What is the TACL-4? A guide to the Test for Auditory Comprehension of Language

The TACL-4 is the Test for Auditory Comprehension of Language, Fourth Edition, a norm-referenced measure of "receptive language" (what a child understands when someone else speaks) for children aged 3 years 0 months through 12 years 11 months. It is administered one-on-one by a qualified examiner, usually a speech-language pathologist, school psychologist, or educational diagnostician, and takes 20 to 30 minutes. The TACL-4 is not an IQ test. It measures a narrow slice of language comprehension rather than general reasoning ability, and is normally used alongside cognitive testing rather than in place of it. This article covers what the test contains, how it is scored, what its published evidence shows, and how its results relate to an intelligence score.

The test was written by Elizabeth Carrow-Woolfolk and published by PRO-ED in Austin, Texas, with a 2014 copyright. The publisher describes it as a measure of receptive spoken vocabulary, grammar, and syntax, organized into three subtests.

• Vocabulary: the meanings of nouns, verbs, adjectives, and adverbs, along with words that represent basic percepts and concepts.

• Grammatical Morphemes: the meaning of function words such as prepositions, pronouns, and determiners, and of inflections such as noun number and case, verb tense, and derivational suffixes.

• Elaborated Phrases and Sentences: the meaning of syntactically based word relations and longer constructions, including questions, negatives, active and passive voice, embedded sentences, and conjoined sentences.

The three subtests combine into a single composite, the Receptive Language Index, which is the most reliable score the test produces.

Every item pairs a spoken stimulus with a picture plate showing three full-color drawings. One drawing illustrates the word, morpheme, or structure being tested; the other two are contrasts or decoys. The examiner reads the stimulus aloud and the child points to the picture that best matches it. PRO-ED states plainly that no oral response is required, which makes it usable with children whose speech is limited or unintelligible.

Items within each subtest run from easier to harder, and the test uses age-related entry points together with "basal" and "ceiling" rules. A basal is the point below which items are assumed correct; a ceiling is the point at which testing stops after a run of errors. The purpose is to keep administration short and target items to each child's level.

Purchasing the test requires PRO-ED qualification Level B, which the publisher's policy sets at a relevant master's degree plus formal training in ethical administration and interpretation, or an equivalent professional credential. The test is not sold to the general public.

Raw scores convert to subtest scaled scores, and the sum of the three converts to the Receptive Language Index, a percentile rank, an age equivalent, and a descriptive term. The profile chart in PRO-ED's free TACL-4/TEXL scoring supplement lines a subtest scaled score of 10 up against an index score of 100, and a scaled score of 20 against an index of 150. That places subtest scores on the mean-10 scale with a "standard deviation" (a measure of how spread out scores are) of 3, and index scores on the mean-100, standard-deviation-15 scale used by most cognitive and language batteries.

The supplement defines the average band as scaled scores of 8 to 12 and index scores of 90 to 109, and tells clinicians that scores below that band should be treated as areas of concern. Its descriptive ladder runs from impaired or delayed at the bottom to gifted or very advanced at the top. Those labels describe a score on one test on one day. They do not describe a child.

For the combined composites covered in the supplement, PRO-ED reports a "standard error of measurement" (the typical amount of noise in an observed score) of 3 points, so an index of 100 carries a reasonable range of roughly 97 to 103. Reporting a single number without that range overstates how precisely any test pins down an ability, a caution that applies equally to IQ scores and the norm groups behind them.

PRO-ED normed the fourth edition on 1,142 children in the United States, with data collected in 2012 and 2013 according to PAR, which distributes the test. The publisher reports that the sample was stratified by age relative to region, gender, ethnicity, and socioeconomic factors, and matched to the Statistical Abstract of the United States 2013.

• Internal consistency: PRO-ED reports "coefficient alpha" values (how consistently items in a scale measure the same thing) of .94 to .96 for the subtests and .97 for the Receptive Language Index.

• Criterion validity: average correlations with established receptive-language measures range from .67 to .71, according to the publisher. That is a substantial relationship, and it falls well short of interchangeability, as you would expect from tests that sample comprehension in different ways.

• Classification evidence: the publisher reports median "sensitivity" (the share of affected children a measure flags) of .77, median "specificity" (the share of unaffected children it correctly passes over) of .84, median classification accuracy of .80, and a median area under the receiver operating characteristic curve of .88. Figures in this range support the test as one input into a clinician's decision.

Elizabeth Carrow published the original version in 1973, a revised edition followed in 1985, and PRO-ED issued the third edition in 1999. The fourth edition made several changes worth knowing. The upper age limit was extended to cover 10-0 through 12-11. Easier and harder items were added to address floor and ceiling effects present in the third edition, which PRO-ED now describes as minimal in the subtests and absent in the composite. Every item was screened with differential item functioning analyses, a statistical check for items that behave differently across groups matched on overall ability. The reliability and validity samples included children with learning disabilities, ADHD, hearing loss, articulation disorders, language impairment, autism spectrum disorder, and intellectual disability, plus children identified as gifted.

