Academies- anyone know much about them?

[EnglishRose1320]

Just have a load of questions about academies basically, how much do they change schools? I know they vary a fair amount but feel a bit in the dark about them and seeing as by 2020 in at least the county I am in we will no longer have an LEA and only have academies I feel I ought to wise up on them. What experiences have people had of them so far both as staff and as parents, I'm looking at it from both view points. Do people think they are a good idea or Not? Sorry bit rambly but basically any info and thoughts appreciated.

You don't appear to be treating all research with scepticism at all though, as you refer to the select committee stuff to support one of your points

It isn't the select committee's own research (and the select committee contains both supporters and opponents of academies). But I was responding to someone who was asking for UK research that showed improvements linked to academy conversion. I was not saying that the research was conclusive, just that it exists and some of it is mentioned in the select committee report. I would regard all of it critically.

And you seem to have changed your explanation of HACP's admission anomalies to the fact that they don't adjust the bands to reflect the local area, rather than families with lower and middle ability children not applying

It is actually the same possible explanation phrased differently.

HACP don't normalise their tests for the local population. They simply divide the applicants into 9 groups based on their test score. If 890 high achievers apply and only 10 low achievers (a deliberately unrealistic scenario) the bottom band will contain 90 high achievers (the 90 getting the lowest scores) and 10 low achievers. Depending on how the low achievers rank against the admission criteria they may or may not get places. Since place allocation involves a random draw it is possible that none of the low achievers get places.

However, if HACP normalised their tests the bottom band would only contain low achievers regardless of the profile of applicants. So their admissions would reflect the spread of achievement in the area rather than the spread of achievement in those applying. And in the scenario I outlined above all 10 low achievers would be admitted.

If all that is a bit opaque I put the blame on fair banding! While the system itself is straightforward it can have unexpected side effects which are difficult to explain. I can see why some schools want to use fair banding but I'm not convinced it is the best approach.

My point about LAs was that some departments now don't exist. In the process of investigating maladministration or whatever, how on earth can a sensible investigation be held when there is no-one to respond to e-mails etc.

Sticking with the example of admissions, someone in the LA must be responsible for it even if they outsource the whole thing. It may be someone fairly high up - Head of Children's Services, for example. The LGO has investigated cases relating to services that LAs have outsourced and has not, as far as I am aware, found investigation any more difficult than it is for non-outsourced services.

Then when a frock of calico is produced, they're held responsible for why it's not a ball gown

That's not quite how I would put it but I don't disagree with the point you are making.

Many LAs are still very inefficient and could do the same job for less. But clearly there is a minimum level of funding needed. To go to extremes, if you give the LA 1p per year to provide rubbish collections there is no way they can meet their legal responsibilities.

"While the system itself is straightforward it can have unexpected side effects which are difficult to explain"

Exactly. This is the 'tinkering' that minifingerz describes.

If a person with expertise and experience in the field like yourself can't explain the 'unexpected side effects' then it's very difficult to see how these systems can be independently investigated.

Why on earth is it called fair banding? How incredibly disingenuous. I bet it would be challenged by more people if it didn't have such a misleading name!!

"Many LAs are still very inefficient and could do the same job for less"

they could always let go of loads of staff and rehire them on really shit terms and conditions. Especially those people at the bottom of the pay scale. Parcel out services to voluntary bodies who have many staff working on short term contracts, or perhaps even classed as 'self employed' so not eligible for pensions or sick pay. This is a great way to save money and will be tolerated because council staff and people working for voluntary organisations are more likely to be women, and we know they all have husbands to pay the mortgage.

LAS often are very inefficient, I agree, but in their favour they have a degree of accountability and responsibility that other bodies don't have.

Exactly. This is the 'tinkering' that minifingerz describes.

No, the unexpected side effects are not "tinkering". They are just results of the way fair banding works. Manipulating test results so by underscoring high ability pupils to put them into the bottom band is tinkering. Drawing more pupils from high ability bands is tinkering. Rigging the random draw so that the highest scoring pupils in each band get places is tinkering.

