What is the TOMAGS-2? The Test of Mathematical Abilities for Gifted Students explained

The TOMAGS-2, or Test of Mathematical Abilities for Gifted Students, Second Edition, is a standardized, "norm-referenced" test published by PRO-ED in 2020 that schools use to identify students in grades 1 through 6 who are gifted in mathematics. Written by Gail R. Ryser and Susan K. Johnsen, it measures a child's ability to use mathematical reasoning and problem solving, with items aligned to the Common Core State Standards for Mathematics. A trained examiner gives it to one child or to a small group in roughly 30 to 60 minutes. The TOMAGS-2 is not an IQ test. It samples a single academic domain, mathematics, while an intelligence test estimates general reasoning ability across many kinds of content. Districts typically use it as a domain-specific companion to an ability test and other evidence when deciding who qualifies for advanced mathematics services.

According to PRO-ED, the TOMAGS-2 is a standardized, norm-referenced instrument for identifying mathematically gifted students in grades 1 through 6. "Norm-referenced" means a child's raw score is compared with the scores of a representative sample of students at the same grade, so the result describes how unusual the performance is relative to peers rather than whether a fixed list of skills has been mastered. The construct the publisher names is mathematical reasoning and problem solving, and the items are written to line up with the Common Core State Standards for Mathematics.

That focus matters because "mathematical giftedness" is a narrower idea than high general intelligence. A child can reason unusually well about numbers without being equally strong in vocabulary or spatial tasks, and the reverse is also true. The TOMAGS-2 is built to find the first kind of child, and the validity evidence PRO-ED reports concerns separating students who are gifted in mathematics from those who are not.

The second edition also extended the test's reach. PRO-ED's catalog lists the original TOMAGS booklets with a 1998 copyright and two levels. The 2020 edition adds a third:

• Primary level: for grades 1 and 2.

• Intermediate level: for grades 3 and 4.

• Advanced level: new in the second edition, for grades 5 and 6.

The TOMAGS-2 is a professionally administered paper test. PRO-ED lists testing time as 30 to 60 minutes and states that it may be given to individuals or to small groups, which makes it more practical than a one-on-one IQ battery when a district is screening a whole grade. Each level has its own student booklet and examiner record form.

PRO-ED classifies the TOMAGS-2 at qualification level B. Under the publisher's policy, level B tests may be purchased by people with a master's degree in psychology, education, or a closely related field plus formal training in the ethical administration, scoring, and interpretation of assessments, or by people holding certain professional certifications, licenses, or supervised training in child assessment. In a school this usually means a school psychologist, educational diagnostician, or gifted coordinator administers or supervises the testing.

The examiner totals the raw score and, using the norms tables in the manual, converts it into a "percentile rank," the percentage of the norm sample the child outscored, and an "index score," a standard score on a common scale so that performance can be compared across children and levels. The examiner's manual is the authoritative source for reading any individual result.

The publisher reports that the TOMAGS-2 was normed on a weighted, demographically representative sample of 1,137 students from 16 states and Washington, D.C., collected between 2014 and 2018. The second edition reports a single set of norms for general education students, which PRO-ED notes includes a small percentage of students with exceptionalities and disabilities.

"Reliability" describes how consistent a score is. PRO-ED reports internal consistency coefficients of .84, .87, and .84 for the Primary, Intermediate, and Advanced levels, and test-retest coefficients of .90, .80, and .79. Those figures are acceptable for a single-domain test, but a score with reliability near .80 carries a wider band of measurement error than a long multi-subtest composite would, so small differences between one child and a cutoff should not be treated as decisive.

The most distinctive part of the technical summary is its "diagnostic accuracy" evidence, which PRO-ED describes as the most rigorous way to establish validity for a classification decision. "Sensitivity" is the proportion of truly gifted students the test correctly flags; "specificity" is the proportion of non-gifted students it correctly leaves out. The publisher reports:

• Primary level: an index cutoff of 108 gave a sensitivity of .86 and a specificity of .81, with an area under the curve of .90.

• Intermediate level: the same cutoff of 108 gave a sensitivity of .78 and a specificity of .82, with an area of .89.

• Advanced level: a higher cutoff of 115 gave a sensitivity of .89 and a specificity of .76, with an area of .91.

The area under the "receiver operating characteristic" (ROC) curve summarizes how well a test separates two groups across all possible cutoffs, with 1.0 meaning perfect separation and .50 meaning chance, so values around .90 are good. The sensitivity and specificity figures also show the trade-off: at the published cutoffs the test misses roughly one in ten to one in five mathematically gifted students and flags roughly one in five who are not. That is one reason a single score should never be the sole gatekeeper for a gifted program.

PRO-ED also reports that the test has consistently excellent ceilings even for the oldest and brightest students who take it. Ceilings matter more for a gifted test than for most others. In a review published in Roeper Review, Russell T. Warne traced the history of "above-level testing," the practice of giving a gifted child a test built for older students, and found that raising the test ceiling was the most consistent reason researchers gave for it, because grade-level tests measure the middle of the ability range most precisely and can run out of hard items for the top few percent. A test built with headroom at the top sidesteps part of that problem. The publisher additionally reports item-level and subgroup studies indicating little or no bias by gender, race, or ethnicity.

