Phonics Screeners and Phonics Inventories: What Each One Buys You
If you have ever tried to work out whether a phonics screener and a phonics inventory are the same thing, the honest answer is that the words do not tell you.
One instrument with the word "screener" in its title is described by its own publisher as a diagnostic. One instrument with the word "inventory" in its title is described by its own publisher as a universal screener. The names are, in the two best-documented cases, exactly backwards.
That matters more than a labelling quibble, because the two categories license different decisions. A screener in the technical sense sorts students into risk groups against a criterion. An inventory tells you which phonics elements a particular student can and cannot read. You can act on the second without ever having the first, and you cannot substitute one for the other.
This article compares the free and widely used phonics surveys and inventories against the timed, norm-referenced instruments that sit next to them, on what each publisher claims, what independent evidence exists, and what the result actually licenses you to say.
What this article covers, and what it does not
This is about instruments built around real words and element-by-element coverage: the CORE Phonics Survey, Dr. Jan Hasbrouck's Quick Phonics Screener, Really Great Reading's Diagnostic Decoding Surveys, and the computer-administered Phonics Inventory now published by HMH.
It deliberately does not re-cover the nonword measures. Pseudoword and nonsense-word tasks have their own piece on this site, and that is where the Nonsense Word Fluency family, the Test of Word Reading Efficiency and England's statutory Phonics Screening Check are treated at length. Where a phonics survey uses pseudowords inside a real-word subtest, that is noted here, but the nonword instruments themselves belong there.
It also leaves the adjacent constructs where they are. Phonological and phonemic awareness measures are a different construct with a different boundary problem. Spelling and encoding instruments are treated separately, which matters here because one of the four instruments ships with a spelling survey attached.
Nothing here is diagnostic guidance, and nothing here tells you which instrument to choose for a described presentation. This piece compares what the publishers say the instruments measure, who gathered the evidence, and what each score permits. The decision sequence is yours.
The names are backwards, and both publishers are consistent about it
Read Naturally publishes Quick Phonics Screener. Its own product description calls QPS "an informal phonics assessment that can be used to diagnose a student's strengths and instructional needs in phonics and decoding skills." Not to screen. To diagnose.
Scholastic published, and HMH now publishes, the Phonics Inventory. Scholastic's own product overview calls it "a validated screener and placement test for the most challenged readers," and the System 44 sampler describes it as "a reliable, computer-based decoding universal screening, assessment and progress-monitoring tool."
One instrument says "screener" on the box and "diagnostic" in the manual. The other says "inventory" on the box and "universal screening" in the manual. If you are choosing between them on the strength of the noun in the title, you are choosing on the one piece of information that is wrong.
So the useful question is not what an instrument is called. It is whether the instrument produces a cut score validated against an external criterion, or a skill profile validated against nothing but its own content. Those are the two products on offer, and almost every argument about phonics assessment in practice turns out to be an argument about which one somebody wanted.
The four instruments, and what each publisher says it is
CORE Phonics Survey
The CORE Phonics Survey is published inside Assessing Reading: Multiple Measures, a Consortium on Reaching Excellence in Education title issued through Arena Press, a division of Academic Therapy Publications. It is a criterion-referenced mastery measure, administered one to one, and it is untimed.
The second edition consists of twelve subtests, lettered A through L. Subtests A to D cover alphabetic knowledge, with a single scored item each: uppercase letter recognition, lowercase letter recognition, consonant sounds and long and short vowel sounds.
Subtests E to L cover reading and decoding across thirty scored items, moving from consonant-vowel-consonant words through blends, digraphs and trigraphs, r-controlled vowels, long vowel spellings, variant vowel spellings, low-frequency spellings and finally multisyllabic words. Each decoding subtest asks the student to blend letter sounds to pronounce a mix of real and pseudowords.
Administration takes roughly ten to fifteen minutes per student. There is no basal and no ceiling rule.
