Class size matters... but money is tight.

Class size matters... but money is tight.

The resistance to smaller class sizes in state schools boils down to one thing - economics.

Are smaller class sizes better for students in schools?
Surely, it's a no-brainer, right? They must be better!

Not so fast! I've had managers tell us that WE, the teachers were the most significant factor in the kids learning. Translate that as "if the kids don't do well it's YOUR fault".

Now that means, that all the other factors, class size, background, class, language etc etc don't matter as much as what the grunt at the front of class does, and the results are ultimately THEIR fault. Not systemic, not structural, organisational or socially contextual. Sure, teachers do matter, but we see kids for 2-3 hours a week, in a class of 20-30 usually, meaning we average a few minutes a week 1 to 1 attention with a kid. Meanwhile there are 168 hours in a week, what happens in the 165 hours a week where I don't see a kid, who knows?

A recent government study suggests the evidence in favour of smaller classroom size is not a strong enough argument for doing anything about it in a classroom.

Why? Money stupid!

The report concludes that:

"class size reduction policies are not the best option in terms of value for money to raising pupil attainment, compared to others such as increasing teacher effectiveness" [my emphasis]

We just need better teachers!!!

But wait, I'll get on to the evidence (never mind the common sense) in a minute. It's worth noting that in private schools, class sizes are STILL decreasing as indicated by falling pupil teacher ratios. So in the 'market for education' the 'best' schools (and private schools undoubtedly have better attainment scores by just about every measure possible) have smaller class sizes.

Why? Because it obviously matters - "class size is the third most common reason for parents to choose to send their child to an independent school".

The context.
Ok. In the UK there hasn't been a lot of research. The context of the study was that the birth rate in the last 7-8 years has risen, so in the next decade there'll be a demand for school places. Contrast this against the fact that pupil numbers have fallen steadily in roughly the same period while class sizes have stagnated.

The government wants 'value for money'. Rather than use the fall in pupil numbers (primary and secondary; by almost 200,000 each) since 2004 as a chance to have smaller classes, they've kept class size the same and tried to save a bit of moolah.

But what of the actual evidence?
Lets get the common sense aspect out of the way. It's obviously much easier for a teacher to relate to a pupil in a smaller class. To give them attention, to explain things, to chat, have fun and share ideas with, that simply cannot be done in a class size of 30 or more. Even in a class of 20 or more it's hard to give pupils, especially those that need a bit more help, the attention they'd like.
If pupil teacher ratio doesn't matter why do affluent parents speed a fortune on private tuition?
If it doesn't matter why not just double up and have classes of 70?

Any education worker can tell you it's much easier to actually get on with teaching and sharing in a small class.

Actual class sizes.
In 2010 in primaries, 75% of class were 26 pupil per teacher or above, with 30 pupils per class being the modal (most frequent) value (22% in 2010, up from 12% in 1996).
In early secondary (key stage 3) 60% were in classes 26 or above, again with 30 being the modal value (11%).
In key stage 4, modal class size in 2010 was 25. with 64% of classes less than 26 pupils. That same year, I was in the 8% who had classes over 30 - and it was very hard work!

But back to the evidence. Let's look at the study. I'll share the summaries of interest:

The evidence base on the link between class size and attainment, taken as a whole, finds that a smaller class size has a positive impact on attainment and behaviour in the early years of school, but this effect tends to be small and diminishes after a few years. The most recent research supports this.

Research findings from England show that in smaller classes, individual pupils are the focus of a teacher’s attention for more time; there is more active interaction between pupils and teachers; and more pupil engagement. In larger classes, there is more time spent by pupils interacting with each other; more time spent by teachers teaching the substantive content of the subject knowledge; and more time spent on non-teaching tasks like taking registers.

Smaller classes have been found to lead to a small increase the number of years a student spends in post-compulsory education. A study from Denmark estimated that a reduction in class size during the whole of compulsory schooling by 5% (from an average class size of 18) provides a rise in post-compulsory education by approximately 8 days.

Research on parental opinion on class size found that 96% of parents believed that the number of children in a class affects the quality of teaching and learning. In the same study teachers and head teachers were also found to consider class size to be an important issue.

Continually falling Pupil Teacher Ratios in independent schools implies demand for smaller class sizes in the market for education. Class size is the third most common reason for parents to choose to send their child to an independent school.

