TN 6 (07-23)

DI 24583.055 Using Intelligence Tests to Evaluate Cognitive Disorders, Including Intellectual Disorder

A. What are intelligence tests?

Intelligence tests are standardized tests that measure a person’s intellectual performance in multiple aspects of cognitive functioning. They may also be referred to as intelligence quotient or IQ tests.

B. How do we use intelligence tests?

We use the results of intelligence tests to assess a person’s general intellectual functioning. Intelligence tests are most commonly used to evaluate cases involving intellectual disorder, specifically to determine whether the person has significantly subaverage general intellectual functioning as required by listings 12.05 and 112.05. However, intelligence tests may also be used when evaluating other cognitive disorders, such as neurocognitive disorder or borderline intellectual functioning, under listings 12.02, 12.11, 112.02, and 112.11. For the mental disorders listings, see DI 34001.032 and DI 34005.112.

NOTE: We do not endorse, prefer, or require any specific intelligence test. Tests are programmatically acceptable when they satisfy the requirements in DI 24583.050B, DI 24583.050C, DI 24583.055C, and DI 24583.055D.

NOTE: The information provided in this section applies generally to intelligence tests for both adults and children. There are additional guidelines specific to intelligence tests for children. For more information, see DI 24583.060.

C. What are our program requirements for intelligence tests?

If the obtained IQ scores will be used to meet or medically equal the IQ requirements of listing 12.05 or 112.05, the intelligence test must meet the program requirements for psychological tests described in DI 24583.050B . That is, the test must be individually administered, administered by a qualified specialist, and standardized, and must meet contemporary psychometric standards for validity, reliability, normative data, and scope of measurement.

In addition to the program requirements for psychological tests, intelligence tests must have a mean of 100 and a standard deviation of 15 if the obtained scores will be used to meet or medically equal the requirements of listing 12.05 or 112.05. All intelligence tests published since 2003 (and most tests published prior to that time) satisfy this requirement. In the rare case you encounter an older intelligence test that used a different mean or standard deviation, contact your Regional Office for additional guidance.

Do not purchase intelligence tests that do not meet these programmatic requirements except for nonverbal intelligence tests under limited circumstances. For more information about nonverbal intelligence tests, see DI 24583.055I.3.

NOTE: These requirements apply to intelligence tests used to meet or medically equal the IQ criteria of listing 12.05 or 112.05. For information about how to consider tests that do not meet these requirements, see DI 24583.055H.

D. What information do we need about the intelligence test?

Just as with psychological tests, we require a test report containing information about the particular test administration and a narrative report. Do not find a person has an intellectual disorder that meets or medically equals the IQ requirements of listing 12.05 or 112.05 unless the record contains the test report with the required information and a narrative report. For more information, see DI 24583.050C.

E. How do we consider IQ scores obtained on intelligence tests?

IQ scores are used to satisfy the criteria in listing 12.05B.1 or 112.05B.1. The criteria are satisfied either when the person has a full scale IQ score of 70 or below or when the person has a full scale IQ score of 71 to 75 accompanied by a verbal or performance (or comparable) IQ score of 70 or below.

NOTE: IQ scores only satisfy one part of the criteria needed to find a person has intellectual disorder that meets or medically equals listing 12.05 or 112.05. For the other criteria, see the mental disorders listings in DI 34001.032 and DI 34005.112.

1. Obtained IQ scores

Only consider a person’s obtained IQ scores for 12.05B.1 and 112.05B.1. Do not raise or lower an IQ score or consider IQ scores that are slightly above or below the criteria in 12.05B.1.a, 12.05B.1.b, 112.05B.1.a, or 112.05B.1.b to find medical equivalence. The 12.05B.1.b and 112.05B1.b criteria for significantly subaverage general intellectual functioning already account for the scenario involving an obtained full scale IQ score that is slightly above 70. If the record contains obtained IQ scores that do not satisfy the IQ score criteria in 12.05B.1 or 112.05B.1 but the criteria in 12.05B.2 or 112.05B.2 are satisfied, consider whether the impairment meets or medically equals listing 12.11 or 112.11 for neurodevelopmental disorders or 12.02 or 112.02 for neurocognitive disorders.

NOTE: For more information about determining medical equivalence for intellectual disorder under listing 12.05 and 112.05, see DI 24583.010B.

2. Full scale IQ "accompanied by" verbal or performance IQ scores

A full scale (or comparable) IQ score is “accompanied by” a verbal or performance part IQ score when the IQ scores were obtained on the same administration of the same intelligence test on the same day. Do not combine obtained IQ scores from different intelligence tests to meet or medically equal the requirements of 12.05B.1.b or 112.05B.1.b.

