How Spelling Should Be Taught: What the Research Actually Says

Most teachers inherit their spelling routine rather than choose it. A list goes home on Monday. Children copy the words, write them in rainbow colors, put them in alphabetical order, use them in sentences. On Friday there is a test, the scores go in the book, and a new list arrives the following Monday. Last week's words are never seen again.

Many teachers have a quiet suspicion that this does not work very well, usually because they have marked the Friday test on one day and read the same children's writing the next. The research suggests that suspicion is justified, and it also suggests what to do instead. The good news is that the better approach is not more expensive or more time-consuming. It mostly rearranges time you are already spending.

This is a plain summary of what the evidence says about teaching spelling. It covers what the major reviews found, why English is more teachable than its reputation, why the familiar routine fails to transfer into writing, and what a well-designed week looks like.

What the big reviews found

The most useful single source is a 2014 meta-analysis by Steve Graham and Tanya Santangelo, which pulled together the experimental studies on spelling instruction across the school years. Their headline finding was clear: explicitly teaching spelling improved students' spelling, and the gains held up over time and showed up in students' own writing, not only on tests. Spelling instruction also improved reading skills and phonological awareness, the ability to hear and work with the sounds in words. And more formal, systematic teaching did better than approaches that expected children to pick up spelling incidentally through reading and writing.

The key word is explicit. Explicit teaching means the teacher shows the pattern, explains why it works, and guides practice until students can use it independently. It is the opposite of distributing a list and hoping the patterns will be absorbed. Many students do absorb them, which is why a list-and-test routine can look fine in a strong class. The students who do not absorb them are exactly the ones who need teaching most.

The National Reading Panel reached a similar conclusion from another direction in its 2000 report. Systematic phonics instruction, teaching the relationships between sounds and letters in a planned sequence, improved young children's spelling as well as their reading. The effect on spelling was strongest in kindergarten and first grade, which is itself an argument for continuing to teach spelling explicitly as words get longer, rather than assuming early phonics will carry children through.

English is more teachable than its reputation

A great deal of spelling practice rests on the assumption that English spelling is chaotic, so each word has to be memorized as a separate item. That assumption is wrong, and it matters.

In a large study published in 1966, Paul Hanna and his colleagues analyzed about 17,000 English words and found that roughly half could be spelled correctly using regular sound-to-letter patterns alone. Many of the rest were off by only one sound. When you also take word parts and word origins into account, the proportion of truly unpredictable spellings falls further still. Louisa Moats has made this case for years: most English spelling can be taught as patterns, and only a relatively small share needs to be learned word by word.

Meaning is the part that is most often left out. "Sign" has a silent g that makes sense only when you see "signal" and "signature". "Muscle" has a c that appears in "muscular". A 2010 review by Peter Bowers, John Kirby and Hélène Deacon found that teaching morphology, meaning the meaningful parts of words such as roots, prefixes and suffixes, helped students' literacy. Some of the benefits were largest for students who were struggling. Word-part teaching is not an enrichment topic for strong spellers. It is often the missing explanation for weak ones.

Spelling is not visual memory

For much of the twentieth century, spelling was treated as a visual skill: good spellers "see" the word in their mind. That idea still survives in some classroom practices, such as drawing boxes around the shapes of words, or tracing letters in colors.

The research on how words are stored points elsewhere. Linnea Ehri's work on orthographic mapping shows that a word becomes fixed in memory when the learner connects each sound in it to the letters that spell it, and links that to the word's meaning. Children who can spell well are not remembering pictures. They are remembering sound-letter connections, patterns and word parts, which is why a good speller can usually tell you why a word is spelled as it is.

The practical consequence is significant. Activities that draw attention to the look of a word without its sounds or meaning, such as shape boxes or decorative copying, give the mapping process nothing to work with. Activities that ask students to say a word, segment its sounds, connect each to letters and notice the tricky part give it exactly what it needs.

The same logic applies to the high-frequency words that are often taught as "sight words" to be memorized whole. Most of them are far more regular than they look. In "said", the s and the d are entirely predictable; only the "ai" is unusual. Teaching students to map the regular parts and flag the one irregular part, sometimes called learning it "by heart", turns a word that seemed arbitrary into a word with one thing to remember.

Why the Friday test does not transfer

The weekly list-and-test routine has three built-in weaknesses, each of which research on memory can explain.

First, it masses practice into one week and then stops. Memory for new spellings fades quickly unless it is revisited. A large body of research on the spacing effect, summarized by Nicholas Cepeda and colleagues in 2006, shows that practice spread across weeks produces far more durable learning than the same practice crammed into a few days. A word practiced Monday to Friday and then dropped is at the peak of its strength during the test, and fading from then on.

Second, most lists are blocked by pattern. When every word on the list ends in "-tion", students never have to decide which ending a word takes. Studies of interleaving, in which different types of problems are mixed together, have repeatedly found that mixed practice produces better long-term learning, even though it feels harder while you are doing it.