Why on earth is it called fair banding?

It is called fair banding because it is designed to give pupils of all abilities an equal chance of admission. In essence you test the applicants, split them into groups based on their test results and admit an equal number from each group. Depending on the approach taken the profile of abilities admitted will reflect the profile of the applicants, or the profile of all children in the area, or the profile of all children in the country. Unlike other forms of selection it does not exclude pupils of low ability.

they could always let go of loads of staff and rehire them on really shit terms and conditions

Alternatively they could buy equipment of the same quality from cheaper suppliers, figure out more efficient ways of doing things, cut the salaries of senior staff (the pay of some Council CEOs really cannot be justified) and so on. There is no need to mistreat staff in the way you describe in order to save money.

I would say that having an obtuse admissions system which no-one can properly explain the 'unexpected affects' of is tinkering tbh.

Ditto, using measures like those which don't reflect the local population is deliberately attempting to increase the numbers of 'high ability' children aka tinkering.

A final (I hope) thought on fair banding...

Fair banding works best when all secondary schools in the area are using it and all use the same test, so that children take the test in their primary schools. If the children have to go to the secondary school to take the test some children will miss out because their parents don't take them to sit the test. Unfortunately the evidence suggests that those who miss out will be disproportionately from the most deprived and disadvantaged groups. This is something that was not foreseen by those who came up with the fair banding system.

no-one can properly explain

Plenty of people can properly explain the unexpected effects. They may be difficult to explain in a couple of sentences but they can be explained.

using measures like those which don't reflect the local population is deliberately attempting to increase the numbers of 'high ability' children

Rubbish. It could equally work the other way. The way HACP's system works means that the profile of those admitted will reflect the profile of those who take the test. If that is skewed towards low ability children admissions will be too. Indeed, there are some schools using the same approach to fair banding as HACP with exactly that outcome - a higher proportion of low ability children than the local population.

It's not rubbish at all. You've said in the post immediately before that in areas where not all schools use 'fair banding' its use effectively discriminates against children from the most deprived and disadvantaged groups. This may not have been 'foreseen' by those who came up with the fair banding system, but it is most certainly known about by schools that continue to use it when the schools around them don't.

As the profile of the the children admitted reflects the profile of those who take the test and,as you've said, evidence indicates that children from deprived and disadvantaged families are less like to take the test, using 'fair banding' is deliberately 'tinkering' with admission. Perfectly legal, yet most definitely tinkering.

Which are these schools using fair banding which admit a higher proportion of lower ability children than the local population, by the way?

Coopers School in Chislehurst has fair banding but because there are other more popular schools nearby its most able top two bands usually go out to 'all offered' whereas the other bands have catchments of a couple of miles. As school places are tight the spaces in the more able bands end up going to lower banded pupils. This doesn't really reflect the geographic of Chislehurst where the most able children will travel to nearby grammars in neighbouring Bexley or choose other schools that are more popular like Bullers which doesn't have fair banding. That is an example of a fair banding school which admits a higher proportion of less able children compared to the local demographic.

It seems to me it is the antithesis of fair to set up a system where you dangle a school that gets good results under the noses of a poor local population and then set up an admissions system skewed against local people getting many of the places. And that result was surely predictable when the school chose "fair banding" in an area where children in the local population fall predominantly into the lower achieving bands? ie surely it was known not to be "fair" from the very beginning? Unless, of course, you take the view the school couldn't possibly be any good in a poor area if it could only cater for locals... Which begs the question what makes a good school, the teachers or the students they are given?

In any event, what is fair about polluting and clogging up local roads with school traffic coming from miles away, while locals bus in the opposite direction, consequently also wasting lots of resources? Surely that is only vaguely "fair" if it is impossible for a school to be any good unless it has the whole spread of intellectual ability to cater for, all in reasonable numbers? In which case, the "fair banding" system still isn't fair if actually it can end up with a huge proportion of high ability children... It all seems like a silly fiddle that works for nobody and serves to confuse people not inclined to sit up reading school admissions criteria at night.

