What is the problem with “sight words”?

My son just entered kindergarten in our local NYC public school. We received a folder from his teacher with two sets of materials: an overview of the Fundations phonics program (good!), and a list of sight words that he would be expected to memorize each week (um).

This is how the sight word overview began:

The contrast between this information, and the printout from the Fundations program (an explicit, systematic, sequential Orton Gillingham based phonics program) was stark, and caused a strong reaction in me, as I knew that this information was inaccurate. Yet the reality is that most people—including all too many kindergarten teachers—are not aware of how this can be problematic, most especially for students who may struggle with word-level reading. So I write this post to try to clarify why this definition of sight words and the associated belief that they all must be memorized is a problem.

Locating the Source of the Problem

First of all, let me be clear that I do not blame my son’s kindergarten teacher for this misunderstanding. She has provided clarity on what the expectations are for my son’s reading based on her experience and materials that are available to her, and provides resources for us to work with my son at home in alignment to these expectations.

This misinformation about “sight words” is rather perpetuated by education publishers, self-proclaimed gurus, and consultants large and small. As one example, you can see echoes of the question “Did you know about 75% of words we read are sight words?” on this Scholastic page: “Sight Words 101“. Yes, that Scholastic, one of the most well known publishers of educational materials. They explain the distinction between “sight words” and “high frequency words,” and mention the Dolch high frequency word list, which “are a list of 220 words that are used so often in print that together they make up an estimated 75% of all words used in books.

They then go on to state what is most problematic in how we talk about these words, whether we term them “sight words” or “high frequency words”:

“You might think that these words are so common that kids would just learn them organically through reading and other everyday print. But many of the words also defy standard phonetic conventions, meaning they are impossible to sound out.”

Not all the information on this Scholastic page is problematic. They helpfully explain that in fact, ALL words that a child can read with automaticity are actually sight words. They also explain that high frequency words, such as Dolch words, are often the ones most often referred to as sight words. All useful!

But you can see how most people would read something like this and say, “OK, so the words that make up to 75% of the words used in books are impossible to sound out. Therefore, these words must be memorized by sight!”

Except that it’s simply not true. Because most words in English, as “opaque” as our orthography may be, are still phonetically decodable. And that includes high frequency words on the Dolch list.

Most Words in Written English Are Phonetically Decodable

It is blatantly false information to state that most high frequency words are impossible to sound out. It is quite the other way around: most high frequency words are either entirely decodable, or at the very least, have a good portion of letter patterns that regularly match phonemes. A good example of the latter that is often used is “said.” The /s/ and the /d/ are regular grapheme-phoneme correspondences, whereas the /eh/ paired to the “ai” spelling is irregular and must simply be memorized. But by anchoring that irregularity alongside the regular pairings, it still is grounded in a phonetic approach rather than blanket memorization.

So the question thus becomes: how do we teach high frequency words that follow phonetic patterns and those that are less regular more strategically, and when? And what about those words that are just completely irregular?

Rather than pretend to be an expert on this matter, I will point you to some useful resources and guides:

What are Sight Words, Then?

Sight words, as even that problematic page by Scholastic referenced earlier acknowledged, are not high frequency words. Sight words are rather any word that a given reader can read with accuracy and automaticity. If we are fluent readers, then most words we encounter in print are sight words, except for those rarer words we haven’t yet encountered frequently. In which case, we then leverage those strategies we have gained as skilled readers: connecting sounds to the letter sequences, or looking for meaningful parts within the word.

Once we’ve encountered the word in print a few times and made the bonds in our mind between the sounds, the spelling, and the meaning of the word, we will thereafter recognize the word at “the speed of sight,” as Mark Seidenberg puts it.

What’s the Problem, Again?

Right, OK. There remains the fact that some high frequency words are highly irregular and must simply be memorized. In fact, when we start learning the alphabet, we do learn via memorization, which is termed paired-associate learning or associative learning in the literature. So what’s the big deal? Kids still need to memorize, right?!

Yes, they do. But when the message is that MOST words in written English are not decodable and thus must be memorized, we set many kids at a great disadvantage. For students struggling to internalize the cipher of the written code, most especially for those who may have dyslexia, teaching them to use everything BUT phonetic decoding strategies robs them of one of the most reliable strategies for gaining fluency with word-level reading.

This problem then gets compounded when schools use an ELA curriculum like TCRWP Reading Units, as my son’s school does and as all too many do, which perpetuates the idea that contextual information like pictures should be used as a FIRST resort to word-level recognition, rather than as a LAST resort. So even when there may be a phonics program used by the school, it’s counteracted by all the sight word memorization and the messages given during the core ELA reading block to use all those cues instead of decoding.

So what does this result in? Too many kids who show up in upper grades who cannot recognize the majority of words in print with any level of the accuracy nor automaticity needed because they have not been taught the phonological, morphological, nor orthographic skills and patterns explicitly enough to become fluent readers.

So yes, a few initial words still need to be memorized via good old flashcard style practice. But the message to our children needs to be clear that most words are not learned this way because we can learn how to read most words by articulating the discrete sounds we can pair to letters (phonemes <-> graphemes) or, as we begin to read more complex and multisyllabic words, by recognizing the meaningful patterns and parts of words (morphemes <-> graphemes).

I invite you to share any further resources you may have encountered on this subject that are useful, or any advice you may have in supporting schools and teachers in making the shift from this approach.

6 responses to “What is the problem with “sight words”?”

  1. Sending sight-word lists home to parents is parent abuse–asking parents to do what experienced teachers know is not an easy task. First of all, why are we expecting kindergarten children to do what once was reserved for first grade? There is now some “age-flation”–earlier age cut-off dates– but not enough to justify pushing academics on very young children. As for “flashcard” words, I can think of nothing but “one,” “once,” and “eye” that can’t be at least partially decoded. (“Eye” is not a problem–it can be turned into a funny face.) My list of real sight words (for first grade not kindergarten) would be: “one,” “eye,” and “the” (since “the” would be the key word for voiced “th” and “once” should go on the back burner). As for the paired associate task of memorizing letter-sounds– the REAL work of kindergarten– having key-word pictures embedded in the shapes of the letters (see Ehri, Deffner, and Wilce, 1984) converts abstract symbols with abstract sounds to familiar pictures with names already memorized (the miraculous achievement of toddlers). My post script to all this is that color-coding the vowels makes almost all words decodable (not “one,” “once,” and “eye”). I don’t use my color code with any but precocious kindergarten children, who should be in first grade and comfortable with CVC words before attempting to translate the irregular vowels in words like “said.” When the time is right, the “ai” matches the picture of a gray elephant for the short “e” sound.


  2. I recently found my 1st grade certificate for Dolch List. I wonder if I was able to do this because my speech therapist helped me with phonemic awareness, and my older brother had learned phonics and taught me? Literacy is a social just issue.


    • Great question. Two places to start: check out that first link to the Reading Rockets post from Linda Farrell et al. Then also check out Devin Kearn’s updated list


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: