Lately, I’ve noticed how often the phrases “evidence-based” and “research-based” appear in conversations about literacy instruction. They show up in articles, on social media, and in professional learning sessions. They seem to have become a shorthand for meaning, “This is the right way to teach.” And I understand – as teachers, we want to do what works. We want to do what is best for our students. When something is labeled this way, it can feel reassuring.
But I’ve also been wondering, what does the research actually say? I’ve felt compelled to look more closely because I am hearing strong and sometimes conflicting messages about literacy instruction. I’m finding that engaging with research isn’t about blindly following it, but about staying curious about what the findings suggest and reflecting on how they connect to today’s classrooms.
My curiosity led me back to a practice many of us know well and may not have questioned in a long time – using syllable division patterns for solving multisyllable words. I’ve taught it. You’ve probably taught it, and many of us were trained to teach it. You know the routine. Find the vowels. Count the consonants. Divide here or there. Decide if the vowel is long or short. Blend it back together and hope the word sounds right.
For a long time, this felt like effective instruction to me. It felt systematic. But recently, I spent time with a research study by Devin Kearns that made me stop and really think about what I was asking readers to do when instructing them to rely on syllable division patterns, especially the two most common patterns VC|CV and V|CV, to solve words with multiple syllables. Kearns (2020) points out that syllable division has long been recommended in reading instruction, but “There has been no study to determine whether the recommended patterns work consistently” (p. S146).
Kearns analyzed nearly 15,000 words that appear in grades 1–8 texts. Not words in isolation. Not carefully selected “words to teach” from phonics or reading curriculums. Actual words students encounter in reading authentic texts. He wanted to know how often the syllable division patterns we teach actually lead to the expected vowel sound.
The answer? Not nearly as often as we might think.
Research Results
Kearns found that as words get longer, the two most common syllable patterns (VC|CV and V|CV) that readers are taught to use to predict vowel sounds become increasingly inconsistent. In other words, these patterns tend to work best under ideal conditions, such as in carefully selected phonics lesson examples. However, when students encounter longer, more complex words in authentic texts, those ideal conditions are less common, making rigid syllable division rules less reliable.
In two-syllable words, the VCCV pattern (like rabbit) follow the expected pattern fairly often. But two-syllable VCV words (like tiger) only follow the “rule” half the time. Once words become longer than two syllables, reliability drops even more (Kearns, 2020, p. S153). Kearns found that the V|CV pattern in words with three or more syllables is least consistent with the vowel sound readers would expect (the long vowel sound) and instead the vowel sound is the opposite of what the pattern predicted. For example, the word culinary is divided into syllables as cu / li /nary. Based on that division, readers expect the u in the first syllable to make a long vowel sound. Instead, the u makes a short vowel sound which directly contradicts what the syllable pattern suggests. Kearns (2020) data show that “The exceptions to the expected V l CV pattern are more the rule” (p. S153).
Kearns (2020, p. S151) also notes that one major reason vowel sounds become less reliable in longer words with both syllable patterns is the presence of unstressed vowels. Kearns’ results point out much of the inconsistency comes from the fact that vowels in longer words are often reduced to the schwa sound, not clearly long or short, especially in VCV words with three or more syllables.
What does this mean for instruction?
A major point Kearns (2020, p. S154) makes is that asking readers to rely on and apply syllable division rules requires significant cognitive demand without offering consistent payoff. It requires students to stop reading for meaning, shift their attention to a multi-step procedure, hold syllable rules in working memory, and then apply a pattern that may or may not work. Kearns argues that this kind of effortful, attention-heavy processing may interfere with fluent reading and comprehension, especially for students who already find reading demanding.
Kearns is clear that students still need support reading multisyllable words. The question is whether syllable division patterns are the most helpful way to provide that support when students are reading words with three or more syllables.
The research points toward more flexible approaches to support students in reading longer words. Kearns (2020, p. S154) suggests the following word-solving strategies, which have been successfully tested in other experimental studies:
- Instruction should focus on helping students flexibly divide words rather than teaching rules for how syllables must be divided (Bhattacharya & Ehri, 2004, as cited in Kearns, 2020).
- When breaking words apart, help readers apply the idea that every syllable contains at least one vowel (O’Connor et al., 2015, as cited in Kearns, 2020).
- Support readers in flexibly trying different vowel sounds, knowing that one vowel can represent more than one pronunciation (Lovett et al., 2000, as cited in Kearns, 2020).
- Teach students to adjust mispronunciations until the word clicks with one they know, a process referred to as set for variability (Gibson, 1965, as cited in Kearns, 2020).
If you’ve ever watched a child say something like “min / i / mize” and then self-correct to “minimize,” you’ve seen this flexibility in action. That moment is powerful. It connects decoding, oral language, and meaning in a way no “rule” ever could.
Instead of relying on rigid syllable rules, Kearns (2020, p. S154) highlights another strategy with stronger evidence: teaching students to notice meaningful word parts such as prefixes, suffixes, and roots. When a reader spots a familiar affix or root, they can use the word part they know, which is usually larger and more meaningful than a syllable. This type of word-solving supports both pronunciation and meaning, rather than requiring readers to stop and apply a complicated set of syllable rules.
Closing
As literacy educators, we must think carefully about cognitive load, instructional payoff, and how word-solving strategies impact readers’ fluency and comprehension. If a strategy requires a lot of mental effort but works inconsistently, it’s worth asking whether it truly supports readers.
Kearns’ research helped me think about how English syllable patterns are not as predictable as I once thought. Most importantly, his findings invited me to rethink long-standing practices and to prioritize strategies that will keep students oriented toward meaning while they solve words. That rethinking feels significant. Being evidence-based means staying curious, even about the practices we once felt sure about.
References
Kearns, D. M. (2020). Does English have useful syllable division patterns? Reading Research Quarterly, 55(S1), S145–S160. https://doi.org/10.1002/rrq.342