So impact and behaviour and attainment is demonstrable.
Smaller classes are considered to positively impact teaching.
Parents like smaller classes.
Private schools have smaller class, and generally do a lot better (why else would anyone pay for schooling?)

But here's the piece of research evidence the government will actually go with:

A study by Hattie (2009) found the impact of reducing class size on attainment to be smaller than the impact of other interventions. Hattie argues that value for money in raising attainment in schools is better achieved through other interventions than class size reduction. This is supported by research from Rivkin et al (2005) that find that increasing teacher effectiveness represents greater value for money than reducing class sizes, while Hanushek (2011) finds that teachers and their quality are, by far, the most important factor in determining student achievement.

What constitutes a 'good teacher' in the eyes of the government and Ofsted is probably pretty far from what many people, certainly radicals, would consider a 'good teacher' to be. The methods by which teachers judged are whatever measures the government and Ofsted deem salient to their needs at a specific point in time. If you can equip students with generic workplace skills like data entry - GREAT TEACHER. If you can tick a bunch of boxes and demonstrate 'learning' in 20minutes - GREAT TEACHER.

Studies like this, commissioned by a government hell bent on reducing a deficit are always going to draw a conclusion that reduces results to 'good or bad teaching' and ignores any other salient factor; social class, background, languages. They're clearly not going to a draw a conclusion that suggests putting more money in schools or supporting students or staff more.

Clearly others might do a better job of interpreting this than me, and if I read the study again I'm sure there's additional stuff I could pick up.
But what I take from all of this is that despite the 'obvious' importance of class size, despite it's demostrable positive impact on attainment, behaviour, relationships, enjoyment of education, all that actually matters is 'value for money'.

We can make all the arguments we wants to change things about schools, but we shouldn't kid ourselves - schools in capitalism are businesses. And business must be 'viable'. If money can be saved, whatever it's social impact, capitalist schools are duty-bound to do so.

Posted By

Dec 22 2011 14:16


Attached files


jef costello
Dec 22 2011 14:48

I remember seeing things about relative 'value' in increasing teacher effectiveness when I was training. Obviously increasing teacher effectiveness is something that requires more work (unpaid) and better work (with no bonus of any kind) whereas decreasing class sizes requires paying more teacher salaries. In my current school they won't pay for any external training so not only does any improvement have to be done in your own time someone also has to give up their own time to prepare and give training.

One of the things I noticed was that in the danish research the drop in class size of 5% was from a base size of 18, which I have only ever seen in bottom set classes that were basically behaviour control. Is there any research based on class sizes that approach the sizes we deal with on a daily basis? I'd be interested to see more in that line.

Dec 23 2011 19:03

research in the Uk has been sparse apparently, but only the Danish study starts from avg sizes of 18 - I imagine the others, like Tennessee Star research etc were in school populations similar to what we'd experience as UK state school teachers

Chilli Sauce
Dec 23 2011 20:30

This is really good Choccy. Your blogs have been spot on lately.

Dec 23 2011 21:20

Cheers, this was kinda rushed. Still trying to strike the balance between 'blogging' and analysis. Will be a while before I get it right!

Dec 24 2011 11:02

No, they've been great.. 'spot on' sums it up perfectly..

Jan 6 2012 11:51

Now, Sutton Council has asked the government to up the legal limit on classroom sizes from 30:

Jan 6 2012 15:57

I know, mental. And although I'd make the argument that they should be reduced across the board, the largest impact of class size is in early years! So such a move would be an exceptionally backward step and massively hurt kids.
And of course it won't be happening in private schools!

the button
Jan 6 2012 16:17


When boys arrive at Eton they usually find themselves in classes of 20 to 24. As their academic study progresses and they make their own choices about what to learn, the class sizes tend to reduce, the size depending on the subject and its popularity in a given year. A typical GCSE class in D block (year 11) would contain about 15 boys learning a second modern language for example. and 20 in mathematics. Once AS-level and A-level choices are made, class sizes reduce again, to about 12 in subjects such as mathematics, or to smaller classes in subjects chosen by fewer boys.

Jan 6 2012 18:01

fuck me, i had 31 in a gcse class last year, AND they were 'set 4'

The amount of times we were told 'class size doesn't matter' - funny enough it was only ever the managers in charge of budgets that said that, not actual classroom teachers