CORE's own assessment sequence in Assessing Reading: Multiple Measures is worth reading, because it tells you how the publisher expects the instrument to be used. In Grades 1 and 2, the Phonics Survey is scheduled "every 4-6 weeks until mastery; more frequently for at risk."
In Grade 3 it drops to "only if indicated," and in the Grades 4 to 12 diagnostic plan every instrument in the battery, the Phonics Survey included, is listed as "only if indicated."
That is a mastery-monitoring schedule, not a screening schedule. A universal screener is administered to everyone at fixed benchmark points. This one is administered repeatedly to the students who have not yet mastered the content, and to older students only when something has already flagged them.
Quick Phonics Screener, fourth edition
QPS is Dr. Jan Hasbrouck's instrument, published by Read Naturally, with the fourth edition carrying a 2025 copyright. Its items are organised into thirteen skill sets arranged from least to most difficult, beginning with letter names and letter sounds and continuing up through four-syllable words. Read Naturally states the range as kindergarten through adult.
The administration logic is adaptive by estimate rather than by rule. The examiner picks a starting skill set based on the student's age, grade and general reading performance, and the student continues through subsequent skill sets until unable to read correctly at least half the items in a set. The publisher is explicit that you do not administer every skill set to every student.
Three parallel forms exist for each assessment, which is how the publisher supports fall, winter and spring comparison. Read Naturally publishes its assessment FAQs openly, which is more than most publishers in this space do and makes the instrument unusually easy to evaluate before purchase.
The fourth edition also includes Quick Spelling Survey, an encoding measure whose thirteen skill sets are aligned to the thirteen in QPS. Read Naturally's own knowledge base describes the time-saving sequence this enables: administer QSS to a group or whole class first, then use the results to decide which students need the one-to-one QPS and which skill set to start them on.
The publisher's reasoning is that spelling a word is harder than reading it, so a student who passes a QSS skill set can almost certainly read the QPS words in the same set.
Diagnostic Decoding Surveys
Really Great Reading distributes the Beginning Decoding Survey and the Advanced Decoding Survey at no charge, as part of a set that also includes a Phonological Awareness Survey, a Letter Knowledge Survey and a Sight Word Survey. The publisher states an administration time of five to seven minutes and permits assessment of an unlimited number of students.
The two decoding surveys split by difficulty rather than by construct. The Beginning Decoding Survey evaluates mastery of high-frequency words, single-syllable words and basic phonics patterns. The Advanced Decoding Survey assesses more complex phonics patterns, multisyllabic words and advanced vowel spellings.
The publisher frames the surveys around three questions: which students are struggling with decoding, what types of words they are struggling with, and what the source of the underlying confusion is. The first of those is a screening question. The second and third are diagnostic questions, and they are the two the instrument is actually built to answer.
Really Great Reading does publish a research page for the surveys, and it is worth reading carefully for what it does and does not contain.
It states that "researchers from the Center for Reading at Pittsburg State University evaluated Diagnostic Decoding Surveys to examine their validity and reliability as measures of decoding performance" and that "the findings supported the surveys' ability to provide consistent, meaningful information about student decoding skills and instructional needs."
What the page does not give is a citation. There is no author list, no title, no date, no journal or report, and no coefficient of any kind. The page also carries an endorsement from Dr. Louisa Moats, quoted as calling the instrument "the most polished little instrument in existence."
An endorsement from a respected authority is a reasonable thing for a publisher to print, and it is not reliability evidence. If the Pittsburg State evaluation is load-bearing for a decision you are making, ask the publisher for the report itself.
HMH Phonics Inventory, formerly Scholastic Phonics Inventory
The Phonics Inventory is the outlier in this group and the most useful contrast in it. It is computer-administered rather than examiner-administered, it is timed rather than untimed, and it is built on nonwords rather than on element-by-element real-word coverage.
Scholastic's product overview states a range of Grades 3 to 12 and above, an administration time of fewer than ten minutes, and a purpose of determining which students need the most intensive intervention. What it measures is described as "fluency of non-word phonological decoding and sight-word identification of older, struggling readers."
