Your new experience awaits. Try the new design now and help us make it even better

ORIGINAL RESEARCH article

Front. Neurol.

Sec. Pediatric Neurology

Volume 16 - 2025 | doi: 10.3389/fneur.2025.1648991

This article is part of the Research TopicNew Insights into Pediatric Neurodevelopmental Disorders: Autism Spectrum Disorder and its ComorbiditiesView all 4 articles

Differential Item Functioning in the Children Autism Rating Scale First Edition in Children with Autism Spectrum Disorder Based on A Machine Learning Approach

Provisionally accepted
Kanglong  PengKanglong Peng1*Meng  ChenMeng Chen2Libing  ZhouLibing Zhou2Xiaofang  WengXiaofang Weng2
  • 1Shenzhen Children's Hospital, Shenzhen, China
  • 2Luohu District Maternity and Child Health-care Hospital, Shenzhen, China

The final, formatted version of the article will be published soon.

Purpose: Our study used Rasch Analysis to examine the psychometric properties of the Children Autism Rating Scale First Edition (CARS1) in children with autism spectrum disorder (ASD).Methods: The Partial Credit Model (PCM) was used to test reliability and validity. The GPCMlasso Model was used to test the differential item functioning (DIF).Results: The response pattern of this sample showed acceptable fitness for the PCM. This analysis supports the unidimensionality assumption of the CARS1. Disordered category functions and DIF were found for all items in CARS1. Performance can be related to age group, gender, symptom classification, and autistic symptoms.Conclusion: Rasch analysis provides reliable evidence to support the clinical application of the CARS1. Some items may produce inaccurate measurements originating from unreasonable category structures. Differences in age group, sex, and symptom classification can be related to test performance and may lead to unnecessary bias. Hence, clinical applications may require additional consideration of population characteristics to draw reliable conclusions.

Keywords: CARS1, Rasch model, Category function, Differential Item Functioning, ASD, machine learning

Received: 18 Jun 2025; Accepted: 13 Aug 2025.

Copyright: © 2025 Peng, Chen, Zhou and Weng. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

* Correspondence: Kanglong Peng, Shenzhen Children's Hospital, Shenzhen, China

Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.