Orthographic Processing: A Composite With No Settled Construct
Updated: 2 days ago
The short version
If you are shopping for an orthographic processing assessment, the first thing to know is that the three major publishers do not agree on what one is. Pearson sells an Orthographic Processing composite on the WIAT-4. PRO-ED sells a Test of Orthographic Competence. Riverside Insights, revising its battery after both of them, shipped the WJ V with no orthographic cluster at all.
That is not a small difference in emphasis. It is a disagreement about whether the thing exists as a separate score.
And when you open the two instruments that do sell it, the components are mostly spelling. The WIAT-4's composite is built from its Spelling subtest plus a timed irregular-word reading subtest. PRO-ED tells you in print that the score to use if you are "specifically interested in orthographic processing" is a composite of four spelling subtests, and the composite is called Spelling Efficiency.
Meanwhile the 58,915-word federal practice guide on foundational reading skills does not contain the string "orthograph" once, and neither does either of the two federal tools charts.
This post is about what you are actually buying, what the score licenses, and what it does not.
Two publishers print the word, and they define it from different sources
Start with the definitions, because they are both in print and they are not the same.
Pearson's WIAT-4 technical manual defines the Orthographic Processing composite as "an overall measure of orthographic processing, including the size of an examinee's orthographic lexicon and the quality of orthographic representations." That wording is quoted from the manual at page 114 in Beaujean and Parkin's 2022 evaluation of the instrument, published in the Journal of Intelligence under DOI 10.3390/jintelligence10020030.
Pearson's own launch flyer for the fourth edition lists Orthographic Processing among the new composite scores and says the edition "goes beyond the what to uncover the why."
PRO-ED defines it differently, and attributes the definition. Its TOC-2 product page describes orthographic processing as "the ability to acquire, store, and use both mental graphemic representations, as well as knowledge of orthographic patterns," and cites Apel (2011).
Those two definitions overlap but they are not interchangeable. One is about the size and quality of a store of word spellings. The other is about mental graphemic representations plus pattern knowledge, which is a two-level construct.
Hold on to the Apel citation. It comes back later in this post, and when it does it undercuts the thing it was cited to support.
The WIAT-4 composite is two subtests, and one of them is Spelling
Here is the composition, from Pearson's own Canadian product flyer, which lays the composite structure out as a table.
The Orthographic Processing composite comprises Spelling and Orthographic Fluency. Spelling is the same subtest that sits inside the written expression domain. Orthographic Fluency is new to the fourth edition and Pearson describes it, in one line on its WIAT-4 store page, as measuring "speed of irregular word reading."
So the processing composite is a dictation spelling test plus a timed sight-word reading test.
That is worth sitting with for a moment. Nothing in either task is a measure of a process in the sense a practitioner usually means — something upstream of reading and spelling that explains them. Both tasks are reading and spelling. The composite is an aggregate of two academic skills, filed under a processing heading.
Beaujean and Parkin make the structural point plainly: both orthographic composite scores "involve the Orthographic Fluency subtest as well as the Spelling subtest, the latter of which is part of the writing domain."
The third subtest exists only if you bought the digital platform
There is a third orthographic subtest, and whether you have access to it depends on a purchasing decision you probably made for unrelated reasons.
Orthographic Choice shows the student a word and a same-sounding misspelling and asks which is correct. Pearson describes it as measuring "recognition spelling skills." Add it to the other two and you get the Orthographic Processing Extended composite.
It is available on Q-interactive only. Pearson's store page marks the subtest with an asterisk and prints the footnote directly underneath it. The subtest line reads "measures recognition spelling skills"; the footnote reads "Q-interactive only." The benefits section says the same thing in full: "Expand the Orthographic Processing composite into an extended composite, using the digital Orthographic Choice subtest."
The practical consequence is that two practitioners can both report a WIAT-4 Orthographic Processing score and mean different composites. One is reporting two subtests. The other is reporting three.
The construct a score represents should not depend on which platform licence your practice happens to hold. On the WIAT-4 it does.
