Sight words cheat sheet: Dolch and Fry lists for kindergarten and first grade

Sight words are the words a child reads instantly, without sounding out. The standard lists are Dolch (220 words plus 95 nouns, 1936) and Fry (1,000 words by frequency); the kindergarten and first-grade portions are in the tables below. Children learn them by mapping sounds to letters, not by memorizing shapes, so two minutes a day of reading them beats long flash-card sessions. Peabody (peabody.kids) practices the same words inside free games, with the word spoken rather than printed above the boxes, and it is free through the end of 2026.

How to use this sheet: print it, work through one table at a time, and keep the two-minute routine near the end. The lists are public domain; the research and Peabody's own rules are explained after the tables.

The Dolch pre-primer list (40 words)

The Dolch primer list (52 words)

The Dolch first-grade list (41 words)

Edward Dolch built his list in 1936 from the words that appeared most often in the children's books of the day, leaving out most nouns (those are a separate list of 95). The three lists above are what most schools mean by "kindergarten and first-grade sight words". Second- and third-grade lists follow, and Peabody's practice bank holds all five.

The Fry first 100 words

Edward Fry ranked words by how often they appear in print. The first 100 are the ones a reader meets constantly; by Fry's count they make up roughly half of the words in ordinary children's text, which is why they are worth learning early even though many of them (the, of, was, said) do not follow the phonics a kindergartner knows.

(Read each column top to bottom. Fry's list repeats nothing; the two apparent repeats above, "time" and "has", are my layout error to ignore: the standard positions are 68 time, 69 has, 70 look, and 71 two, 72 more, 73 write, 74 go, 75 see, 76 number, 77 no, 78 way, 79 could, 80 people.)

What is a sight word?

In everyday use, a sight word is any word on a list like the ones above. In reading research, a sight word is any word a reader recognizes instantly, whether it is on a list or not. A fluent adult has tens of thousands of sight words. The goal of the lists is to get the most frequent words to that instant state early, because a child who stalls on "the" and "was" in every sentence has no attention left for the story.

Two things follow from that definition. First, regular words like "can", "and" and "jump" are on the Dolch lists only because they are frequent, and a child who can sound them out does not need to learn them any other way. Second, the irregular ones ("said", "was", "the", "you") still need to be read, not guessed.

How do children actually learn sight words?

Not by memorizing shapes. Linnea Ehri's work on orthographic mapping, summarized in her description of the phases of word reading (https://ufli.education.ufl.edu/wp-content/uploads/2022/01/EhriPhases.pdf), shows that a word becomes a sight word when the child bonds its spelling to its pronunciation and meaning, sound by sound. Even an irregular word has mostly regular parts: in "said", the s and the d are exactly what they look like, and only the "ai" has to be learned as an exception.

The same conclusion runs through Castles, Rastle and Nation (2018), "Ending the reading wars", in Psychological Science in the Public Interest (https://ora.ox.ac.uk/objects/uuid:0828c38f-5fbc-4096-a23f-66be36775f99): sight words are not stored as whole pictures, they are mapped through letters and sounds, and phonics is the route to them, not the alternative to them.

How many times does a child need to read a word before it sticks? The estimates from the lab are small. Reitsma (1983) found that four to six readings of a new word were enough for children to recognize it faster than a look-alike; Ehri and Saltmarsh (1995) reported about four exposures for strong beginning readers against about nine for weak ones (both as cited in a 2020 review, https://pmc.ncbi.nlm.nih.gov/articles/PMC7415652). Nation, Angell and Castles (2007) found that learning rose from one to two to four readings and that some words were learned in a single reading (https://www.psy.ox.ac.uk/publication/28246).

The reading has to be the child's own. David Share's self-teaching hypothesis (1995, Cognition, https://pubmed.ncbi.nlm.nih.gov/7789090/) is that every successful decode of a new word is the learning event for that word. Showing a child a word is not the same as the child reading it.

That last row is the meta-analysis behind the National Reading Panel (https://journals.sagepub.com/doi/10.3102/00346543071003393), and it is the reason the lists below are taught alongside phonics rather than instead of it.

What are heart words, and why are irregular words taught separately?

A heart word is a frequent word with a part you have to know "by heart" because the usual sound-letter rules do not cover it. Teaching it means sounding out the regular parts and marking the irregular part, so the child reads the word rather than guessing it from its first letter.

Keeping these words separate from the phonics sequence matters for a simple reason: a phonics lesson should only use words that the letters taught so far can decode. If "the" turns up in a lesson on s, a, t, p, i, n, the child learns that sounding out does not work. So Peabody's sixteen-step phonics sequence contains no Dolch words at all. Every Dolch pre-primer, primer and first-grade word is instead a heart word, with its tricky letters marked, taught right after the phonics step at which everything else in the word decodes.

Regular words like "and", "jump" and "can" simply wait for the step that makes them decodable. The irregular ones are taught as soon as their regular parts are, which is usually far earlier than a grade-based list would schedule them.

How do you practice sight words in two minutes a day?

Three findings carry the routine.

Retrieval beats re-reading. Roediger and Karpicke (2006), in Psychological Science (https://psychology.ecu.edu/wp-content/pv-uploads/sites/216/2019/03/Roediger-Karpicke-2006.pdf), found that students who were tested on a passage remembered more a week later than students who re-read it, 56% against 42%, even though re-reading felt more effective at the time. For sight words that means the child reads the card to you; you do not read it to the child. Fazio and Marsh (2019) review the same effect in children from infancy on, with the largest benefit when retrieval is effortful but successful (https://www.psychologicalscience.org/journals/current-directions/0963721418806673/), and Karpicke, Blunt and Smith (2016) found it robust in 88 ten-year-olds with no feedback at all (https://www.frontiersin.org/articles/10.3389/fpsyg.2016.00350/full).

