Test Review: Gilliam,J. E. (2015). Attention-Deficit/Hyperactivity Disorder Test (2nd ed)

Test Description

The Attention-Deficit/Hyperactivity Disorder Test–Second Edition (ADHDT-2) is published through Pro-Ed in Austin, Texas. It was formally published in 2014, following critical revisions of the ADHDT, the reportedly popular initial version of this test that was published in 1995. The ADHDT-2 purports to act as a screener for individuals exhibiting patterns of inattention, hyperactivity, and impulsivity, symptoms commonly associated with attention-deficit/hyperactivity disorder (ADHD). Specifically, the ease and simplicity associated with this test target use in the school environment, targeting individuals 5 to 17 years of age. The ADHDT-2 is a rating scale to be completed by an individual who has observed the child for a substantial amount of time. The test manual suggests a classroom teacher, parent, or other caregiver who has frequent interaction with the child for at least 2 weeks. Selecting the most competent individual, specifically in regard to familiarity with the child, to complete the rating scale will likely provide the most valid results. Five primary purposes are elaborated upon as identification, assessment of severity of symptoms, assistance with the Individualized Education Program (IEP) for the student, documentation, and research. The ADHDT-2 was constructed for individual administration, though several rating scales may be requested from different individuals to gain more reliable and valid reporting of potential ADHD-like behaviors.

In total, the ADHDT-2 is a brief rating scale consisting of 33 Likert-type scale statements to which the rater must describe the child. The scale ranges from 0 to 3. A score of “0” means never observed, a score of “1” means occasionally observed, a score of “2” means often observed, and a score of “3” means very often observed. Further direction is provided for the rater, defining each score as the frequency a behavior is exhibited within a 6-hr period. “Occasionally observed” is when a behavior occurs 1 to 2 times per 6-hr period, “often observed” is when a behavior occurs 3 to 4 times per 6-hr period, and “very often observed” is when a behavior occurs at least 5 to 6 times per 6-hr period. The Summary/Response Forms are divided into two primary sections: Inattention and Hyperactivity/Impulsivity. The Inattention subscale is composed of 14 Likert-type scale statements, and the Hyperactivity/Impulsivity subscale is composed of 19 Likert-type scale statements, for a total of 33 statements. The subscales represent the primary behavioral factors commonly associated with ADHD, per the Diagnostic and Statistical Manual of Mental Disorders (5th ed.; DSM-V; American Psychiatric Association [APA], 2013). From each subscale, the scores are added to obtain an inattention raw score and a hyperactivity/impulsivity raw score. Age and gender norms are presented for conversion of raw scores of inattention and hyperactivity/impulsivity to percentile ranks and scaled scores. Age norms are divided into a 5 to 7 year age range and an 8 to 17 year age range. Gender norms are provided to convert sum of scaled scores to percentile ranks and ADHD indexes.

Overall, the test materials and stimuli appear very user friendly. The kit is provided in a sturdy box for storage. The materials appear to be appropriately designed. The test contains minimal materials allowing for an easily accessible and transportable test. The protocol is brief, appropriate for a screener-based test, minimizing the potentially burdensome task of one that is more lengthy or intensive. The Examiner’s Manual is clearly organized and user friendly. Explicit sections thoroughly explain the different components of the test. The manual is brief, though comprehensive. Adequate attention is given to score analysis and interpretation. Appropriate emphasis is provided regarding the interpretation of results in that the test should complement other data and not alone identify an ADHD diagnosis. No evidence of race or sex bias, or efforts to detect such bias, are presented.

Technical Adequacy

The test items for the ADHDT-2 were generated largely based on the DSM-5. Within the formal diagnosis of ADHD, the two primary subtypes are inattention and hyperactivity/impulsivity; therefore, the items were created based on behaviors that support these categories. The items are presented as a rating scale, and all items are expected to be answered to create scores for an individual’s likelihood of ADHD. The ADHDT-2 was normed on a sample of 1,591 students. Three brief inclusionary criteria were noted—a formal diagnosis of ADHD, being in the age range of 5 to 17 years, and residing in the United States. Comorbid conditions, gender preferences, and racial and ethnicity prevalence were noted; yet overall, the sample is reported to appropriately represent diversity within the ADHD population. No explicit rationale was provided regarding why the norm sample was based solely on children with a diagnosis of ADHD, though the decision to base norms on a clinical sample rather than a more general sample could be related to the purpose of the test.

