With another grading season peaking in the UK and elsewhere, academics across the country are busy adding up their students’ scores and converting them into degrees. But with so much traditional practice amid the disruptive effects of the pandemic, it is worth pondering whether the system is really fit for purpose.
A percentage scale is routinely used in UK higher education. Values between 70 and 100 percent conventionally mean first-class performance, while values between 0 and 39 are classified as “failed”. Taken together, the first-class and fail areas make up 70 percent of the scale.
Most students, however, get grades crammed into the remaining 30 percent. That in itself makes little sense. But the problem goes deeper.
In the past, some markers tended to set implicit, subjective ceilings for the grades they were willing to give. This meant that some students failed because their work was rated with a “hard marker”. To counter this, universities (rightly called for by generations of external auditors) have put in place structures to encourage the use of the full range of available results.
This often includes the indication of the grades that can be given for different levels of achievement, so that if two different markers make the same criteria-oriented judgments about work of equal value, they award the same number of points. The “2-5-8 system” is a common example. In the upper second class this provides, for example, that a 65 denotes a “medium” 2 (i) quality, with 62 and 68 being available for smaller adjustments upwards or downwards. If this is translated into the first-class range, the top grade of 78 would stand for a “high” premiere.
But why have a 100 point scale that ends at 78 (or that starts at 30?) That would be fair but stupid. Instead, things get stretched on the high end. Typically, 72 is awarded for a “low” first, 80 for a “medium” first, 90 for a “high” first and 100 for one or the other exceptional work.
Leaving aside the fact that every university does this differently, so students who do at the same level in different institutions will get different grades. The bigger problem is what happens within individual universities.
Take Student A, who averages a high top second of 68 so far. For their next job, they improve and get a high first for which they get a 90. All well and good. We need to appreciate the improvements students make in their work.
Meanwhile, Student B averages a high lower second on the same course, which equals 58. For their next work they make a leap in improvement of the same order of magnitude to a high upper second. You will be awarded 68. Well done.
These two students have improved their work in a way that an evaluation system should evaluate equally. But student A gets 22 extra points that go towards his final result, while student B only gets 10. The opposite effect occurs at the lower end, with students being overpunished in ways that don’t happen further up.
That can’t be fair. All students entering a university, regardless of their course of study, should have equal opportunities to learn and improve. For example, a student who does their best in the middle end of the scale should have access to the same rewards for improved performance as those who do at the higher end of the scale.
The 0-100 scale is so familiar that we tend to look right through it as its structural problems are obviously hidden. But, in our opinion, it would not survive a proper equality test.
One could argue that improvements at the higher end of the scale are more difficult and therefore deserve a greater reward. But this argument is usually not made explicit in rating systems – and for good reason. It is a rather spurious, ex-post justification of an essentially arbitrary decision made some time ago in our distant past to step on the threshold of honor first class.
From our point of view, the large numerical scope for notes above this threshold is also one of the drivers of note inflation. The answer to both problems is to change the scale to a linear one. For example, in a 0-16 grade point structure, all levels up and down have the same reward or cost for improvement or deterioration in performance.
In the examples above, Students A and B would both receive three additional points. This is fair. It’s not inflationary either. If we are working on improving the assessment to make it more equitable for all students, this would be a good place to start.
Andy Grayson is Associate Professor of Psychology, Susannah Lamb is Director of Academic Quality, and Chris Royle is Academic Standards and Quality Manager at the School of Animal, Rural and Environmental Sciences at Nottingham Trent University.
source https://collegeeducationnewsllc.com/the-uks-percentage-scale-is-unfair-and-fuels-grade-inflation/
No comments:
Post a Comment