Heber Primary Ofsted

[MarkE]

I don't have a vested interest in Heber, but find the topic interesting. Here's a recent Guardian article noting a not dissimilar case, and also providing some data on the number of schools awarded the different ratings by OFSTED.

Strikes me from a quick peruse that the number of schools awarded 'Good' is actually increasing rather than decreasing. (from 44% three years ago to 56% now), whilst the proportion in the 'Requires improvement / Satisfactory' category dropped over the same period from 39% to 31%.

This would suggest that overall schools are either improving or that the assessments are in fact less stringent than they used to be. It remains plausible that Heber could have improved over that timeframe but have been out-improved by other schools that stayed in the 'Good' bracket whilst the criteria were toughened up.

Likewise, it's entirely possible that one set of assessors viewed the same school differently, or saw different perspectives of the same school. It happens, whether one likes it or not.

Alternatively, it could genuinely have worsened. It presumably doesn't take a huge change to drop from, say, the bottom end of the 'Good' bracket to the top end of the 'Satisfactory' bracket. That happens too (although it might be going some to convince most on this thread, it would seem).

[LondonMix]

Why? It's the National Literacy Trust-- is that acceptable?

I'm sure there are a lot of people very happy at Heber and we are lucky to have so many quality establishments in the local area. I really didn't start out for this to be me against Heber! I think Heber is fine with some issues that need to be addressed potentially - which is all I've ever said.

Whatever you think of the report, and even if Heber wins its appeal, at the very least it will lead to some positive changes and deeper reflection on certain practices.

Though people think schools are being downgraded willy-nilly the truth is that 78% of schools are judged as Good or Outstanding now which is 9 precentage points higher than last year. Ultimately, I think Heber will recover its rating and this is just a temporary set-back.

ETA-- x-posted! My point exactly regarding overall increase in higher ratings...

KestonKid Wrote:

-------------------------------------------------------

> May we know for which educational charity

> LondonMix volunteers?

[Magpie]

I am less concerned about relative measures of progression, than absolute levels of teaching and performance. It is easier to improve performance from a low base than a high base, and the percentaqe improvements are likely to be more significant, particularly when you get more cash due to higher levels of deprivation and more children who speak English as a second language. The stats can also be skewed by focusing effort on really low performers. The real question for me, as a parent whose child has just started at Heber, is whether Maths teaching is any good, not whether it is better or worse at improving relative levels of performance.

[DaveR]

For me the key point is a simple one - if your immediate reaction on hearing this news is 'it must be wrong' then basic psychology suggests you are then going to devote your efforts to establishing why it is wrong, rather than looking for evidence on both sides. It sounds like the school are taking a more measured approach, which is good news.

[KestonKid]

Thank you to LondonMix for the info about the National Literary Trust. No problems at all with what sounds a thoroughly good thing. Forgive me for being slightly suspicious. I am sure there is a hidden agenda circling Heber School - the situation sounds so similar to that at Downhills Primary School in North London. My interest is that of a Heber grandparent (Year 5 now). Zoe Williams in today's Guardian says it all far better than I could.

[Scribble]

"That's fair but its also worth noting that if the cohort is too small to analyse they don't. For instance the number of students at Bessemer who started out as high achievers was too small to do any statistical analysis on, so the DoE state this and as a consequence don't present any disaggregated stats on that cohort's progression."

This is inaccurate. Figures are suppressed when there are so few pupils in a given group that the anonymity of those pupils risks being compromised. This has nothing to do with a threshold value for valid statistical analysis, and implies nothing about the validity of figures for small groups who fall just above the cut-off point.

https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/190768/Confidentiality_Policy_v4.pdf

[bluesuperted]

Excuse the pedantry but I believe it's the National Literacy Trust that LondonMix means?

Interesting thread. Only slightly tangential but my comment is this: I hate Michael Gove.

[DaveR]

"I am sure there is a hidden agenda circling Heber School - the situation sounds so similar to that at Downhills Primary School in North London"

Perhaps you'd like to explain further, maybe with some evidence?

[LondonMix]

Yes, sorry that was a typo. Literacy.

Scribble you are right that privacy within the coherts is part of it. However, there are also guidelines regarding not doing analysis for numbers below certain thresholds for statistical reasons. For schools entire number of students in a school I think its 10 or 11 as the threshold. Privacy is a different consideration within cohorts.

[LondonMix]

Also to Kestonkid-- I think the government's Academy program at best could be described as overly zealous and ideologically driven (in a bad way). I don't have anything against free schools-- sometimes they are great but the devil is in the detail and execution is key. Some things that have happened are shocking. Still, I wouldn't go so far as thinking that there was a conspiracy against Heber.

[KestonKid]

If you google Downhills Primary School there are several items explaining what happened there. History now, as it was given by Mr Gove to Harris Federation to run. It all depends whether or not you think that state schools should be run by an elected local authority, or by a private organisation. The Guardian today has an article by Rajeev Syal about a free school and its troubles.

[Scribble]

LondonMix - do you have a link please? - because I suspect that would refer to inferential statistics, not descriptive statistics like these.

