Before this week’s University admissions statistics were released, we already knew that state school admissions had increased for the 2012 cycle, thanks to UCAS data released last year. This year’s figure of 63.3% of entrants being state educated is up four and a half percentage points from last year and is above the target range set by the University. If we compare it to last year’s data, provided by Student Beans, that’s a higher percentage of state-educated students than at Oxford, Durham, Bristol and Imperial College London.

But there’s a lot more information in the new data, and not all of it shows an ideally accessible Cambridge. Here are some points to consider:

The median student admitted to Cambridge last year had three or more A* grades.

Just under 60% of students accepted in the 2012 admissions cycle got at least A*A*A* at A Level. There were big gender differences in grades (68% of male entrants got three A* grades compared to 50% of female entrants), but this is likely to be due to gender variation in the subjects studied and applied for, as arts students tend to enter with lower grades than science students.

There were also big differences between the grade profile of the overall admissions pool and those who were accepted – just 30.5% of those who applied got A*A*A* or above, suggesting that the Cambridge admissions system does a good job of distinguishing between those who will achieve the highest grades at A Level and those who end up falling short. It remains to be seen whether it will be possible to do this if, as expected, the University can no longer base decisions on information from AS Levels.

There are persistent gender gaps in most subjects.

60% of arts students who were accepted this year were female while 54% of science students were male. Women were the majority in all arts subjects except Economics, History and Music, while the number of men and women admitted to Land Economy was equal. Meanwhile, male applicants were a majority in all science subjects except Veterinary Medicine.

The gender gap was particularly large in Maths, where only 25% of applicants were female, but they were also significantly less likely to take up a place - women only made up 15% of those accepted. (However, there are extra complications in the form of the STEP exam which might affect this gap. Without data on offers as well as acceptances, we can’t really try to explain this.)

While the number of state school students has gone up, this doesn’t mean that everyone has equal access.

Using POLAR2 data on university participation by postcode, we can see that 53% of this year’s accepted students came from areas in the top fifth of university participation, while only 3.6% came from the lowest fifth. On this metric, Cambridge affords less upward mobility than the average UK university, which admits 12% of its students from the bottom quintile and 28% from the top – a much less steep drop-off.

This isn’t merely a result of fewer students from lower quintiles applying to Cambridge – students in the top fifth were ten percentage points more likely to be accepted than those living in the bottom 40% of neighbourhoods for participation. That said, it’s worth noting that the grade profiles of individuals in different neighbourhoods are likely to differ, so we probably shouldn’t expect the same admissions profile as other universities nationally.

More to come

One caveat is that we don’t really know why the number of state school entrants increased so much this year - compared to last year, a greater percentage of applications were from state school students, but state educated students also had a higher success rate than the previous year. These factors could be down to the University’s access work attracting more qualified students, students applying to more competitive universities to justify paying the higher tuition fees, an increase in grade requirements at other universities, more generous assessment of students from state schools, or any combination of these and other factors. While it’s good news that the University is meeting its targets, be wary before attributing it to any specific cause or causes. (And similarly, the same for gender and participation data – it is likely that there are multiple factors involved, so it’s worth avoiding knee-jerk explanations).

There’s much more to look at in the data. Watch this space in a couple of weeks for more, as well as discussion of what we don’t know.