An IQ test estimates general cognitive ability by sampling several domains of reasoning and combining them. The TACL-4 samples one domain, comprehension of spoken English, through one response mode. A child who scores low on the Receptive Language Index may have a language disorder, a hearing difference, or a broader cognitive difficulty, or may be an English learner tested in a language they are still acquiring. The score alone does not distinguish among those possibilities.

The relationship runs the other way as well. Receptive language and nonverbal reasoning are correlated, yet they are far from the same thing, and the field has moved away from requiring a gap between the two before identifying a language problem. The CATALISE Delphi consensus study, published in PLOS ONE by Dorothy Bishop, Margaret Snowling, Paul Thompson, Trisha Greenhalgh, and a multinational panel, concluded that language impairment should be identified regardless of whether it mismatches nonverbal ability. The panel noted that discrepancy scores are statistically unreliable for classification and that restricting services to children with a large gap denies help to those who need it most. Practitioners call that older approach "cognitive referencing".

What a TACL-4 score does well is describe how a child handles spoken vocabulary, grammatical markers, and sentence structure compared with same-age peers. Pairing it with a cognitive measure answers a question neither one answers alone.

The TACL-4 was co-normed with the Test of Expressive Language, or TEXL, by Carrow-Woolfolk and Elizabeth Allen, published in 2014. The TEXL covers the same ages and the same three language forms on the production side, and its norms come from 1,205 children. When a clinician gives both, the free scoring supplement yields Vocabulary, Morphology, Syntax, Receptive Language, Expressive Language, and Oral Language indexes from one summary form. PRO-ED sets the threshold for a statistically significant gap between the receptive and expressive indexes at 7 points and the threshold for clinical usefulness at 21 points.

Broader batteries answer other questions. The CELF-5 samples receptive and expressive language across many more tasks in a longer sitting, while the Peabody Picture Vocabulary Test uses a similar pointing format and covers only single-word receptive vocabulary. The TACL-4 sits between them.

The American Speech-Language-Hearing Association states that speech-language pathologists carry out screening, assessment, diagnosis, and treatment of spoken language disorders, drawing on formal and informal measures together. It also warns that standard score interpretations are not valid when the norming sample does not represent the child being assessed, a caution that applies to any standardized score.

Adults curious about their own reasoning ability need a properly normed instrument rather than a quiz. The Reasoning and Intelligence Online Test was developed by RIOT IQ with psychometrician Dr. Russell T. Warne for adults aged 18 and over. It runs 15 subtests across six cognitive indices covering verbal reasoning, fluid reasoning, spatial ability, working memory, processing speed, and reaction time, takes about 52 minutes, and reports scores on the mean-100, standard-deviation-15 scale described above. It does not replace an individually administered diagnostic evaluation conducted by a qualified professional.

The TACL-4 is normed for children aged 3 years 0 months through 12 years 11 months. The fourth edition raised the upper limit, adding norms for ages 10-0 through 12-11.

PRO-ED lists testing time at 20 to 30 minutes for an individual administration. Entry points, basal rules, and ceiling rules keep each child from working through items far above or below their level.

No. Each item asks the child to point to one of three pictures after hearing a word, phrase, or sentence. The publisher states that no oral response is required, which is part of why the test suits children whose speech is limited.

No. It measures comprehension of spoken vocabulary, grammatical morphemes, and sentence structure, while an IQ test samples multiple reasoning domains to estimate general cognitive ability. A low TACL-4 score has several possible explanations that the test alone cannot separate.

PRO-ED classifies the TACL-4 at qualification Level B, which covers professionals with a relevant master's degree and formal assessment training, full membership in a body such as the American Speech-Language-Hearing Association, or a healthcare license.

The TACL-4 measures what a child understands; the TEXL measures what a child produces. They were co-normed, cover the same ages and the same three language forms, and can be combined to yield an Oral Language Index plus a receptive-expressive discrepancy analysis.

1. Carrow-Woolfolk, E. (2014). TACL-4: Test for Auditory Comprehension of Language, Fourth Edition [Product page]. PRO-ED. proedinc.com

2. Carrow-Woolfolk, E., & Allen, E. A. (2014). TACL-4/TEXL comprehensive scoring supplement. PRO-ED. proedinc.com

3. Carrow-Woolfolk, E., & Allen, E. A. (2014). TEXL: Test of Expressive Language [Product page]. PRO-ED. proedinc.com

4. PRO-ED. (n.d.). Qualifications policy. proedinc.com

5. PAR. (n.d.). Test for Auditory Comprehension of Language, Fourth Edition, TACL-4. parinc.com

6. Bishop, D. V. M., Snowling, M. J., Thompson, P. A., Greenhalgh, T., & CATALISE Consortium. (2016). CATALISE: A multinational and multidisciplinary Delphi consensus study. Identifying language impairments in children. PLOS ONE, 11(7), e0158753. journals.plos.org

7. American Speech-Language-Hearing Association. (n.d.). Spoken language disorders [Practice Portal]. asha.org

Hero photo: A teacher writing English parts of speech and verb tenses on a whiteboard. Photo by Preply.com Images, via Wikimedia Commons, CC BY 2.0 (cropped).