Three estimates of reliability are provided within the ADHDT-2: internal consistency, test–retest, and interrater. Coefficient alpha was utilized to test for reliability among the test items. Coefficient alphas were calculated for the two subscales, inattention and hyperactivity/impulsivity, as well as the ADHD Index. Furthermore, coefficient alphas were calculated for scores of each of the 13 age groups (5-17 inclusively) and also subgrouped by gender (male and female) and race (White, Black/African American, Asian/Pacific Islander, American Indian/Eskimo/Aleut, two or more races). Overall, the individual and average coefficient alphas are reported at .90 and higher, signifying an excellent display of internal consistency. The coefficient alpha reliability scores for Hyperactivity/Impulsivity appear to be slightly higher than the scores for Inattention and ADHD Index, though to no significant degree. There were no significant differences in the coefficient alphas for age, gender, or race, though the scores for American Indian/Eskimo/Aleut appear to fall slightly lower than other subgroups across the Inattention, Hyperactivity/Impulsivity, and ADHD Index scores. Still, the scores are reported as .90, .95, and .91, respectively, still suggesting strong internal consistency. Test–retest reliability, otherwise acknowledged as stability reliability, was conducted on 70 children. The test was given twice approximately 2 weeks apart. A variety of relationships was represented through the raters. The test–retest reliability was reported as excellent, with corrected correlation coefficients ranging from .80 to .89. Interrater reliability was assessed using a sample of 65 children as rated by two separate individuals each. Interrater reliability was calculated through intraclass correlation (ICC). ICC revealed excellent agreement on all values—Inattention, Hyperactivity/Impulsivity, and ADHD Index. Overall, each estimate of reliability supports the notion that scores presented for the standardization sample of the ADHDT-2 have excellent reports of reliability.

Three tests of validity were conducted for the ADHDT-2: content-description validity, criterion-prediction validity, and construct-identification validity. The items were chosen based on author review of the Diagnostic and Statistical Manual of Mental Disorders (4th ed., text rev.; DSM-IV-TR; APA, 2000) and DSM-5, as well as additional measures focused on ADHD symptoms. Items that overlapped DSM criteria with ADHD diagnostic measures were compiled. Content-description validity assessed the ADHD domains represented by the items, selection and appropriateness of the items, and overall validity via classical item analysis. Through classical item analysis, item discrimination was studied. Overall, acceptable content-description validity was evident. Content-prediction validity was determined by analyzing the items in regard to predicting the presence of ADHD symptoms. For this aspect of validity, test creators first analyzed diagnostic criteria of ADHD and matched these criteria with the 33 total items included in the test. Correlations with other criterion measures of ADHD—Conners’ Teacher Rating Scale–Revised: Long (CTRS-R:L; Conners, 2001) and Vanderbilt ADHD Diagnostic Teacher Rating Scale (VADTRS; Wolraich, Feurer, Hannah, Baumgaertel, & Pinnock, 1998—were calculated. Overall, scores on the ADHDT-2 correlated very closely with scores on the other constructs evaluating ADHD, proposing that the scores calculated from the ADHDT-2 present appropriate criterion-prediction validity. Diagnostic accuracy was also evaluated, further supporting criterion-prediction validity, with a sensitivity index of .90 and a specificity index of .82. Finally, construct-identification validity was measured. Gender and age differentiations were evaluated, as it has been understood that ADHD can present differently between genders and across age spans. In addition, differentiations among diagnostic groups with individuals with no reported ADHD diagnosis though identified as having another behavioral or developmental problem such as a learning disability or autism spectrum disorder were made. Conclusively, differentiations for age, for gender, and among diagnostic groups reported good levels of validity. Exploratory factor analysis was also conducted, arriving at similar conclusions with strong levels of construct-identification validity.

Commentary and Recommendations

Overall, this test appears to be an adequate initial screener to identify an individual potentially presenting with the dominant ADHD behaviors of inattention and hyperactivity/impulsivity. The brevity and ease of the test would be especially useful in the school environment. Reliability and validity score estimates are also significant strengths of the test. It would be particularly important to note, as the test emphasizes, that the purpose of the test is to primarily act as a screener for ADHD symptoms, and further testing and evaluation would be necessary, specifically in regard to testing the items for potential biases. Still, the application of this test could be especially useful for the school-age population. In conclusion, the ADHDT-2 is an appropriate screener for ADHD, though caution is encouraged in regard to a formal diagnosis of ADHD. Supplementing this screening measure with complementary information such as classroom observations, developmental history, and additional cognitive assessment measures is highly recommended to gather the most reliable and valid interpretation of the individual’s ADHD-like behaviors.