Time goes by fast, and there is a lot to write. But I do not have much time anymore for writing things down here: lots of work at the Uni, and at home the usual routine with shopping and house-related work on weekends instead of travelling.
At the Uni, the fall semester has ended. Student assessment is not quite easy: over time the assessor has to apply a common standard and cannot get influenced - the same standards need to be applied to all students. When I assess written homework, I go through the whole pile of submissions twice: the 2nd time is to ensure that I have applied consistent standards to everyone. But it is more difficult to keep this consistency in the individual examination. Last week, when the students of one course were assessed by all of us tutors, a few students had submitted some not so great works - so when I saw something decent, then I was quite enthusiastic and gave high marks. However, as the moderation process started (comparison of all student's grades, of marks given by all tutors, to ensure consistency), it turned out that I had given a few marks too high. I am not really sorry about this - after all, I think that marks should be encouraging to the student, and I do not like this marking scheme that we have to use here, in which no student ever gets a 100%, because no work is ever perfect. But I feel sorry for the few students whose work initially had a higher mark, and now gets a lower one.
Just for information to those who are not familiar with the marking scheme employed here: students' work is graded between 0% and 100%. In order to get a "pass", the mark must be at least 40%. 40%-50% is a "3rd grade" mark, 50%-60% a "2.2", 60%-70% a "2.1", and above 70% there is a "1st". This is a highly non-linear scale, and I find it difficult to determine an exact translation of what the student did into these percentages. I can quite easily determine "by feeling", in which category a student work fits. And so does every other tutor: the whole grading is done "by feeling". If I try to add up points and translate the students' work directly into precentages, then there are some big discrepancies:
- from a bare "pass" to a "1st", the step is just 75%. What I mean: the student has 40% for a pass, and for a "1st", (s)he just needs to add 3/4 of that work to get the highest possible grade category. This is somewhat not intuitive. And if the students puts much more effort in, the grade remains the same: a "1st". In order to remove this distinction, the marking schemes have often "stretched" this regime between 40% and 70%: suddenly there is not a linear correlation between the students work and the mark, but the marking category is given in weird worldy descriptions. And in the "1st" caegory, 75% is the most given mark. To me this means that the student just fulfilled 3/4 of what was asked, but got a 1st! This is not logical. And therefore, the tutors never give high 1st marks: even if the student fulfilled everything, (s)he just gets what the tutor feels would be appropriate. 75% for everything is fine, 80% for excellent ingeniuity, 90% for outstanding - can be used in real world production, and 100% never. This sounds a bit unfair to students...
I prefer a directly quantifyable marking scheme: the student would collect points for each question / criterion, and the sum of all the points would translate into a percentage. 100% would then be the mark which would given to the students when everything was correct and was there in the assignment. Of course, this would mean that the "1st" would have to be redifined: 100%-90% would be a "1st", as is in the US an "A".
I wonder if this marking scheme is customary at all UK universities. If so, then I think something has to change drastically, as this current scheme is unfair to the student and introduces a lot of "variability" by the tutor marking, since the non-linearity between student points and marking percentage grade prevents from establishing a reproducable marking scheme.
Something for those responsible for student education to think about.