Validation of the Parent-Report and Teacher-Report Social Responsiveness Scale (SRS) in the Netherlands

Background: The Social Responsiveness Scale (SRS) is a widely used Autism Spectrum Disorder (ASD) questionnaire that can be completed by parents and teachers. In several countries, the SRS has been found to be able to discriminate between children with ASD and children with other psychiatric disorders. However, the validity of the Dutch version of the SRS in a high-risk sample remains to be investigated. In addition, few studies have investigated the validity of the teacher-report SRS. 

Objectives: The aims of the present study were to investigate 1) agreement between the parent-report and teacher-report SRS scores and 2) correspondence of the parent-report and teacher-report SRS scores to ASD classifications on standardised diagnostic instruments: the Autism Diagnostic Observation Schedule (ADOS) and the Developmental, Dimensional and Diagnostic Interview (3Di).

Methods: Our sample consisted of 283 children aged 4-10 years who had been referred to one of six mental health care centres in the Netherlands. For each consecutive referral, both a parent and a teacher were asked to complete the SRS. For research purposes, children with a positive screen on the parent-report SRS (total raw score ≥75, n = 390) and a random selection of children with a negative screen (n = 202) were invited for in-depth diagnostic assessments including the ADOS and the 3Di. Of these, 203 screen positives and 80 screen negatives participated in at least one diagnostic assessment. For n = 255 (90%) we also had a completed teacher-report SRS. Pearson correlation was used to investigate parent-teacher agreement on the SRS. Receiver Operating Characteristic (ROC) analyses were used to examine the ability of the parent-report and teacher-report SRS to predict ASD classifications on either the ADOS or the 3Di or on both. 

Results: In the total sample of n = 283, 43% (109 out of 251) of the children were classified as having ASD on the ADOS, 48% (115 out of 241) on the 3Di and 26% (55 out of 209) on the ADOS as well as the 3Di. The correlation between the parent-report and teacher-report SRS scores was r = .28 (p < .01). For the parent-report SRS the Area Under Curve (AUC) of the ROC was .63 for predicting an ASD classification on the ADOS, .83 for the 3Di classification, and .82 for a classification on both the ADOS and the 3Di. For the teacher-report SRS the AUCs were .68 (ADOS), .63 (3Di), and .71 (ADOS and 3Di).

Conclusions: The parent-report and teacher-report SRS were only moderately related, suggesting that both informants provide some unique information. We found further support for the cross-cultural validation of the SRS. However, our preliminary results indicate that the parent-report SRS may correspond better to ASD classifications on standardised diagnostic instruments than the teacher-report SRS. We are currently investigating how the teacher-report SRS can contribute to the diagnostic assessment of ASD.