DIBELS explained for parents

DIBELS is a set of short timed reading checks that many US schools give three times a year, from kindergarten upwards, to decide which children need a closer look. It stands for Dynamic Indicators of Basic Early Literacy Skills. In kindergarten it samples four things: naming letters, breaking spoken words into sounds, sounding out made-up words, and reading real words.

A DIBELS score is a sample of one skill on one morning, and it is built to sort rather than to measure how well your child reads. That is not a criticism of the test, it is what a screener is for. The benchmark goals are published openly by the University of Oregon, which owns DIBELS, and the kindergarten chart is below with each subtest explained in the order your report probably lists them.

What is DIBELS, and why does your child’s school give it?

It is a screener, not a diagnosis. Each subtest takes about a minute, an adult scores it live, and the point is to catch children who need extra help early enough for the help to work. Nothing in a DIBELS result names a condition or identifies a learning difficulty.

Schools give it because a growing number of states require reading screening three times a year. Minnesota’s READ Act is a worked example: its screening guidance requires students in kindergarten through third grade to be screened in fall, winter and spring for phonemic awareness, phonics, decoding, fluency and characteristics of dyslexia, and it lists DIBELS 8th Edition among the approved tools. Other states have similar laws with their own approved lists, which is why these reports started arriving in folders.

What do the kindergarten subtests ask your child to do?

Four subtests, and the useful way to understand each is what it is not. Reports usually abbreviate them, which is most of why they are hard to read, so here are the abbreviations first.

On your report	What it is	What it is NOT
LNF	Naming letters, one minute	Not letter sounds
PSF	Saying a spoken word’s separate sounds	Not reading, and no letters involved
NWF	Sounding out made-up words. Reports two numbers	Not a vocabulary test
WRF	Reading real words from a list	No goal at the start of kindergarten

Letter Naming Fluency (LNF) asks for letter names, not sounds. A child sees a page of mixed-case letters and names as many as they can in a minute. It is the subtest parents most often misread, because knowing that m is called “em” is a different thing from knowing it says /m/, and the second is what reading actually needs. If this one was low, the skill underneath it is letter knowledge: teaching the alphabet and letter sounds is the method, and what order to teach letters covers sequence.

Phoneme Segmentation Fluency (PSF) is pure listening, with no letters involved at all. The adult says a word and the child says its separate sounds: mat becomes /m/ /a/ /t/. If this was low, the skill is hearing the sounds in words, and Elkonin boxes is the one activity that makes it visible to a five-year-old.

Nonsense Word Fluency (NWF) is sounding out, and it reports two numbers from one task, which almost no parent-facing explanation mentions. Correct Letter Sounds counts the individual sounds your child produced correctly; Words Recoded Correctly counts how many whole made-up words they read as words. A child can score reasonably on the first and zero on the second, and that gap is informative: it means the sounds are there and the blending is not. The skill is sounding out short words.

Word Reading Fluency (WRF) uses real words, read from a list in a minute. The publisher sets no goal for it at the start of kindergarten. If this was low, the skills are sounding out and a small store of words known on sight: how to teach sight words.

Why does the test use made-up words?

Because a made-up word cannot have been memorized, so the only way to read it is to sound it out. That isolates decoding from recall. A child who recognizes cat as a familiar shape tells you nothing about whether they can decode; a child who reads mip has to have used the letters.

It is the same reasoning behind the argument that guessing at words is the problem to catch early: a reader leaning on pictures and context looks fluent until the context runs out. Made-up words remove the context on purpose.

Testing with them and teaching with them are completely different things, and this is the one place a well-meaning parent can do real damage. Timothy Shanahan is blunt about it in On Teaching Nonsense Words: “Please do not teach these nonsense words to the kids. It is harmful to kids.” His reason is the mechanism, not the morality: “If kids are memorizing pronunciations for those nonsense words, then the tests no longer can tell how well the kids can decode.”

The University of Oregon says the same thing in the license on its testing materials: “Assessment materials should not be used for student practice or coaching. Practice and coaching on the materials will invalidate the results.” That is the publisher of the test telling you not to practice the test.

What are the DIBELS benchmark goals for kindergarten?

These are the scores at or above benchmark, from the publisher’s own chart. The source is the University of Oregon’s DIBELS 8th Edition benchmark goals, with goals updated July 2020. B, M and E are beginning, middle and end of the kindergarten year.

Kindergarten subtest	B	M	E
Letter Naming Fluency (LNF)	25	37	42
Phoneme Segmentation Fluency (PSF)	15	43	53
Nonsense Word Fluency: Correct Letter Sounds	20	36	49
Nonsense Word Fluency: Words Recoded Correctly	no goal	9	13
Word Reading Fluency (WRF)	no goal	10	18
DIBELS Composite Score	332	393	450

Two things in that table are worth reading twice. The blanks are real, and they are not missing scores: at the start of kindergarten the publisher sets no goal for whole words recoded or for real-word reading, and reading even one word puts a child in the green core-support range. And the jumps are uneven. Breaking words into sounds goes from 15 to 43 between fall and winter, which is the publisher saying that this skill is expected to take off over that stretch.