The item design is documented in the same overview. Letter recognition items represent all twenty-six letters, in lowercase only, on the publisher's reasoning that lowercase is more challenging and more appropriate for older readers.
Sight word items draw targets from the first three hundred words on the Dolch and Fry lists, with visually similar common words as distracters, and from the first five thousand words in the American Heritage Word Frequency Book, with misspellings of the target as distracters. Decoding items use answer choices that are all nonwords following English conventions.
The System 44 sampler states that the instrument was validated against the Test of Word Reading Efficiency and the Woodcock-Johnson. Scholastic's launch announcement positioned it from the outset as a measure of the foundational reading skills of adolescent struggling readers. That is an external criterion, which is exactly what the untimed surveys do not have, and it is the whole reason this instrument gets called a screener while they get called diagnostics.
What NCII does with these instruments, and why that is the clearest evidence available
The National Center on Intensive Intervention maintains the Academic Screening Tools Chart, which is the closest thing the field has to an adjudicated register of screening instruments. What the chart's bubbles actually mean is a separate piece on this site.
Tools are reviewed by a Technical Review Committee against classification accuracy, reliability, validity, sample representativeness and bias analysis. NCII is careful to say that posting a tool "does not constitute endorsement" and that the chart "represents all tools that were reviewed, not those that were 'approved.'"
Read the chart as it stood at its August 2026 update and one thing stands out. Every reading sub-construct reviewed on it is a rate measure.
Letter Naming Fluency and Letter Sounds
Phonemic Segmentation Fluency
Nonsense Word Fluency, and Nonsense Word Fluency Correct Letter Sounds
Pseudoword Decoding - Fluency
Word Identification Fluency, Word Reading Fluency and Word Recognition Fluency
Oral Reading Fluency, Oral Reading Fluency Words Correct and Passage Reading Fluency
composite scores, comprehension subtests and vocabulary
There is no untimed phonics survey on it. There is no entry whose construct is phonics element mastery. The chart's own filter set offers Reading or Mathematics as the subject, and its shortest administration band is one to five minutes.
Now read NCII's separate Example Diagnostic Tools page, and "Phonics Inventory" appears immediately, listed under Examples of Common Diagnostic Data Sources for literacy alongside error analysis in word reading and spelling, running records, word list reading and curriculum-specific diagnostic tools. NCII attaches a caveat to that page that it does not attach to the charts:
"Unlike our behavior progress monitoring tools chart and academic progress monitoring tools chart, these diagnostic tools have not been independently evaluated by an NCII Technical Review Committee."
So the organisation that adjudicates screening evidence files the phonics inventory in a different place, for a different decision, and explicitly declines to rate it. That is not an oversight and it is not a criticism of the inventories.
It is a statement about what kind of claim each instrument can support. A classification-accuracy review requires a cut score and an external criterion. An untimed criterion-referenced survey has neither, so there is nothing for a review committee to compute.
If you have ever been asked to justify an instrument choice to a school team that only recognises what is on the screening chart, this is the distinction to make, and NCII's own two pages are the citation.
The one independent psychometric study, and what it found
There is a single peer-reviewed psychometric investigation of a phonics survey in this group, and it is worth knowing in detail because it is the only evidence of its kind.
Reutzel, Brandt, Fawson and Jones published "Exploration of the Consortium on Reading Excellence Phonics Survey: An Instrument for Assessing Primary-Grade Students' Phonics Knowledge" in The Elementary School Journal in September 2014. The study examined the second edition of the CORE Phonics Survey across test-retest, internal consistency and interrater reliability, and face, content, construct, consequential and criterion validity.
Start with what the researchers found before they collected any data. They consulted the CORE handbook in which the survey is published, found no reliability or validity evidence, and then telephoned both CORE and the publisher:
"Both sources, CORE and the publisher, confirmed they had no data on the validity and reliability of the CORE-PS and could not provide any technical information for this assessment that is being used to assess knowledge, to monitor mastery, and to design instruction for literally thousands of students."
