Studies involving ‘sentence combining’ show that children get better at sentence combining-related activities and tasks. Whether they go on to write better compositions of their own depends on something the headline claims overlook: the conditions under which sentence-level knowledge is taught and used.
Sentence combining has had a remarkable second life. Teachers are told it is one of the best-evidenced writing practices available. They are told that practising the joining of short ‘kernel’ sentences into longer ones will lift the quality of the writing their children produce.
This article takes that claim at its word and goes back to the studies.
Three things that are often conflated
Studies and programmes described as ‘sentence combining’ often bundle together quite different things. It helps to separate them.
- Condition 1: Sentence-combining practice. Children manipulate sentences supplied to them, typically joining short kernel sentences into longer ones.
- Condition 2: Sentence-level craft instruction. Children learn how writers make sentence-level choices to influence the meaning, emphasis, rhythm and cohesion of their piece.
- Condition 3: Using sentence craft during real composition. Children make those choices in a piece of writing they are crafting, not as part of an exercise.
These are not the same. A programme can offer the first without the other two. The distinction between knowing about grammar and using it to make meaning has long been central to research on contextualised grammar teaching (Jones et al., 2013; Myhill et al., 2012). The rest of this article reads the sentence-combining evidence through this lens.
Where the claim comes from
Much of the recent enthusiasm for sentence combining traces back to the Writing Next report. Graham & Perin’s meta-analysis of writing instruction for students in Grades 4–12 reported an average effect of +0.50 on writing quality for sentence combining (Graham & Perin, 2007a, 2007b). Interestingly, the same analysis found a negative effect of −0.32 for traditional grammar (structural linguistics) instruction.
However, to judge the claim, we need to look beneath the headline figures at the individual studies.
The early studies
The first studies on sentence combining were carried out in the 1960s and 1970s. At the time, researchers were interested in whether teaching children to join sentences together would help them perform better on their resulting sentence-combining activities/tasks (Miller & Ney, 1967, 1968; Perron, 1974; Vitale et al., 1971; Stoddard & Renzulli, 1983). Most of these studies looked at changes in children’s sentence structure rather than the overall quality of their writing. It’s important to remember that writing long sentences doesn’t mean that your writing is better. A longer or more complicated sentence is not automatically a better sentence. Indeed, Stotsky’s (1975) early review already warned against assuming that more complex sentences means better writing.
The flagship primary study
Saddler & Graham (2005) is the study most often cited for primary-aged children. It is one of the five studies behind Writing Next’s sentence combining figure. Its abstract states that the children’s new sentence combining skills led to improved story writing and to the use of those skills when revising their manuscripts (p. 43). However, the results section tells a narrower story.
● No clear advantage in first drafts. The children wrote a 15 minute story before and after the sentence-combining programme, and then revised their writing in a second session. Overall, the sentence-combining group did not produce better-quality writing than ‘the grammar group’. The one difference the researchers found was that, after the programme, children who had practised sentence combining improved their writing when they were given time to revise their stories. The grammar group did not show the same improvement. In other words, sentence combining did not appear to make children’s first drafts better, but it may have helped them improve their writing when they went back and revised it.
● Very few combining revisions. However, after instruction, children actually made on average fewer than one actual sentence-combining revision. The authors concluded that other aspects of the programme must also have contributed to children’s improved writing quality – not sentence combining.
● Far from real writing. There were 44 fourth graders, taught in pairs outside of class by six teachers. They were asked to write a one-off story in 15 minutes in response to a picture prompt.
● A weak comparison. The comparison group was taught formal grammar, an approach that is particularly ineffective or even harmful to children’s writing quality. Despite this, the sentence-combining children’s first drafts were no better.
In fairness, the children were not prompted to use their taught sentence-combining skills at posttest, so the revision gains were genuinely spontaneous. The authors themselves described their quality prediction as only partially fulfilled and called for the need for further studies.
Read through the three conditions, this was mainly Condition 1, delivered as peer-assisted practice, with some opportunities for pairs to apply the skills when writing and revising a text.
The UK study built to test transfer found no reliable effect
Walter and colleagues’ (2021) study is one that most directly tests the claim about future writing. It was a UK randomised study of 71 struggling writers aged 7–10, drawn from the lowest-attaining fifth of their year groups. Children were matched in threes and randomly assigned to either an intensive sentence combining intervention, a morphological spelling intervention or a ‘waiting list’ group.