If you are reading someone else's report, this is a thing to check rather than assume. The composite name does not tell you which version you are looking at, and the subtest list does.
PRO-ED's answer to the same question is a composite called Spelling Efficiency
PRO-ED's TOC-2 yields five composites. Four of the five have "Spelling" or "Conventions" in the name. The fifth is Orthographic Knowledge, which pools all six subtests.
Then the product page does something unusually direct. It tells you which composite to use if orthographic processing is specifically what you are after, and it is not the one called Orthographic Knowledge:
"The Spelling Efficiency composite was created for use by those who are specifically interested in orthographic processing (i.e., the ability to acquire, store, and use both mental graphemic representations, as well as knowledge of orthographic patterns [Apel, 2011])."
Spelling Efficiency combines four subtests: Sight Spelling, Homophone Spelling, Word Scramble and Letter Choice. All four are spelling tasks. Two are timed.
So the publisher of the field's dedicated orthographic instrument tells you, in print, on the page where it sells the test, that the orthographic processing score is a spelling composite.
That is not a criticism of the TOC-2. It is a well-documented instrument and the candour is to its credit. It is a statement about the construct: when two independent publishers operationalise orthographic processing, both of them arrive at spelling.
What counts as orthographic at PRO-ED that does not count at Pearson
The six TOC-2 subtests, and what each one asks a student to do, are set out in our spelling assessment post, which has a section on them. There is no point restating it here.
What that post does not do, and what matters for this one, is compare where the two publishers draw the boundary of the construct.
PRO-ED draws it wider. Its Orthographic Knowledge composite includes Punctuation and Abbreviations, which form a Conventions composite that PRO-ED itself describes as "the non-spelling (and sometimes not even alphabetic) aspects of the English writing system."
Read that against the Pearson composite and the gap is stark. At PRO-ED, knowing that "Dr." abbreviates Doctor and that a sentence takes a question mark falls inside orthographic knowledge. At Pearson, orthographic processing is irregular-word reading speed and spelling, and punctuation is nowhere near it.
Both are defensible as construct definitions. They cannot both be the same construct.
There is a second asymmetry worth naming. PRO-ED's Spelling Fluency composite exists to be compared against Spelling Accuracy, and the page says what the comparison is for: to determine "whether a student has developed accurate orthographic images but does not process these visual images quickly or automatically." That accuracy-versus-speed dissociation has no equivalent on the WIAT-4 composite, which blends an untimed and a timed subtest into one number.
The WJ V has no orthographic cluster, and where it files spelling instead
Now the control case, and it is the strongest single piece of evidence in this post.
Riverside Insights revised the Woodcock-Johnson after the WIAT-4 appeared. Its support site publishes a complete list of new and updated clusters in the WJ V, separated into clusters that are entirely new and clusters carried over with changes.
There is no orthographic cluster on that list. There is no orthographic cluster anywhere in the battery.
Counted today, the string "orthograph" appears zero times in that cluster document, zero times in the WJ V brochure, and zero times in Riverside's WJ V achievement-to-CHC mapping document, which exists specifically to tell you which broad ability each achievement test measures.
What Riverside did add is a new achievement cluster called Spelling Skills, and it tags that cluster Grw, Ga — reading and writing, plus auditory processing. Not visual. Not orthographic.
And the test a practitioner might reach for as the WJ V's orthographic proxy, Letter-Pattern Matching, sits in Cognitive Processing Speed (Gs). Riverside files it as a speed test, not a literacy one.
Two publishers sell orthographic processing as a composite. The third revised its battery, declined to create the cluster, and filed spelling under auditory processing instead.
That is a three-way disagreement about what kind of thing orthographic processing is, between the three companies whose instruments populate most psychoeducational reports in the United States.
Pearson runs the same task under two different headings, in two of its own products
This one is a single-publisher finding and it is the cleanest illustration of the problem.