EXAMPLE: An adult person obtained a full scale IQ score of 72 on the Wechsler Adult Intelligence Scale, Fourth Edition (WAIS-IV), administered in 2018. On a previous administration of the WAIS-IV in 2013, the person obtained a verbal comprehension index score of 68. Do not combine these scores to find the requirements of 12.05B.1.b are satisfied.

NOTE: Do not purchase an additional intelligence test simply to obtain verbal or performance part scores. Additional testing may create an inconsistency that must be resolved or yield scores affected by the practice effect, see DI 24583.055F.2. If the evidence of record contains multiple intelligence tests with scores that vary by more than a few points, see DI 24583.055I.8.

3. "Comparable" scores

A comparable score is an IQ score that reflects the same type and scope of a person’s intelligence, even though a different name for the score may be used.

a. Full scale IQ scores

A full scale IQ score reflects a wide range of cognitive abilities, such as verbal comprehension, perceptual reasoning, memory, and processing speed. A score that is comparable to a full scale IQ score will also represent a person’s general intelligence. Some IQ tests refer to these scores as “total,” “overall,” or “general intellectual ability” scores, which are comparable to a full scale IQ score.

b. Verbal IQ scores

A verbal IQ score represents a person’s verbal intelligence, which measures verbal ability, comprehension, and knowledge and the ability to think abstractly. A score that is comparable to a verbal IQ score will also represent a person’s verbal abilities. Some IQ tests refer to these scores as a “verbal comprehension index,” which is comparable to a verbal IQ score.

c. Performance IQ scores

A performance IQ score represents a person’s nonverbal intelligence, which is the ability to analyze information and solve problems using visual and spatial reasoning, sometimes requiring good eye-hand coordination and speed. A score that is comparable to a performance IQ score will also represent a person’s nonverbal intelligence. Some IQ tests refer to these scores as a “perceptual reasoning index,” which is comparable to a performance IQ score.

d. Other part scores

Not all IQ tests produce part scores that are comparable to verbal and performance IQ scores. Some IQ tests produce part scores that are not comparable to full scale, verbal, or performance IQ scores. Do not consider part scores that are not comparable to full scale, verbal, or performance IQ scores. For example, “working memory” and “processing speed” indices are part scores that are not comparable to full scale, verbal, or performance IQ scores.

4. IQ scores at the margin

IQ scores are at the margin when they are within a few points on either side of the scores specified in the listings. Only consider the obtained IQ scores included in the report to determine whether the score meets or medically equals the requirements of listings 12.05 and 112.05. The obtained full scale IQ score and relevant part scores, where applicable, must be at or below the value specified in the listing. Do not consider other factors when determining whether the reported IQ score satisfies the requirements of 12.05B.1 or 112.05B.1. Do not raise or lower an IQ score by assuming it was within the standard error of measurement.

The narrative report is particularly important when IQ scores are at the margin. While only qualified specialists may determine an obtained IQ score is not an accurate reflection of the person's typical intellectual functioning, adjudicators must consider intelligence test results in the context of the entire medical record.

F. What other considerations may affect obtained IQ scores?

1. Standard error of measurement

The standard error of measurement is a statistical confidence interval within which the person’s true IQ score falls. The standard error of measurement estimates the significance of an IQ score that improves on retesting with the same instrument. It is a standard deviation unit defining the degree of chance error for any particular score and varies by intelligence test. Generally, we expect a person's true score to fall within approximately 3 to 5 points both above and below the reported score.

Only consider obtained IQ scores. Do not raise or lower an IQ score or consider IQ scores that are slightly above or below the criteria in 12.05B.1.a, 12.05B.1.b, 112.05B.1.a, or 112.05B.1.b by assuming they are within the standard error of measurement.

2. Practice effect

The practice effect refers to gains in IQ scores on tests of intelligence that result from a person being retested on the same instrument. The size of the practice effect decreases as the time between test administrations lengthens. Use best efforts to obtain a copy of the test report for previously administered intelligence tests to determine the test version and date of the previous administration. When purchasing intelligence tests, avoid purchasing the same intelligence test within 12 months from the first administration. When multiple administrations of the same intelligence test are present in the record and show an improvement in IQ scores over time, consider the time between test administrations. If intelligence testing is administered as part of a consultative examination (CE) and the person is retested on an instrument they previously completed, make the CE provider aware of the previous testing and ask them to consider its effect. Only qualified specialists, medical and psychological consultants (MC or PC), and other contracted medical and psychological experts may conclude the obtained IQ scores are not an accurate reflection of the person's general intellectual functioning.

G. Who is responsible for conclusions based on testing?

Presume the obtained IQ scores are an accurate reflection of the person’s current general intellectual functioning at the time the test was administered, unless evidence in the record suggests otherwise. Only qualified specialists, MCs or PCs, and other contracted medical and psychological experts may conclude the obtained IQ scores are not an accurate reflection of the person’s general intellectual functioning. Administrative law judges and other lay adjudicators are not qualified specialists and cannot make this determination. For more information about conclusions based on testing, see the mental disorders listings in DI 34001.032 and DI 34005.112.

