Read on Twitter
How to sort #sqaresults debacle, an exasperated thread from someone teaching stats for the past 20 years @MrMcEnaney @LucyHunterB @DJohnsonMSP
First principles:
1 no detriment
2 student performance not that of cohort or socio-ec circs should be relevant
1/n
First principles:
1 no detriment
2 student performance not that of cohort or socio-ec circs should be relevant
1/n
3 assume teachers are capable of professional judgement
4 if estimation errors, assume underestimation is more damaging to prospects
2/
4 if estimation errors, assume underestimation is more damaging to prospects
2/
Solutions:
1 whatever modelling produces, if lower than prelims, use prelim mark
2 model exam mark on basis of
a) student& #39;s perf across all subjects last 2 years
b) perf in same or similar subjects last 2 years
c) teacher estimated mark
d) available coursework marks
3/
1 whatever modelling produces, if lower than prelims, use prelim mark
2 model exam mark on basis of
a) student& #39;s perf across all subjects last 2 years
b) perf in same or similar subjects last 2 years
c) teacher estimated mark
d) available coursework marks
3/
3 one-off expansion of funded places to uni, college, training to correct for possible damage to life chances due to covid, exam cancellation
As an aside, I know PISA scores aren& #39;t perfect but it& #39;s clear pre-covid approach problematic and entrenching inequality
4/
As an aside, I know PISA scores aren& #39;t perfect but it& #39;s clear pre-covid approach problematic and entrenching inequality
4/
In addition, and I say this w some feeling, the UK (yes I know educ is devolved), at both secondary and university, suffers from an over bureaucratisation of teaching and assessment that devalues the professional judgement of teachers & lecturers, stifles creativity ...
5/
5/
... and is so time consuming it comes at the expense of breadth in the secondary and uni curricula.
This is a very (tho not uniquely) British problem but layering that onto modelling assumptions that entrench inequality across classes should make one weep
6/6
This is a very (tho not uniquely) British problem but layering that onto modelling assumptions that entrench inequality across classes should make one weep
6/6