The PAL-II Reading and Writing, authored by Virginia Wise Berninger and still listed by Pearson as current, includes a subtest called Word Choice. Pearson's own applying-PAL-II-to-practice paper describes it as "a measure of recognizing a correct word-specific spelling among choices all pronounced the same."
That is the same task as WIAT-4 Orthographic Choice. Word, same-sounding foil, pick the real spelling.
On the WIAT-4 it is an orthographic processing subtest. On the PAL-II it is a spelling screen. The same paper, in its grade-level screening tables, pairs "PAL II RW Word Choice (spelling by recognition)" with "WIAT 4 Spelling (spelling from dictation)" as two ways of getting at spelling.
So Pearson's own instructional guidance treats the recognition task as a spelling measure, while Pearson's own achievement battery sells it as orthographic processing.
The PAL-II paper also contains a sentence that the orthographic processing framing tends to obscure, and it comes from the instrument's author: "Although phonological and orthographic skills are necessary, they are not sufficient for English spelling, which is a morphophonemic orthography." Morphology is the missing third, and no orthographic composite on either instrument contains a morphology subtest.
An independent factor analysis of the WIAT-4 did not find an orthographic factor
Pearson's theoretical framework splits word reading into cipher skills and word-specific knowledge. The orthographic composites are the word-specific knowledge side of that split. It is a coherent model and it is the reason the composite exists.
Beaujean and Parkin tested it, and the paper is free in full text both at PMC and as an ERIC record. Working from the correlation matrices in the WIAT-4 technical manual, they ran their own exploratory factor analyses across four age groups: 4 to 7, 8 to 11, 12 to 19 and 20 to 50. Their conclusion on the reading side is one sentence long:
"The word decoding/reading subtests do not dissociate into cipher skills and word-specific knowledge, but instead all hang together along with the Oral Reading Fluency subtest."
The loadings say the same thing. In the five-factor oblique solution, Orthographic Fluency loads .99, .76, .23 and .79 on the first factor across the four age bands. Word Reading loads .98, .77, .76 and .99 on that same factor. Pseudoword Decoding loads .72, .86, 1.07 and .87. Orthographic Choice loads .81, .58, .35 and .64.
They are all on one factor. The orthographic subtests do not separate from the decoding subtests at any age.
Two honest qualifications. The authors describe the resulting structure as "messy" rather than clean in the other direction, and one loading above 1.0 is flagged in their own notes as a Heywood case, which is a sign of an unstable solution rather than a real value.
Their broader recommendation is cautious about nearly every WIAT-4 composite, not only this one: they advise psychologists to limit interpretation "to just a few in the mathematics and reading domains."
But the specific finding holds, and it is the finding that matters here. On the instrument's own standardisation data, the orthographic subtests behave as reading subtests.
The orthographic choice task: what the research actually tested
The recognition task has been studied directly, and the study is the most useful thing a practitioner can read on this subject.
Compton, Gilbert, Kearns and Olson (2020), in Annals of Dyslexia, volume 70, page 243, under DOI 10.1007/s11881-020-00202-0, took 296 participants aged 8 to 18 from the Colorado Learning Disabilities Research Center twin sample, gave them an 80-item orthographic choice task, and separately asked them to read the same 80 target words aloud in a list.
Then they asked a precise question: how much of performance on the choice task is explained by whether you can already read that specific word?
The answer, from their crossed random-effects item-response models:
When a participant read the target word correctly in isolation, the probability of picking the right spelling was .90.
When a participant read the target word incorrectly, the probability of picking the right spelling was still .79.
Chance on a two-option forced choice is .50. So .79 is well above guessing, and the authors read that as evidence the task is not purely a word-reading test. Their own framing of the .90 figure is blunter: it suggests "item-specific word recognition and performance on the OC task are one in the same."
Their overall interpretation is a both-and. The task "taps both item-specific orthographic knowledge and more general orthographic knowledge." Receptive spelling, phonemic awareness, word frequency and orthographic neighbourhood size all accounted for significant additional variance after target-word reading was controlled.