Crucially, the main outcome was a short piece of independent narrative writing. It was used specifically to judge whether the intervention generalised to children’s writing. Children were assessed at baseline, immediately after the eight-week out-of-class intervention (two 25–30 minute small-group sessions a week), and again three months later.
- Sentence-combining students got better at sentence-combining tasks. The sentence combining group did better on a sentence combining test than the other groups immediately after the intervention. They continued to perform better on the test than their ‘waiting list’ peers three months later.
- Children’s compositions did not improve. There was no significant effect on the quality of children’s compositions, and no differences in productivity or accuracy.
Reliable transfer of sentence-combining to children’s own writing was not found. Any quality differences were considered to be not statistically significant. Here, children who received sentence-combining instruction got better at sentence-combining tests. This was Condition 1, delivered out of class in about 400 minutes of teaching.
The strongest positive result: a replication in Turkey
A replication of Saddler and Graham’s (2005) programme was conducted in Turkey (Tavsanli et al., 2023). Six teachers of Grades 2 to 4 were randomly assigned, one per grade to each condition: 88 children received the programme and 83 continued with their usual writing lessons. Over 10 weeks, the class teachers delivered 30 hours of instruction, two weeks on sentence types followed by eight weeks of sentence combining. Each week, children practised combining, then worked in pairs to revise a paragraph, then planned, wrote and revised a short story using the structures they had learned. They were also asked to use the taught structures at other times of the day.
Immediately afterward the intervention, the children’s opinion essays outscored the comparison group on sentence fluency, length and overall quality. However, two features limit what this study can tell us.
- A different writing programme, not just extra sentence-combining instruction. The comparison classes followed a product-based textbook, and the researchers observed that their teachers rarely encouraged children to plan, revise or edit their writing. In contrast, the sentence-combining children were specifically asked to plan, draft and revise their story with a partner every single week. The researchers suspect that some of the gain in quality may well have come from that, rather than from the sentence-combining activities.
- No follow-up. The essays were written immediately after the programme ended, so we don’t know whether the gains lasted.
The meta-analytic headline and the primary evidence base
The +0.50 figure in Writing Next rests on just five studies, with individual effects actually ranging from +0.21– +0.66 (Graham & Perin, 2007a). Only two of the studies involved primary-aged pupils. One was Saddler & Graham (2005), discussed above. The other was a study of ‘high-ability’ fifth and sixth graders compared against an unspecified control condition. The remaining three studies were undertaken by pupils who were a lot older.
For primary-aged children, then, the research base is incredibly thin. When Graham and colleagues conducted a meta-analysis of writing instruction specific to the elementary grades, they (rather sensibly) required at least four studies. Sentence combining did not meet that threshold (Graham et al., 2012).
A Dutch study compared three ways of teaching nine-year-olds (van de Gein, 1991). Pupils were randomly assigned to a traditional grammar course, a sentence-construction course or a ‘direct writing’ course, in which they drafted and revised their own texts with teacher feedback. Compared with direct writing, the sentence-construction course had almost no effect on the quality of children’s writing. Drawing on studies like this, Koster and colleagues concluded that ‘when grammar is taught in isolation, and not in a “real” writing context, it may not be clear to students how to apply what they learned when writing a text’ (Kostger et al., 2015, p. 268).
When sentence knowledge is applied while composing
The study that comes closest to the later conditions is Limpo & Alves (2013). Portuguese fifth and sixth graders were taught sentence-level craft moves through Self-Regulated Strategy Development (SRSD) instruction. After five introductory lessons of combining exercises, they were taught to use it while writing their own opinion essays, through teacher modelling and independent writing. The strategy was framed around meaning-making. Compared with a control group receiving standard instruction, the children’s sentence construction improved and their essays were of higher quality and longer. Importantly, most of the sentence-level gains in the children’s essays only emerged after receiving lessons on using the strategy while writing, not as part of the exercise phase. This study is the clearest evidence that Conditions 2 and 3 matter.
A recent trial with much younger children points the same way. In SRSD Plus, Grade 1 and 2 teachers in the United States taught sentence structures through combining activities and then guided children to use them in the information texts they were writing (Kim et al., 2025). Compared with usual instruction, children’s sentence proficiency improved as did the quality of their information texts. There was also some transfer to their later opinion writing.
Sentence-level knowledge appears more likely to become usable when it is taught as a strategy and applied while children compose their own texts.
Sentence-level teaching within class writing projects
The studies share a pattern. Where sentence-level instruction showed an effect on the quality of children’s writing, children were using it in their ‘own writing’ (Kim et al., 2025; Limpo & Alves, 2013; Saddler & Graham, 2005; Tavsanli et al., 2023).
