2 ms·

Well as a point of fact, they didnt. Their requested, "accurate" grades represent a 20% increase in general attainment. That has never happened in the history o

Well as a point of fact, they didnt. Their requested, "accurate" grades represent a 20% increase in general attainment. That has never happened in the history of modern education, and occurs only when, eg., civil wars end. As a comparative reference, we'd expect an at-most, 3% increase in IQ in the general population.

ppod 6y ago
I have no problem with the idea of down-weighting the entire national cohort to correct for this. The problem is that they actually weighted the results on a per-school basis.
- corin_ 6y ago
  Even down-weighting the entire national cohort... some kids have seen grades go from C (or B/C in one case) to U! A fail grade for an exam they weren't allowed to take. That would be an improvement if your luck of the draw was national rather than biased in favour of private schools, but it's still a pretty awful outcome. There may not have been any solution available that would be 100% fair to both this year and other year cohorts, but at the very least they should avoid making any student feel actively screwed over.
mantap 6y ago
This is desirable and it's easy to explain why. For any student there will be a range of grades they could expect to achieve. The fairest grade to assign is the best case grade in that range (not the midpoint), because there is no evidence that they would not have studied hard and achieved that grade in the exam. If every student is given their best case grade then you will indeed see an overall rise, this simply reflects a lack of knowledge of who would have studied hard and who would not have.
- sukilot 6y ago
  That's only fair in an environment where education is not a scarce good. But college education needs some way to artificially pick winners because the schools are underfunded and then economy is classist.

2 ms·

Well as a point of fact, they didnt. Their requested, "accurate" grades represent a 20% increase in general attainment. That has never happened in the history o

by mjburgess 6y ago

ppod 6y ago
I have no problem with the idea of down-weighting the entire national cohort to correct for this. The problem is that they actually weighted the results on a per-school basis.
- corin_ 6y ago
  Even down-weighting the entire national cohort... some kids have seen grades go from C (or B/C in one case) to U! A fail grade for an exam they weren't allowed to take. That would be an improvement if your luck of the draw was national rather than biased in favour of private schools, but it's still a pretty awful outcome. There may not have been any solution available that would be 100% fair to both this year and other year cohorts, but at the very least they should avoid making any student feel actively screwed over.
mantap 6y ago
This is desirable and it's easy to explain why. For any student there will be a range of grades they could expect to achieve. The fairest grade to assign is the best case grade in that range (not the midpoint), because there is no evidence that they would not have studied hard and achieved that grade in the exam. If every student is given their best case grade then you will indeed see an overall rise, this simply reflects a lack of knowledge of who would have studied hard and who would not have.
- sukilot 6y ago
  That's only fair in an environment where education is not a scarce good. But college education needs some way to artificially pick winners because the schools are underfunded and then economy is classist.