Third, the test checks words in isolation, with full attention. Writing requires students to spell while also planning ideas, forming sentences and managing handwriting. Only spellings that have become automatic survive that load. That is why so many parents see a perfect Friday score and the same words misspelled in a story on Monday.

Copying is not practice

Copying a word out five or 10 times feels productive, and it fills time. Unfortunately it can be done almost entirely without attention: many children copy the first letter down a column, then the second, in a way that bypasses the word altogether. Rainbow writing and similar activities focus attention on colors and letters rather than sounds and meaning.

What works better is retrieval. In a 2013 review of common study techniques, John Dunlosky and colleagues rated practice testing and distributed practice as the two most effective methods they examined, and rated rereading and highlighting among the least effective. For spelling, the classroom version is simple: students hear the word, write it from memory, and then check it themselves against the correct spelling. The familiar "look, say, cover, write, check" routine works precisely because the "cover" step turns copying into recall.

What a good week looks like

Here is one way to build the research into a week of about 15 to 20 minutes a day. It works with or without a Friday test.

The features matter more than the specific days. The pattern is taught, not merely listed. Students retrieve words from memory several times, never simply copying them. Earlier words return on a planned schedule. And every week includes spelling inside sentences, not only in isolation. For a short, ready-made version of the review element, see a five-minute spaced retrieval routine.

Teach to the stage, not only the grade

Students in one class can be several years apart in spelling development. The widely used Words Their Way approach, developed by Donald Bear and colleagues, describes a series of stages. Children move from spelling with letter names, through vowel patterns within single-syllable words, to syllables and suffixes, and finally to spelling based on meaning and word origin. A short spelling inventory, a list of words chosen to reveal which features a student controls, can place each child in a few minutes.

Knowing the stage lets you group students so that each is working on the patterns they are ready for. A child who still writes "rane" for "rain" needs vowel team work, not a list of "-tion" words. The errors themselves are the best diagnostic tool you have; we explain how to read them in what a spelling mistake tells you.

Measure retention, not Friday

If the Friday score is the only data, the routine will look like it is working. A better measure is a cold check on words taught two or three weeks earlier, and a look at how often taught words appear correctly in students' independent writing. Those two numbers tell you whether spelling instruction is actually changing students' writing, which is the whole point. They are also easy to collect: a five-word cold check once a fortnight takes three minutes, and a quick scan of one piece of writing per student per month shows you which taught words are really in use.

Practice software can carry some of this load, especially the spaced review of old words, which is hard to manage by hand for a whole class. Spelling.School, for example, reads each word aloud, marks the exact letters a student got wrong, and schedules each word to return before it is forgotten. But the explicit teaching of patterns, meaning and sounds is the teacher's job, and no software replaces it.

Where to start

If you change only one thing, add review to the Friday test. Include three to five words from earlier weeks on every test, plus one dictated sentence. That single change brings spacing, mixing and writing in context into the routine, the three things the usual approach leaves out.

When you are ready to go further, teach one pattern per week explicitly, with its reason, and use the table above to replace copying with retrieval. For a clear starting set of patterns, see the seven most important English spelling patterns.

Questions people ask

What is the most effective way to teach spelling?

The most effective approach is explicit, systematic teaching of how words work: sound-letter patterns, word parts such as roots and suffixes, and the reasons behind tricky spellings. Students should practice by writing words from memory and checking them, with earlier words returning on a spaced schedule. A 2014 meta-analysis by Graham and Santangelo found that this kind of instruction improved spelling, reading and writing.

Do weekly spelling tests work?

Weekly spelling tests measure whether students can spell words right after practicing them, but they do little to make those spellings last. Most lists are practiced for one week and then dropped, so the words fade and do not appear correctly in writing. Tests work much better when they include review words from earlier weeks and a dictated sentence, turning them into spaced retrieval practice.

Should spelling be taught separately from reading?

Spelling should be taught explicitly, with its own dedicated time, but closely linked to reading. Reading and spelling draw on the same knowledge of sounds, letters and word parts, and research shows that spelling instruction also improves reading. However, many children who read well still spell poorly, so reading alone does not reliably teach spelling.

How much time should spelling get each week?

There is no single research-based number, but many structured programs give spelling around 15 to 20 minutes a day in the primary grades. How the time is used matters more than the total: explicit teaching of patterns, word sorts, retrieval practice from memory and spaced review of earlier words. Time spent on copying or decorative activities adds little.

Does copying words out help?

Copying words out repeatedly does little for spelling, because it can be done without paying attention to the sounds, patterns or meaning of the word. Writing a word from memory and then checking it is far more effective. The "look, say, cover, write, check" method works because the cover step turns copying into recall.

Fix the words you actually get wrong

A few minutes a day, spaced repetition, built only for spelling. Try five words without signing up.