Thanks PettsWood, that's interesting.

To my knowledge, 'fair banding' was established to distribute a mix of abilities across a range of schools. Unless all schools in an area use it, it seems to contribute to a schewed intake of either higher ability or lower ability children.

The dominant theme from research eg The Sutton Trust is that is harms poorer pupils.

using 'fair banding' is deliberately 'tinkering' with admission

What I was calling rubbish was your comment that fair banding was a way of deliberately increasing the numbers of high ability children. It isn't. It can reduce the numbers of children from deprived backgrounds but that is not the same thing. This touches a nerve with me because far too many people equate being from a deprived background with being low ability. They are not the same.

The dominant theme from research eg The Sutton Trust is that is harms poorer pupils

I'm not sure I would go so far as to say it harms poorer pupils. The alternative used by most schools is to award places based on distance. The result is that pupils from disadvantaged backgrounds can't get places because they can't afford to move close to the school. I would say that, unless all schools in the area use fair banding and agree a common test which is then administered by primary schools (thereby ensuring that all pupils take it regardless of background), it is no better and no worse than the common alternatives for pupils from disadvantaged backgrounds. This is one of the reasons I would like to see more schools take advantage of the ability to give priority for admissions to pupils qualifying for pupil premium.

set up an admissions system skewed against local people getting many of the places. And that result was surely predictable when the school chose "fair banding" in an area where children in the local population fall predominantly into the lower achieving bands

But the admissions system at HACP was not and is not skewed against local people getting many of the places. Indeed, its admission criteria are designed to ensure that 90% of places go to pupils living within 2 miles of the school. Choosing fair banding will only skew admissions towards the higher achieving pupils if parents of lower achieving pupils either fail to apply to the school or fail to get their child to the test. If they apply and ensure their child attends the test there is no disadvantage.

www.bbc.co.uk/news/education-35261625

This article from a couple of days ago is timely and seems to suggest that it is not only 'fair' banding that is more likely to be problematic in the hands of academies.

On my phone so hope link works.

Being from a deprived background equates with reduces opportunities, not reduced ability. Like the opportunity to sit a fair banding test to increase the number/quality of schools that you can apply to.

You outlined this disadvantage a few posts ago.

Agree that most admission systems disadvantage already disadvantaged families. Maybe Brighton's random allocation is the way to go.

Are 90% of the children at the academy in question local children or aren't they?? What % of local children who apply to the school are getting a place there??

Are 90% of the children at the academy in question local children or aren't they

They are. Their admissions criteria ensure that, once LAC and siblings have been allocated places, 90% of the remaining places go to children living within 2 miles of the school.

Croydon is a large borough. HACP is located in the north-east of the borough and primarily services that area. This area is not as affluent as the south of the borough but nor is it particularly deprived. Some other Harris academies in that part of South London are situated in more deprived areas. The local primary schools were, at the time in question, outperforming both national and local averages.

What % of local children who apply to the school are getting a place there

I'm afraid that information does not appear to be available. I know that last year there were 1986 applicants for 180 places but Croydon don't give any more details. An FoI request to either Croydon or HACP would be needed to find out.

So, how many eligible children are living within a 2 mile radius of the school? And can 10% of the children come from as far away as they like? What sort of application to available place ratios are there in other nearby schools? Why allow 10% of children to come in from elsewhere in the first place?

And is it correct that fair banding only applies to 10% of the places? Or do all children have to take a test to get in? Who sets the test? And what does the test involve? Presumably it isn't actually a full-blown IQ test?

Oh, and who marks the tests?

To guard against any foul play, you would surely need one hell of a lot more checks on schools than is actually affordable, particularly if every school can set a different test. And why are SATS test results used to predict minimum GCSE results if they are not a test of ability or potential? And if they are a good predictor, then why are further tests required? And why are children required to sit so many tests when nobody entirely trusts the results? And what is done about children who under-perform in tests?