Bound that carefully. This is one sample of 296, roughly a third of whom had a school history of reading difficulty, on one version of one task, and it is a construct-validity study rather than an evaluation of any commercial subtest. It does not tell you what a WIAT-4 Orthographic Choice scaled score means for the student in front of you.
"Difficult if not impossible to distinguish from tests of word recognition"
The Compton paper also does the field a service by collecting the criticism in one place, with citations, and the summary line is from Castles and Nation.
"Measures of orthographic processing skill are difficult if not impossible to distinguish from tests of word recognition."
That is quoted in Compton et al. from Castles and Nation (2017), page 152. The same passage reports that orthographic choice measures correlate with measures of word reading in the range of .50 to .70, and names Burt (2006), Castles and Nation (2006) and Vellutino and colleagues (1994) as having criticised the task extensively for tapping word reading rather than orthographic knowledge.
A correlation of .50 to .70 is not identity. Two measures correlating at .60 share a bit over a third of their variance. There is room for a distinct construct in what is left over, which is roughly the argument Compton and colleagues make.
What there is not room for is a clinical inference that treats a low orthographic composite as an explanation of poor word reading, when the two measures are that closely related and the direction of the relationship is unresolved.
The authority the publisher cites later published that the construct is inconsistently measured
Recall that PRO-ED attributes its definition of orthographic processing to Apel (2011) — What Is Orthographic Knowledge?, published in Language, Speech, and Hearing Services in Schools, volume 42, issue 4, pages 592 to 603.
Seven years later, Apel returned to the question with Victoria Henbest and Julie Masterson in Orthographic knowledge: clarifications, challenges, and future directions, in Reading and Writing, under DOI 10.1007/s11145-018-9895-9. The abstract states the problem in its own words:
"The terms used to describe the two levels of orthographic knowledge, lexical and sublexical orthographic knowledge, have been variably defined and inconsistently measured, potentially contributing to discrepancies in research findings."
The paper goes on to discuss "dissimilarities in how orthographic knowledge has been operationally defined and measured and the associated differences in tasks used to assess that construct," and offers suggestions for what researchers might do "to help move the field toward a better understanding of orthographic knowledge."
That is the author of the definition on the product page, writing that the construct is variably defined and inconsistently measured, and that the field needs to move toward a better understanding of it.
There is nothing improper in a publisher citing a 2011 definition. But if you are relying on that definition because it looked authoritative on the product page, the same authority's later and more cautious position is the part you want.
The word does not appear in the federal evidence base
This is the pattern this site keeps finding, and it is unusually clean here.
The IES and What Works Clearinghouse practice guide Foundational Skills to Support Reading for Understanding in Kindergarten Through 3rd Grade is the federal government's recommendation document for exactly this territory. Extracted and counted today, it runs 58,915 words.
The string "orthograph" appears in it zero times. Not as a whole word, not as a substring. "Orthography" zero. "Visual memory" zero. "Word-specific" zero. "Lexical" zero.
Against that: "phonics" 88 times, "spelling" 80 times, "phonological" 49 times.
The guide is not silent about the behaviour. "Irregular word" appears 8 times, "sight word" 6 and "whole word" 6. It discusses teaching children to read words that do not follow regular patterns. It simply never names a construct called orthographic processing, and it names no instrument for measuring one.
The two federal tools charts tell the same story. The academic screening chart and the academic progress monitoring chart at the National Center on Intensive Intervention both contain zero occurrences of "orthograph." The screening chart's Subject filter offers two options, Reading and Mathematics. There is no orthographic domain to filter on.
The construct is a product category, not an evidence category. It is sold by publishers and it is absent from the federal documents that rate the tools.
Norms, age bands and what each instrument actually covers
If you have decided you want one of these scores anyway, the practical facts differ and they are easy to check.
The WIAT-4 was published in 2020, covers ages 4:0 to 50:11, and is Qualification Level B at Pearson. The orthographic composite is available from the print kit in its two-subtest form; the extended three-subtest form is Q-interactive only.
