What Is the SPELT-3? The Structured Photographic Expressive Language Test, Third Edition
The SPELT-3 uses 54 photographs to elicit a child's grammar, ages 4-0 to 9-11. It is not an IQ test. What it measures, who gives it, edition status.
Dr. Russell T. WarneChief Scientist
Share
The SPELT-3 is the Structured Photographic Expressive Language Test, Third Edition, a language assessment published in 2003 by Janelle Publications for children ages 4-0 through 9-11. It measures expressive "morphology" (the word endings that signal plurals, possession, and tense) and "syntax" (how words are arranged into sentences), using 54 full-color photographs paired with short spoken prompts. Speech-language pathologists administer and score it. The SPELT-3 is not an IQ test, and it produces no IQ score of any kind. It sits next to cognitive testing rather than replacing it: when a school team wants to know whether a child's grammar is behind where it should be, the SPELT-3 answers that narrow question, while a separate cognitive measure describes reasoning ability.
What the SPELT-3 measures
The construct is expressive morphosyntax, a compact way of saying "the grammar a child actually produces out loud." The child looks at a photograph of an ordinary scene, hears a brief question or statement from the examiner, and replies. Each item is built so that a correct, natural answer requires one specific structure. A prompt about what two boys are doing pulls for a present progressive verb form; a photograph of something that already happened pulls for past tense.
According to the publisher, the items target four families of structure.
• Inflectional morphology: plurals, possessive nouns, and verb tense endings, the endings that carry grammatical meaning without being separate words.
• Verb form: present progressive, regular and irregular past, future, and the contractible and uncontractible copula and auxiliary (the "is" in "he is happy" and in "he is running").
• Pronoun usage: subject pronouns, possessive pronouns, and reflexive pronouns.
• Syntax: negatives, conjoined sentences, "wh" questions, interrogative reversal, negative infinitive phrases, propositional complements, relative clauses, and front embedded clauses.
Janelle Publications makes a specific argument for this elicited format. A child's spontaneous conversation may never create the occasion to use a relative clause or an irregular past tense, so a language sample can leave a clinician unsure whether a structure is absent because the child cannot produce it or because the topic never called for it. Structured elicitation manufactures the occasion and removes that ambiguity.
What the SPELT-3 does not cover is worth stating. It leaves out receptive language, vocabulary breadth, pragmatics, articulation, and reading. Clinicians who need those domains reach for other instruments, often a broad-band battery such as the CELF-5 or a paired receptive and expressive measure like the OWLS-II. The SPELT-3 is deliberately narrow.
How the test is given and scored
Administration is one examiner with one child, face to face, using a bound album of fifty-four 4" x 6" photographs. The publisher lists administration time at 15 to 20 minutes, one of the shorter figures among standardized language measures in common use. There is no subtest structure to negotiate; the examiner works through the album in order.
Scoring is where the clinical judgment lives. The examiner writes down exactly what the child said and then decides whether the target structure was produced correctly. Janelle Publications includes a manual chapter on African American English with a scoring guide, written by Nola T. Radford, so that forms which are grammatical in a dialect are not scored as errors. A second chapter by Laura Barnes covers the technical qualities of the standardization sample scores.
The test yields standard scores, confidence intervals, percentile ranks, percentile bands, and test-age equivalents, reported separately for females and males across the age range. Normative tables rest on a national sample of more than 1,800 children drawn from the four geographical regions the U.S. Census Bureau defines. That size is respectable for a single-purpose language probe, if smaller than the samples behind major cognitive batteries. Our explainer on what a norm group is covers the same logic that applies here.
Why the SPELT-3 is not an IQ test
The distinction matters because parents regularly receive an evaluation report containing several standard scores on the familiar mean-100 scale and reasonably assume they all measure the same thing. They do not. A SPELT-3 standard score of 82 describes one child's production of grammatical structures relative to same-age peers in the norm sample. It says nothing directly about reasoning, working memory, or processing speed. A low score is therefore not evidence of low cognitive ability, and a high one is not evidence of high cognitive ability. The link between language and cognition is real but partial, so the two kinds of score often diverge.
The field has moved away from defining language disorder by its relationship to IQ. The CATALISE consensus panel, a multinational Delphi study led by Dorothy Bishop and colleagues and published in 2017, concluded that a child with a language disorder may have a low level of nonverbal ability and that this does not preclude a diagnosis of developmental language disorder. Older practice required a discrepancy between verbal and nonverbal scores before a child could qualify for services, and that requirement was dropped because it excluded children with real language impairment.
Cognitive testing and language testing answer different questions about the same child, and a full evaluation usually involves both. The American Speech-Language-Hearing Association's guidance on spoken language disorders describes comprehensive assessment as combining formal and informal measures across phonology, semantics, morphology, syntax, and pragmatics, alongside case history, hearing screening, and cognitive assessment. The SPELT-3 supplies one piece of that picture, and diagnosis rests with the qualified clinician who reads all of it together.
What the research says about its accuracy
The SPELT-3 has an unusually well-documented accuracy record for an instrument of its size. Perona, Plante, and Vance published a diagnostic accuracy study in Language, Speech, and Hearing Services in Schools in 2005, studying 42 four- and five-year-olds with specific language impairment and 43 children with typically developing language across an exploratory and a confirmatory sample. They reported 90% sensitivity and 100% specificity at a cutoff standard score of 95, and concluded that the data support using the SPELT-3 to differentiate between children with normal language and children with impaired language.