We keep being told that improving exam results year on year have not resulted in better educated young people. Are we sure that more tests and exams are going to fix that? Doesn't it all just bring the whole concept of exams and league tables into disrepute? Why should we believe that schools genuinely are doing better now, but that they were not genuinely doing better, before?

So, how many eligible children are living within a 2 mile radius of the school

I don't have that information. However, the primary schools within 1 mile of HACP (some of which are in Bromley) between them have around 540 pupils per year. Having looked at numbers of pupils in schools further away my guess would be 1,000-2,000.

And can 10% of the children come from as far away as they like

Yes. HACP cannot legally put any limit on that.

What sort of application to available place ratios are there in other nearby schools

HACP is the most oversubscribed school in Croydon. Looking at all Croydon secondary schools, in 2015 13592 applications were received for 4226 places so the average secondary school in Croydon has 3.2 applications per place.

Why allow 10% of children to come in from elsewhere in the first place

They cannot legally allocate all places to pupils living within 2 miles - that was made clear by the Greenwich judgement. Given their admission number, allocating 10% of places to pupils living over 2 miles from the school means that, after LACs and siblings have been admitted, they will generally only have 1 place per band for these pupils. They therefore cannot legally go below 10%.

If they wanted to ensure that all pupils were local they would have to abandon the random draw they currently use and instead use distance as a tie breaker. That undermines the purpose of fair banding in that it allows more affluent families to get places by buying up properties close to the school.

So, do all children have to take the test to get in and is this done in the school itself, not their primary school? Do other secondaries in the area also have an entrance test? Is it really possible to establish fair entry criteria if different schools have different methods for deciding what they think is fair? And who is the final arbiter of whether an admissions process is fair or not? And who checks up on it and how often? And if house price premiums are the reason for the 10%, then why only 10%? And is it a mixed demographic background of children who are applying from amongst the 10%? Is the system chosen not a little bit confused as to what it is trying to achieve?

"Croydon is a large borough. HACP is located in the north-east of the borough and primarily services that area. This area is not as affluent as the south of the borough but nor is it particularly deprived. Some other Harris academies in that part of South London are situated in more deprived areas. The local primary schools were, at the time in question, outperforming both national and local averages."

The immediate area that HACP is in is at the bottom of the top half of the 2015 deprivation index (ie, it's not particularly deprived, but it's also not massively affluent), but it's not densely populated (large Victorian houses, many a good few detached with large gardens). 80% of the other wards within 2 miles of HACP fall in the bottom half of the deprivation index, including 3 wards in the bottom 10%, and 5 in the next decile up. There are several big estates within 2 miles of the school, including one on Anerley Hill which is on its doorstep. These estates are obviously densely populated compared to the roads immediately surrounding the school, which are full of million pound houses, and which drag the entire ward from the bottom half of the deprivation index to just above the middle. I'd be interested to know how many children from these houses are actually educated in non-selective state schools. Very few I should think. Many children from this part of SE London go to grammar schools in Bromley and Sutton, and there are a good number of big, thriving private schools such as Sydenham High, Dulwich College and Alleyns a short bus ride away.

Anyway - I'm going to C&P a blog post I came across: here

"Last year, the Sutton Trust report prompted me to dig around a bit in the Harris statistics. Here’s some findings which don’t quite fit with the predominant conservative world view about Harris being the saviour of education.

here

This is Harris Crystal Palace’s DFE dataset. Harris Crystal Palace is the chain’s flagship, which is one reason why I focused on it. Give credit to the LibDems who insisted on this lot being published. It’s really quite comprehensive. Obviously pinches of salt all round, accepting that issues like 5 A*-C rating isn’t the most solid of performance indicators once schools have figured out how to game the system. Nevertheless, a few things really stand out:

Harris Crystal Palace is based in a fairly deprived area in Norwood. Its nearest three primaries (1), (2), and (3) have the following indicators :

FSM = (1) 37.6%; (2) 20.5%; (3) 40.4%

EAL = (1) 44.6%; (2) 19%; (3) 39.4%

SEN or School Action Plus = (1) 19.9%; (2) 6.6%; (3) 19.7%

Other local schools have similarly high levels, so we can see that this is a tough area with a high proportion of FSM, SEN and EAL. We’d expect to see Harris Crystal Palace with a similar set of figures. Here are theirs :

FSM = 11.3%

EAL = 15.5%

SEN or school action plus = 3.9%

So as you can see, Harris take far fewer children with FSM, EAL or SEN than those prevalent in their local area. What a remarkable feat for a non-selective school. How do they manage this ?

