The league tables are out - let's talk Progress 8

[noblegiraffe]

The secondary league tables were published this week:

www.compare-school-performance.service.gov.uk/schools-by-type?step=default&table=schools&for=secondary

Some schools at the top have fab progress 8. Harris Battersea has done amazingly with 80% PP

I saw on Twitter this interesting breakdown of Progress 8 by type of student. Girls make more progress than boys, students with EAL make more progress than those who have English as a first language. Disadvantaged pupils make less progress, students with SEN even less.
When you look at the league tables, you then notice at the top several girls schools, and London schools with a high proportion of EAL.

Is progress 8 really a measure of progress, or a measure of school intake? Does a student being EAL outweigh them being disadvantaged?

And why are boys doing so much worse than girls?

And why are boys doing so much worse than girls?

They always have. And in IMHO while they can get in to college/uni and eventually jobs with poorer grades it won't change.

What is interesting is that schools with high %SEN pupils will continue to have poorer progress than those with very low % ... and those with very high proportions of disadvantaged pupils (particularly those in areas which are not highly ethnically diverse, e.g. coastal towns) will also have poorer progress.

Which is pretty much exactly the same pattern as for

• Raw results
• Ofsted grades

(as i have said many times before, sorting schools by %PP, particularly for schools outside London, is depressing, in that schools with the lowest %PP are almost universally outstanding, those with the highest %PP are most commonly RI or Inadequate)

So Progress 8, raw results and Ofsted grades - unsurprisingly, because they are strongly causally linked - are not independent of intake.

Noble, remembering your instructive post long ago on the impact of a small number of outlying results on Progress8, are there ranges given for each of the figures used to construct the graph?

There was a good piece in Schools Week on this yesterday - will try to link to it

schoolsweek.co.uk/progress-8-is-biased-towards-grammar-schools-heres-the-solution/

Needless to say my son at GS - great pg8 results. My son's comp was below average, more so for middle and high ability kids though.

The problem with Performance 8 tables is that this is only the second year this measure has been used and the qualifications with the greatest weighting i.e. English and Maths have completed changed specification since last year.
middle your sons schools may fit the pattern, but can you really draw a conclusion from just 2 schools.
Interestingly in my city LEA (outside London) there are a few surprises in the new Progress 8 tables with several schools with very high %PP performing better than some very sought after schools with more priveleged intakes.

Don't grammar schools actually often struggle to get good P8 scores, given the high starting points?

Certainly my school's P8 (leafy semi rural , fairly affluent) is 'below average'.

The other thing is that Performance 8 this year was calculated differently from last year. Last year, the 'points' given to each letter grade GCSE went up in uniform steps - 1 point per grade, IIRC. This year, there were more points to be gained by getting from an A to an A* (1.5) than from an F to an E (0.5 I think).

Again, this benefits schools with higher performing intakes, as the points available for squeezing A*s out of pupils 'meant' to get As are 3x greater than getting an E rather than an F.

www.gov.uk/government/uploads/system/uploads/attachment_data/file/561021/Progress_8_and_Attainment_8_how_measures_are_calculated.pdf

I just read the link middle but I didn't really understand it...

Piggy, that was the case with last year's calculation - and with my cynical hat on, that's why it was changed, because grammars didn't look 'good enough' using it.

'Don't grammar schools actually often struggle to get good P8 scores, given the high starting points?'

They shouldn't struggle because children of above average ability should have greater potential for learning.

'This year, there were more points to be gained by getting from an A to an A* (1.5) than from an F to an E (0.5 I think)'

How is that not biased? Why was it changed?

Yes, I imagine so cant : might make sense.

They shouldn't, no jammy but at my school we have always struggled against the very high targets set for academically good, but not brilliant, students - hence our weak P8. Our attainment 8 score is much much better.

jammy, at least the problem will go away once all subjects are on numerical grades (so from summer 2019 results, which is when e.g. the more obscure languages are first examined).

Piggy, my DC's comp is as leafy as they come, has a very high attaining cohort, and loses a % of its highest achievers to superselective grammars. It has a very low %PP and %SEN for a comp, by virtue of its catchment. It also has a well above average Progress8, with the highest Progress8 being for the middle attainers.

I think the key is that it loses its highest attainers, though? They're the ones that can bugger P8. As can the PPs so if you don't have many of them, and the ones you do have fare well, you'll be fine!

It depends what you mean by highest attainers. ]

Yes, they possibly lost the top 0.5 - 1% - but there was a full maths set who pretty much all got L6s in year 6 and Level 8 or 9 in their GCSEs, so it isn't a large number IYSWIM?

Yes, I do. I think ours did, too. But it wasn't good enough it seems!!

Interesting - just been digging a little further for DC's school.

High prior attainers was the biggest group numerically, and got above average P8. Middle attainers next largest group, really exceptionally high P8. Low prior attainers, counted as average P8 but numerically higher P8 than the higher attainers.

What was interesting was that the definition of 'average' moved, in a way I wasn't expecting. The reporting at prior attainment level is subtle enough to compensate for the fact that lower prior attainers have higher P8 in general, so the 'numerical P8' which got the school 'average' for lower attainers was significantly higher than the P8 that got them 'above average' for the higher attainers.

Blimey : that is detailed digging! I didn't realise these things and I thought I had a good handle on P8!

I just clicked options just above the table (having selected 4 local schools to compare) to see P8 by prior attainment.

I would have expected (on no evidence at all) that these 'sub P8s' would have used the same definition of 'average' as the P8 as a whole. However, instead it seems to use the 'average for similar students nationwide', which is probably sensible but means that a school with very high numbers of LA students could be in the odd position of being 'average' in all 'sub P8s' but 'above average' overall.

Hmmmm. that does explain rather a lot...

Please could someone explain in layman jargon. In my area the schools with most disadvantaged pupils did really badly, way below average on p8. What does it mean? The school with tiny and richer catchment came out well above average.

Our local, superselective boys' grammar has a 'below average' P8 listing for this year.

Which makes sense, of course, if you know that IGCSEs aren't counted when they tot up the scores.

How many parents do know that, I wonder?

And how many parents know that it's a child's first attempt at an exam in a particular subject that is counted, not the grade they actually leave school with?

And how many parents know that, up until this year, a qualification that can be taught in as little as three days (the ECDL) counted the same as a GCSE in, say, history, for accountability purposes?

I cannot tell you how much I loathe the league tables, and the pernicious impact they've had on our education system since 1992...

Oh, and noble - it's quite possible that at least one reason boys do so much worse than girls using these measures has to do with handwriting bias....