Accidentally stumbling on a measure of intelligence: revelations and implications

For the past few years I have been accidentally corroborating a study that shows reading ability is linked to intelligence. The evidence is fairly dramatic when it reveals itself and I’d be chuffed if other primary teachers also contributed to my informal study (it takes minimal time and can be done without children knowing)! Disclaimer: I am just a primary teacher and not a full time researcher. I can’t do thousands of replicates, but I’m always looking for patterns and understanding of the children I teach. I do read the results of far more serious research, but I’m also curious about what my own class can reveal. Rest assured, nobody in my class gets pigeon-holed and if you take that message from this blogette, then you’re getting the wrong end of the stick.

Who is in the primary classroom but not in the secondary classroom?

In any primary class you will have a huge range of ability, even if the class only has one year group in it (many will be mixed year groups). Added to this, you will also have a higher ratio of children with SEN, particularly those with statements, than in secondary schools. This is because it is very common for those with the greatest needs to either be taught at home from the age of 11, or to go into a special school at that age. At the other end of the spectrum, the most intelligent and hard working will go on to the grammars or perhaps take up a scholarship/assisted place at a good local private school. Secondary schools therefore receive a year 7 cohort with a narrower range of intellectual ability, and then of course the range of ability in any class will be further narrowed by the use of streaming.

What can the primary teacher do?

Although I have been told not to arrange the group tables in my class by ability, I find that some subjects seem to necessitate it, otherwise how would I be able to work with a group, direct the TA to work with a group or easily give out the different worksheets? Mixed ability seating is sometimes virtually impossible when you consider that at KS2 you will have a few children who are so far behind and so unable to cope on their own that you have to have an adult with them, and my conscience would not cope well if I used higher ability pupils to look after them. My argument is that if ability banding happens in secondary schools, but on a bigger scale, why should the primary teacher be prevented from occasionally arranging the children in this practical way?

How do I manage with such a huge range of ability?

On taking a new class, I had, as usual, received a bunch of data for each child and I normally used this to put children into groups. The groups might only exist in my planning so that I could make sure the teaching/learning/work was differentiated and for regularly updating target tracker (half termly) or various spreadsheets (weekly), but for guided reading I did (and still do) actually work with groups on a carousel system as per school policy. Of course, children change as they develop and I was starting to think maybe my reading groups weren’t quite right. I didn’t have the time or the energy to have the children sit a comprehension test and then spend hours marking and deriving levels from the data generated, so I needed to do something quickly. I had read about various studies showing that reading ability and amount of time spent reading was positively correlated with intelligence and was also painfully aware that the children in my class had different levels of reading fluency. It struck me that there were so many who were, despite the system saying ‘They can read’, not truly fluent readers, but had been ‘moved on’ and I wondered what the real scale of this problem was. Maybe I should just measure the fluency?

To my mind, the lack of reading fluency had parallels with a common lack of basic number bond and times tables knowledge that so many children come up to KS2 with. Just as a child struggles to understand and calculate KS2 maths skills because they have to spend so much time using their fingers to count, it seems that so many children struggle with comprehension because they are stuck with slow reading that prevents them from getting the whole sense of a sentence and the storyline. My solution for the children who lack maths prowess is to make them practice using and applying the basic knowledge until it is in the long term memory, ready for instant recall, and this has proven extremely effective (and to any mathematician this makes perfect sense, but not to those in primary education who overtly detest ‘rote learning’). The standard solution in primary ed for lack of comprehension skills is for the teacher to ‘teach’ higher order thinking skills, but I was thinking that perhaps the slow readers simply needed to do more reading and become more fluent?

The revelation

So, I asked the TA to find a text appropriate to the age group I taught and to photocopy it just once. On one copy she marked the number of words in each sentence, and the unmarked copy was to be given to a child to read. One by one, the children were asked to read for a minute out of earshot of the rest of the class and the children were not told what exactly we were looking for or measuring, other than ‘We need to check your reading’. There was no stopwatch either as the TA surreptitiously looked at her watch and immediately afterwards she would note how many words that child had read on a simple list of names. This was also an opportunity for the children to let the TA know about anything that is troubling them, so it served a social purpose too (TAs are very important for this). She then gave me the data and I popped it into a spreadsheet.

