Does teaching consonant clusters systematically aid decoding?

Consonant clusters (henceforth referred to as clusters) exist in English morphology and impact on how sounds blend together and how words may be segmented into syllables (Grainger et al., 2012). Their usage links to how vowels (single vowel graphemes and digraphs) operate in words, and this knowledge then has a bearing on how children learn to read and spell words throughout their education. Educational research into clusters is limited and much of the work that has been done is based on linguistic studies (Groff, 1972); (Gregová, 2010)), which have not always translated into phonics in educational practice.

This article aims to provide some experimental evidence on refining Groff’s (Groff, 1972) sequence of clusters so that systematic phonics teaching is further developed. Groff (Groff, 1972) proposed that there was a need to reconsider the order of teaching clusters, and proffered a sequence that needed to be further refined through research. Groff (Groff, 1972) suggested that clusters could be ranked by their reading or spelling simplicity. Yet some of Groff’s clusters were not clusters but digraphs (‘sh’, ‘ch’), and there were missing clusters (‘sc’, ‘sm’, ‘sn’, ‘scr’, ‘squ’, ‘spr’, ‘spl’ and ‘shr’). The aim of this research was both to develop Groff’s (Groff, 1972) cluster sequence and to provide research into an area of phonological development that has an impact on the emerging morphological awareness of our Year 1 pupils.

Research design

In our school we use the ‘Letters and Sounds’ systematic synthetic phonics scheme (DFES, 2007). After phase 5, we cover clusters, and wanted to introduce more clarity and a systematic way of teaching them. Our research design centred on the research question ‘Does teaching consonant clusters systematically aid decoding?’ To explore this, we designed a five-week explorative ‘Lesson Study’ to investigate how our proposed order of clusters was supporting our Year 1 pupils’ learning.

We used extended preparation Lesson Study (Mynott, 2017), spending the autumn term 2017 observing phonics, planning lessons, devising lesson sequences and discussing phonics research. In December 2017, we met and planned our approach to teaching clusters systematically.

Using our cluster sets – Figure 1 – we planned Lesson Study alongside pupil assessments to evaluate the effectiveness of our systematic cluster teaching. Each of our planned Lesson Study lessons focused on pupils’ learning of clusters: their recall and usage of clusters in decoding words, including multi-syllabic words. After each lesson, we transcribed and analysed our discussions by thematically codingIn qualitative research, coding involves breaking down data into component parts, which are given names. In quantitative research, codes are numbers that are assigned to data that are not inherently numerical (e.g. in a questionnaire the answer 'strongly agree' is assigned a 5) so that information can be statistically processed. them. This article will focus on the recall and decoding elements we observed as a means to support our exploration of our research question.

Ethical consent was sought in line with the school’s ethics policy. All pupils consented, alongside their parents, for their data to be used anonymously, and all teachers involved in this research are named authors and are fully consenting participants.

Case

The study was undertaken in a Year 1 class of 30 pupils. Twenty-seven pupils in this class were English as an additional language (EAL) learners. All pupils chosen as case pupils were pupils performing lower than age-related expectations during the autumn teacher assessments.

We conducted initial assessments in December 2017, with pupils reading a text that included a range of graphemes, digraphs and clusters with multi- and mono-syllabic words, enabling pupils to demonstrate all of their decoding skills. The case pupils were chosen as they did not demonstrate knowledge of clusters and it would be possible to see whether the systematic teaching of clusters supported their recognition and decoding. Pupils in the initial assessment made cluster errors and other decoding errors, which are recorded alongside their reading time in Figure 2.

For three consecutive weeks, we held research lessons on teaching clusters, with a slightly different focus for each lesson. Lesson One focused on whether pupils were recalling and using taught clusters to decode words. Lesson Two focused on whether pupils were recalling clusters and starting to discern where and how clusters worked in relation to root words, but also whether pupils could do this while discerning between digraphs and clusters. Lesson Three continued the focus from week two, while exploring whether pupils could see root words within multi-syllabic words using their emerging understanding of clusters.

Assessment data was reviewed in February 2018 following the five-week teaching sequence.

Findings

There was an improvement in both pupils’ reading times and accuracy, as indicated in Figure 2. Of course, the size of the sample was small and there was no control group, so we cannot know whether these improvements came about through the general teaching and learning in the Year 1 class, or through this particular intervention.

However, assessment observation notes suggest that, while clusters had been a problematic aspect of the initial assessment, clusters were now no longer impeding pupils’ reading speed or accuracy; and that this improvement may have been due to improved ‘chunking’ of clusters – using them as a single sound – which enabled faster decoding combined with greater morphological awareness, particularly around syllable segmentation.

