CoChat
Sign Up
Cited by 148Writing and Handwriting EducationReading and Literacy DevelopmentEFL/ESL Teaching and Learning

Grammar for writing? An investigation of the effects of contextualised grammar teaching on students’ writing

Open access
Jones, Susan, Myhill, Debra, Bailey, Trevor·2013·Reading and Writing
DOI 10.1007/s11145-012-9416-1
Open this paper in CoChat — free accountOpen PDF

Abstract

The role of grammar instruction in the teaching of writing is contested in most Anglophone countries, with several robust meta-analyses finding no evidence of any beneficial effect. However, existing research is limited in that it only considers isolated grammar instruction and offers no theorisation of an instructional relationship between grammar and writing. This study, drawing on a theorised understanding of grammar as a meaning-making resource for writing development, set out to investigate the impact of contextualised grammar instruction on students’ writing performance. The study adopted a mixed-methods approach, with a randomised controlled trial and a complementary qualitative study. The statistical analyses indicate a positive effect on writing performance for the intervention group (e = 0.21; p < 0.001); but the study also indicates that the intervention impact differentially on different sub-groups, benefiting able writers more than weaker writers. The study is significant in being the first to supply rigorous, theorised evidence for the potential benefits of teaching grammar to support development in writing.

Integrity signals

Registry

Crossref · 10.1007/s11145-012-9416-1

Cites

74 of 74 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 9 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    The analysis reported in Table 3 shows a significant positive effect on writing performance for students in the intervention group whose baseline data indicated that they were above average in writing attainment.

    “The analysis reported in Table 3 shows a significant positive effect on writing performance for students in the intervention group whose baseline data indicated that they were above average in writing attainment.”
    Quote in the paperSubstantial claim
  • Findingown data

    Taken overall this model is very highly significant (F 7,36 statistic = 10.42, p-value = .00001).

    “but low overall explanatory value is not atypical in educational studies and taken overall this model is very highly significant (F 7,36 statistic = 10.42,p-value = .00001).”
    Quote in the paperNot a substantial claim
  • Findingown data

    The least experienced teachers often followed the teaching materials more rigidly and inflexibly.

    “Conversely, the least experienced teachers often followed the teaching materials more rigidly and inflexibly.”
    Quote in the paperNot a substantial claim
  • Findingown data

    A student with above average writing ability in a low-performing school with a teacher with low subject knowledge and more than ten years or less than five years' experience is less likely to benefit from the intervention.

    “For example, a student with above average writing ability in a low-performing school with a teacher with low subject knowledge and more than ten years or less than five years' experience is less likely to benefit from the intervention”
    Quote in the paperNot a substantial claim
  • Findingown data

    Students in classes with teachers with lower GSK made less improvement than those with teachers with higher GSK.

    “Students in classes with teachers with lower GSK made less improvement than those with teachers with higher GSK.”
    Quote in the paperSubstantial claim
  • May combine two claims. 25
  • Classified as probably not a substantial claim. 22
  • The quote only partly states this claim. 1

  • May combine two claims.

    Findingown data

    The school inspection outcomes and the percentage of students with Special Educational Needs (SEN) were significant factors.

    “The analysis indicates that the school inspection outcomes and the percentage of students with Special Educational Needs (SEN) were significant factors.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Lesson observations repeatedly show teachers struggling to cope with student questions on grammar and sometimes communicating incorrect information to students.

    “Indeed, the lesson observations repeatedly show teachers struggling to cope with student questions on grammar and sometimes communicating incorrect information to students.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The length of the teacher's experience and the quality of the teacher's subject knowledge of grammar were both significant in influencing student writing outcomes, regardless of whether students were in the comparison or intervention group.

    “The length of the teacher's experience and the quality of the teacher's subject knowledge of grammar were both significant in influencing student writing outcomes, regardless of whether students were in the comparison or the intervention group.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    More able writers in the intervention group improved significantly more than less able writers who received no significant benefit from the intervention.

    “Overall then, more able writers in the intervention group improved significantly more than less able writers who received no significant benefit from the intervention”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The mean value of the outcome variable over all 744 participants was 9.24%, indicating an overall improvement between pre- and post-writing tests in both the comparison and intervention groups.

    “The mean value of this variable over all 744 participants was 9.24%, indicating an overall improvement between pre-and post-writing tests in both the comparison and intervention groups.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Analysis of the raw scores by sub-components of the mark schemes indicates that the greatest improvement was made in Composition and Effect, though similar improvements were made at sentence and text level.

