Research Edition
For schools: sits beside your school's existing literacy screener — not instead of it. Covers oral phonological awareness only (no letter naming, nonsense words, or ORF). Informal screening aid for SLP / early childhood caseloads; not a universal district screener and not for eligibility decisions.
Screen a child's phonological awareness. Build a book around the one skill that needs practice. Read it together. Screen again in six to eight weeks.
Bamm is built to close that loop, and it tells you honestly how certain it is at every step — including the step that still depends on a person.
Open the Research Edition screenerJoin the pilotKDP book tools (separate)
What this is: an informal screening aid for educators and clinicians, plus a book engine that turns results into practice material. What this is not: a diagnostic instrument. No screener can diagnose a reading disability, and this one doesn't claim to. It tells you where to look — not what to conclude.
If you work with three-to-seven-year-olds, you already know the shape of this problem. Here is how specific it gets.
Screening has to reach the settings where young children actually are. Many early-literacy screening tools are built around school systems and district purchasing. Bamm is built so a family childcare program or a small preschool can screen phonological awareness too: it runs in a browser, the screener is free, and there is no district contract.
A label alone leaves out the uncertainty. A screening result is often summed up as a risk band or a yes/no: "at risk" or "not at risk." On a few items per skill, how sure that call is matters as much as the call itself, so Bamm prints how confident each result is right beside it. The person deciding whether to refer a child gets the result and the uncertainty together.
A result should lead to practice. When a screening flags initial sound isolation, the useful next step is practice material for that child. Many tools focus on letters; Bamm focuses on how young children hear and say sounds — rhyming, initial sound, blending, and segmenting. Phonological awareness is oral. It comes before print.
And a parent needs a next step. Bamm produces a screening report a parent can hand to a pediatrician or a speech-language pathologist to start a real conversation about an evaluation.
Screen, intervene, re-screen is the loop every reading specialist already runs in their head. Bamm is built to run it with them.
There are two screen lengths, and the short one is the default. It runs two parts chosen for the child's age band — five items each, ten items in all, and it does not include the instruction-comprehension check. The pair changes with the band: syllable blending and rhyme recognition at 3–4, syllable segmentation and rhyme recognition at 4–5, first sounds and sound blending at 5–6, and blending and segmenting from 6 up. The setup screen names the two before you start. It is a between-evals check, not a profile, and every result it produces is stamped short mode.
The full PA profile is the other option, and it is the one that covers all eight skills — syllable blending, syllable segmentation, rhyme recognition, rhyme production, initial sound isolation, sound blending, final sound isolation, phoneme segmentation — forty items, about 20 to 25 minutes. The parts run in that order, earliest-developing first, so a child who stops half way through has finished the ones most likely to tell you anything. Choose it on the setup screen when you want the whole picture rather than a spot check.
Every item comes with a verbatim script, so an SLP and a preschool teacher giving a first screening say the same words — whether that produces the same results is one of the first things the pilot will measure. In the full profile you can also add an optional instruction-comprehension check, five items that run before the skills, because a three-year-old who misses rhyme items may not struggle with rhyme — she may not have understood the directions. If that check is failed, the report downgrades the skill findings and says so. What it does not do is tell you which explanation is the right one: it flags the doubt, and a person resolves it.
Each skill gets a plain-language finding and a certainty band: how confident the result is, given how few items a young child can reasonably sit for. Wide error bars are shown, not hidden. The report says where to look. It never says what a child has.
Say a five-year-old's results flag initial sound isolation — she can clap syllables and hears rhymes, but can't tell you that ball starts with /b/. The screener pairs that finding with a book recipe — for initial sounds, an alphabet-style book that repeats a few target sounds at the start of words on every page. You write or adapt the story around that recipe, and the book engine handles everything print-related, delivering print-ready files you print or have printed. If her name is Mara and her target is /m/, her name can carry the sound — Mara and the mittens. Mara meets a mouse. If her name is Jaxon, the sound lives in the words around him instead. Her world, the sound she's practicing, on paper she can hold. Not a worksheet, not a screen — a book with her in it.
Today, a person checks that the finished story really does exercise the target sound. Doing that check automatically is the hard part — see what isn't solved yet.
The book is practice and exposure, woven into something a child asks to read again. It is not a treatment, and we won't call it one. After six to eight weeks of reading together, run the screener again and load the saved session file: the tool shows both screenings with their intervals and says plainly when it cannot tell whether anything changed — which, on five items a side, is often. Now you have two honest observations instead of one label. It is not a measure of growth or of response to intervention, and it does not attribute a change to the book; what it gives you is a second look, and a clearer basis for deciding whether to bring in a specialist, with paperwork ready to hand over.