The TOC-2 covers ages 8:0 through 24:11 and takes 30 to 40 minutes. It has a single form usable individually or in groups, which is unusual in this cluster. PRO-ED states its norms are based on 1,512 students tested "from Summer of 2017 through early Spring of 2020," stratified to the 2021 US school-age population.
Two things follow from that norming window. Those norms are now five to nine years old, which is well inside the usable range but worth knowing. And the sample was collected entirely before the spring 2020 school closures, which is a point in its favour for stability and a point against it for currency, depending on what you are arguing.
One cross-seller discrepancy, which this site keeps finding and which is always worth a look. PRO-ED's complete kit carries a 2023 copyright line. Hogrefe, which distributes the TOC-2 in the United Kingdom, lists its publication date as 2022 and its own qualification requirement as Level 2, on a scale that has nothing to do with PRO-ED's letters.
If you need the edition year for a report, take it from the examiner's manual rather than from a storefront.
Hogrefe's listing also contains a phrase worth noticing: the TOC-2's age and grade equivalents "translate subtest raw scores into orthographic ability ages." An orthographic ability age is a strong construct claim resting on a score composed of four spelling subtests. For what is wrong with age and grade equivalents generally, our post on grade equivalent scores is the longer answer.
Can an orthographic processing score identify dyslexia or dysgraphia? No
Flatly, no, and this is the question that brings most practitioners to this subject.
No score on either instrument identifies a disorder. The WIAT-4's own stated uses include "supporting diagnoses of specific learning disabilities," and the operative word in that phrase is supporting. A composite contributes to a decision made by a qualified professional operating inside their own scope, using multiple sources of evidence.
There is a sharper version of the point available here, and it comes from the measurement evidence above rather than from an ethics code. If the orthographic subtests load on the same factor as the decoding subtests, then a low orthographic composite in a student who also reads words poorly is not independent evidence of anything. It is the word reading problem measured a second way.
That matters for report writing. "Weak orthographic processing" reads like a cause. On this evidence it is closer to a restatement.
For the related and equally common error on the phonological side, see our post on pseudoword decoding measures, which found that the WJ V's Word Attack is labelled part auditory processing and behaves, in its publisher's own factor data, as a reading measure.
What a low score does license you to do
It licenses a small number of useful things, and they are worth stating positively.
A within-instrument comparison. The TOC-2's Spelling Accuracy against Spelling Fluency comparison is designed for one specific question, and the publisher states it: has the student built accurate spellings that are not yet fast? That is an instructionally actionable distinction.
An untimed-versus-timed read. On the WIAT-4, Spelling is untimed and Orthographic Fluency is timed. A gap between them is informative in the same direction, and the composite hides it. Look at the subtests.
A recognition-versus-production read. Recognising the correct spelling among same-sounding options is an easier task than producing it from dictation. Pearson's own PAL-II screening tables pair the two deliberately for this reason.
A reason to teach specific words. None of this argues against teaching irregular word spellings explicitly. It argues against treating a composite as a diagnosis.
What a low score does not license is an inference about an underlying visual or memory weakness. The word "visual memory" appears nowhere in the federal practice guide, nowhere in the Riverside CHC mapping document, and nowhere in either publisher's definition of the construct.
What AET's Code actually binds you to here
The AET Code of Ethics, September 2025, runs 2,978 words. Counted today, it contains:
"orthographic" — zero times
"orthography" — zero
"spelling" — zero
"test" and "tests" — zero
"processing" — exactly once, and that instance is about the electronic processing of client information, not about a cognitive process
Against that, "assessment" appears 7 times. The AET Fact Sheet, about 761 words of readable page text, contains "orthograph" zero times, "spelling" zero and "process" zero.
This is now the eighth consecutive post on this site to find the Code silent on the specific subject the post is about, and the conclusion is the same each time and it is not a criticism. The Code governs conduct, not instrument selection.
It binds you through its general assessment and competence provisions, which is where the obligation genuinely sits: operate inside your training, do not administer what you are not qualified to administer, and do not make claims a score does not support.
