Monday, July 30, 2007

That Drop-Out Report In "Brief"

Breathless stories about this report from the National Audit Office assessing dropout rates from UK universities, have led people to believe, yet again, that there's something badly wrong with our HE system in the UK.

Yet again, this view is largely the result of people not actually reading what they see fit to comment on.

Briefly, the report looks at the number of non-completers from university between 1999 and 2005, and deduces a probable non-continuation rate of something around 20 per cent for students, with certain factors, such as background, being indicators of likelihood of leaving university.

Cue predictable laments about the decline of the HE system.


The dropout rate is not really a dropout rate - as the authors themselves say:

There are particular difficulties with data about part-time students due to the inherent flexibilites in patterns of study
Or, in other words it's an overestimate because some part-timers take longer to complete than expected.

The next problem for the doom-mongers is on page 17 - a graph taken from the OECD's Education At A Glance, showing, er, that the UK's dropout rate is actually rather low - worse only than Korea, Greece, Ireland and Japan. Ireland's case is especially interesting - as a country whose HE system has grown very rapidly, and now sees greater participation amongst young people than the UK, the lower dropout rate deals a real blow to the popular theory that trying to send 50 per cent of 18-30 years olds into HE is the cause of this 'high' dropout rate.

The most interesting part of this is the historical context, though. Some reports have made a couple of interesting assumptions - firstly that this dropout rate is high - which it may be, but it's lower than almost everyone else.
The second assumption is that it has gone up. This is actually not supportable. The report starts with 1999 because that's when this data first started being collected in a systematic way. And the dropout rate has actually fallen (albeit in a marginal and probably insignificant way) since then. Before that, well, I am not at all sure that systematic, country-wide university non-completion data is really available. And I have looked.

This begs a number of questions. Firstly, and most importantly, what drop-out rate is acceptable? The report itself notes that you can't get, and shouldn't want, a 100 per cent completion rate because many dropouts are actually quite rational. A student learning a series of particular modules that they needed for a career plan. Someone getting a job that they wanted. Children. Someone just deciding that now is not the best time to go to university after all. Japan has the lowest dropout rate at just under 10 per cent, whilst the US runs at nearly 50 per cent. Perhaps we're actually doing rather well. Perhaps our dropout rate is too low? (I don't personally believe that, by the way, but it's worth considering).

And the second is - were things really better back in the day? I can't find any evidence that they were, and I reserve the right to be very sceptical about anyone who says otherwise without any evidence. And that's something that's happening a great deal at the moment.

Technorati tags: ,,,,,

Saturday, July 14, 2007

How To Get Your Graduate Survey Covered In The Press

Blimey, has it really been that long since I last posted? I am a very bad and lazy blogger (who has been flat out on assessing the postgraduate labour market and having comedy run-ins with a national newspaper, thanks for asking.)

This week has seen a kind of distillation of how the media reports stories on graduate employment, graduate salaries and graduates in general.

Two major surveys came out - the Association of Graduate Recruiters Survey, and the Higher Education Statistics Agency Destinations Of Leavers of Higher Education Survey.

These are all hideously long names, so like everyone else in the field, I'm going to refer to them as the AGR survey and the DHLE survey.

I've covered the AGR survey at length before, so suffice to say, it's not a bad survey as such, it's just not representative of the graduate labour market - just that bit of it that pertains to big, private sector employers, mainly in London and largely associated with the finance industry. It mainly asks what employers aim to pay new graduate recruits this year. It found that the number of vacancies were up, but that the average starting salaries that their employers expected to pay in London might be down this year. It usually covers in the region of 20,000 graduates.

The DHLE meanwhile, is a census survey of everyone who graduated from university in 2006. It asks graduates what they were doing six months after they get their degree. It's a hugely important survey which is the basis of a lot of performance indicators for universities, and a great deal of careers information and so we get a response rate of around 80%. More than 200,000 first degree graduates (it also covers graduates at other levels, but let's set those aside for a moment) replied to the survey.
It found that everything was pretty much as normal. Graduate unemployment levels remain around 6% (it was 6.2% last year - it may have budged by 0.1%, but I don't have the detailed figures yet). Graduate salaries haven't fallen. Graduate underemployment seems, if anything, to have gone down - although, again, I'd prefer to do the calculations myself before I'm sure. As an aside - it's not collected by Government - HESA are an independent charity - and gets played with by all kinds of awkward independent researchers keen to find where things are going wrong. So it's not a cover-up exercise.

So, two surveys, one non-representative, but with a slightly alarmist (although not terribly significant) message, and one which is representative and tells students and graduates that actually, graduate jobs are in reasonable supply and graduate salaries aren't actually falling really.

Guess which one gets all the press coverage, and which one gets largely ignored?

Technorati tags: ,,,,,