimes visible scientific mistakes, but they are less likely to ask interesting and challenging questions during supervisions”.

An English student at Christ’s told Varsity: “in terms of preparing you for the wider range of subjects you may come across in exams I prefer the more experienced academics. The difference is those academics have been writing the exams for years and years”. Similarly, one Trinity law student complained that among graduate student supervisors “there was a distinct lack of knowledge about how to approach exam questions or structure exam answers within the time limit.

However, a large number of respondents to the survey suggested that graduate students often made better supervisors, telling Varsity that their communication skills, teaching technique and understanding of student life tended to surpass that of fellows and professors.

“The graduate supervision are as good as, if not better then the experienced academics”, a second year scientist at Magdalene said. “They seem to be more structured, and a lot of relevant experience is given by people who have only recently sat the same exams.”

A Classics student at Robinson said they found their graduate supervisors to be “extremely informative and in many ways much better at communication than the experienced academics who supervise me”, while an engineer at the same college said PhD supervisors were “much more friendly, down to earth, and understanding of student issues”.

Varsity’s investigation made colleges reveal data on supervisions drawn from CamCORS, the Cambridge college’s online reporting system for supervisions, and colleges noted that in some cases data may be partially inaccurate or incomplete due to incorrect or differing methods of recording supervision data on the system. The figures may also have been influenced by actions on the part of students rather than colleges, for instance if undergraduates degrade or fail to attend supervisions which are provided for them.