The principles of SRSD offer a well-evidenced structure for doing just that. Applied at the sentence level, SRSD-style teaching would look like this.
- The teacher introduces a sentence ‘craft move’, such as subordination, and discusses why they, as a writer, like to use it.
- The teacher then models how they are going to use it in their writing that day, sharing their thinking so the choice is shown as a decision about making meaning and the effect it will have on their readers.
- Children are then asked to do the same thing for themselves that day. They use the craft move in the piece they are currently working on.
- The teacher evaluates children’s use of the craft move during writing time and offers feedback and additional responsive instruction.
The sentence-level craft move is taught for a purpose, at the moment of need, and applied straight away to writing children care about. This addresses all three conditions. Contextualised grammar teaching offers some support for this kind of approach too (Jones et al., 2013; Myhill et al., 2012; Young & Ferguson, 2024).
Where to start
For teachers who want to move away from sentence-combining exercises, or who want to start teaching at the sentence level, two Writing For Pleasure Centre resources are designed around all three conditions.
Sentence-Level Instruction: Lessons That Help Children Find Their Style and Voice for 3–11 Year Olds (Young & Ferguson, 2023) organises its lessons by what children are trying to achieve in their writing: focused, balanced and developed sentences. Each follows a ‘teach, then invite’ routine. The teacher names a sentence craft move, shows it in their own writing or in a mentor text, and invites children to use it in the piece they are writing that day, monitoring its use through pupil-conferencing.
Sentence-Building Mini-Projects (Ferguson & Young, 2025) is for children who need more sustained sentence-level work. Instead of combining ready-made kernel sentences, children make short little books of their own and revise the sentences they have composed within it. Drawings help make the effect of each sentence choice visible to the children. The projects are designed to sit alongside and support whole-class writing projects.
In summary
Sentence combining reliably improves children’s ability to perform well on sentence-combining activities. Evidence that it improves the quality of children’s writing is modest and inconsistent.
Perhaps a fair objection is that sentence-level teaching should be judged by sentence-level activities, not by the overall quality of children’s texts. On those terms, sentence combining clearly works. But the claim teachers are often given is that it will improve the quality of their children’s writing. A sentence skill that does not show up in children’s own texts has not yet done the job it is being promoted for. ‘Does sentence combining work?’ is therefore the wrong question. The studies suggest that what matters is the conditions under which sentence-level knowledge is taught and used.
Practice with sentence-combining produces gains on sentence-combining tests. Gains in composition have appeared mainly where sentence craft was taught as a strategy and applied while children were actually composing their own texts (Kim et al., 2025; Limpo & Alves, 2013; Tavsanli et al., 2023).
What the more successful studies have in common is that sentence-level craft was taught explicitly, modelled during real composing, and applied at once in texts children were working on.
References and further reading
- Andrews, R., Torgerson, C., Beverton, S., Freeman, A., Locke, T., Low, G., Robinson, A., & Zhu, D. (2004). The effect of grammar teaching (sentence combining) in English on 5 to 16 year olds’ accuracy and quality in written composition. Research Evidence in Education Library. EPPI-Centre, Social Science Research Unit, Institute of Education, University of London.
- Andrews, R., Torgerson, C., Beverton, S., Freeman, A., Locke, T., Low, G., Robinson, A., & Zhu, D. (2006). The effect of grammar teaching on writing development. British Educational Research Journal, 32(1), 39–55. https://doi.org/10.1080/01411920500401997
- Berninger, V. W., Nagy, W., & Beers, S. (2011). Child writers’ construction and reconstruction of single sentences and construction of multi-sentence texts: Contributions of syntax and transcription to translation. Reading and Writing, 24(2), 151–182.
- Ferguson, F., & Young, R. (2025). Sentence-building mini-projects. The Writing For Pleasure Centre.
- Graham, S., McKeown, D., Kiuhara, S., & Harris, K. R. (2012). A meta-analysis of writing instruction for students in the elementary grades. Journal of Educational Psychology, 104(4), 879–896. https://doi.org/10.1037/a0029185
- Graham, S., & Perin, D. (2007a). A meta-analysis of writing instruction for adolescent students. Journal of Educational Psychology, 99(3), 445–476. https://doi.org/10.1037/0022-0663.99.3.445
- Graham, S., & Perin, D. (2007b). Writing next: Effective strategies to improve writing of adolescents in middle and high schools [A report to Carnegie Corporation of New York]. Alliance for Excellent Education.