[civilservant]

Some confusion in LondonMix's advice on publication of small numbers in cohorts i.e. "Scribble you are right that privacy within the coherts is part of it. However, there are also guidelines regarding not doing analysis for numbers below certain thresholds for statistical reasons. For schools entire number of students in a school I think its 10 or 11 as the threshold. Privacy is a different consideration within cohorts."

The latest DoE guidance on publication of small number cohorts is here http://www.education.gov.uk/schools/performance/primary_12/Publication_of_small_numbers.pdf

This is a common standard for good practice across all statistical reporting, and applies to any data publisher, whether MORI or BMRB or Dept for Education

[DaveR]

"If you google Downhills Primary School there are several items explaining what happened there. History now, as it was given by Mr Gove to Harris Federation to run. It all depends whether or not you think that state schools should be run by an elected local authority, or by a private organisation."

How does this support your suggestion of a hidden agenda concerning Heber school?

[LondonMix]

Scribble you may be right but the guidance suggests to me that statistical significance is one of the considerations as I highlighted. As you will see below, the guidance discusses declining statistical significance as you approach 11 total pupils for value add analysis and then cuts off publication data at 10 pupils. However, the logic behind suppressing the information at 10 pupils may not actually be linked to a cut of in statistical reliability.

Please note that all of the below only concerns the value-add data. The progression data is suppressed for privacy reasons if there are less than 6 pupils in a cohort (which is a non-statistical consideration). I thought I said that in my last post but if it wasn't clear then I am happy to reiterate that.

Edited for typos---

Statistical Significance

The degree of significance that can be attached to any particular school's value added measure depends, among other factors, on the number of pupils included in the value added calculation. The smaller the number of pupils, the less confidence can be placed on the value added measure as an indicator of whether the effectiveness of a school is significantly above or below average.

Mainstream Schools

As a guide:

?at KS1 to KS2, for schools with 30 or more pupils in the value added measure, measures of 99.1 to 100.9 represent broadly average performance, while for schools with 50+ pupils, measures of 99.3 to 100.7 are broadly average.

When comparing the measures for two schools a similar effect holds:

?at KS1 to KS2 when comparing schools with cohorts of about 30 pupils, differences of up to 1.3 should not be regarded as significant, while for schools with about 50 pupils, differences up to 1.0 should not be regarded as significant.

Special schools

Particular care should be taken with the results for special schools and other schools where cohort sizes are small. For example, for special schools with 11 pupils included in the value added measure, only measures of below 98.5 or above 101.3 are likely to be significantly different from the average. When comparing special schools with cohorts of about 11 pupils, (remembering that the VA measure and KS2 results are not published for schools with 10 pupils or less) differences of up to 1.7 should not be regarded as statistically significant.

http://www.education.gov.uk/schools/performance/archive/ks3_04/k4.shtml

[Scribble]

Yes, so going back to Curmageon's original points:

"Please when considering the data consider the absolute numbers they are representing.

For example the low attainers = 4 or 5 students who could be low attainment for a number of reasons including learning disability which may make the targeted progress within the timescale all but impossible - or not (I have no knowledge of this cohort or data)

However there are so many issues with the league tables and analysing data which is standardised when we are actually talking about individuals and cohort to cohort can change dramatically"

- they remain valid.

[LondonMix]

Yes, a very small low-attaining cohort would be a mitigating factor. We know the cohort of low attainers at Heber is at least 6 (as the data isn't suppressed) but if it is 6 that would be small and it could be unfortunate circumstances that led to 4 out of 6 not progressing 2 levels. I think everyone has agreed with that from the outset.

Heber overall though is a large primary school and the size of its total pupils makes its total value add score reliable statistically. I am not saying that to be argumentative but just to be clear.

[????]

Still, half the teaching staff going on strike next week is bound to help........

[mrs.lotte]

Is this a general teacher's strike or specific schools?

nb re the Ofsted report - I read it and to my untrained eye the grade of 3 for management was largely due to insufficiently rigorous tracking, and (on the face of it) that seems borne out by the stats of progress between KS1 and KS2.

Tracking should be documented so if Ofsted have got it wrong doubtless that would be easy to establish.

It seems to be to be the sort of thing that is pretty invisible to parents (unlike, nice atmosphere, good behaviour etc which was praised in the report) but it is the bread and butter of Ofsted. Consequently if the findings had a legitimate basis then it must be better that it is identified and tackled, which it sounds like the school are doing.

[rachel83]

17th October is a strike day for some schools in Southwark but not all.

[Ann]

I think it's a general teachers's strike but on a region by region basis. The Charter have closed for years 8,9 and 10 but other years to attend as usual.

Tracking is important. A teacher may be hardworking, dedicated and inspirational; I've experienced some very good teaching with my own children and some not so good. Ignore OFSTED at your peril, would you really want a school with no objective assessment?

[canela]

It is a union-led strike, so the teachers belonging to certain unions are striking. that is why not all the classes at Heber are closed on 17th. It is wider than Southwark - not sure if just London or nationwide.