    “Analysis of the raw scores provided by the markers, broken down into the sub-components of the mark schemes, indicates that the greatest improvement was made in Composition and Effect, though similar improvements were made at sentence and text level.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The use of multiple regression modelling reveals that the intervention was differentially experienced by learners, with more able writers benefiting strongly whilst weaker writers benefited less.

    “However, the use of multiple regression modelling reveals that the intervention was differentially experienced by learners, with more able writers benefiting strongly whilst weaker writers benefited less.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In the comparison group (n = 332) the mean outcome was 6.41%, lower than in the intervention group (n = 412) where mean outcome was 11.52%.

    “In the comparison group (n = 332) the mean outcome was 6.41%, lower than in the intervention group (n = 412) where mean outcome was 11.52%.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The intervention benefitted some writers more than others, and prior writing attainment was a significant factor in this.

    “The data analysis indicates that the intervention benefitted some writers more than others, and that prior writing attainment was a significant factor in this.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    All the non-intercept terms in the final model are very highly statistically significant (at a 0.1% level) with the exception of the teacher GSK score which nevertheless is still significant at the 5% level.

    “All the non-intercept terms in this model are very highly statistically significant (at a 0.1% level) with the exception of the teacher GSK score which nevertheless is still significant at the 5% level.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    A cross-over design was used to minimise any possible test effect.

    “To minimise any possible test effect, a cross-over design was used: half the sample took Writing Task 1 as the pretest, and the other half took Writing Test 2; this was reversed for the post-test.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    The sample was first stratified at teacher level according to their GSK scores to ensure that the two groups were matched.

    “The sample was first stratified at teacher level according to their GSK scores to ensure that the two groups were matched, given that GSK is known to be a factor in the teaching of grammar.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Teachers' grammatical subject knowledge is a factor in the teaching of grammar.

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductown data

    Both the intervention and comparison group were taught the three types of writing over the same period, addressing the same curriculum teaching objectives, and producing the same written outcomes.

    “Both the intervention and comparison group were taught the three types of writing over the same period, addressing the same curriculum teaching objectives, and producing the same written outcomes (in other words columns 1, 2, and 4 of Table 1 were the same for both groups).”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Study design and conductcitation

    A complementary qualitative data set was collated alongside the experimental study to provide in-depth understanding of the theoretical, pedagogical and contextual implications of the statistical data.

    “a complementary qualitative data set was collated alongside the experimental study to provide in-depth understanding of the theoretical, pedagogical and contextual implications of the statistical data.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductnone given

    Both groups were given the same set of stimulus materials and resources, but only the intervention group had detailed teaching units, planned at lesson level, in which grammar was explicitly taught.

    “Both groups were given the same set of stimulus materials and resources, but only the intervention group had detailed teaching units, planned at lesson level, in which grammar was explicitly taught.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductreasoning

    The topic was selected to avoid known gender preferences in writing and to avoid the need of any specific topic knowledge or experience.

    “In order to avoid any possible bias created by the precise choice of writing task, the topic was selected to avoid known gender preferences in writing and to avoid the need of any specific topic knowledge or experience.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductown data

    Both tests were independently marked by Cambridge Assessment, an independent organisation.

    “Both tests were independently marked by Cambridge Assessment, an independent organisation who were responsible for setting and marking the national test of writing for fourteen year olds until 2006.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S4 (based).

    Premisenone given

    There have been no systematic studies of whether making meaningful connections between particular linguistic structures and particular writing tasks supports the development of students' writing.

    “However, there have been no systematic studies of whether making meaningful connections between particular linguistic structures and particular writing tasks supports the development of students' writing.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    The study was blind as participant teachers were not told the research focus was grammar; instead they were told the focus was on the teaching of writing.

    “The study was blind as participant teachers were not told the research focus was grammar; instead they were told the focus was on the teaching of writing (see also Ethical Considerations 2.7).”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductnone given

    The classes were randomly assigned, using a random number generator, to either a comparison or an intervention group.

    “The classes were then randomly assigned, using a random number generator, to either a comparison or an intervention group.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisereasoning

    Existing research on grammar instruction is limited because it only considers isolated grammar instruction and offers no theorisation of an instructional relationship between grammar and writing.

    “However, existing research is limited in that it only considers isolated grammar instruction and offers no theorisation of an instructional relationship between grammar and writing.”
    • Also stated in: § 1.1 The effectiveness of grammar teaching · plain
      “But none directly address the interrelationship of grammar and writing, or offer a theoretical account of such an inter-relationship.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductcitation

    The school sample was secured by using local authority data to compile a numbered list of all mixed comprehensive schools in the South-West and the Midlands, with a random number generator used to determine a rank order, to avoid selection bias.