Two caveats belong with those numbers. The study sampled four- and five-year-olds, the bottom of the test's range, so the evidence is strongest at the younger end. And the cutoff of 95 sits above the conventional one standard deviation below the mean, so a score in the high 80s or low 90s is less reassuring than a clinician might assume.
Janelle Publications reprints that conclusion on its product page and reports further reliability and validity studies in the manual, without publishing the coefficients online.
Editions: the SPELT-3, the SPELT-P 2, and a SPELT-4 in progress
The edition picture is easy to get wrong, so here it is plainly.
• SPELT-3 is current. The third edition, authored by Janet Dawson, Connie Stout, and Julia Eyer, remains the version Janelle Publications sells, listed at $299.00 for the kit. The publisher's own About Us page states that it continues to serve clinicians today.
• SPELT-P 2 is a companion, not a successor. The Structured Photographic Expressive Language Test-Preschool 2 covers ages 3-0 through 5-11 using 40 items and 44 photographs, with norms based on more than 1,700 children. Both are current at the same time because they serve different ages. Greenslade, Plante, and Vance published an accuracy study of the preschool version in 2009, reporting 90.6% sensitivity and 100% specificity at a cutoff standard score of 87.
• A SPELT-4 is being standardized but has not been published. Janelle Publications' field testing page is recruiting speech-language pathologists to collect normative data for a fourth edition covering the same 4-0 through 9-11 age band, with examiners receiving a copy of the published version when it appears. No release date is given. Until it ships, the SPELT-3 is the edition to cite.
• A Spanish SPELT-3 is a separate instrument. Authored by Henriette W. Langdon, it uses 13 large photographs and carries its own norms rather than being a translation.
Where a language probe fits next to a cognitive test
A short, targeted instrument like the SPELT-3 earns its place by doing one thing well and quickly. It tells a clinician whether a child's grammatical production falls outside the typical range, with enough published accuracy evidence to justify the inference. It does not say how a child reasons or solves novel problems.
Cognitive assessment answers those other questions, and the standards there are just as demanding: a real norm sample and published reliability and validity evidence, with scores reported as ranges rather than single points. To see what a properly normed cognitive measure looks like from the inside, the RIOT IQ test is available for adults 18 and over. The Reasoning and Intelligence Online Test was developed by RIOT IQ with psychometrician Dr. Russell T. Warne and runs 15 subtests across six cognitive indices covering verbal reasoning, fluid reasoning, spatial ability, working memory, processing speed, and reaction time. It takes about 52 minutes and reports scores on the mean-100, standard-deviation-15 scale. Because it is built for adults, it is not an option for the age band the SPELT-3 serves, and like any online measure it does not replace an individually administered diagnostic evaluation. For a child whose language or learning is a concern, the route is an evaluation by qualified professionals.
Frequently asked questions
Is the SPELT-3 an IQ test?
No. The SPELT-3 measures expressive morphology and syntax in children ages 4-0 through 9-11 and reports language standard scores only. It contains no reasoning, memory, or nonverbal problem-solving items and yields no IQ score.
Who can administer the SPELT-3?
Speech-language pathologists, working one on one with the child. Scoring requires judging whether a spoken response contains the target structure, including decisions about dialect variation, which is why the manual carries a scoring guide for African American English.
What ages does the SPELT-3 cover, and how long does it take?
Ages 4-0 through 9-11. Janelle Publications lists administration time at 15 to 20 minutes.
Is there a SPELT-4 yet?
Not as of this writing. Janelle Publications is recruiting speech-language pathologists to help standardize a fourth edition for the same age range, but no published version or release date has been announced.
What is the difference between the SPELT-3 and the SPELT-P 2?
Age range and item set. The SPELT-P 2 covers ages 3-0 through 5-11 with 40 items and 44 photographs targeting earlier-developing structures, while the SPELT-3 covers 4-0 through 9-11 with 54 photographs and more advanced syntax. It is a companion instrument rather than a newer edition.
Does a low SPELT-3 score mean a child has a language disorder?
Not on its own. A single test score is one piece of evidence with measurement error attached to it. Diagnosis rests with a qualified clinician who combines standardized results, language sampling, case history, hearing status, and observation across settings.
References
1. Janelle Publications. Structured Photographic Expressive Language Test 3 (SPELT-3). janellepublications.com
4. Perona, K., Plante, E., & Vance, R. (2005). Diagnostic accuracy of the Structured Photographic Expressive Language Test: Third Edition (SPELT-3). Language, Speech, and Hearing Services in Schools, 36(2), 103-115. doi.org
5. Greenslade, K. J., Plante, E., & Vance, R. (2009). The diagnostic accuracy and construct validity of the Structured Photographic Expressive Language Test-Preschool: Second Edition. Language, Speech, and Hearing Services in Schools, 40(2), 150-160. doi.org
6. Bishop, D. V. M., Snowling, M. J., Thompson, P. A., Greenhalgh, T., & the CATALISE-2 consortium (2017). Phase 2 of CATALISE: A multinational and multidisciplinary Delphi consensus study of problems with language development: Terminology. Journal of Child Psychology and Psychiatry, 58(10), 1068-1080. pmc.ncbi.nlm.nih.gov
7. American Speech-Language-Hearing Association. Spoken language disorders [Practice Portal]. asha.org
Hero image: “Albo illustrato - libri.jpg” by Lunetta2000, licensed CC BY-SA 4.0, via Wikimedia Commons (https://commons.wikimedia.org/wiki/File%3AAlbo_illustrato_-_libri.jpg).
Take our professional IQ test
Want to know your IQ? Try the first ever professional online IQ test.