Phonological awareness is oral. A practice book works because an adult reads it aloud, so what counts is what the child hears: whether the words on a page really carry the target sound, in the right place, often enough, in words short enough for a young child to attend to each sound. Those are properties of spoken language, not spelling — and checking them automatically means knowing how every word on the page is pronounced.
Personalization makes that harder, because the book is about one child, by name, and standard English pronunciation dictionaries don't cover many of the names children actually have. We checked the most widely used one, CMUdict, against the 2,060 names on the Social Security Administration's top-1,000 lists for boys and girls born in 2021, 2022, and 2023: 28.8% have no entry, including 13 names that ranked in the top 100 on at least one of those lists — among them Camila, Gianna, Jaxon, and Thiago. That measures how often a system has to guess at a pronunciation, not how often it guesses wrong. Either way, a guessed pronunciation is a page no one can automatically confirm.
The difficulty isn't uniform, either: how often ordinary children's text happens to meet a sound target varies widely from one sound to the next, and the screener — not the book — chooses the sound. So two kinds of work sit ahead. One is building a verification layer that can confirm a personalized page meets its target. The other is measuring the screener in real hands: whether two examiners reach the same result, and how its results compare with established measures.
Every result carries a Wilson 95% interval on five items — wide by design — and a band you can read without a statistics course. These are the four bands, in the words the report actually prints: Narrower interval (still small n — not a normed score), Wide interval (few items — not a normed score), Provisional — do not treat as definitive, and Insufficient data. They name the width of the interval, not confidence in the child, and none of them is a normed score.
Beside the band, each part carries one reading, and the report, the email summary and the exported file all print the same one: Practice focus (below provisional cutoff); Below cutoff · not yet expected at this age · not a target; At or above provisional cutoff (small sample — interpret cautiously); or, when the screen could not read the part at all, No answer on any item · cannot tell, Same answer every time · cannot tell, Too few items · cannot tell, or Not administered. A part the screen could not read is never reported as a target, and never seeds a book or a practice page. We put the error math on the report, because you're the one making the decision.
This is a book built around the child — her name, her world, the sound she's practicing — delivered print-ready, printed into something she keeps, rereads, and shows people. Practice a child asks for is practice that happens.
When results suggest a closer look, the screener generates a screening report: what was screened, what was found, how certain, and what it means in classroom language. A parent can hand it to a pediatrician, the school district's Child Find office, or a certified SLP and start a real conversation. The packet also records who gave the screen and under what conditions — self-reported, and printed beside the certainty band so the reader can weigh it.
Family childcare programs and independent preschools are where a large share of American three-to-five-year-olds spend their days, and Bamm is built with those settings in mind. It runs in a browser, requires no district contract, no credential gate, and no child data ever leaves the device.
Speech-language pathologists. A quick, scripted, informal probe between formal assessments — and a way to send a family home with individualized practice material instead of a photocopied word list. Your formal instruments do what this never will; this fills the weeks between them.
Reading specialists and interventionists. A pre-K–to–grade-2 progress loop for the oral skills that come before decoding, with re-screening built into the design rather than bolted on.
Early-childhood educators and family childcare providers. If you've wondered about a child's ear for sounds, this is built for your setting first — scripted administration, plain-language results, and a clear map of when and where to refer.
Researchers and pilot partners. The item bank, cutoffs, and book mappings are all revisable data, and the platform is entering its pilot phase. If you study early literacy or screening in non-district settings, we want to work with you — see below.
Most product pages don't have this section. On a page about screening young children, we think the limits are the product information.
It is not diagnostic. No screening tool — this one included — can diagnose dyslexia, a language disorder, or any other condition. Diagnosis takes a qualified professional, a comprehensive evaluation, and far more evidence than any 20-to-25-minute screen can gather. If results here raise a concern, the right next step is your pediatrician, your school district's Child Find / early intervention program, or a certified speech-language pathologist. The screening report exists precisely to make that handoff easier.
It is not norm-referenced — yet. There are no national norms behind these scores. A single provisional cutoff — 60% of the items administered — applies to every skill, and the age at which each skill is expected is informed by the developmental research literature. Both are provisional. A pilot can't produce norms — that takes a large, representative sample — but it is the first step toward the evidence norms require. Until then, results are a structured, honest observation — not a percentile.
Five items per skill means wide error bars, on purpose. A three-year-old will not sit for a fifty-item battery, and pretending otherwise produces garbage data politely formatted. We chose brevity and then refused to hide its cost: the confidence intervals on every result show you exactly how much five items can and cannot tell you.
It is entering its pilot. The pilot is recruiting now. We are looking for practicing SLPs as co-design partners, and their review of the item bank is one of the pilot's first jobs. Item content and cutoffs will change as pilot data comes in.