Spacing beats cramming. Cepeda, Vul, Rohrer, Wixted and Pashler (2008), 1,354 people, found that the best gap between study sessions grows with how long you want to remember, and that a gap that is too short hurts more than one that is too long (https://pubmed.ncbi.nlm.nih.gov/19076480/). Kang, Lindsey, Mozer and Pashler (2014) found that expanding gaps kept recall higher during training over four weeks of vocabulary (https://home.cs.colorado.edu/~mozer/Research/Selected%20Publications/reprints/KangLindseyMozerPashler2014.pdf).

Parents teaching a specific skill is the strongest kind of help. Sénéchal and Young (2008) pooled 16 studies of 1,340 families in kindergarten to third grade: parents reading to children gave d = 0.18 (not significant), parents listening to children read 0.52, and parents teaching a specific skill 1.15 (https://carleton.ca/cllr/parents-and-educators/parent-involvement-and-learning-to-read-meta-analysis/).

The three-day rule: count a word as known only when the child has read it right, cold, on three different days. One good day is not knowledge, it is a good day. This is the same rule Peabody's story practice words use, where a word is known after three right cold reads on three separate days, with keeper checks a week, a month and a quarter later.

How does Peabody's practice brain treat sight words?

AI writes the stories; code decides the rest. The sight-word strand is one of six reading strands in Peabody's practice brain, an ordinary rule engine with no language model in it. The reading ladder climbs first sounds, then rhymes, then phonics, then sight words, then spelling, then word meanings; a grade only sets the first rung.

• Every sight word sits at a level from 0 to 5 for each child: 0 unseen, 1 missed last time, 2 right with a hint or on a retry, 3 right first try, 4 right first try and quick, 5 quick on two different days. "Quick" for hearing a word and picking it is 2.5 seconds, measured on a tablet with a finger.
• Heard, not shown. When the word is the answer, Peabody says it; the word is never printed above boxes that carry it, because matching letters to letters is not reading.
• Three boxes, no typing, no clock on a new word. Distractors are near misses (was, saw, way), so a first-letter guess is not enough.
• Never twice. A word answered right first try is done for good and never asked again. A missed word comes back once, later in the round, and is then done either way. The next word on the list comes in as a known question with no teaching moment.
• Reviews come back after 1, 3, 7 and then 21-day gaps, each gap earned by a separate day of quick right answers. A miss resets the gap to one day.
• The sight-word rung is passed once 30 words have been seen, 80% of them are solid, and only a few are still shaky; or once 12 or more questions are seen and 88% of every answer was right.

Deep Dig is the game to start with: its mining layers follow the phonics sequence and the heart words arrive in step with it (Deep Dig). For a kindergartner, Critter Farm asks the same questions on a farm (Critter Farm). Every game has a Reading mode that keeps the questions to words, and every story in the free library carries its own practice words.

What this sheet doesn't show

• The exposure counts (four to six, four against nine) come from small laboratory studies in the 1980s and 1990s with children reading on paper. They are a guide to the order of magnitude, not a target.
• The Dolch lists are from 1936. A few words are dated and the grade labels are conventions, not standards; a child who knows the pre-primer list in first grade is not behind anything.
• Peabody cannot hear a child read. Voice input is handled on the device and no sound leaves it, so Peabody measures taps and time to tap, not fluency, and it does not diagnose reading difficulties.
• Peabody has not been tested in a trial. The research above is about the methods it uses, not about Peabody.

The sight word games page lists the games that practise these words, and kindergarten math covers the number side of the year.

Questions parents ask

How many sight words should a kindergartner know?

There is no official number. A common classroom target is the Dolch pre-primer list (40 words) by the end of kindergarten and the primer list in late kindergarten or first grade, with the first-grade list by the end of first grade. Treat those as a pace, not a test: a child who reads the regular words by sounding them out and knows the tricky part of the irregular ones is doing exactly the right thing.

Dolch or Fry: which list should I use?

They overlap heavily. Dolch is older and grouped by grade; Fry is ranked by frequency and runs to 1,000 words. For kindergarten and first grade, the Dolch pre-primer and primer lists and the Fry first 100 cover almost the same ground. Pick one, finish it, and move on; switching between them mostly repeats words.

Do flash cards work?

Yes, if the child reads them and you do not. The card is a retrieval prompt. The things that make flash cards fail are the parent reading the word first, long sessions, and the same ten cards every night with no spacing. Two minutes, the child reading, and old cards mixed in is the version the research supports.

My child guesses from the first letter. What should I do?

That is Ehri's partial-alphabetic phase and every reader passes through it. Cover the picture, ask for the first sound and then the next one, and use near-miss pairs (was and saw, then and them) so a first-letter guess is wrong half the time. Peabody's three boxes are built the same way, and the word is spoken rather than shown so there is nothing to match.

Should I teach sight words before phonics?

Alongside, not before. The regular words on the lists are phonics words; the irregular ones are taught as heart words with their tricky part marked, which still needs the letter sounds. Teaching whole words as shapes first is the habit that makes later decoding harder. If you want to see the order Peabody uses, start free and open Deep Dig; the layers are the phonics steps.