The absence is informative in one direction only. There is no professional-body rule requiring or forbidding an orthographic processing measure. The decision is yours and the justification has to be yours too.
A decision sequence, not a decision
We do not tell you which instrument to buy. We never have. But the order of the questions is itself useful, and this one has an unusual first step.
1. Ask what you would do differently with the score. If the answer is "teach irregular word spellings explicitly," you can reach that decision from a spelling sample and an error analysis. The composite is not the shortest path.
2. Decide whether you want the construct or the comparison. If you want the accuracy-versus-fluency dissociation, the TOC-2 is built for it. If you want a number inside an achievement battery you are already administering, the WIAT-4 composite is already there.
3. Check the age band before anything else. The TOC-2 starts at 8:0. For a six-year-old there is no TOC-2 option at all, and that single fact resolves a great many of these decisions.
4. Check which composite your platform gives you. Two subtests on paper, three on Q-interactive. Know which one you are reporting.
5. Look at the subtest scores, not only the composite. Everything useful in this cluster is in the spread between an untimed spelling task and a timed one.
6. Write the finding as a skill, not as a process. "Spells irregular words accurately but slowly" is defensible, specific and teachable. "Orthographic processing deficit" is none of those.
Where this sits next to the other assessment posts
This post is one piece of a larger argument that runs through the assessment writing on this site, and the pieces are worth reading against each other.
Our spelling assessment post covers the TOC-2's six subtests task by task and compares the spelling instruments on what each publisher says they are for. Our post on pseudoword decoding measures found four non-convertible scoring units for one construct and two publishers scoring the identical student response 1 and 0.
Our post on achievement test qualification levels established that the letter on a storefront is a purchasing rule rather than an industry scale.
The standard error of measurement post is the one to read before reporting any composite as a point value. Our post on outdated test norms is the one to read before relying on a norming sample's currency. Our grade equivalent scores post explains why an "orthographic ability age" is a weaker number than it looks.
On the construct side, our post on rapid automatized naming is the closest parallel in the whole set: another processing measure, another contested relationship to reading. The phonological awareness assessment post covers the other half of the WIAT-4's new processing pair. And our dyslexia screening tools post is where the screening-versus-diagnosis boundary is worked out properly.
For the instructional side, our post on reading intervention programs and our syllable division post both document the same pattern on the teaching end: a vocabulary the field shares and the publishers define differently. Our post on what "Orton-Gillingham based" means is the original instance.
If you are newer to this and want the role map first, start with educational therapist versus learning specialist versus EF coach and then what educational therapy is. For the ethics framing used above, our informed consent post and our psychoeducational evaluation post are the relevant pair.
Our reading comprehension assessment post, our written expression assessment post and our morphology instruction post complete the literacy assessment picture, the last of these covering the third component the PAL-II author named as necessary for English spelling.
Morphological awareness is the neighbouring construct with the opposite history: no dedicated instrument for decades, then two of them in two years, each described by its own publisher as the first. Morphological Awareness Assessment: Two New Tests, One Contested Claim is that story, with every morphology subtest in the batteries compared alongside it.
Where the published record runs out
Several things in this area could not be established from publicly readable sources, and saying so is more useful than filling the gaps.
The WIAT-4 technical manual is not public. Everything above about the manual's definition of the composite and its internal structure is quoted at second hand from Beaujean and Parkin, who had the manual.
The reliability coefficients for the Orthographic Processing composite, its intercorrelations with the Basic Reading composite, and the manual's own factor-analytic evidence are all behind a purchase. If those numbers are load-bearing for a report you are writing, you need the manual.
The TOC-2 examiner's manual is likewise not public. The composite reliabilities, the confidence interval widths and the validity evidence for the Spelling Efficiency composite as a measure of orthographic processing are in it and not on the product page.
PRO-ED's qualification level for the TOC-2 is not stated on the product page we could read. Hogrefe gives a Level 2 on its own UK scale. Those are different systems and neither translates to the other.
