The everyday blog of Richard Bartle.
RSS feeds: v0.91; v1.0 (RDF); v2.0; Atom.
Previous entry. Next entry.
11:11am on Friday, 18th August, 2006:
The A-Level results are out, and I was interested to learn that girls outperformed boys in every subject except Modern Languages.
Isn't language supposed to be something at which girls excel? Clearly, there is something wrong with the Modern Languages course if boys are actually better at it.
Also, sixth form students are flocking to take Mathematics. This is a result of a curriculum change, which while not entirely a "mathematics for the non-mathematician" approach, nevertheless softens some subject matter that was added because too many university students were lacking it. Students don't want to take an exam unless they'll do well in it.
In my day, A-Level results were normalised. The top 10% got an A, then next 15% got a B, the next 10% got a C, the next 15% got a D, the next 20% got an E. 20% were given an O-level instead, and 10% failed outright. It didn't matter if the exam was easier or harder from one year to the next, because the (altogether reasonable) assumption was that across a cohort, the distribution of ability would be pretty much the same from year to year. Is it really likely that one year group of 250,000 students would work harder or have a significantly greater number of geniuses than the group that came before or went after?
Well, yes. This was a good scheme in that it identified the best students, but a bad scheme in that 30% of students were guaranteed to fail. This meant that the no-hopers never took the subject, so that over time the only people who did any particular A-level were the ones who were good at it — and 30% of those would still get a fail. Thus, in 1988, the UK switched from a normative model to an absolute model. Rather than saying the top 10% of students got an A, we now say that you get an A if you score more than 80% on the uniform mark scale (which is generated from the raw marks, although I'm not entirely sure how). Whatever, the result is that someone who got an E grade in Mathematics in 1988 would have got a B for the same mark in 2004.
The problem with giving nearly a quarter of all A-Level students an A grade is that you can't tell which of these are the better ones. This makes it harder for the top universities to pick the students they want. In the olde days, if you had an over-subscribed course such as medicine, you asked for 3As and this sorted the wheat from the chaff (or at least the public school entrants from the state school entrants). Nowadays, if you ask for 3As you're still over-subscribed. Thus, there's talk of adding an A* grade, which is what they have in GCSEs. This, of course, would lead to the same kind of grade inflation we have at the moment, as successive governments put pressure on schools and examiners to show how their successful education policies translate into an improvement in examination results.
Except, I heard yesterday a solution that wouldn't involve some 90%=A* tinkering. The idea, from the Conservatives of all people, was to give the * grade to the top 10% of candidates. Thus, if you get an A it means you have an excellent grasp of the subject in all areas, and potential employers know you've reached a certain high standard. If you get an A* (or, if academic achievement were suddenly to decline, I guess conceivably a B*), it means you were in the top 10%. This means universities can get along with using the normative system they want, and the government can continue to boast about its achievements in its drive to everyone-gets-an-A perfection (which was the reason given by my Physics teacher back in 1977 as to "why you should never put a socialist in charge of education").
I'd tell you what grades you need to get onto my course at Essex University, but the entire domain name essex.ac.uk is currently unreachable so I can't check. Just what you want when students whose grades weren't as expected are rushing to the clearing system to try find a place...
About this blog.
Copyright © 2006 Richard Bartle (firstname.lastname@example.org).