Grade inflation
Widmann writes about grade inflation in the English GCSE. The raw numbers are actually pretty scary reading. In pretty much all subjects, the grades have steadily gotten better over the years. In many cases more than doubled.
Unless students are getting smarter or the teaching much better, this should be a cause for concern…
–
239-244=-5
August 27th, 2009 at 8:34 pm
Even if students are getting smarter or the teaching is getting better, grade inflation is still a problem. Users of grades, e.g., universities selecting students to accept, or employers recruiting recent graduates, need to be able to tell which of two students to prefer. If, say, 20% of students get the best mark in all subjects, there is no way to distinguish between them, and universities and employers have to use other means instead.
August 27th, 2009 at 9:10 pm
I disagree that it is a problem if, indeed, students are getting smarter or teachers getting better!
In that case, you do have grades reflecting the student’s capabilities, so that is fine with me. It is only a problem if you have to take into consideration the year of the exam to properly judge the grade…
August 27th, 2009 at 9:30 pm
But what if all the students eventually ended up getting the same grade?
August 28th, 2009 at 7:07 am
If that reflects that they all know the same thing, then what is the problem? ;)
August 28th, 2009 at 4:19 pm
Imagine all the students got between 90% and 100% in a test. They’re clearly not equally good, because there is a difference between getting 90% and getting 100%. However, if the grading rules says 90% and above gives you an A, all of these students will get the same mark, and you won’t be able to tell them apart. That’s a problem if you want to find the very best student in the group.
September 13th, 2009 at 4:12 am
There’s bigger problems with grade inflation than just the fact you can’t differentiate between a 98% and a 95%. Firstly, there’s random error, so when the range of possible scores becomes quite small (i.e. 90-100%), most of the difference between any two scores can be totally random. So that smart kids may actually get worse than dumb kids half the time.
There’s some other problems associated with educational pedigrees (i.e. going to Oxbridge), that this guy from the US discusses: http://higher-ed-reform.blogspot.com/2009/09/inflation.html
September 13th, 2009 at 6:56 am
Thanks for the link, Sam. A nice read, and yes, I think educational pedigree is working as a substitute for grades since the latter are not easily compared over time or space…
As for differentiating grades, yes that is a problem if there are too many and they are trying to capture too fine nuances.
With our grades here, that have just been reformed, we used to have three grades in the middle of the spectrum that would let us capture “just below” and “just above” average (and average as well, of course). Now we just have a single grade there.
That, I think, is too coarse while the previous was about right.
Any finer, though, and the grade would indeed end up a bit arbitrary…