ASHA's journal platform could not be read. The ASHA publications site returns an automated-client block, so Apel (2011) was confirmed through the University of South Carolina's institutional repository record rather than from the article itself. The citation details are from that record.
No independent factor analysis of the TOC-2 was found. The WIAT-4 has Beaujean and Parkin. Nothing equivalent surfaced for the TOC-2, so the construct claim behind the Spelling Efficiency composite rests on the manual.
Nothing here is a statement about what any law or regulation requires. It is a comparison of what publishers publish and what the research literature reports. Eligibility determinations, evaluation obligations and anything touching a student's legal entitlements belong with your district, your state agency and, for independent practice, an attorney.
Frequently asked questions
Is orthographic processing a real construct?
The research literature treats it as real and contested. The strongest critical position, quoted in Compton and colleagues from Castles and Nation (2017), is that measures of orthographic processing skill are "difficult if not impossible to distinguish from tests of word recognition."
The strongest supportive position, including Compton and colleagues' own, is that choice tasks tap both word-specific and more general orthographic knowledge. What is not in dispute is that the construct has been, in Apel, Henbest and Masterson's words, "variably defined and inconsistently measured."
What does the WIAT-4 Orthographic Processing composite consist of?
Spelling and Orthographic Fluency. The Orthographic Processing Extended composite adds Orthographic Choice, which is available on Q-interactive only.
Which TOC-2 composite is the orthographic processing score?
PRO-ED says Spelling Efficiency, which combines Sight Spelling, Homophone Spelling, Word Scramble and Letter Choice. The composite named Orthographic Knowledge pools all six subtests, including Punctuation and Abbreviations.
Does the WJ V have an orthographic processing cluster?
No. Riverside's published list of new and updated WJ V clusters contains no orthographic cluster, and the string "orthograph" does not appear in that document, in the WJ V brochure, or in Riverside's achievement-to-CHC mapping document. Its new Spelling Skills cluster is tagged Grw and Ga, and Letter-Pattern Matching sits in Cognitive Processing Speed.
Can a low orthographic processing score explain weak word reading?
Not on the available evidence. In Beaujean and Parkin's factor analyses of the WIAT-4, the orthographic subtests did not separate from the decoding subtests at any age band. A low orthographic composite in a student who reads words poorly is more likely to be the same difficulty measured again than an explanation of it.
Does the federal practice guide recommend assessing orthographic processing?
It does not mention it. The 58,915-word IES guide on foundational reading skills contains zero occurrences of "orthograph." It does discuss irregular words and sight words, and it names no instrument for assessing an orthographic construct.
Is there an orthographic domain on the NCII tools charts?
No. Both the academic screening chart and the academic progress monitoring chart contain zero occurrences of "orthograph," and the screening chart's Subject filter offers only Reading and Mathematics.
What should I write in a report instead?
Describe the skill and the conditions. "Spelled 14 of 20 irregular words correctly from dictation, and read the same words accurately but at roughly half the rate of grade peers" is specific, defensible and points at instruction. "Orthographic processing deficit" asserts a mechanism the measurement evidence does not support.
Where this leaves your next intake
The honest summary is that orthographic processing is a well-intentioned construct with a product problem. Two publishers sell it, operationalise it as spelling, and define it from different sources. A third declined to build it. The federal evidence base does not use the word. The academic authority one publisher cites has since written that the construct is inconsistently measured.
None of that means you should never administer these subtests. The TOC-2's accuracy-versus-fluency comparison is genuinely useful, and the WIAT-4's Orthographic Fluency subtest gives you something a dictation spelling test does not.
It means the composite is the wrong unit. Look at the subtests, write about the skills, and keep the word "processing" out of the sentence where a skill description would do.
If this is the kind of thing you want more of, we publish one practitioner-facing piece a day on exactly these questions: what the publishers actually say, what the research actually tested, and what you can defensibly write down.
Join the Illuminate educator network for new posts as they go up.




Comments