    “In order to avoid selection bias (Cook and Wei 2002), the school sample was secured by using local authority data to compile a numbered list of all mixed comprehensive schools in the South-West and the Midlands.A random number generator was used to determine a rank order, and each school was approached in rank order until the desired sample of 32 was reached.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Cook and Wei 2002 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    Final model selection was checked using full residual analyses and standard diagnostic procedures.

    “Final model selection was then checked using full residual analyses and standard diagnostic procedures.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Stark, Kopp and Fischer argue that extraneous load is attributed to sub-optimal instructional design and interferes with productive learning.

    “Similarly, Stark, Kopp and Fischer (2011) argue that extraneous load is”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Stark, Kopp and Fischer (2011) source

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

    Interpretationreasoning

    It might be the case that the level of conceptual thinking required to understand grammatical concepts and transfer that learning into writing was too high a cognitive challenge or created too high a cognitive load for less able writers.

    “It might also be the case that the level of conceptual thinking required to understand grammatical concepts and transfer that learning into their writing was too high a cognitive challenge or created too high a cognitive load for this group.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Study design and conductown data

    The qualitative study indicates that all bar one teacher used the teaching units throughout the research period, and there was a high level of commitment to the project.

    “The qualitative study, particularly the lesson observations and teacher interviews, indicates that all bar one teacher used the teaching units throughout the research period, and there was a high level of commitment to the project.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Interpretationreasoning

    The study is the first to supply rigorous, theorised evidence for the potential benefits of teaching grammar to support development in writing.

    “The study is significant in being the first to supply rigorous, theorised evidence for the potential benefits of teaching grammar to support development in writing.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Interpretationown data

    The results and predictive scenarios indicate the complexity of interventions in educational settings, which defy simplistic causal explanations.

    “As anticipated in the research design, the results and the predictive scenarios indicate the complexity of interventions in educational settings, which defy simplistic causal explanations.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisereasoning

    High inspection grades do not correlate with 'high ability' students in this study, because all the schools were comprehensive and inspection assessments take into account baseline performance on entry.

    “It is important to note that high inspection grades do not correlate with 'high ability' students: all the schools in the study were comprehensive schools, which do not select students on ability, and the inspection assessments take into account the baseline performance of students on entry to the school.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductnone given

    The markers did not know from which treatment group the writing was derived, and the research team was not involved in the process of marking and assessment at all.

    “The markers did not know from which treatment group the writing was derived, and the research team was not involved in the process of marking and assessment at all.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationreasoning

    It is possible that more able writers have clearer communicative and rhetorical intentions for their writing than less able writers, enabling them to make more appropriate use of their grammatical understanding to shape text appropriately.

    “It is also possible that more able writers have clearer communicative and rhetorical intentions for their writing than less able writers, enabling them to make more appropriate use of their grammatical understanding to shape text appropriately.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Cognitive load theory acknowledges constraints on working memory capacity when tackling cognitively complex tasks such as writing, and argues that teaching materials need to be designed to manage cognitive load appropriately.

    “Cognitive load theory (Sweller 1988) acknowledges the constraints that can be placed upon working memory capacity when tackling cognitively complex tasks, such as writing, and argues that teaching materials need to be designed to manage cognitive load appropriately.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Sweller 1988 source

  • May combine two claims.

    Interpretationown data

    The study demonstrates that more able students' learning about writing is enhanced by explicit understanding of how grammar choices can be used to shape written text to satisfy writers' rhetorical goals, but it also indicated no parallel benefit for less able writers.

    “The study demonstrates that more able students' learning about writing is enhanced by explicit understanding of how grammar choices can be used to shape written text to satisfy writers' rhetorical goals, but importantly it also indicated no parallel benefit for less able writers.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    One drawback of the RCT was the requirement for standard teaching materials to be used across all groups, making it difficult to tailor the teaching materials to particular classes' needs or interest.

    “One drawback of the RCT was the requirement for standard teaching materials to be used across all groups, making it difficult to tailor the teaching materials to particular classes needs or interest.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationreasoning

    The lack of impact of the intervention upon less able writers may be because the grammatical features addressed in the teaching materials were poorly matched to their learning needs as writers.

    “The lack of impact of the intervention upon less able writers may be because the grammatical features addressed in the teaching materials were poorly matched to their learning needs as writers.”
    Quote in the paperNot a substantial claim

  • May combine two claims. The quote only partly states this claim.

    Interpretationown data

    The analysis indicates that both hypotheses are confirmed: contextualised grammar teaching is positively related to students' writing performance, and the quality of teachers' grammatical subject knowledge mediates the impact of contextualised grammar instruction.

    “The analysis indicates that both hypotheses are confirmed.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductown data

    The final sample for data analysis was 31 teachers and their classes, and the final student sample was 744.