A subsequent literature review located no published evidence either. That is the baseline: one of the most widely used phonics assessments in American schools reached 2014 with no technical documentation at all. CORE now hosts the study itself on its own site, which is to the publisher's credit and is also the only technical material it offers for the instrument.
What the study then found was mostly reassuring, and specifically unreassuring in one place.
Test-retest reliability was high. With 170 students in Kindergarten through Grade 3, the Pearson correlation for total scores across grade levels was .98, with grade-level coefficients of .95 in Kindergarten, .91 in Grade 1, .94 in Grade 2 and .95 in Grade 3.
Internal consistency split sharply by section. Cronbach's alpha was .94 for the reading and decoding subtests E to L and .92 for the full instrument, with individual decoding subtests ranging from .86 for blends with CVC to .97 for multisyllabic words. Alphabetic knowledge, subtests A to D, came in at .64, below the .70 the authors treat as the acceptability threshold.
Interrater reliability showed the same pattern. Generalizability and phi coefficients were .73 and .73 for the alphabetic section, which the authors call acceptable, and .96 and .95 for the reading and decoding section, which they call strong.
Criterion validity was estimated against DIBELS, using Nonsense Word Fluency in Kindergarten and Oral Reading Fluency in Grades 1 to 3, with 129 students. The measures used in that study have since split into two separately maintained families, the University of Oregon's DIBELS and Acadience Reading, which is worth knowing before you compare a 2014 coefficient to a benchmark published today.
Total CORE-PS scores correlated at .66 in Kindergarten, .78 in Grade 1, .86 in Grade 2, .67 in Grade 3 and .84 across Kindergarten through Grade 3, all significant at p less than .01.
Broken out by section, the decoding subtests held up at .69, .79, .86, .67 and .86. The alphabetic section did not, returning .37 in Kindergarten, minus .02 in Grade 1, .13 in Grade 2, .37 in Grade 3 and .42 overall.
The authors' own interpretation of the alphabetic finding is blunt, and they attribute it to ceiling effects and to the single-item design rather than to the construct:
"The less than acceptable alpha associated with CORE-PS subtests A-D raises serious questions about the use of subtests A-D."
They recommend that subtests A to D "may not be an appropriate measure from midyear kindergarten or for students who have already mastered the content," that the section be revised to use more than one scored item per subtest, and that if it is administered at all it be given on two occasions with two raters to reach adequate generalizability. They stop short of recommending that the section be dropped.
Two things follow for practice. The first is that the decoding half of a phonics survey is a good deal better evidenced than most practitioners assume, on reliability as well as on concurrent validity. The second is that the alphabet half is not, and the reason is a design choice rather than a measurement mystery: four subtests with one scored item each cannot do much work.
Note also the direction of the criterion study. The researchers correlated the survey against fluency measures because, as their consequential validity interviews found, that is how teachers and coaches actually use it: to troubleshoot a low oral reading fluency score.
Between 44 and 71 percent of the variance in oral reading fluency was accounted for by phonics knowledge on the survey. The instrument earns its place as an explanation of a rate problem, which is a different job from identifying one.
National guidance has treated it that way too: Student Achievement Partners' early-reading assessment guidance lists the CORE Phonics Survey among the instruments a teacher might use to work out what underlies a weak reading profile.
The Institute of Education Sciences practice guide Foundational Skills to Support Reading for Understanding in Kindergarten Through 3rd Grade, released in July 2016 and revised in December 2019, is the standard reference for the instruction such a profile would point to.
Timing is a decision the examiner makes, not a property of the instrument
The cleanest illustration that the timed and untimed categories are softer than they look comes from QPS, and from its author.
