Item reduction of the “Support Intensity Scale” for peoplewith intellectual disabilities, using machine learning
| Título | Item reduction of the “Support Intensity Scale” for peoplewith intellectual disabilities, using machine learning |
|---|---|
| Año | 2024 |
| Autor | Félix González Carrasco, Felipe Espinosa, Izaskun Álvarez-Aguado, Sebastián Ponce, Vanessa Vega Córdova, Miguel Roselló-Peñaloza |
| Filiación | PUCV/UDLA/UV |
| Tipo de Publicación | Artículo en Revista Académica |
| Revista | British Journal on Learning Disabilities |
| Indexación | ISI |
Background: The study focuses on the need to optimise assessment scales for support needs in individuals with intellectual and developmental disabilities. Current scales are often lengthy and redundant, leading to exhaustion and response burden.The goal is to use machine learning techniques, specifically item‐reduction methodsand selection algorithms, to develop shorter and more efficient scales.Methods: A data set of 93 participants was analysed using the Supports NeedsScale. Five feature‐selection algorithms were evaluated to create a shortenedquestionnaire. For each algorithm, a Random Forest model was trained, and per-formance was assessed using metrics like accuracy, precision, recall and F1‐score tomeasure how well each model predicted support needs.Findings: The "Select from Model" algorithm successfully identified key items thatcould predict the level of Support Needs using the Random Forest model. Only51 variables, out of the original 147, were needed to maintain predictive accuracy.The reduced questionnaire maintained good reliability and internal consistencycompared to the original instrument, with a strong F1 score indicating excellent predictive performance.Conclusions: The study demonstrates that machine learning techniques are effectivein reducing the length of support needs questionnaires while preserving their psy-chometric properties. These methods can help institutions provide more efficientaccess to information about support needs without compromising validity or reliability, potentially leading to better resource allocation and improved care for individuals with intellectual disabilities.