The answer is that they have a “banded” admissions policy. Here’s the words from their own website :

Ten percent of places will be reserved for students based on their aptitude for Technology, which is one of the specialisms of the Academy.

The remaining places will be allocated by placing students, based on the results of their Non-Verbal Reasoning Test, into one of 9 ability groups of approximately equal size. The assessment is not a pass or fail test. It is designed to ensure that students of all abilities have an equal chance of gaining a place at the Academy.

So Harris assess all applicants, and take children from across the ability range. Super. Except….

Have a look further down their stats at the section called “Cohort Information”. This tells you the nature of the intake based on prior achievement. So those students achieving less than level 4 at the end of primary school are “low attainers”, those with level 4 are “middle attainers”, and those with above level 4 are “high attainers”.

Harris’s figures for the 2012 cohort are as follows :

Low attainers – 1% Middle attainers – 29% High attainers – 70%

Well, well. It turns out that Harris’s banded system designed to give access to students of all abilities, only seems to identify those students of average or – the great majority – above average ability. Harris Crystal Palace is a de facto grammar school. It is selective on both social and academic grounds, excluding a hugely disproportionate number of disadvantaged or less able students from its locality. In other words, its admissions policy is a fiction. Its much-praised results have nothing whatsoever to do with its academy status, or the wondrous abilities of Mr Carpet Warehouse. Rather, its results are due to excluding the local population and cherrypicking clever middle class children from much further away (it’s a matter of local amusement in my wealthy white middle-class area of Beckenham that kids from here who apply to Harris Crystal Palace always get in, despite it being miles away and theoretically banded – our kids always get in the top 10% of each band, somehow…).

Now that is already an admissions investigation in the making. However, there’s more. With that sort of intake, you’d expect astonishing results.

My own school (comprehensive in a grammar school area) takes in 10% low ability, 50% middle and 40% high. Good, but not Harris standard. Harris select an overwhelmingly above-average ability cohort. So let’s compare results.

Ebacc : Harris (43%); my school (45%)

Average point scores : Harris (383.5); my school (407.5)

Value Added : Harris (998.3); my school (1046.9)

So despite fiddling their admissions, despite selecting an overwhelmingly high-ability cohort, despite receiving more money, having smaller class sizes, more support staff, even a higher average teacher salary, Harris deliver worse results than my poor old, bog-standard comprehensive which doesn’t even get the grammar school kids.

Perhaps the staff in my school should be Gove’s heroes instead ?

Update

I pulled those figures a year ago from the 2012 data. In the interests of fairness, I had a revisit of Harris Crystal Palace’s 2013 figures, so that we could see if they did better in 2013 than they did in 2012.

The statistics remain odd :

Harris continues to have a disproportionately able intake (46% higher ability, 45% middle ability, 9% lower ability). This has rather more lower ability than their previous year, but still a very high-ability intake overall.

SEN remains very low at 5%. Obviously parents of children with SEN don’t choose Harris for some reason – possibly since Harris refused a place to a student in a wheelchair on the grounds that it was a health and safety risk to other students – a story which gained lots of coverage in this area.

They also have a low EAL (for London) of 12%. In other words, we’re still looking at an intake which is of higher ability, and has fewer challenging students, than both other secondary schools in the area, and – oddly – the local primary schools which surround it.

So what have they done with this privileged intake ?

Their average points score has dropped to 323 (still lower than my comp, with its less able intake).

