Why putting children as young as seven in "streams" doesn't work

Francis Gilbert's picture
A report from the University of London's Institute of Education has found that, across the UK, assigning students to certain classes based on academic capability happens to one-in-six youngsters by the age of seven. I appeared on Radio 5 Live Breakfast talking about this issue with Nick Seaton, who is secretary for the Campaign for Real Education. It was an entertaining and heated discussion which can be listened to here. I point out that streaming doesn't work because it tends to label children in the bottom sets as failures. Usually, these bottom sets are based more on "social class" and "gender" -- and in some schools according to ethnicity --  than ability, with the bottom sets being often full of naughty boys of working class origin, with parents who are not engaged with school at all. I know because I've taught my fair share. In my current school, we teach mixed ability and it works much better than setting and streaming (which is putting children in one stream for ALL subjects) because you can pair up able students with the less able, and because children don't feel like they're labelled as failures before they've even begun.
Share on Twitter Share on Facebook

Be notified by email of each new post.


Fiona Millar's picture
Wed, 15/06/2011 - 18:39

I think the real problem with streaming is that there is little opportunity for movement so it stifles aspiration ( rather like the 11plus). No one wants to move down and those at the bottom can't move up unless there is a space. I wonder whether the schools that are streaming it are doing so for the benefit of the teachers rather than the pupils.
Good primaries can manage mixed abilities in primary schools, and some setting maybe in the later years. There is a place for setting by subject in secondary schools but it isn't right for all subjects.

Francis Gilbert's picture
Wed, 15/06/2011 - 19:52

Yes, I think you're right Fiona that setting can be effective in some subjects -- but even then the sets tend to get split according to social class/gender/ethnicity lines.

Allan Beavis's picture
Wed, 15/06/2011 - 22:40

In my experience of primary schools, mixed ability classrooms work exceptionally well. The more able took great pride and felt a great sense of responsibility in helping classmates who needed more support and the latter benefited from a natural and inclusive environment in which they were inspired to improve their learning alongside a classmate. This would not happen if primary school children were streamed. I think it's quite cruel to stream off the "clever" children to another class - what motivation is left for the ones who would feel abandoned and excluded? Children - especially primary school children - develop at vastly different rates and ages. Also crucial is that they should be instilled with a love of learning and going to school at that age. Being siphoned off into the lower stream would surely do more to de-motivate a child from enjoying school? Hardly the most positive frame of mind to enter secondary school and aspire to do well in GCSEs.

Janet Downs's picture
Thu, 16/06/2011 - 13:42

I know I've said this before but it's worth repeating. OECD evidence says this about selection:

"The data from PISA show that creating homogeneous schools and/or classrooms through selection is unrelated to the average performance of education systems, but clearly associated with larger variation in student achievement and a significantly larger impact of socio-economic background on learning outcomes. In particular, the earlier in the student's career the selection occurs, the greater the impact of socio-economic background on learning outcomes. That suggests that selection tends to reinforce inequalities as students from disadvantaged backgrounds tend to be exposed to lower quality learning opportunities* when compared to their peers from more advantages socio-economic backgrounds."

*Lower quality learning opportunities in this context means "access to more teachers" and spending on education, as well as such things as class size, participation in after-school lessons and lack of material resources. The first two factors account for more of the variation in student performance than the other factors, according to OECD.


Ben Taylor's picture
Fri, 17/06/2011 - 00:24

Some thoughts.

You say mixed ability is a win win but I am not sure how. I'd like to see you teach a class of 30 where some are illiterate and others want to practice iambic pentameter.

I realise you have said that you set different work in a mixed set but can you make it absolutely clear how? What was the spread of grades over your class you spoke about of your own students?

How do you address Nick Seaton's point of the overall system deficit? It's ok if it works for you but what about other situations?

Once again we have the issue of labelling failures - who is issuing the label of failure? I admit this is complex but you certainly are keen to reinforce the idea rather than focus on administering to objective need.

PS Birbalsingh disagrees, see also comments at her blog of many teachers whose personal experience is at odds with you. Whose anecdote should we believe? Maybe the parents can be left to decide as Nick Seaton said.


I' d like to add anecdotally I agree peer group assistance can work. But my experience was giving help to people at a lower level of learning in addition to my own study, rather than being co-taught.

Allan Beavis's picture
Fri, 17/06/2011 - 03:04

I don’t think using such an extreme example of illiteracy and iambic pentameter in one primary school classroom is reasonable and just deliberately confuses the issue.