H. How do we consider intelligence tests that do not satisfy program requirements?

The record may contain results of intelligence tests that do not satisfy program requirements. Do not use those results to meet or medically equal the requirements of listings 12.05 and 112.05 except when using nonverbal intelligence tests. For more information about nonverbal intelligence tests, see DI 24583.055I.3.

Consider intelligence tests that do not satisfy program requirements for the information they provide. For example, if an intelligence test was not administered by a qualified specialist and the results indicate potentially listing-level IQ scores, then purchase a programmatically acceptable test.

I. Additional considerations

1. Use of interpreters

Do not use an interpreter for standardized intelligence testing. The tests are intended to be administered with a standard protocol, which often includes specific wording used for presenting test instructions and test items. Any deviation from the standard protocol can impact performance, scoring of the test, and interpretation of results. For example, test items presented in a different language may not be equivalent to how the item is presented in English. For more information on intelligence testing options that do not rely on fluency in a particular language, see DI 24583.055I.3.

2. Spanish language intelligence tests

There are Spanish language versions of some intelligence tests. Most of these tests do not meet programmatic requirements due to the lack of normative data based on Spanish-speaking individuals who reside in the United States. Additionally, many of the Spanish-language versions of intelligence tests are based on an edition of the English-language test other than the current publication. A limited number of Spanish language intelligence tests may be acceptable under very specific circumstances. Do not purchase Spanish language intelligence tests that do not meet programmatic requirements. If an intelligence test is needed, instead purchase a nonverbal intelligence test.

a. Spanish language tests that may be programmatically acceptable

  • Wechsler Intelligence Scale for Children-Fifth Edition (WISC-V) Spanish: This test may be used for Spanish-speaking children who have not completed more than 5 consecutive years in the United States school system. It is not programmatically acceptable for children with more than 5 years of schooling in the United States.

b. Spanish language tests that are not programmatically acceptable

  • Batería III Woodcock-Muñoz: This test is not programmatically acceptable. It uses norms substantially based on populations outside of the 50 United States and the District of Columbia (for example, Central America, South America, Spain, Puerto Rico).

  • Batería IV Woodcock-Muñoz: The test was co-normed with the Woodcock-Johnson IV with a small calibration sample of Spanish-speaking individuals. The resulting normative sample reflects a mixed sample of English- and Spanish-speaking individuals and is not representative of the population for whom the test is intended.

  • Escala de Inteligencia de Wechsler para Adultos-Tercera Edición (EIWA-III): This test is not programmatically acceptable. It uses norms based on the Puerto Rican census.

  • Escala de Inteligencia Wechsler Para Niños-Revisada de Puerto Rico (EIWN-R PR): This test is not programmatically acceptable. It uses norms based on the Puerto Rican census and is anchored to an outdated edition of the WISC.

NOTE: Older versions of these tests also are not programmatically acceptable, as are Spanish-language versions of the WISC other than the Spanish versions of the WISC-IV and WISC-V.

For more information on considering intelligence tests that do not meet programmatic requirements, including Spanish-language intelligence tests, see DI 24583.055H.

3. Nonverbal intelligence tests

Many nonverbal intelligence tests will not satisfy the requirements for standardized tests of general intelligence found in section 12.00H of the mental disorders listings. Although many nonverbal tests measure multiple areas of intelligence, they do not include language and thus are not comprehensive tests of general intelligence.

Nonverbal intelligence tests are sometimes the only way to assess a person’s intellectual functioning. In particular, nonverbal intelligence tests are used when the person is unable to undergo standard administration of traditional intelligence tests that include both verbal and performance components. Common situations where nonverbal intelligence tests are acceptable include testing of people who do not speak English as their first language or who have significant language impairments. In cases where a nonverbal intelligence test is the only way to accurately measure a person’s intellectual functioning, medical equivalence may be appropriate.

Purchase a nonverbal intelligence test when the person is unable to undergo standard administration of a traditional intelligence test. The results of nonverbal intelligence tests used in the clinical setting to diagnose intellectual disorder may be used to medically equal the IQ requirements of listing 12.05 or 112.05. Examples of nonverbal intelligence tests used in the clinical setting include, but are not limited to, the Comprehensive Test of Nonverbal Intelligence, Second Edition (CTONI-2), Leiter International Performance Scale, Third Edition (Leiter-3), and Wechsler Nonverbal Scale of Ability (WNV). Do not use the results of a nonverbal intelligence test to meet the IQ requirements of listing 12.05 or 112.05.

For more information about medical equivalence, see DI 24508.010 and DI 24583.010.

