A Non-parametric Analysis of Gender Differences With the Impacts of Applied Behavior Analysis on Autistic Individuals

Abstract

IntroductionConsidering the scarcity of research that directly investigates the differences between genders in their response to applied behavior analysis (ABA) therapy for individuals diagnosed with autism spectrum disorder (ASD), the objective of this study is twofold. First, it aims to reinforce the male-to-female ratio reported in existing scientific literature, thereby contributing to a broader understanding of gender distribution in ABA therapy for ASD. Second, it seeks to identify gender-based differences in aggregate target behaviors at various time intervals using three distinct datasets. The goal is to determine if gender influences the effectiveness of ABA therapy for ASD, which could inform future therapeutic strategies. Ultimately, this study strives to enhance our understanding of gender disparities in ABA therapy response among ASD individuals and hopes to improve therapeutic outcomes for all, regardless of gender. Materials and methodsThree to five behavioral technicians per child collected daily general target mastery data for 263 individuals with autism. This data was gathered using a large N design through retrospective chart reviews within the "Catalyst" tracking software (DataFinch Technologies, Atlanta, USA). Three separate datasets were collected from June 7, 2023 to January 7, 2024. Behavior analysts employed a mixed model of discrete trial training, mass trials, and naturalistic environment treatment over seven months. General target mastery data was assembled for 259 children and four adults, with seven data instances missing. Descriptive statistics encompassed central tendency and dispersion measures, including the data distribution's mean, standard deviation, median, and range. Non-parametric inferential analysis was performed with the Mann-Whitney U test. ResultsMann-Whitney U computations resulted in non-significant gender differences on all gender comparisons for the three datasets covering the seven-month timeframe.