- Hillocks, G., Jr. (1986). Research on written composition: New directions for teaching. National Conference on Research in English & ERIC Clearinghouse on Reading and Communication Skills.
- Jones, S., Myhill, D., & Bailey, T. (2013). Grammar for writing? An investigation of the effects of contextualised grammar teaching on students’ writing. Reading and Writing, 26(8), 1241–1263. https://doi.org/10.1007/s11145-012-9416-1
- Kim, Y.-S. G., Harris, K. R., Goldstone, R., Camping, A., & Graham, S. (2025). The science of teaching reading is incomplete without the science of writing: A randomized control trial of integrated teaching of reading and writing. Scientific Studies of Reading, 29(1), 32–54. https://doi.org/10.1080/10888438.2024.2380272
- Koster, M., Tribushinina, E., de Jong, P. F., & van den Bergh, H. (2015). Teaching children to write: A meta-analysis of writing intervention research. Journal of Writing Research, 7(2), 249–274.
- Limpo, T., & Alves, R. A. (2013). Teaching planning or sentence-combining strategies: Effective SRSD interventions at different levels of written composition. Contemporary Educational Psychology, 38(4), 328–341. https://doi.org/10.1016/j.cedpsych.2013.07.004
- Mellon, J. C. (1979). Issues in the theory and practice of sentence combining: A twenty-year perspective. In D. A. Daiker, A. Kerek, & M. Morenberg (Eds.), Sentence combining and the teaching of writing (pp. 1–38). University of Akron & University of Central Arkansas.
- Miller, B. D., & Ney, J. W. (1967). Oral drills and writing improvement in the fourth grade. Journal of Experimental Education, 36(1), 93–99.
- Miller, B. D., & Ney, J. W. (1968). The effect of systematic oral exercises on the writing of fourth-grade students. Research in the Teaching of English, 2(1), 44–61.
- Myhill, D. A., Jones, S. M., Lines, H., & Watson, A. (2012). Re-thinking grammar: The impact of embedded grammar teaching on students’ writing and students’ metalinguistic understanding. Research Papers in Education, 27(2), 139–166. https://doi.org/10.1080/02671522.2011.637640
- Perron, J. D. (1974). An exploratory approach to extending the syntactic development of fourth-grade students through the use of sentence-combining methods
- Saddler, B., & Graham, S. (2005). The effects of peer-assisted sentence-combining instruction on the writing performance of more and less skilled young writers. Journal of Educational Psychology, 97(1), 43–54. https://doi.org/10.1037/0022-0663.97.1.43
- Saddler, B., Asaro, K., & Behforooz, B. (2008). The effects of peer-assisted sentence-combining practice on four young writers with learning disabilities. Learning Disabilities: A Contemporary Journal, 6(1), 17–31.
- Saddler, B., Behforooz, B., & Asaro, K. (2008). The effects of sentence-combining instruction on the writing of fourth-grade students with writing difficulties. Journal of Special Education, 42(2), 79–90. https://doi.org/10.1177/0022466907310371
- Stoddard, E. P., & Renzulli, J. S. (1983). Improving the writing skills of talent pool students. Gifted Child Quarterly, 27(1), 21–27.
- Stotsky, S. L. (1975). Sentence-combining as a curricular activity: Its effect on written language development and reading comprehension. Research in the Teaching of English, 9(1), 30–71.
- Tavsanli, O. F., Graham, S., Kaldirim, A., & Collins, A. (2023). The effect of sentence combining instruction with second- to fourth-grade children: A replication study in Turkey. Educational Psychology Review, 35, Article 93. https://doi.org/10.1007/s10648-023-09806-5
- Vitale, M. R., King, F. J., Shontz, D. W., & Huntley, G. M. (1971). Effect of sentence-combining exercises upon several restricted written composition tasks. Journal of Educational Psychology, 62(6), 521–525.
- Walter, K., Dockrell, J., & Connelly, V. (2021). A sentence-combining intervention for struggling writers: Response to intervention. Reading and Writing, 34(7), 1825–1850. https://doi.org/10.1007/s11145-021-10135-8
- Young, R., & Ferguson, F. (2023). Sentence-level instruction: Lessons that help children find their style and voice for 3–11 year olds (2nd ed.). The Writing For Pleasure Centre.
- Young, R., & Ferguson, F. (2024). Grammar mini-lessons for 3-11 year olds (2nd ed.). The Writing For Pleasure Centre.