Does traditional grammar instruction improve children’s writing ability?


Alice Sullivan and Dominic Wyse. 

Children in England have recently taken their statutory tests at age 10-11 (commonly known as Key Stage 2 SATs). The results, published today, show that the pass rate has plummeted compared to last year. This is because the nature of the tests changed dramatically in 2016. We focus here on why the new English tests have been so difficult for children to pass – and why most parents would struggle to pass the tests too.

The key change in the English writing SATs is that there is a new focus on formal grammatical terminology. So, for example, eleven-year-olds have been faced with questions such as this:

Tick the option which shows how the underlined words in the sentence below are used.

The insect-eating Venus flytrap is a carnivorous plant.

Tick one:

  • as a main clause
  • as a fronted adverbial
  • as a subordinate clause
  • as a noun phrase

In preparation for these tests, children have been spending substantial classroom and homework time on grammar exercises such as the one shown below. This is an example of homework from an eight-year-old child:

Homework

If you are curious to know how you would fare on the Key Stage 2 English assessment, you can test yourself with questions adapted from the sample English question papers for 2016 published by the DfE Standards & Testing Agency. The actual grammar paper used in the 2016 test is here.

If you can’t answer these questions, you are not alone. One interesting feature of these tests is that most adults will be utterly stumped by them, including those who write perfectly grammatically. Most of the parents of the eleven-year-olds taking these tests have never learned to use the technical grammatical terminology required, because it wasn’t taught at school when they were kids. So, if people can write grammatically without knowledge of the grammatical terminology that describes what they are doing, why are primary schools now teaching children this terminology?

Traditional instruction in grammar was commonplace in the 1950s classroom, but was largely abandoned by the 1970s, partly due to a research evidence that formal grammar teaching of this sort was ineffective. The consensus was that the grammar of one’s native tongue is best learned by extensive reading and writing practice supported by appropriate teaching, rather than by being taught a set of formal rules and technical grammatical terms. Educators also noted that children found traditional grammar lessons dry and boring.

So, what changed? You might imagine that such a radical shift in the primary curriculum could only have occurred because some new evidence emerged to show that the teaching of formal grammatical terminology is the best way to improve children’s writing skills. You would be wrong. Reviews of the evidence still concur that this is far from the case – see, for example, this EPPI review, a review from the EEF, this study, and this one.

One source for the new thinking on grammar is E. D. Hirsch, a US academic who has had a profound influence on the thinking of former education minister Michael Gove and the Department for Education. There is much that appears commonsensical in Hirsch’s work, but it is not always evidence based. Hirsch states that “Parents and teachers should ignore the “expert” advice that holds that it is unnecessary to teach children grammar.” (p.54 Hirsch 2006 “The Knowledge Deficit: Closing the shocking  education gap for American children” Houghton Mifflin Books, Boston). He asserts that, in order to learn to write grammatically, children need to be taught to name the parts of speech. Hirsch’s view appears to be driven by his concern that failing to teach standard English usage to disadvantaged children does them a serious disservice. It would be hard to argue with that. But it does not follow that the best way to teach children to write grammatically is to teach them technical terms such as “subordinating conjunction”.

This matters because classroom practice has changed dramatically in line with the new curriculum. Because SATs are high-stakes assessments, teachers have no choice but to devote a great deal of time to preparing children to pass these tests. Clearly, there is a huge opportunity cost. Time spent on formal grammar is time that cannot be spent on other activities such as reading and creative writing. Even the criteria for teacher assessment, which are supposed to focus on the composition of writing, are dominated by grammar, punctuation, etc.

The stated goals of the primary curriculum place a laudable emphasis on the importance of language fluency for all. We agree that “All the skills of language are essential to participating fully as a member of society; pupils, therefore, who do not learn to speak, read and write fluently and confidently are effectively disenfranchised”. (p.3) The UK has a longstanding problem with poor basic literacy skills. But there is no evidence that formal grammar teaching will ameliorate this problem. No assessment has been made of how this new curriculum will affect children who are struggling with reading and writing. And the research evidence clearly shows a range of better ways to improve children’s writing.

It seems to us that the Key Stage 2 grammar curriculum is an egregious example of non-evidence-based policy. It is important to note that we are not arguing against teaching children to write grammatically, nor are we making an argument against testing in general. We are arguing in favour of bringing evidence to bear on education policy. The alternative is for the ideological fads and fancies of politicians to drive what happens in schools, at the expense of children’s learning.

Tagged with: , , ,
Posted in curriculum & assessment, Education policy, Evidence-based policy, Literacy, Schools
11 comments on “Does traditional grammar instruction improve children’s writing ability?
  1. Philip Reilly says:

    Looks horrifying for children to be attempt in such questions. All of the grammatical terminology has been visible as part of teaching and learning for a number of years (see Grammar for Writing documents from mid 2000’s). However, in my experience (primary) a small proportion of children within a year group do use the grammar and as you say aren’t aware of the terminology associated.
    When I have attempts to discuss things such as main and subordinate clauses it just goes over their heads. Identifying in reading examples of such grammar is a minefield because there are so many exceptions within the English language- spelling rules is probably the classic.
    All this links to the demands of PISA tests. When school leaders feedback on their “analysis” of all the data they invariably focus on phonics/spelling and maths because these are ‘measurable’.