    “Thus the final sample for data analysis was 31 teachers and their classes.The initial student sample was 855: over the year of the study, due to students leaving schools or being absent for the post-test, and due to the loss of one whole class, as outlined above, the final student sample was 744.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    The estimated effect for above-average baseline writers in the intervention group is strong in an educational intervention context and provides robust evidence for the beneficial impact on this group of receiving contextualised grammar teaching.

    “The estimated effect for this group is strong in an educational intervention context (Cohen 1969, 23) and provides robust evidence for the beneficial impact on this group of students of receiving the contextualised grammar teaching.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Cohen 1969, 23

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

38 of 38

  • The paper relies on this without support.

    Teachers' grammatical subject knowledge is a factor in the teaching of grammar.

    “given that GSK is known to be a factor in the teaching of grammar”

  • The paper relies on this without support.

    The comparison group did not receive contextualised grammar teaching of the kind under investigation.

    “Because it is not possible in a naturalistic educational setting to prevent any teaching of grammar naturally occurring, we have consistently used the term 'comparison' rather than 'control' group.”
2 of 2

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

  • A scenario estimate falls outside its stated confidence interval

    Minor

    The stated interval from −10.7 to −1.3 is centred on −6.0 and does not contain the reported estimate of −0.6. The point estimate appears to have lost a digit.

    Reported -0.6 · recomputed -6

1 of 1

Cites

Flagged first, then by citations

1 of the 75 references couldn’t be matched to a known paper.

Research works printed but not linked: 9

  1. British Educational Research Association (BERA). (2004). Revised ethical guidelines for educational research: 2004. Nottingham, UK: BERA.
  2. Cajkler, W., & Hislam, J. (2004). Teaching and understanding grammar: How trainee teachers make sense of a new curriculum? In N. Bartels (Ed.), Applied linguistics in language teacher education (pp. 295–312). Dordrecht: Kluwer.
  3. Department for Education and Skills (DfES). (2001). Framework for teaching English: Years 7, 8 and 9. London: DfES.
  4. Laurinen, I. (1955). The development of sentence sense in the light of the results attained in the teaching of writing in Finnish primary schools. Unpublished PhD thesis.
  5. Locke, T. (2005). Writing positions and rhetorical spaces. In B. Doecke & G. Parr (Eds.), Writing = learning (pp. 75–95). South Australia: AATE, Wakefield Press.
  6. Myhill, D. A. (2010b). Harnessing grammar: weaving words and shaping texts. Better: Evidence-Based Education, Winter 2010, pp. 12–13.
  7. National Commission on Writing in America’s Schools and Colleges (NCW). (2003). The neglected ‘R’: The need for a writing revolution. New York: College Board.
  8. OFSTED. (2009). English at the crossroads. London: OFSTED.
  9. Qualifications and Curriculum Authority (QCA). (1998). The grammar papers. London: QCA.

Cited by

Sorted by citations

Related

Authors
Jones, S., Myhill, D., & Bailey, T.
Affiliations
  • University of Exeter (GB)
  • University of Wollongong (AU)
Corresponding author
Myhill, Debra
ORCID
Venue
Reading and Writing
Publisher
Springer Science and Business Media LLC
Published
2012-09-14
Volume
26
Issue
8
Pages
1241-1263
ISSN
  • 0922-4777 (ISSN-L)
  • 1573-0905
DOI
10.1007/s11145-012-9416-1
OpenAlex
W1971469592
Cited by
148
References
75
Citations by year
2026: 11 · 2025: 14 · 2024: 12 · 2023: 16 · 2022: 14 · 2021: 15 · 2020: 15 · 2019: 14 · 2018: 5 · 2017: 5 · 2016: 7 · 2015: 11 · 2014: 3 · 2013: 5 · 2012: 1
FWCI
31.13
Type
journal article
Language
en
Open access
Licence
www.springer.com/tdm (tdm)
Locations
  • Reading and Writing (publishedVersion)
  • Open Research Exeter (University of Exeter) (submittedVersion)
  • Research Online (University of Wollongong) (submittedVersion)
  • Figshare (submittedVersion)
Funders
Economic and Social Research Council (ES/J00037X/1, ES/F015313/1)
Topics
  • Social Sciences › Education › Writing and Handwriting Education
  • Social Sciences › Psychology › Developmental and Educational Psychology › Reading and Literacy Development
  • Social Sciences › Arts and Humanities › Language and Linguistics › EFL/ESL Teaching and Learning
Keywords
grammar instruction, writing instruction, writing performance, meaning-making, writing development, English as a second language

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-11.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-11.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review