Deep Dives

The Issue

As students progress to decoding multisyllabic words, they require a deeper understanding of English orthography— particularly how pronunciation, spelling, and structure interact in complex words. While English graphemes have never perfectly mapped into one-to-one sound correspondences (take, for example, all of the sounds the grapheme <ea> can represent), attempting to apply learned phonics rules becomes especially dicey when encountering multisyllabic words. There have long been varying instructional pedagogies for how to best provide instruction in decoding multisyllabic words, with a popular method incorporating teaching students to divide words using a set of syllable division patterns and subsequently labeling the syllable types before attempting to read the word. Proponents of this method (which I will henceforth refer to as ‘rule-based’ instruction) claim that this instruction provides students with the structure and rules they need to find success in decoding, and is a component of explicit strategy instruction that is recommended by the National Reading Panel (2000).

It could be said that the rigidity of teaching syllable types and division principles conflict with the structural properties of English. In writing, English is a morphophonemic, deep orthography, in which spelling reflects both sound and meaning. When spoken, English is a time-stressed language, where rhythm is based on stress patterns rather than syllables. The rigidity of teaching syllable types and division patterns subsequently fails to provide students with the flexibility needed to arrive at accurate, recognizable pronunciations of the word (and thus the word’s meaning). 

Opponents to rule-based multisyllabic instruction argue that syllable types and syllable division rules have too many exceptions to be considered a valuable use of instructional time, and students should instead be taught linguistic flexibility in sounding out multisyllabic words, and to recognize the morphologic structure that shapes many multisyllabic words’ spelling, pronunciation and meaning.

The Research

You may be starting to pick up a pattern with some of these columns. Not unlike what we’ve found with decodable texts, we simply don’t have strong research that isolates teaching syllable types and syllable division patterns to students as an individual component of instruction or intervention. Instead, we’re left to look at what research we have that teaches syllables as part of a larger curricular package, as well as analyze the theoretical basis on which this teaching method is based.

At the center of the debate is whether it is worthwhile and effective for teachers to provide rules-based instruction in dividing syllables. Although this isn’t always clearly defined in the research, this is generally taken to mean that students are taught to divide words based on patterns such as VC | CV division patterns (sometimes known as rabbit words) or V | CV division (sometimes known as tiger words), with additional provisions made for words ending in consonant-<le> patterns (such as 'purple'). Additionally, students are taught to identify the discovered syllables as one of six(ish) syllable types: open, closed, vowel team, r-controlled, vowel-consonant-<e>, or consonant-<le>.

Multiple problems arise from this instructional strategy, however. First of all, it doesn’t always (or even almost always) work. From investigating almost 15,000 words used in grades 1-8 texts, Kearns (2020) found that the VC | CV (rabbit) type syllable division pattern fit less than ¾ of applicable words, at 70.6%. Moreover, dividing words into V | CV (tiger) syllable patterns only fit 30.5% of words – less than ⅓ of the time. While these patterns of syllable division worked a bit better when isolating for bisyllabic words only— rather than looking at all multisyllabic words— the truth of the matter is that students aren’t only reading words with a maximum of two syllables, so whether rigid division principles are instructionally useful remains unclear.

Students can be introduced to these exceptions, such as learning to divide words as VC | V (sometimes known as camel words) if the V |CV division did not create a known word. But even these provisions underemphasize the impact that vowel reduction on unstressed syllables plays in multisyllabic English words. For example, neither V |CV nor VC | V division placements will help students arrive at the unstressed /ə/ as the first vowel sound in ‘above’. Shortcomings of this strategy increase when considering the expectation to label syllable types. What type is the first syllable in ‘above’? Or the last for that matter? What syllable type is the last in ‘patient’ or even ‘action’?

Experimental research contributes to this discussion in a limited but meaningful way. What we do have suggests that instruction in how to approach multisyllabic words is better than no instruction at all— and clearly better than relying solely on rote memorization or whole-word strategies.

For instance, Diliberto et al. (2008) found that a rule-based approach to multisyllabic word instruction led to gains in both real and nonsense word reading, as well as in overall reading comprehension for struggling middle school students, although it’s worth noting that these gains did not extend to oral reading fluency (see additional notes on study limitations, below). On the other hand, a study by Bhattacharya and Ehri (2004) demonstrated that flexible syllable division strategies also produced measurable gains for struggling readers in grades 6–10. Importantly, both approaches yielded growth relative to no intervention or to whole-word strategies. Teachers should use caution when considering the benefits of these strategies, however. Filderman and Toste (2021) found that while students who were taught strategies to read multisyllabic words did improve in this skill relative to a control group, they actually performed worse on measures of overall reading proficiency, including both fluency and comprehension, implying that the syllable intervention students received may have been at the expense of more valuable instruction.