  2. Sandy Leaton Gray says:

    I am the same age as Michael Gove. I attended four different primary schools between 1971 and 1979 and contrary to popular opinion, in each of them, we were indeed taught grammar, and spent plenty of time parsing sentences. I consider that this set me up well for work as a writer and sub-editor on various national publications, and later as an author of books and journals in my own right. It also helped me learn several languages to a high standard, including Latin. However the grammar I was taught in the 1970s was significantly less technicist and linguistically pedantic, and apparently much more useful in identifying patterns in language that would allow clear expression when replicated, and ready transfer of knowledge amongst languages. In other words, it was a shrewd and concise framework for understanding language and expression that, in my case anyway, proved its worth. What baffles me is that we now have a reductive model of grammar on display that is likely to inhibit writing rather than facilitate it. It makes no sense to me as a former sub-editor and I imagine it seems arbitrary to anyone other than postgraduate students of linguistics and their teachers. ironically the time spent on acquiring this faux lexicon of pedantry will probably be drawn from time that might otherwise be spent learning modern foreign languages. This deserves a hollow laugh.

  3. John Hodgson says:

    You write in your last paragraph of ministers’ ideological fads and fancies. Ideology may seem fanciful but it is related to the exercise of power. The effect of these abstruse grammatical tests is to put severe pressure on teachers and children in maintained schools. If teachers are found wanting in terms of their pupils’ performance, schools can be made into academies. If children fail tests in English that don’t, as you rightly say, actually test their communicative capabilities, the goods of skilled employment and of further and higher education can be withheld from them. This is an exercise of governmental control.

  4. “Non-evidence-based policy”? You just don’t mention the evidence base: Debra Myhill’s research which shows very clear positive effects of explicit grammar teaching on children’s writing. No doubt we need even more research evidence, and especially on the effects of grammar teaching at primary school (Myhill’s research has been in the secondary sector); but it’s certainly not the case that there’s no research evidence.

    “Difficult to pass”? Not so – 72% achieved the expected standard in SPaG (spelling, punctuation and grammar), which is actually higher than the percentage for Maths and Reading. Yes, the tests would be hard for adults, but you explain why that is. The results are actually remarkable, given the increased difficulty of the test compared with last year (and the shameful lack of government support for primary teachers).

    “Because SATs are high-stakes assessments, teachers have no choice but to devote a great deal of time to preparing children to pass these tests”? Not so. What’s being tested is just what’s in the National Curriculum, so if teachers are following the curriculum, they won’t need to teach anything extra. Moreover, the curriculum only requires children to learn on average six grammatical terms through primary school – not a big deal, surely?

    “People can write grammatically without knowledge of the grammatical terminology that describes what they are doing”? True, some people can; but what they can’t do is to talk about the grammar of what they read and write. Grammatical terminology provides a metalanguage for reviewing alternative constructions, unpacking complexity, understanding ambiguities, and much more.

    “The UK has a longstanding problem with poor basic literacy skills.” Precisely, so there’s no evidence to support “the consensus … that the grammar of one’s native tongue is best learned by extensive reading and writing practice supported by appropriate teaching”. It’s time to try something different, and explicit grammar is a promising avenue. But of course there are plenty of other avenues, such as explicit study of vocabulary and word meaning.

    Dick Hudson (http://dickhudson.com/the-spag-tests/)

    • Andrew McCallum says:

      Debra Myhill’s work shows some link between grammar teaching and improvements in writing – but a very particular kind of grammar teaching and one unlike that promoted by testing at Key Stage 2, and unlike most of the teaching that prepares pupils for these tests. Her own meta-analysis of research into teaching grammar found that unless grammar is taught in contextualised fashion it has no discernible link to raising levels of literacy:

      https://ore.exeter.ac.uk/repository/bitstream/handle/10036/15272/PrincipledUnderstanding.pdf?sequence=2&isAllowed=y

      KS2 tests, most definitely, do not offer a contextualised approach to grammar teaching.

      • You say Myhill’s research has been at secondary level. However Grammar for Writing has been tried in primary schools. The EEF evaluation concluded:
        ‘The overall effect size of the class-level intervention compared to the ‘business as usual’ control was small (+0.10, estimated two additional months’ progress) and statistically insignificant (i.e., may have occurred by chance). A larger effect was observed when children were taught in small groups (+0.24, estimated three additional months’ progress), although this is very similar to the difference in effect size between the small group intervention and the whole class approach. This suggests that the observed gains in writing outcomes are likely to be as a result of teaching pupils in small groups, rather than any intrinsic benefit of teaching Grammar for Writing. There is, therefore, little evidence that the intervention provided additional gains in writing outcomes.

        The study does demonstrate that small group teaching is an effective strategy to increase writing skills among pupils between Level 3c and 4b, which is reasonably consistent with the emerging evidence base for small group tuition.’

      • You’re confusing testing with teaching. The SPaG test isn’t a method of teaching, any more than a test of arithmetic is a method of teaching arithmetic. And you’re quite right: grammar has to be applied to writing in order to improve writing. No surprise there.

  5. Rosemary Davis says:

    How stultifying for 11 year olds to be asked such unimportant grammar questions in their SATs. Such questions do nothing for creativity, fluency or motivation to write. I passed but much of it was guesswork even though I read English

  6. fish64 says:

    The new grammar test is beneficial for children’s future learning of modern languages. One of the reasons so many children find MFL difficult is that their grasp of English grammar is shaky – no detailed knowledge of English tenses, for example. Children in France and Germany are better equipped for learning MFL, as they have a better grasp of the grammar of their own language. See my post here:

    https://fish64.wordpress.com/2016/04/27/spag-a-foreign-language-teachers-perspective/

  7. […] As a matter of urgency commission a review of English in the national curriculum, including investigating the damaging effects of grammar teaching as currently configured. […]

  8. […] As a matter of urgency, commission a review of English in the national curriculum, including investigating the damaging effects of the way grammar is currently taught. […]

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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

@IOE_London

Enter your email address

Want to keep up with IOE research?