4. Screening and other abbreviated IQ tests

Screening tests are used to provide only gross determinations of levels of functioning. They are frequently used to determine whether a person needs more comprehensive evaluation and do not replace more comprehensive versions of intelligence tests.

Abbreviated IQ tests are quick measures of intelligence used in a variety of situations when a comprehensive assessment is not required. Most abbreviated IQ tests are essentially screening tests. For example, the Wechsler Abbreviated Scale of Intelligence (WASI) is an abbreviated test used to estimate IQ based on some, but not all, of the subtests contained in the Wechsler series of intelligence tests.

Do not use the results of abbreviated tests to determine whether the person has a mental impairment that meets or medically equals the requirements of listings 12.05 or 112.05. Abbreviated or screening tests may be used to rule out a medically determinable impairment but cannot be used to establish a medically determinable impairment. Purchase abbreviated tests or screening tests only in rare circumstances. If the record only contains results of an abbreviated test and those results suggest the person may have a cognitive impairment, then consider purchasing a programmatically acceptable intelligence test.

5. Updated editions of intelligence tests

Most providers will transition to updated editions within a reasonable period of time following the release of the updated edition due to State and national ethical codes and a need to keep current with advances in psychological testing. The provider is allowed to exercise professional judgment in selecting the most appropriate test instrument for a particular person. We expect this will usually be the most recently normed edition of the test. When considering tests for purchase as part of a CE, do not purchase tests more than two years after the publication of the new editions.

6. School records

School records, including transcripts and individualized education programs (IEP), sometimes include the results of intelligence tests. Do not use the results of intelligence tests reported in school records if only the score is documented. Only use IQ scores reported in school records when the documentation includes the required test administration information and narrative report, see DI 24583.055D.

NOTE: Intelligence test results contained in school records may not be current. For example, school transcripts from elementary school that contain otherwise acceptable intelligence test results cannot be used to meet or medically equal the IQ requirements for an adult person. For more information, see DI 24583.055I.7.

7. Current IQ scores

IQ scores must be current to be used to meet or medically equal the IQ requirements of listings 12.05 and 112.05. IQ scores stabilize after age 16 and are generally considered current after that time. For younger children, consider the child's age at the time of testing and the test results themselves when determining whether IQ test results obtained before age 16 are sufficiently current for adjudication. The general guidelines are:

Age at time of testing

IQ less than 40

IQ 40 or greater

Before attainment of age 7

Current for 2 years

Current for 1 year

From age 7 to attainment of age 16

Current for 4 years

Current for 2 years

Intelligence test results obtained at age 16 or older may be considered current indefinitely, provided they are not inconsistent with the person’s current functioning.

For additional guidelines regarding current IQ scores in children, see DI 24583.060.

NOTE: IQ scores that are not current may still provide useful information about whether a person’s intellectual disorder began during the developmental period. For example, a full scale IQ of 88 obtained when the person was age 15 may provide evidence the intellectual disorder did not begin during the developmental period

8. Multiple or serial intelligence tests

The record may contain multiple intelligence tests administered to the same person. Generally, use the results from the most recently administered intelligence test if that test was programmatically acceptable. While only qualified specialists may determine an obtained IQ score is not an accurate reflection of the person's typical intellectual functioning, adjudicators must consider the extent to which intelligence test results support their determination whether a mental impairment meets or medically equals a listing in the context of the entire medical record. Important considerations include, but are not limited to, the following:

  • Was there a statement regarding validity included in the narrative report?

  • Was there an intervening event, such as a head injury, that might account for a significant decrease in obtained IQ scores?

  • Was the same version of the same intelligence test administered in close succession so that the practice effect might be a factor?

  • How old was the person at each intelligence test administration?

  • Are the most recent intelligence test results and the person's customary behavior and daily activities consistent with one another?

For information about considering and resolving inconsistencies, see DI 24501.016.

NOTE: When purchasing an additional intelligence test, provide a copy of the previous intelligence test results and report to the qualified specialist administering the new test.

J. References

  • DI 24501.016 Evidence Evaluation

  • DI 24508.010 Impairment or Combination of Impairments Equals a Listing – Medical Equivalence

  • DI 24583.010 Determining Medical Equivalence for Mental Impairments

  • DI 24583.050 Using Psychological Tests to Evaluate Mental Disorders

  • DI 24583.060 Additional Guidelines for Using Psychological Tests to Evaluate Mental Disorders in Children

  • DI 34001.032 Mental Disorders (Listing of Impairments – Adult Listings (Part A))

  • DI 34005.112 Mental Disorders (Listing of Impairments – Child Listings (Part B))


To Link to this section - Use this URL:
http://policy.ssa.gov/poms.nsf/lnx/0424583055
DI 24583.055 - Using Intelligence Tests to Evaluate Cognitive Disorders, Including Intellectual Disorder - 07/26/2023
Batch run: 07/26/2023
Rev:07/26/2023