Why say all this on a marketing page? Because the entire premise of this platform is that the people making decisions about children deserve the uncertainty, not just the verdict. That has to start here.
No. This is an informal screening aid, not a diagnostic instrument. It can suggest that a particular skill deserves attention and practice; it cannot determine why, and it cannot identify dyslexia or any other condition. Screening tells you where to look. Diagnosis — which requires a comprehensive evaluation by a qualified professional — tells you what's there. If you have a concern about a child, talk with your pediatrician, contact your school district's Child Find / early intervention program, or consult a certified speech-language pathologist.
Ages 3 through 7, with expectations adjusted by age band. What's just emerging for a three-year-old (rhymes beginning to click) is very different from what's expected of a six-year-old (segmenting a word into individual sounds), and the results are framed accordingly.
The item bank covers eight distinct phonological awareness skills: syllable blending, syllable segmentation, rhyme recognition, rhyme production, initial sound isolation, sound blending, final sound isolation, and phoneme segmentation. The default short screen runs two of them chosen for the child's age band — ten items — and the full PA profile runs all eight. Note what that list is not: it isn't phonics (letters and print), and the broader set isn't the same thing as phonemic awareness (the phoneme-level subset). Phonological awareness is oral — hearing and playing with the sound structure of spoken words, before print enters the picture.
The default is a short between-evals screen: two parts, five items each, about 5 to 8 minutes, with no instruction-comprehension check. The full PA profile covers all eight skills in about 20 to 25 minutes, and you can add the optional five-item instruction-comprehension check to it. Any part can be ended early when a child has had enough, and the report says which parts that happened to. Results are stamped with the mode that produced them.
One page per child for the examiner, and a separate one-page Family Page that says in plain words what was listened for, what the child already does well, what will be practiced and how sure we are, and when we plan to listen again — never a percentage, never a label. The family page carries a DRAFT watermark: its wording has not been reviewed by a licensed speech-language pathologist.
On the examiner's page each part carries one reading — Practice focus (below provisional cutoff), Below cutoff · not yet expected at this age · not a target, At or above provisional cutoff (small sample — interpret cautiously), or one of the "cannot tell" readings when the screen could not read the part — plus a certainty band in the words listed above, the confidence interval behind it, and a sentence about what the skill looks like in a classroom ("may not notice sound patterns in songs and chants," not a jargon code). Nothing is normed and nothing is for eligibility decisions. If an articulation or hearing concern was noted at setup, every scored part stays Provisional. If the optional instruction-comprehension check was run and failed, findings are downgraded and the report says so plainly.
The screener is free. No account, no licence, nothing to buy — it runs in your browser and you can screen as many children as you like. Printing the report, the family page and the practice pack is free too.
The one thing that costs money is a book: $19 one-time per book, which buys the print-ready files and unlimited re-downloads of that book. You build and preview a book for free and only pay at download. There is no subscription, no school licence, and no site or seat pricing — those do not exist, and if you need something other than $19 a book, write to us and say so.
Yes, though it's designed for educators and administration matters — the scripts should be read exactly as written, in a quiet moment, without coaching. Treat the results as a conversation starter, never a conclusion. And a result that looks fine does not override a parent's gut: if you're concerned about your child's speech or language, skip the screener entirely and talk to your pediatrician, your school district's Child Find / early intervention program, or a certified SLP. Trust your concern — it's the best screener in the building.
Structurally, not by promise. The screener runs entirely in your browser: no account, no login, no child names — sessions use participant codes you assign (a code is still private student information). Nothing is transmitted to us or to anyone, and analytics is switched off on the screener pages. When you export or print a report, the file goes to your device and nowhere else. We never see a child's data because the system is built so we can't.
Run a screening. It's free, the default short screen takes about 5 to 8 minutes (the full profile about 20 to 25), and there's no account to create. Try it on a colleague first to get comfortable with the scripts — then screen a child whose ear for sounds you've been wondering about.
Or help us build the evidence. We're recruiting pilot partners — SLPs, reading specialists, preschool and family childcare programs, and early-literacy researchers — to refine the item bank, test the screen-build-re-screen loop in real settings, and build toward proper validation. There is nothing to buy either way: the screener is free to everyone, and a pilot partner's part is reviewing the items and telling us what breaks in real use.
Bamm Research Edition is built in Springfield, Oregon by Sara Larson — director and lead teacher of a licensed family childcare program, with 25 years in early childhood education, and a published children's author. It comes from her own work with young children in a family childcare setting. The screener is an informal screening aid, not a diagnostic instrument, and this page will always say so.
New to this topic? Start with our plain-English guide: Phonological awareness for ages 3–7 — what's typical, what to play, and when to ask for help.