If your report names a different edition, this chart does not apply to it. Goals, subtest names and timing differ between editions, and the 8th Edition is the current one. Some schools report Acadience Reading instead, which descends from the earlier DIBELS Next and has its own goals.

What do the colors and the composite score mean?

The bands describe how much support a child is likely to need, in the publisher’s own words, and they are more careful than the colors suggest. The chart’s legend defines four:

Band	What the publisher says
Blue	Core support, negligible risk. Nearly all students in this range score at or above the 40th percentile on the criterion measure
Green	Core support, minimal risk
Yellow	Strategic support, some risk
Red	Intensive support, at risk. About 80% of students scoring below the 20th percentile fall in this range

Read the word “risk” as “likelihood of needing extra teaching”, because that is what it is measuring. The bands are defined by percentile ranks on a separate criterion measure, so they are statements about how a group of children at that score tended to go on to do. They are not statements about what your child can do.

One detail that will not match your expectations: Letter Naming Fluency has no blue band at all. Every other subtest’s at-or-above-benchmark figure sits on the blue row. LNF’s 25, 37 and 42 sit on the green row. If you are comparing rows across a chart, that is the one that does not line up.

A score just under a blue goal is not a failing score. Below the blue figure but inside the green range is still core support with minimal risk, in the publisher’s own words. For breaking words into sounds at the start of kindergarten, green runs from 5 to 14, under the blue goal of 15.

The composite score combines the subtests into one number, which is why it is in the hundreds while the subtests are in the tens. It is the figure most reports lead with, and it is the least diagnostic thing on the page: a composite tells you there is something to look at, and only the subtests tell you what.

What does a low score not tell you?

It does not tell you your child cannot read. It tells you they produced fewer correct items in a minute than the benchmark, on one day, on one skill.

It does not diagnose anything, and it specifically does not identify dyslexia. Some state screening laws mention characteristics of dyslexia, which is what makes this confusing, but a screener flags children for a closer look and a diagnosis is a separate process with separate people.

And because every subtest is timed, speed and accuracy are mixed into one number. A child who reads carefully and correctly but slowly can land below benchmark on a skill they genuinely have. That is a real limitation of a one-minute measure, and it is worth asking the teacher about directly if the result surprised you.

If the word “behind” has entered the conversation, what “behind in reading” actually means covers the vocabulary schools use and what each term actually commits to.

How can you help at home without teaching to the test?

Work on the skill underneath the low subtest, and ignore the test’s own format. The mapping is one to one:

Subtest that was low	What to practice
Letter Naming Fluency	Letter names and sounds together, a few at a time
Phoneme Segmentation Fluency	Saying words slowly and counting their sounds, out loud, no letters
Nonsense Word Fluency	Sounding out real short words, such as CVC words
Word Reading Fluency	Sounding out, plus a small set of common words on sight

Do not drill made-up words. The publisher says it invalidates the result, and Shanahan adds in a later post that teaching phonics with them “misses (and distracts kids from) the whole point of phonics.” In the same post he says that if you are only giving one test, and you want to monitor phonics progress in kindergarten through second grade, he would choose a real word reading test. Real words are the right material at home.

Ten minutes a day on one skill is enough to start, and the National Reading Panel found that explicit, systematic work on phonemic awareness and phonics is what moves these skills, which is exactly what the first two rows of that table are.

How does DIBELS compare with Acadience, i-Ready and MAP?

They are different instruments and their scores are not interchangeable. Name the one your school used and match any chart to it.

Instrument	What it is
DIBELS 8th Edition	Short timed one-minute measures of specific early reading skills, from the University of Oregon
Acadience Reading	A separate product descending from the earlier DIBELS Next, with its own goals and its own publisher
i-Ready	A longer computer-adaptive test covering reading more broadly, not a one-minute measure
MAP Growth	A computer-adaptive test reporting a growth score across a wide range of reading skills

A score on one of these says nothing about where a child would land on another, so do not convert between them or compare a sibling’s number from a different school.

What should you ask the teacher?

Bring the report and ask about the subtests rather than the composite. Four questions that only work with the paper in your hand:

• Which subtest was lowest, and was it accuracy or speed?
• Was it the correct-letter-sounds number or the whole-words number on the nonsense word task?
• What support follows from this, and when will you check again?
• What is the one thing we should practice at home?

Is my child on track with reading? has a broader set of conference questions that work without a report, and they are worth reading alongside these rather than instead of them.

How can you check the same skills yourself, at home?

You cannot run DIBELS at home, and you can check the skills underneath it. The test is timed, given by a trained adult and scored against the University of Oregon’s published goals, so none of it transfers to a kitchen table and no home version produces a comparable number.

What does transfer is an untimed look at the same skills that ends on what to practice rather than on a score. The five-minute reading check follows a kindergarten report closely. Its first check works the listening skill PSF samples, putting spoken sounds together where PSF pulls them apart. Its second asks for letter sounds rather than the names LNF counts, its third and fourth are the sounding out NWF measures, and its fifth is the sight-word half of WRF. Stop at the first one that wobbles. That is the skill to practice, and it is the same one the teacher will check again.