And then, when the children were out at break, I arranged the numbers low to high.


I invited the TA to have a look as I arranged the numbers and it was at this point we were stunned into silence. Essentially, what was revealed was an exact order of children by intelligence. It was uncanny and it was horrifying in its implications.

How do I know how intelligent the children are anyway?

Oh come on. You can tell! We all can. Surely? I think where some primary teachers might find it difficult to tell is with the highly intelligent children; I’ve had conversations about astrophysics, going to Cambridge, how Latin feeds into our language and the finer points of medieval church architecture with some 10 year olds that really brighten up my day! But I’m not sure many primary teachers would be able to talk about astrophysics? Also, I think that if many primary teachers go into teaching wanting to promote wellbeing rather than academic prowess, they therefore might be more likely to tune into measures other than intelligence? However, I find that the ecosystem of teacher and pupils morphs together such that the teacher unconsciously modifies types of questioning for each child and I use my knowledge of the type of conversation (and questions) I can have with that child (whether it be in lesson or at breaks) as a proxy measure of intelligence. Primary teachers do get to know each child very well; I could identify in a fraction of a second children who should be given the opportunity to go to a grammar, an overtly academic school such as MCS  or to take up a scholarship place at a private school.


What this data showed was that if reading fluency was correlated with intelligence as well as reading comprehension skills, and if reading fluency was determined by amount of time practising reading, then I needed to get the slow readers to be fluent readers by getting them to read more. Subsequent checks over the years have revealed that the most intelligent children are prolific readers and this kind of makes sense: if you’re reading more, you are spending more time quietly concentrating and focusing the mind on one thing (good training for being able to concentrate in general) at the same time as increasing the rate of vocabulary, idiom and general knowledge assimilation relative to other children who read less per unit of time. I’ve also found that children who read more and who are therefore more fluent also tend to be good at spelling, regardless of the spelling lessons I teach. This is probably because the more fluent readers are inadvertently studying the spelling rules of the English language without even realising it.

The simple data set, which could (and should) have been improved with more replicates to make it statistically significant, also showed me that of reading comprehension goes hand in hand with reading fluency and time spent reading, then my guided reading ‘lessons’, where children read for a mere minute or two each to be followed by questioning linked to Bloom’s Taxonomy, were/are possibly pointless? Have a little chew on that, fellow primary teacher!


I think more needs to be done in primary schools to tackle lack of reading fluency. As I have said on twitter, nobody got better at anything by doing nothing about it. Children need to read more and they need to be forced to read until they are truly fluent because this is the point where true joy of reading, and choosing to read for pleasure, really takes off. When children reach this stage, lots of different kinds of books need to be made available and children need to be given/encouraged to read quietly for a good half an hour a day (or more!). Also, I have found whole class reading, where I read and the children follow using a reading strip and pointing with their finger (asking questions about vocab along the way) to be incredibly useful, particularly for non-fiction texts and when I ask slower readers to re-read paragraphs that I have just read (going over the same words helps with speed of recall and instant decoding).

It is horrifying to consider that children just don’t read enough. It is also horrifying when we consider that children read less in primary school than they used to; the use of iPads, complete lack of textbooks, lack of silent concentrating during periods of writing or reading coupled with lack of reading opportunities and habits at home conspire to produce a generation of children who not only cannot comprehend, but it seems that their average intelligence is affected too.

Children need to read more.

Who’s with me?


14 thoughts on “Accidentally stumbling on a measure of intelligence: revelations and implications

  1. Studies have shown that reading comprehension is highly related to nonverbal intelligence, so your hunch is probably correct. But the way you “measure” intelligence (by the kinds of conversations you can have with children, if I understand it correctly?), you are bound to pick out the children who are good at language and do a lot of reading as the most intelligent, so the observation is circular.


