Miles, K.P., Rubin, G.B., & Gonzalez-Frey, S. (2017). Rethinking sight words. The Reading Teacher, 71(6), 715-726.
This article will help teachers rethink the ways they have taught elementary school students about sight words (aka “high-frequency” words). That rethinking is long overdue.
The most common current practice has been around for many years and continues to be seen in many elementary schools. Young readers in the early grades, as well as struggling older readers, are typically provided with one of the popular high-frequency word lists, such as the venerable “Dolch lists” which have been used for many years and are still out there (try an Internet search to see just how popular that list still is) or the Fry lists (almost as venerable and popular), or some other more recent list (there are several out there).
The typical instructional approach tends to be the same, no matter which list is used. Students practice the words on the list, often on flash cards, and are told that these words cannot be sounded out using the graphophonemic relationships they know, so the words must be memorized. The words are then presented in all sorts of contexts, including on the ubiquitous word walls.
Miles et al ask whether this “all or nothing” (or “decodable vs. non-decodable”) approach to high-frequency words is either realistic or valuable. They propose that there may be a middle ground between words that are regularly spelled (and thus easy to decode) and irregularly spelled (and thus difficult or even impossible to decode). They suggest that there may be three categories: 1) words that are regularly spelled, like “cat”, and so have an easy phoneme to grapheme correspondence, 2) words that are “temporarily” irregularly spelled, like “gate” or “then” that can become regular once children are taught the frequently occurring patterns in those words, and 3) words that are “permanently” irregularly spelled, and may indeed need to be memorized, like “have” or “great”.
Miles et al state that there are fewer permanently irregularly spelled words than teachers think, and even those words do have some letters that can be sounded out and that fit expected patterns. They conclude that direct instruction in phoneme-grapheme relationships is beneficial for all kinds of words.
Two helpful tables are provided on pages 719 and 720, showing examples of all three kinds of words. As a teacher educator, I can envision a teaching activity where teacher candidates research the lists currently being used at their field experience schools and do some classification to see where the words on the lists they are using fit.
Miles et al conclude with a detailed description of an instructional intervention where children are taught to segment words into phonemes and then to spell those words. Several kinds of sensory input are used; children are asked to say each phoneme in a word, tap it out on their arms, and watch as the teacher spells the word on a whiteboard. They then practice using plastic “counters” and a grid of boxes (reminiscent of an Elkonin box), and then write the words using markers. I can envision having teacher candidates design a lesson focusing on a word or a set of words. I could see teaching words with similar patterns in a lesson, similarly to the way that onsets and rimes are taught. As students advance, more difficult (but common) patterns could be used (for example, words like “light” and “night”, which could even be compared to words that sound like them (for example, “kite”). Even a difficult word like “thought” could be taught with words like “brought”.
The intervention suggested here is really nothing that will seem unfamiliar to experienced elementary teachers, but the reconsideration of how we teach sight words/high-frequency words is refreshing and needed. Instruction in sound-symbol correspondences can empower readers. English spelling does have its irregularities, but maybe it’s not as impossible to cope with as we have thought, and effective instruction can make it less troublesome. The authors present some limited research (small sample size and no control group) in support of the proposed intervention. It will be interesting to see if deeper research will support their conclusions, and if teachers will rethink the old ways of teaching and try the intervention with their own students.
