Reliability & Validity of the ASRS

The Autism Spectrum Rating Scales (ASRS) measure specific behaviors associated with Autism Spectrum Disorder (ASD). The importance of a rating scale such as this can help prevent under or over diagnosis which can lead to improper treatment and poor continued development.

To diagnose ASD early means that the proper care and interventions can be implemented to help the individual with ASD lead a relatively normal life with the tools they need to manage their disorder.

The ASRS is used for children aged 2 to 18 years. As with any diagnostic assessment, it is critical that reliability and validity be maintained in order to properly diagnose and treat those with ASD.

Reliability

Reliability comes in the forms of internal consistency and test-retest reliability.

Internal consistency measures how all items on a scale can consistently measure the content. This assessment is typically examined using a range from 0.0 to 1.0. A high level of consistency is demonstrated by a 1.0 rating.

Test-retest reliability is measured by the stability of two test scores after the assessment is administered on two different occasions. These scores are computed over a two- to four-week interval.

Reliability testing has demonstrated an internal consistency for the ASRS of 0.92 or higher in all areas, with an overall score of 0.97. Test-retest reliability for the ASRS is 0.90 to 0.92 for parent ratings and 0.78 to 0.81 for teacher ratings, depending on the age range of the child being tested.

Validity

Validity is measured by discriminative validity and convergent validity.

Discriminative validity measures the ASRS’s ability to tell the difference between participant groups. For instance, the assessment should be able to distinguish between a group with Autism Spectrum Disorder and a group comprised of the general population. It should also be able to distinguish between a group with ASD and another clinical group.

In order to establish discriminative validity, a series of function analyses were conducted to see how well the scores could predict a group of individuals with ASD or a general population group.

The ASRS can predict group membership and correctly classify the group with 92.1% accuracy. The assessment also rates ASD groups higher than general population groups, revealing that the differences between the two groups are statistically significant.

Convergent validity is defined by a correlation between the results of the measure of interest and the results of the theoretically-related measures. To establish convergent validity in the ASRS, parents and teachers completed the assessment using at least one other measure that assesses ASD.

Some of the other measures used include the Gilliam Autism Rating Scale Second Edition (GARS-2), the Gilliam Asperger’s Disorder Scale (GADS), and the Childhood Autism Rating Scale (CARS).

While the correlations have to take into account the differences in metrics for each rating scale, the scores from each assessment when compared to a general population group demonstrate convergent validity in the ASRS.

Accurate assessment of individuals with ASDs for proper diagnosis and treatment is critical for the individual and the caregiver. Under or over diagnosis results in improper care and courses of continued treatment for individuals with ASDs.

The reliability and validity of the ASRS ensures that every individual assessed can be distinguished from an individual of the general population and can be diagnosed correctly for proper courses of treatment.

For more information about the ASRS, visit WPS Publish online.