An IQ test estimates general cognitive ability by sampling many kinds of reasoning, typically verbal, quantitative, spatial, and memory tasks, and combining them into a full-scale score. The TOMAGS-2 samples one domain. A child at the 98th percentile on the TOMAGS-2 has shown unusual mathematical reasoning for their grade; the score says nothing directly about their vocabulary, working memory, or overall IQ. The two kinds of tests tend to correlate, because general ability contributes to every academic domain, yet they are not interchangeable. The RIOT IQ article on how gifted testing works covers the ability-test side of that picture.

The TOMAGS-2 is also easy to confuse with a sibling PRO-ED product. According to the publisher, the TOMA-3, the Test of Mathematical Abilities, Third Edition, is used to identify and quantify mathematical deficits in children aged 8 through 18. The TOMAGS-2 looks for students who are far ahead, so a district would not use one in place of the other.

A strong TOMAGS-2 score is therefore evidence for a specific placement decision, such as accelerated mathematics, rather than for a global "gifted" label. Neither the TOMAGS-2 nor any other single instrument diagnoses a learning condition; that judgment belongs to qualified clinicians working from a full evaluation.

Professional guidance in gifted education has long pushed districts away from single-score decisions. In its position statement on the role of assessments in identification, the National Association for Gifted Children states that identification must rest on defensible measurement practices, that the assessments chosen must match the definition of giftedness the district has adopted, that instruments should be administered and interpreted by people with appropriate credentials, and that the association advocates the use of multiple assessments. The same statement notes that one-on-one testing may produce more accurate data for very young children, which is relevant to a test that can be given either individually or in groups.

Under that framework, the TOMAGS-2 suits a district whose definition of giftedness includes specific academic aptitude in mathematics, and it can be combined with an ability test, classroom performance, and teacher or parent observations. The Study of Mathematically Precocious Youth at Vanderbilt University, founded by Julian Stanley in 1971, has followed more than 5,000 participants identified mostly by age 13, and illustrates why finding quantitative talent early is worth the effort. The RIOT IQ article on child giftedness covers what parents can expect from that wider process.

Some honest limitations apply. A norm sample of 1,137 students is modest when spread across six grades, so percentile ranks at the extreme top rest on relatively few children. And because the items are aligned with a particular set of curriculum standards, students with unusual exposure to advanced mathematics, or unusually little, may score in ways that reflect opportunity as much as aptitude. None of this makes the test unsuitable; it is a reason to read a TOMAGS-2 score as one well-constructed piece of evidence rather than a verdict.

The TOMAGS-2 is designed for children and for one domain of talent, so adults curious about their own reasoning ability need a different instrument. The RIOT IQ test, the Reasoning and Intelligence Online Test, was developed by RIOT IQ with psychometrician Dr. Russell T. Warne for adults aged 18 and older. It consists of 15 subtests across six cognitive indices: verbal reasoning, fluid reasoning, spatial ability, working memory, processing speed, and reaction time. It takes about 52 minutes and reports scores on the familiar scale with a mean of 100 and a standard deviation of 15. Like the TOMAGS-2, it does not replace an individually administered diagnostic evaluation. Readers who want a full-scale estimate of general reasoning ability rather than a single-domain talent score can learn more at RIOT IQ.

No. The TOMAGS-2 measures mathematical reasoning and problem solving in one academic domain. An IQ test estimates general cognitive ability by combining several kinds of reasoning tasks into a full-scale score.

Grades 1 through 6, in three levels: Primary for grades 1 and 2, Intermediate for grades 3 and 4, and Advanced for grades 5 and 6. The Advanced level was added in the 2020 second edition.

PRO-ED lists testing time as 30 to 60 minutes. The test can be given to one student or to a small group.

PRO-ED classifies it at qualification level B, which generally requires a master's degree in a relevant field with training in ethical test use, or an equivalent professional credential. In schools it is usually handled by a school psychologist, diagnostician, or trained gifted coordinator.

There is no single universal cutoff. PRO-ED's accuracy studies used index cutoffs of 108 for the Primary and Intermediate levels and 115 for the Advanced level, but each district sets its own criteria, and professional guidance recommends combining the score with other evidence.

The test is meant to measure reasoning and problem solving rather than rehearsed procedures, so coaching on specific items would undermine its purpose. Ordinary strong mathematics instruction is the appropriate preparation.

1. PRO-ED. (2020). TOMAGS-2: Test of Mathematical Abilities for Gifted Students, Second Edition, Complete Kit. proedinc.com

2. PRO-ED. (1998). TOMAGS (previous edition) Primary Level Student Booklets. proedinc.com

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

4. PRO-ED. (n.d.). TOMA-3: Test of Mathematical Abilities, Third Edition, Complete Kit. proedinc.com

5. National Association for Gifted Children. (2008). The role of assessments in the identification of gifted students [Position statement]. nagc.org

6. Warne, R. T. (2012). History and development of above-level testing of the gifted. Roeper Review, 34(3), 183-193. eric.ed.gov

7. Vanderbilt University. (n.d.). Study of Mathematically Precocious Youth. my.vanderbilt.edu

Hero photo: A wooden multiplication-table block set. Photo by Onderwijsgek, via Wikimedia Commons, CC BY-SA 3.0 NL (cropped).