Overview
Interview-Revised (ADI-R), and the GARS found that the ADOS and ADI-R led to approximately 75% agreement with team diagnoses, but the GARS was generally ineffective at discriminating between children with various team-determined diagnoses and consistently underestimated the likelihood of ASD. Regarding the GARS-2, Volker et al. (2016) found that factor structure was similar to that of the GARS but that more than one-third of GARS-2 items were assigned to the wrong subscale; however, these researchers also pointed out that internal consistency estimates met or exceeded standards for screening and were generally higher than those in previous studies. Importantly, in a test review, Karren (2017) noted that the GARS-3 is significantly different from its earlier versions, retaining only 16 items from the previous versions and adding 42 new items. In addition, this reviewer noted that the GARS-3 has undergone study that resulted in a conclusion that it is a psychometrically sound instrument; moreover, a noted strength is its attempt to link assessment with intervention by including a booklet that includes recommended instructional objectives for each item. However, Karren (2017) further noted that no psychometrics have been assessed for items in which raters cannot accurately answer, a practice that also has questionable pragmatism. In addition, the norming sample for the GARS-3 consists only of individuals with ASD diagnoses, meaning that target children’s ratings can only be compared with an ASD sample. Finally, the sample has limited racial and ethnic diversity and may be limited in utility for individuals between 20-22 years of age.
Summary
Age: 3 years - 22 years
Time to Administer: 5-10 minutes
Method of Administration: Rating scale of 58 items on 6 subscales based on the DSM-5 ASD diagnostic criteria, completed by parent, teacher, or clinician; it can also be completed as a structured interview Yields standard scores, percentile ranks, severity level, and probability of autism
Subscales: Overall Composite Score: Autism Index Score
Subscale Scores: Restrictive, Repetitive Behaviors; Social Interaction; Social Communication; Emotional Responses; Cognitive Style; Maladaptive Speech
Screening/Diagnosis: D
Autism Related Research
Although no research was located that elucidates the utility of the GARS-3 (aside from the manual [Gilliam, 2014]), there have been findings relevant to previous versions of the GARS that users may consider. Mazefsky and Oswald (2006) found in their study of discriminative validity of the Autism Diagnostic Observation Schedule (ADOS), Autism Diagnostic Interview-Revised (ADI-R), and the GARS found that the ADOS and ADI-R led to approximately 75% agreement with team diagnoses, but the GARS was generally ineffective at discriminating between children with various team-determined diagnoses and consistently underestimated the likelihood of ASD. Regarding the GARS-2, Volker et al. (2016) found that factor structure was similar to that of the GARS but that more than one-third of GARS-2 items were assigned to the wrong subscale; however, these researchers also pointed out that internal consistency estimates met or exceeded standards for screening and were generally higher than those in previous studies. Importantly, in a test review, Karren (2017) noted that the GARS-3 is significantly different from its earlier versions, retaining only 16 items from the previous versions and adding 42 new items. In addition, this reviewer noted that the GARS-3 has undergone study that resulted in a conclusion that it is a psychometrically sound instrument; moreover, a noted strength is its attempt to link assessment with intervention by including a booklet that includes recommended instructional objectives for each item. However, Karren (2017) further noted that no psychometrics have been assessed for items in which raters cannot accurately answer, a practice that also has questionable pragmatism. In addition, the norming sample for the GARS-3 consists only of individuals with ASD diagnoses, meaning that target children’s ratings can only be compared with an ASD sample. Finally, the sample has limited racial and ethnic diversity and may be limited in utility for individuals between 20-22 years of age.