      • No, I mean: You argue there’s a link between reading and intelligence. Then you say you “measure” intelligence by just knowing which kids are smart. One way you say you see this is by the kinds of conversations you have with the children. That means what you consider “intelligence” is probably language + general knowledge. So your observation is probably “language ability, general knowledge and reading are linked”. Which is also true, but doesn’t say much about intelligence.

        Liked by 1 person

      • Well, intelligence is one of those impossible-to-define-concepts, but I think most would agree that there is a strong nonverbal component to it too: Students can be great at reasoning but poor at language – with poor comprehension as a result. Or the other way around, great at language, poor at reasoning, also not a good sign for comprehension.


  2. Interesting read but I think you are wrong in your assumptions that there is a narrower range of ability among students in secondary schools than primary. Data that I have seen show that the spread between lower and higher ability students widens continually as students get older – this was from a paper analysing big data sets of thousands of students.


    • Ah yes, of course I completely agree on spread of ability increasing over time; a natural consequence of entrenching disadvantage through ‘play’ in EYFS all the way through to differentiation up till year 6. Is the spread still more so given that the lowest contingent (the ones who are severely disabled or with severe SEN) goes elsewhere? I have only seen children who are both deaf and blind in primary schools.


  3. There is absolutely no question but that reading fluency develops with practice, and that the more children read the faster their cognitive development. This was Stanovich’s conclusion in his 1986 paper on the ‘Matthew Effect’, which is one of the most commonly cited research papers ever.

    However, the ability to learn to decode is only weakly related to general intelligence, especially when decoding skills are poorly taught (or virtually ignored, as they often were a generation ago). Persistent failure can indeed lead to poor performance on verbal intelligence scores, but this has little if any effect on non-verbal intelligence. We used to routinely test our intake with the non-verbal section of the Children’s Ability Scales (our feeder schools were appalling) and we found that these scores correlated weakly with scores on the NFER-Nelson reading test and the Suffolk Scale. One year, the pupil with the highest score on the CAS had one of the lowest scores on both reading tests. By the time he caught up in reading and spelling, his formal education was drawing to a close. Needless to say, this boy had serious behaviour problems in primary school, and I have no doubt whatever that his primary school teachers wrote him off as a thicko. Although reading comprehension with older pupils who have good decoding skills tracks oral language comprehension very closely, I would be very worried about using this as a guide with KS1 pupils.

    I also strongly dispute your contention that pupils ‘pick up’ spelling from reading. Once again, our test scores revealed that quite a substantial percentage of pupils who had low scores on Young’s Parallel Spelling Test had average and above-average scores on the reading tests. Spelling seldom gets the attention it deserves, and a lot of bad habits are engrained by premature attempts to get them to write using ‘invented spellings’. I wish I had a pound for every time I’ve corrected a pupil who wrote ‘thay’ for they.


    • I agree with you here on the correlation in KS1: children are still learning to decode using taught phonics knowledge. I will address this in my next post which is about what we can practically do to get children to proper fluency in reading (and good teaching of phonics will feature).

      As for the spelling, yes I will have to agree with you here too! I teach the spelling rules as a completely separate lesson (which will invoke the ire of many in primary ed who believe SPG should be taught as a component embedded in creative writing sequence) and find that children really benefit from this, particularly the poor speller; they just need a rule to follow rather than being asked to look everything up in a dictionary all the time.


  4. Great post and great test. Your comment on good readers being good spellers is slightly erroneous. There is a correlation between good readers and spellers because the good readers tend to have better memories and this is how they have become more fluent more quickly. The better memory also means they tend to be better at spelling. However you won’t improve spelling by improving reading necessarily. On the other hand you can improve reading by teaching spelling. Hope this is helpful. I think the simplicity of your test is genius.


  5. I’m with you-fluency makes a huge difference from the point of view of a pupil being ready for secondary education. The crucial thing is that reading is so much faster than other ways of sharing information. Fluent reading is about 300 words a minute, whereas normal speech is about 100. That’s a huge gap for alternative strategies to make up. Even if not everything that is read consciously “sticks”, a lot of ideas and structures will.

    Liked by 1 person

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s