Anyway on to Nick Seaton. Aside from his extreme pro-selection stance, I can’t take him or the organizations he is involved in seriously because he was at best economical with the truth when we spoke. Here’s why.

I can’t remember if it was you, Ben, who posted up this article (one of a number) that he posted in the Yorkshire Post about Grammar Schools.
In it, he claims that “an ICM opinion poll, carried out for the National Grammar Schools Association, produced results that surprised many people.” The surprise being that so many were supportive of Grammar Schools.

When I checked with the ICM, they told me that no such survey was commissioned by the National Grammar Schools Association but that a general poll had been commissioned by the Campaign for Real Education. They were unable to let me have the precise questions asked, whether the public were surveyed in person or by telephone; if the questions were posed in such a way to suggest any bias; they could not let me have any data or analysis and suggested that I contact Nick Seaton direct for these.

Nick Seaton admitted to me that the CRE commissioned the poll and that only two simple questions about Grammar Schools were asked. When I asked him if he could let me have full copies of the survey – questions, data, analysis etc. – he said they were buried deep in some old computers and would take a long time to dig out. I never did receive them.

Why did Seaton claim in print that the survey was carried out for the National Grammar Schools Association and not by CRE? Was it to make the survey appear more credible and exhaustive about Grammar Schools, because asking only two very simple and leading questions is almost guaranteed to deliver the result you seek. He spun this feeble “research” in his article to make it appear as if the country is clamouring for Grammars and that governments are out of touch with what people want (as I type this I am mindful that this is the phrase Angie Smithers would use).

If Seaton wants to be taken seriously then he should stop making deceptive comments to bolster unreal claims.

Ben Taylor's picture
Sat, 18/06/2011 - 11:10


OK regarding mixed ability is Francis writing about secondary schools, I thought this was his kind of teaching? That's what I am interested in.

I still don't understand this point about labelling people in bottom sets as failures. Who says this and why? Just because it takes longer to learn than someone else, and so people are grouped accordingly to spread the effort, why is that failure?

I teach in a private English language school - most of the punters come from overseas to get fluent in English. Do you think we should put the elementary students in with the people getting ready for university entrance via the academic English course IELTS? You can bet the classes would not work and the students would be very unhappy.
Our regulators the British Council wouldn't think much of that either.

According to Francis' reasoning the beginners class are failures when really they are just at a different stage of development.

I recognise that peer assistance is useful but schools should be able to have sets if it helps them.

Allan Beavis's picture
Sat, 18/06/2011 - 11:42

Its not a question of official labellling but of perception and by extension a self-perception of failure, especially in primary school, where quite frankly it is just as important for schools to instil a love of learning, not hitting personal or school or teacher targets. Important also to learn how to get on with other children and to gain a first understanding of the wider world.

As I'm writing this and scrolling up, I have become aware that at least both Janet and Fiona have given an explanation for what you seem not to understand, so perhaps it would have helped had you read them too? You do have a habit of going round and round in circles chasing your own tail, Ben!

Janet Downs's picture
Sat, 18/06/2011 - 13:30

Ben - selecting children according to ability is not the same as having a class of beginners who are all at the same stage. In your class of adult beginners there will, of course, be those who pick up the subject quicker than others. Presumably, these will eventually go on to intermediate and so on, while others may repeat.

But we're not discussing adults - we're talking about children even as young as 7. As usual, I'll consult the oracle - OECD:

"[School] Systems that show high performance and an equitable distribution of learning outcomes tend to be comprehensive, requiring teachers and schools to embrace diverse student populations through personalised educational pathways. In contrast, school systems that assume that students have different destinations with different expectations and differentiation in terms of how they are placed in schools, classes and grades often show less equitable outcomes without an overall performance advantage."

This would suggest that dividing children up into "academic", "technical", "vocational" does not result in "an overall performance advantage." (page 15 OECD 2010*)

And, "In countries where 15-year-olds are divided into more tracks based on their abilities, overall performance is not enhanced, and the younger the age at which selection for such tracks first occurs, the greater the differences in students performance, by socio-economic background, by age 15, without improved overall performance." (page 15*)

* http://www.oecd.org/dataoecd/34/60/46619703.pdf

Paul Holbourne's picture
Sat, 15/06/2013 - 06:46

The children MUST be challenged at their level! Now, if that means providing work according to the child's ability within a mixed ability setting, so be it.
If all the children are treated the same, there will be lots of bored as well as difficult children.
I usually provided three levels of work, and further extension for the very bright ones. Yes, it does mean plenty of preparation time and helpful texts.

Add new comment

Already a member? Click here to log in before you comment. Or register with us.