Abstract

BackgroundThe 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 methods and selection algorithms, to develop shorter and more efficient scales.MethodsA data set of 93 participants was analysed using the Supports Needs Scale. Five feature-selection algorithms were evaluated to create a shortened questionnaire. For each algorithm, a Random Forest model was trained, and performance was assessed using metrics like accuracy, precision, recall and F1-score to measure how well each model predicted support needs.FindingsThe Select from Model algorithm successfully identified key items that could predict the level of Support Needs using the Random Forest model. Only 51 variables, out of the original 147, were needed to maintain predictive accuracy. The reduced questionnaire maintained good reliability and internal consistency compared to the original instrument, with a strong F1 score indicating excellent predictive performance.ConclusionsThe study demonstrates that machine learning techniques are effective in reducing the length of support needs questionnaires while preserving their psychometric properties. These methods can help institutions provide more efficient access to information about support needs without compromising validity or reliability, potentially leading to better resource allocation and improved care for individuals with intellectual disabilities. This study helps make a tool called the Supports Intensity Scale shorter and easier to use. This tool measures what kind of help people with intellectual disabilities need. Right now, the tool has many questions, which can be tiring for people to answer. We used a method called machine learning, which is a type of computer programme, to find out which questions are most important. We looked at answers from 93 people to do this. We found that we could use just 51 questions instead of 147 to get the same results. This shorter version of the tool still works well and helps us understand what support people need. It makes it easier for people to answer and for helpers to understand their needs. This can lead to better care and support for people with intellectual disabilities.