Viability Of Machine-Learning Strategies To Solve Psychometric Problems

Validating scales for clinical purposes is a common process in medicine and psychology. Using machine-learning and statistics, we revalidated the Fenigstein & Vanable Paranoia Scale and showed that these kinds of approaches could be used both to achieve construct validity and criterion validity and could thus add an additional layer of evidence to traditional validation approaches. However, there is still a lot of work needed in order to evaluate the whole range of applications, disadvantages, and potential limitations of these approaches when applied for psychometric purposes.