Is there a best way to teach irregular words?
How should we teach those tricky words that occur commonly in reading and spelling, but which don’t have common letter-sound correspondences, such as the words “said” or “one”? There are lots of ways to teach these words, and some ways are better than others. Though there is limited research directly comparing approaches, there are some clear guidelines that have emerged from the research for teaching these words, and from the science of reading more generally.
First, we need to clarify what is meant by “irregular” words. Terminology is important here, to make sure we’re talking about the same thing, and to avoid using terms that mean something entirely different when used within the research.
Irregular words are those which have unexpected letter-sound correspondences within them – such as the “ai” in “said.” An irregular word might be a high-frequency word, or a low-frequency word (or anything in between); the concern is how to teach the high-frequency ones in particular, since they’re going to be encountered a lot and from a young age. The irregularity might be because the letter-sound correspondence has not been taught yet (temporary irregularity), or it might be inherent to the word itself (permanent irregularity). Permanently irregular words are those where only that word or very few words contain that letter-sound correspondence. Only about 15% of high-frequency words are permanently irregular. (If a word had regular and known letter-sound correspondences, they can simply be decoded in the standard way, and would not require special attention in the way being discussed in this post.)
Irregular words are not “sight words,” even though this has been a very common term for referring to these words in the past. They’ve historically been called sight words because it was thought they had to be learned by sight – that they couldn’t or shouldn’t be sounded out, and needed to be taught and learned as whole words.
In the research literature, sight words are something very different: they are words that have already being learned so that we recognise them very quickly and seemingly effortlessly (i.e., by sight). They have become sight words because they have been orthographically mapped; this fancy term refers to an incredibly important idea that is very well-supported by decades of research. A word that has been orthographically mapped has the spelling/letters (the orthography) and the sounds glued together, allowing us to quickly access meaning. This is an enormous hint to what approach we should be using with irregular words.
You can see why we should never be calling irregular words “sight words” in the way they often are in the classroom: In the classroom, it’s commonly used to refer to words the students haven’t learned yet and are expected to learn “by sight;” in the research it refers to words we have already learned well and orthographically mapped, whether regular or irregular.
OK, so how might we teach these irregular words then – the ones that sit outside of the student’s current decoding ability?
Our written language is a code for sounds. Even in irregular words, those letters all represent sounds (just some of them are unexpected sounds). The best way to teach irregular words is by drawing attention to the sounds in that word and the letters that represent those sounds – both the expected and the unexpected. There are numerous ways to effectively do this.
One of the most common modern ways is by the “heart word” method, where the unexpected sound is identified by having a heart drawn around it, being underlined, or in whatever way the teacher chooses. The unexpected sound is then learned “by heart” rather than the entire word being learned by heart. Lots of activities can be used to reinforce this learning. (Or the other sounds in the irregular word could be marked with anchors, to represent the fact they anchor the word with their expected sounds. The idea is the same: identify the unexpected parts and the expected parts.)
Another approach is to simply include the irregular words within the teaching unit for the unexpected sound it contains. For example, you could include “said” within the lessons that introduce and practice the “short e” sound (/s/ /e/ /d/, where “ai” represents /e/). Again, this will work because explicit attention is being drawn to the relationship between the letters/spelling and the sound itself – the student is being supported towards orthographically mapping that word. (Given the high frequency of “said,” and the fact it is an uncommon way to produce the /e/ sound, it would likely need more practice and attention than other /e/ words.)
An important skill that can support the learning of some irregular words, is the ability to “flex the vowel:” when a word doesn’t sound quite right, we can encourage the child to engage in this skill (which is sometimes referred to as the second stage of decoding), to get the word to match one in their existing vocabulary. This is also known as “set for variability” or more simply as “mispronunciation correction.” Mispronunciation correction has been tested and shown to be an effective method for learning irregular words. Relying solely on this approach might present some challenges for particular irregular words, such as “eye” and “one,” especially out of the context of a sentence, so this method can be supported by other approaches (I’ll return to this point below).
Morphology and etymology are also helpful tools in tackling irregular words: By making links to the meaning through related words and to origins of words, we can help students distinguish between two, too, and to, and link two, with twin, and twice. Again, attention is explicitly drawn to the spelling/letters as part of these investigations.
Yet another option for effective teaching of irregular words, is actively spelling the presented irregular word using sounds – presenting the irregular word, and getting the child to spell it as the main form of practice. Again, that explicit link is being made between the written form of the word and the sounds those letters "make:" that is always the key.
What doesn’t work as well is simply presenting the word as a whole and telling the child to repeatedly read it off a flash card (for instance). Some level of intentional and explicit sound analysis is always best. This is clear from experimental studies and from the underlying research that pathed the way to understanding orthographic mapping.
Knowledge of the relationship between sounds and letters is “non-negotiable,” even for remembering irregular words. Phonics skills are the foundation and are necessary for remembering words. Words are not remembered based on visual memory; we aren’t good readers because we memorised the “shape” of thousands of words, and neither should we use that approach to teach the small number of permanently irregular words.
It’s important that whatever method is chosen for learning irregular words, that it is used consistently within that classroom and/or school, to make it easier for the student to understand the task they are engaged in and what is expected of them. To clarify though: this requirement for consistency does not mean we can’t or shouldn’t use a variety of irregular word supports across literacy lessons; we could use morphology and etymology, alongside mispronunciation correction, alongside a heart word approach. What we want to avoid is having (for example) three different ways in which we teach the heart word approach within a single classroom (sometimes using underlining, sometimes an anchor image, sometimes circling the unexpected sound).
The teaching of irregular words does not require extensive classroom time; expending a lot of classroom time on these words, is a poor use of precious instructional time and is not necessary.
In summary, though there is limited research at this stage about precisely which irregular word teaching method is best compared to the others, there is clarity that an effective approach will link sounds to the letters within those words, which supports orthographic mapping. There is evidence that a range of approaches to teaching irregular words that fall within that general category, can be effective, and that approaches which focus on the word as a whole – without drawing attention to the individual letters within that word – are an inferior option. It could be beneficial to use a range of methods to teach and support the learning of irregular words, but we need to be careful to maintain consistency in the classroom as well in the specifics of how we deliver the instruction. And, finally, a reminder that we should avoid referring to irregular words as sight words; this is inconsistent with the way that term is referred to in the literature, and suggests the whole word needs to be learned by sight, which is not what we should be doing.
[The above post was first shared in the private Facebook group, Effective Education Practices NZ, in 2023. I’ve now shared it publicly here, as it’s one of my more popular posts and many have found it useful. I hope you found it useful too!]
***
The most useful resources I found when exploring this topic:
“High Frequency Words What, Why, and How It Pertains to the Science of Reading” (SoR- What I Should Have Learned in College YouTube channel, 2021)
“Teaching irregular words: What we know, what we don’t know, and where we can go from here” (Collenbrader et al., 2020)
“Irregular Words with Dr. Danielle Colenbrander” (Teaching Literacy Podcast, 2022)
“Teaching Children to Read Irregular Words: A Comparison of Three Instructional Methods” (Colenbrander et al., 2022) (Not open access, but is discussed in some detail in the podcast episode shared above)
“David Kilpatrick ‘How We Remember Words, and Why Some Children Don’t’” (UC SDI P20 Literacy Collaborative YouTube channel, 2020)