The Ebacc headline has hurt Harris (as it’s hurt many academy chains which claimed to be focusing on being “academic”, but actually were manipulating the qualification system). They’re down to 40% achieving Ebacc, compared to my own school’s 59%.

An interesting snippet here is that the tone of Gove’s article, and Harris’s propaganda, is that they do marvellous things with disadvantaged students. Yet they entered none of their prior low attainers for the Ebacc subjects. To be honest, this isn’t surprising, as few children not achieving Level 4 by the end of primary school are going to be able to access the ebacc subjects – I don’t have a problem with guiding them into subjects which they might get more out of. However, what’s interesting is that only 40% of their middle attainers, and even only 78% of their prior high attainers were entered for Ebacc – that means a LOT of perfectly able students are being kept away from the more academic GCSEs. As a comparator, my school’s figures were 65% middle attainers and 93% high attainers entered for the Ebacc. There’s a question there about why Harris aren’t entering their very able cohort for the academic qualifications which they claim is their focus. Something doesn’t add up.

There are some stats which I can’t really compare to last year because their method of calculation has changed, so it wouldn’t be comparing like with like. However, suffice it to say that Harris Crystal Palace’s FSM intake of 12% is much lower than the average for the area in which it is situated. The school continues to take an intake which is much more able, and much less disadvantaged than the community which surrounds it. I’m sure Moynihan is investigating this right now, to ensure that the school changes its admissions policy to target those disadvantaged working class kids who Gove cares so much about.

I’m not, by the way, claiming that Harris Crystal Palace is a “bad school”, as Gove and Wilshaw like to classify schools in their simplistic binary way. They do what they do with the students they have. What I AM saying is that they have found a way to consistently admit an intake which is unrepresentative of their community, and what they achieve with that intake is actually nothing very special. Maybe in Gove’s eyes, this makes Harris a “hero”. I set the bar a bit higher for my heroes.

Update 2

I was asked to do a similar study on another Harris school.

I chose Harris S Norwood because it’s also quite close to me, so I know the area, and it’s been open for 7 years according to its website, which means that Harris can be held accountable for everything going on there, whereas some of the others you mentioned, like Beckenham or Bromley, have only been in the Harris gulag for a year or so, so the results/intake is not yet 100% Harris.

Here’s the link

www.education.gov.uk/cgi-bin/schools/performance/school.pl?urn=135249

The key stats are :

Overall SEN 5.9% EAL 16.2% FSM 22.6%

I had a look at the three closest primaries again, who you’d expect to send the bulk of their children to Harris S Norwood. S Norwood is not a wealthy area, so you won’t be surprised to find that the three nearest community primary schools have the following figures :

SEN (1) 5.1%; (2) 8.6%; (3) 8.5%

EAL (1) 32.8%; (2) 33.3%; (3) 17.1%

FSM (1) 41.2%; (2) 41.9%; (3) 23.4%

I had originally omitted local religious primary schools (their intakes are often disproportionately advantaged), but before I am accused of cherry-picking, here’s the two which are within the same distance as the community schools above :

SEN (1) 5.8%; (2) 11.5%

EAL (1) 73.4%; (2) 32.2%

FSM (1) 22.8%; (2) 22%

So what we have emerging in terms of an overall picture, is of a secondary school whose population seems significantly less challenging than its surrounding primary population, particularly in the field of EAL, but also in SEN and FSM terms. Not quite as stark a gap as Harris Crystal Palace, but there’s still a disconnect. Also interesting is that by the time you get to the Y11 cohort measured in 2013, the number of SEN and EAL kids has shrunken quite a lot – to 3% and 11% respectively. So either this cohort has an unusually small number of EAL and SEN kids in a school which already has an unusually small number of EAL and SEN kids, or some EAL and SEN kids have disappeared before Y11. Impossible to say from the statistics, but Croydon council have recently asked for an investigation into Harris’s record of “missing” students, so it’s an interesting finding.

