Friday, April 10, 2020

Why grades and not just a rank order?



As I understand it, the system for awarding grades for GCSE and A-level in summer 2020 will be broadly as predicted

1)      Teachers will be asked to provide predicted grades for each student in the subject (so if there are several classes/teachers these will need to be combined into a single list), and then to rank the students within each grade.
2)      The awarding bodies will look at the past few years of value-added data in a school and the baseline scores for the current students and will move students up or down accordingly, using the rank of the students (so if they decide there need to be 3 more students getting a grade 9 then the top three with a teacher grade of an 8 will be moved up).
3)      The students and teachers will find out the moderated grades.

This leads to the question of why it is necessary for us to put down grades.  If the data will specify a particular number of grade 9s, then a rank order of the students would have allowed the awarding bodies to put in the grades they have determined.

Here are my thoughts:
Teachers have an understanding of what each grade in their subject ‘looks like’.  We don’t have that sort of feel for where someone is in their year group.  This means it will be easier to compare across classes.  It will also be easier as a starting point for deliberations within school.

This leads to an obvious way of ranking students: start with the grade for each student, then decide whether they are a middling version of that grade, or just scraping into the grade, or nearly at the grade above.  Then rank the students within that sub-grade.  (To be clear: I don’t think this will be a quick job and I don’t think it will be easy to get the ‘right’ grade, let alone the right third of a grade, but this seems less bad than any other system.)

It fits with existing data.  We are likely to have some of the data we will use to help us to make and justify our decisions as grades already, such as mock exam results, predicted grades, etc.

It will help us as teachers to gain a better knowledge of grades for the future.  If all of our predicted grade 6s are moderated down to become grade 5s then we might need to redraw our mental grade boundaries in future (or the system used here might be flawed).

It will help us to explain things to pupils and their parents.  I don’t know what sort of comeback there might be after the grades are moderated and published.  For example, might we be called on to explain how we reached particular decisions about grading?  I will, clearly, have evidence I can use to justify my suggestion of a particular grade.  It would be possible for me to present that information to parents and pupils.  If they are then unhappy that their grade has been moderated downwards then that is a matter for Ofqual and the system they have put in place.  It would be much harder for me to justify why I have ranked a particular pupil as being number 94 in the year group if there were no grades involved.

It will be interesting to see how easy this will be to do and what proportion of grades are moderated (and in which direction).

No comments: