Spotlights: Egidio Robusto
Egidio Robusto, a Professor of Psychometrics at the University of Padua since 2006, has a distinguished career in psychological research, particularly in the area of statistical and mathematical modelling applied to psychology. Prior to his current tenure, he served as a researcher at the Second University of Naples (now known as University of Campania Luigi Vanvitelli) and as an associate professor at Sapienza University of Rome. His academic journey reflects a deep commitment to advancing the field of psychometrics. As of January 2024 he is Director of FISPPA (Department Of Philosophy, Sociology, Pedagogy And Applied Psychology).
Research Interests
Prof. Robusto’s research interests are extensive and primarily focus on the application of formal modelling in psychology. This encompasses the development of statistical and mathematical models for the assessment of knowledge and skills, the construction and validation of measurement instruments, and the exploration of item response theory and Rasch models. His work significantly contributes to the understanding of how psychological attributes can be quantified and analysed.
Teaching
In his role as a lecturer, Dr. Robusto imparts knowledge in psychometrics and analysis models for latent variables. His teaching, deeply rooted in his research interests, reflects his expertise in latent variable analysis, showcasing his ability to effectively translate complex statistical concepts into comprehensive learning material.
Highly cited
Anselmi, P., Colledani, D., & Robusto, E. (2019). A Comparison of Classical and Modern Measures of Internal Consistency. Frontiers in Psychology, 10.
This study compares Kuder-Richardson Formula 20 (KR20), Cronbach’s alpha (α), and person separation reliability (R) for internal consistency. KR20 and α may overestimate error variance, while R, based on Rasch measurement, provides a more conservative and linearly accurate measure. The research emphasizes R’s superiority in skewed score distributions and the utility of Rasch-based statistics for addressing random responses.
Book
Robusto, E., & Cristante, F. (2001). Analisi log-lineare di variabili psicosociali: Vol. 1. Introduzione ai modelli fondamentali. LED Edizioni Universitarie.
Log-linear models are essential in social sciences and psychology, evolving since the 1970s to advance beyond chi-square models and bivariate analysis. This book explores estimating effects and investigating their sources under the saturated model hypothesis. It covers concepts like event independence and distributions, bivariate and multivariate model applications, and real-world data interpretation challenges in six chapters.