Let’s look at the actual achievement of the cohort. At Crystal Palace we saw that the Harris school seemed to be taking a much more able intake than a random or comprehensive selection would provide. Will it be repeated at South Norwood ?

Certainly, it’s nowhere near as skewed towards the top end as Harris Crystal Palace, with 19% low attainers, 54% middle attainers, and 26% upper attainers. The figures for the LEA are that 74% of KS2 students are middle or higher attaining students. The figure is 75% nationally. So South Norwood has a higher ability intake than the local average and than the national average. It is not, therefore, struggling manfully with disadvantaged working class kids. It’s taking in a student body which is less disadvantaged than the local average, and more able than the local and national averages.

So what does it do with these students ?

The answer is mixed :

78% 5 A*-C including English and Maths is the headline – significantly higher than the national average, which you’d expect from their higher than average ability intake.

Only 31% getting the Ebacc though, and again we see the pattern we saw at Harris Crystal Palace, which is that most of the low-attainers (9% entered) and middle-attainers (43% entered) are not entered for Ebacc subjects. Only 70% of high-attainers were entered too. That’s fairly low for very able students who you’d expect to be doing more academic subjects.

Average point score was 292 when you strip out the non-GCSE qualifications, which is lower than Crystal Palace, but then you’d expect that because Crystal Palace has a higher prior attainment.

So there you have it. What we can see again is that somehow another Harris school has managed to find an intake which is more advantaged than the locality in which it is situated. What it then achieves with this intake is pretty much in line with what other London schools achieve with similar intakes. There’s nothing outstandingly good, or atrociously bad about it. In other words, it’s a school which does what you’d expect with the students it’s got."

And is it correct that fair banding only applies to 10% of the places

No. Fair banding applies to 90% of the places. The other 10% are allocated based on technological aptitude.

Or do all children have to take a test to get in

Yes.

Who sets the test

A third party. Not sure who they are using at the moment.

And what does the test involve

It is a non-verbal reasoning test.

Presumably it isn't actually a full-blown IQ test

It is very similar to an IQ test. However, unlike the 11+ (which was also an IQ test) it is not used to select only the highest scoring applicants, simply to divide them into ability bands.

Oh, and who marks the tests

I believe it is marked by the third party that supplies them.

To guard against any foul play, you would surely need one hell of a lot more checks on schools than is actually affordable, particularly if every school can set a different test

Not really. The random draw used to allocate places must be overseen independently. That is the only part of the process that can be fiddled to adjust the intake. Every applicant has a roughly 1 in 12 chance of being admitted. That applies regardless of which band they are in and who is in the band with them. All fair banding does is guarantee that the same number of pupils are drawn from each ability band.

And why are SATS test results used to predict minimum GCSE results if they are not a test of ability or potential

SATS are a test of achievement, not of potential. It is therefore not measuring the same thing as NVR tests. Achievement at KS2 is a decent predictor of GCSE results but a child with potential who has scored poorly in SATS may do much better at GCSE.

And why are children required to sit so many tests when nobody entirely trusts the results

You may not trust the results but saying nobody does is a bit extreme. I think you will find that most people do.

And what is done about children who under-perform in tests

In fair banding under-performing in the test doesn't disadvantage you in any way. Looking at HACP, there are around 220 pupils in each band chasing 18 places. That applies regardless of which band a child goes into. So there is no disadvantage to under-performing, nor is there any advantage in over-performing.

Are we sure that more tests and exams are going to fix that

Fair banding is not designed to fix that. It is simply designed to ensure that those admitted to a school are a reasonable cross section of applicants and that parents cannot gain an advantage by buying properties close to the school.

Doesn't it all just bring the whole concept of exams and league tables into disrepute

I don't see how fair banding can bring exams and league tables into disrepute.

Why should we believe that schools genuinely are doing better now, but that they were not genuinely doing better, before

Schools were doing better before but we weren't improving as fast as other countries. The apparent improvement in results was due to grade inflation and academic studies were able to show exactly how much grade inflation was occurring. Indications are that schools are still improving. The question is whether we are now improving as fast as other countries.