QPS was not developed as a timed assessment. The examiner's manual has always instructed examiners to attend to the rate of the student's response, and Hasbrouck's own long-standing practice was to mark whether a student responded slowly or quickly rather than to record a time. Read Naturally quotes her reasoning directly:
"Accuracy only tells us so much. If we factor in rate, which then allows us to look at automaticity, it tells us a lot more."
In 2024 Read Naturally published her change of practice. Having begun timing the assessment in classrooms where she volunteers, she reported seeing "very clear groupings" in the times, and made the point that two students with identical accuracy are not identical students: "Two kids can get everything correct, but if one completes the assessment in 17 seconds and another completes it in 45 seconds, we know those are two really different kids."
The fourth edition acted on this. It carries guidance in the examiner's manual on how to time the assessment, and the score sheets now include places to record students' times.
Hasbrouck also attached a caution that is easy to skip and is the operative part for anyone working one to one. Examiners should be nonchalant about the timing and should tell students the assessment is not about speed. In her words, "Faster recognition is a sign of mastery, but we are not trying to teach students to do anything faster."
Three things are worth drawing out of this.
Recording a time does not turn a criterion-referenced inventory into a norm-referenced screener. There are no normative time bands for QPS. What the time buys is a within-student, within-skill-set read on automaticity, which is clinical information rather than classification information. Decision rules built for rate measures do not transfer to it.
Conversely, the instruments that genuinely are timed screeners are timed because their norms depend on it. A rate measure's cut score is meaningless without the standard administration window, which is why those instruments are far less flexible than an inventory and far more defensible as a sorting device.
And the field's two categories are therefore not a property of the stimulus material. The CORE survey and QPS use overlapping content. What separates a screener from an inventory is whether a cut score was validated against an external criterion under fixed conditions, and nothing about the word list tells you that.
What each result actually licenses you to say
Setting the marketing aside, here is what you can defend from each kind of result.
From an untimed criterion-referenced survey or inventory, you can say which phonics elements the student read correctly and which they did not, on this date, under these conditions.
You can use that to select a starting point for instruction, to decide which patterns to teach next and in which decodable material, and to document change on a later form of the same instrument. With a QSS-style companion you can also say whether the student can spell the patterns they can read, which is a genuinely different question.
You cannot say the student is at risk, is in the bottom whatever percentile, or meets a criterion for anything. There is no norm group and no validated cut.
From a timed screener with published classification accuracy, you can say that the student's score falls in a risk band the publisher has validated against a criterion, and you can say what proportion of similarly scoring students in the validation sample went on to a poor outcome, to the extent the technical documentation reports it.
You cannot say which phonics elements the student does not know, and you cannot read a skill profile out of one the way a dyslexia screening battery is sometimes assumed to deliver. A rate score on nonwords does not decompose into a skill profile, which is precisely why NCII keeps a separate diagnostic list.
And from either one, you cannot say the student has a disorder. Identification is a different process with different requirements, and the Association of Educational Therapists Code of Ethics is specific about the boundary.
Section II.I of the code governs what an educational therapist may and may not do with assessment. The AET fact sheet's list of things an educational therapist does not do includes administering "cognitive, intelligence, or psychological tests (unless otherwise qualified to do so)."
Reading a phonics survey is not on that list. Converting one into a diagnosis is a different act.
Where the published record runs out
Several things in this comparison are simply not publicly documented, and it is better to say so than to fill the gap.
No independent psychometric study of QPS or the Phonics Inventory was located in preparing this piece on a par with the 2014 study of the CORE survey. For the Diagnostic Decoding Surveys, Really Great Reading states that a validity and reliability evaluation was carried out at Pittsburg State University but publishes no reference to it, so the study could not be read and its findings could not be checked.
The HMH research page for the Phonics Inventory renders its content through client-side scripting and served no readable text to an automated fetch, so the specific validation coefficients behind the Test of Word Reading Efficiency and Woodcock-Johnson comparison could not be read directly.