It is also worth noting that this isn’t a simple binary between rule-based and flexible instruction. A third approach outside the scope of this article emphasizes morphological analysis as a foundational strategy for decoding multisyllabic words. At this point, though, we lack empirical comparisons between these different instructional approaches. That leaves us with promising options, but without definitive evidence to suggest that one is clearly superior to the others.

In Sum

Taken together, this leaves us with a landscape of instructional options— rule-based syllable division, flexible decoding strategies, and morphological analysis— but without clear evidence favoring one approach over another. Each has theoretical and practical merit, and each has been associated with improved outcomes compared to no intervention at all.  

Older elementary students, adolescents, and adults who struggle with reading may benefit from a moderate amount (3–15 hours) of syllable-based intervention to support the development of multisyllabic word reading skills. There is, however, some concern that this instruction may come at the expense of instructional practices that promote fluency and comprehension development (Filderman & Toste, 2021), so educators should be cautious about allocating large amounts of instructional time (e.g., 20+ hours) to syllable instruction, particularly if doing so reduces time for fluency or comprehension work.

As with our discussion of Heart Words, the evidence suggests that students need to engage with words beyond simply seeing and reading them in an attempt at rote memorization. What remains unclear is which instructional strategy works best— and for which learners.

Questions we just don’t have the answers to:

1. Is teaching syllable division patterns and syllable types an effective strategy?
   1. How much instructional time should be allocated for this instructional method?
   2. For what grades?
   3. For at-risk? All students? ELs?
2. How does morphological analysis fit into this instructional method? Is it at odds with learning syllable types and division principles? Is it the next transitional step?
3. How and when do students learn to use flexible strategies to decode the words that don’t fit the mold?

Nitty-gritty of some of the research:

Filderman & Toste, 2021

Participants

• 88, 4th and 5th graders with or at-risk for reading difficulties (including those with limited English proficiency)

Study Design

• Multisyllabic word reading lessons were customized at the beginning of the program based on diagnostic assessments (e.g., what vowel patterns, affixes, or types of words students struggled with). Students were given explicit instruction in vowel patterns and affixes, practice decoding multisyllabic words, and repeated readings for fluency. Sessions were 30-40 minutes, provided 4x per week for 8-10 weeks (38-40 total sessions). The control group performed tasks designed by teachers, including but not limited to: independent reading, Leveled Literacy Intervention (LLI), IXL Learning, and homework.

Findings

• Intervention groups performed better on the multisyllabic word reading measure. Gains were enhanced when progress was monitored and intervention was adjusted accordingly. However, the intervention groups performed worse on a combined measure of fluency and comprehension than the control group.

Notes/Limitations

• It is unclear if gains on multisyllabic word reading were sustained after completion of this intervention, and/or if students receiving this intervention made fluency and comprehension gains once returning to classroom instruction, which may have had more focus on reading connected texts.

Diliberto et al., 2008

Participants

• 74, 6th-8th grade students with high-incidence disabilities (e.g., ADHD) & at-risk for reading difficulties

Study Design

• The treatment group was taught rule-based syllable skills for 60x 15-minute sessions of syllable instruction over the course of 6 months.

Findings

• The treatment group performed better on real and nonsense word reading and comprehension. They also performed better on the fluency measure, but this wasn’t statistically significant- meaning, the difference in how much the treatment group grew as compared to the control group could have been the result of chance.

Notes/Limitations

• It is unclear what exactly the control groups were doing during the time that the treatment groups received this intervention. Additionally, it is noted that the treatment and control groups were not tightly matched for skill strengths, and the treatment group began the study with more significant deficits than the control group, which potentially impacted their growth potential. Similarly, the treatment group had both more overall students and significantly more 6th graders than the control (21 students as opposed to 5), so it is unclear how this might have impacted the results in terms of malleability/ growth trajectory. Finally, because the study design did not ensure consistent classroom instruction/curriculum, differences in student performance/growth could be attributed to other confounding variables besides solely the syllable instruction.

Bhattacharya & Ehri, 2004

Participants

• 60 struggling readers in grades 6-10

Study Design

• Split into 3 groups: one group received treatment for 6x 30-minute sessions in syllable-analysis, another in whole-word reading, and a control group did not receive treatment. The syllable analysis group received basic explanations of syllables (that each had a vowel), with flexible syllable division strategies permitted

Findings

• The syllable-analysis group outperformed the other treatment group as well as the control group on most assessed skills, including nonsense word reading, unfamiliar word reading, and spelling, with the students at the lowest grade equivalent (3rd-grade reading level) receiving the most benefit. Other measures (such as reading trained and familiar words) yielded similar results across the groups.

Notes/Limitations

• Because students at 4th- and 5th-grade reading levels were already close to ceiling on the word reading measure used in this study, it was difficult to see gains in their performance. Furthermore, as neither fluency nor reading comprehension was assessed, it is unclear how well this skill growth transferred to application in real texts.