Hudson Golino

My research focuses on developing new quantitative techniques and metrics, integrated into a general approach – Exploratory Graph Analysis (EGA) – which is part of the relatively new area of network psychometrics. This work combines network science, information, and quantum information theory, as well as computational methods to address fundamental problems in psychometrics, with the following goals: 




  • Improve the estimation of the number of latent factors in an automatic (or semi-automatic) way.
  • Develop innovative fit indices for structural analysis and dimensionality assessment/reduction.
  • Improve the estimation and the interpretability of latent factors in intensive longitudinal data.
  • Develop new techniques for item analysis from a network psychometrics perspective (including, for example, network loadings, item parameters, and new reliability metrics).
  • Improve text mining (via new techniques to detect emotions in text and estimate latent topics).
  • Construct general representations of structure built from intraindividual variability, quantifying individuals' homogeneity using new complexity metrics.

In the coming years, I envision a research program that fuses the power of Large Language Transformer Models (LLMs) with psychometric methods to revolutionize various fields of science. My plan encompasses the development of innovative tools, such as the GRC Approach for Automatic Item GenerationMonticello Simulations, and Synthetic Data Generation with LLMs, which will expedite research across domains.

I collaborate with applied researchers from the U.S. and abroad in conducting research on intelligence, cognition, aging, and other topics. He also has an active line of research in text mining, machine learning, and large language transformer models.

My main package, EGAnet for R, developed in partnership with Alexander Christensen (Vanderbilt University) and other collaborators, is one of the leading packages in the area of network psychometrics, with more than 3,000 downloads per month.

I have published in many high-level methodological journals, including Psychological Methods, Psychometrika, Multivariate Behavior Research, Behavioral Research Methods, Journal of Behavioral Data Science, and Assessment, as well as applied journals, such as Nature Scientific Reports, Intelligence, Psychological Test Development and Adaptation, Journal of Intelligence, and European Journal of Personality, and others.

In 2012 I was awarded the International Test Commission Young Scholar Scholarship. In 2015 I received the Sanofi Innovation in Medical Services award for developing a system to improve the prediction accuracy of outcomes in intensive care units using machine learning models.

Golino completed his Ph.D. in March 2015 at the Universidade Federal de Minas Gerais (Brazil), where he studied applications of machine learning in Psychology, Education, and Health. 

Golino also holds an M.Sci. in Developmental Psychology (2012), and a B.Sci. in Psychology (2011), all from Universidade Federal de Minas Gerais. At UVA, he will teach undergraduate and graduate courses on quantitative methods at the Department of Psychology. He expects to offer courses on applied machine learning for Psychologists and on constructing and validating assessment instruments.

Currently, his research focuses on 1) the integration of information theory and network psychometrics to estimate the number of latent factors; 2) the development of non-linear dimensionality assessment techniques based on network psychometrics; 3) the development of new topic modeling techniques based on network psychometrics; 4) text mining and machine learning applied to several fields (including political science and cyber-security).

I am always looking for new collaborators and creative and innovative ideas and applications.