Systematic cluster teaching improves phonology

The development of pupils’ understanding of clusters, particularly beginning clusters, rapidly improved throughout the teaching sequence. Lesson One observations indicated that only six pupils were confident at recognising and decoding clusters within words. By Lesson Two, more pupils were recognising clusters and some were recognising multiple clusters within words like ‘snowflake’. Lesson Three observed most pupils recognising and applying their knowledge when reading, discerning between clusters and digraphs like ‘ch’ and ‘th’ accurately and quickly. The case pupils’ data indicates that, by the end of the cluster sets, pupils were confident with clusters and no longer making errors in decoding clusters within words, including multi-syllabic words. Essentially, the cluster sets had built on their phonological knowledge and enabled them to access more complex phonological structures.

Focus on resources

While the phonology of pupils increased, the team found that clusters are often related to more complex phonological structures and morphological rules. As such, the research lessons indicated a need to develop resources that explicitly link to the cluster sets being taught, and which anticipate misconceptions within words prior to lesson sequences. This is similar to findings found in Lesson Study research on mathematics and the practice of ‘kyozaikenkyu’, the study of instructional aids (Doig et al., 2011).

Our Lesson

One observations found that six pupils correctly identified both clusters in ‘grapefruit’, despite ‘fr’ not being in the Set 1 clusters. ‘Grapefruit’ came to articulate the major learning aspect of this lesson, as it showed the necessity to fully consider not only the sequence of clusters but also the words we wished the pupils to explore. The ‘gr’ cluster in ‘Grapefruit’ was in Set 1 and the split a – e had previously been taught; however, the ‘ui’ is an unusual vowel digraph and, combined with ‘fr’ being in the second set of clusters, which we had not yet taught, it made ‘grapefruit’ a poor selection for our initial set. As a result, the team considered their multi-syllabic word choice, which led to the selection of ‘snowflake’ for Lesson Two, as both clusters and all vowel digraphs had previously been taught. In Lesson Two, more pupils were able to recall the taught clusters (Set 1 and Set 2) and some pupils were starting to categorise words into multiple lists if they contained more than one cluster; ‘snowflake’ was categorised into ‘sn’ and ‘fl’. The careful consideration of word choices remained prominent, as the team had wanted to see whether the pupils were able to distinguish clusters from digraphs within Lesson Three as a means of further assessing how secure pupils’ cluster knowledge was becoming.

Developing morphological awareness

Lesson Two started to explain the morphology of syllables to pupils. Although the teaching focused on identifying the clusters, the team felt that a wider exploration of root words was needed to further pupils’ understanding of segmentation around the root. While clusters often appear at the start of a new syllable, this is not universal, and recognising the root word is important to identify prefixes and suffixes with non-standard phonic pronunciations. Again, word choice formed part of the discussion, as the team wanted to avoid the misconception that clusters always start a new syllable. By Lesson Three, we observed pupils distinguishing between clusters and digraphs, and pupils using and applying this to decoding. The focus on root words supported pupils’ thinking about how words could be segmented. The team felt that to extend this further would mean developing their understanding of morphology so that they would be better equipped to plan for teaching pupils beyond phonics, as they progressed through the school.

Conclusion

Our small-scale intervention suggests that teaching clusters systematically may improve the phonological awareness, reading speed and accuracy of Year 1 pupils. However, this research is in its infancy and our sequence and resources need further development to ensure that all our resources, including words, are carefully chosen to support pupils’ phonological and morphological understanding. By using the sequence of clusters, we ensured that, as with other elements of systematic phonological development, pupils had awareness of a range of clusters, along with how they worked with the morphological arrangements of the words they were decoding.

We found that our exploration of clusters was a useful starting point to further develop pupils’ understanding of the morphology of words, particularly how syllables work. By exploring and developing the work on clusters that Groff (Groff, 1972) had proposed, we found that teaching clusters systematically did appear to aid decoding and improve reading speed and accuracy. Our work also helped us understand more about carefully selecting words in our systematic teaching (in line with Lesson Study research (Fujii, 2016); Doig et al., 2011)) so that we can raise and address misconceptions. Our Lesson Study work had an additional subject knowledge impact, showing the need to further develop our understanding of morphology to build on the work we are undertaking, linking aspects of phonology (like clusters) to syllable awareness and spelling.

Like Groff (1972), we end this article with a need for our work to be further explored, to refine the systematic teaching of clusters. Our research needs to be upscaled, to explore the use of control groups, and to be trialled in different educational contexts to analyse the impact of EAL and other contextual information on cluster acquisition. We also need to consider the sequence of end clusters and the implications of phonology and morphology beyond Year 1 to support the systematic teaching of spelling for all children.