The claim that the comparison was made is quoted above from Scholastic's own sampler page. If a particular coefficient is load-bearing for a decision you are making, request the technical manual from the publisher rather than relying on a product page, including this summary of one.
The 2014 study examined the second edition of the CORE Phonics Survey. Assessing Reading: Multiple Measures has since been issued in a revised second edition, and the sample pages CORE publishes do not state whether subtests A to D were redesigned in response to the published criticism.
Whether the single-item alphabetic design survives in the current printing is a question for CORE, and it is the single most useful question anybody could ask them.
Nothing published by any of these four publishers states a false-positive or false-negative rate for an untimed phonics survey used as a screener, because none of them claims that use. If a school asks you to use one that way, the absence is the answer.
Frequently asked questions
Is a phonics screener the same thing as a phonics inventory?
Not reliably, and the titles are not a guide. Read Naturally describes Quick Phonics Screener as an informal assessment used to diagnose, while Scholastic describes the Phonics Inventory as a validated universal screener. The distinction that holds is whether the instrument produces a cut score validated against an external criterion under timed, standard conditions, or an element-by-element skill profile judged against its own content.
Can I use the CORE Phonics Survey as a universal screener?
The publisher's own assessment sequence does not use it that way. In Assessing Reading: Multiple Measures it is scheduled every four to six weeks until mastery in Grades 1 and 2, and only if indicated in Grade 3 and in Grades 4 to 12. It also appears nowhere on NCII's Academic Screening Tools Chart, which reviews classification accuracy, and "Phonics Inventory" appears instead on NCII's unrated Example Diagnostic Tools page.
How long does each instrument take to administer?
Really Great Reading states five to seven minutes for its Diagnostic Decoding Surveys. The 2014 study reports roughly ten to fifteen minutes per student for the CORE Phonics Survey. Scholastic states fewer than ten minutes for the Phonics Inventory. QPS has no single figure, because the examiner administers only the skill sets a given student needs.
Is the CORE Phonics Survey reliable?
For its decoding half, the available evidence says yes. The 2014 Elementary School Journal study reported test-retest reliability of .98 for total scores across Kindergarten through Grade 3, Cronbach's alpha of .94 for subtests E to L, and interrater generalizability of .96. The alphabetic section, subtests A to D, returned an alpha of .64 and weak criterion validity, and the authors recommend interpreting it with caution and revising it.
Should I time an untimed phonics assessment?
The author of QPS now recommends it for her own instrument, with the fourth edition adding manual guidance and time fields on the score sheets. What a time buys is a read on automaticity within a skill set, not a normative score, because there are no published time bands. Her own caution applies: be nonchalant about it, and do not communicate to the student that speed is the goal.
Which instrument should I use?
That is not a question this article answers, and no comparison of publisher claims can answer it. What the comparison does settle is the decision sequence.
Establish first whether the question in front of you is a sorting question or an instructional question, then check whether the instrument you are considering has the kind of evidence that question requires, and then read what the publisher says the instrument is rather than what the title says.
One more thing
The most useful thing in this whole comparison is also the least glamorous. Two national resources exist, maintained by the same federally funded centre, and one of them rates instruments while the other explicitly does not. Knowing which list your instrument is on, and why, settles most arguments about what you are allowed to conclude from it before they start.
That pattern repeats across the assessment pieces on this site, and the short version of each is worth having in one place. Nonword and pseudoword measures buy a rate score and not a skill profile. Phonological awareness instruments stop where the construct stops. Oral reading fluency norms are a sample, not a standard.
The NCII tools charts rate what was submitted rather than what exists. Grade equivalent scores should not be reported at all.
And the difference between a norm-referenced and a criterion-referenced result is the difference this whole article turns on. The IDA Knowledge and Practice Standards are where the expectation that a practitioner can read these distinctions is written down. If you are working out where instrument selection sits in your own scope of practice, the role comparison is the place to start.
If you are building out an assessment practice and want more of this kind of comparison, join the Illuminate educator list and it will reach you there first.



Comments