Sophia Rabe-Hesketh

About

Sophia Rabe-Hesketh is a distinguished statistician with a focus on multilevel/hierarchical modeling, item response theory, longitudinal data analysis, and missing data. She has authored over 130 peer-reviewed articles across more than 70 journals, including Psychometrika, Journal of Econometrics, Biometrics, and the Journal of the Royal Statistical Society, Series A, and holds an h-index of 74 in Google Scholar. Her notable contribution includes the development of the GLLAMM (Generalized Linear Latent and Mixed Modeling) framework for a wide range of multilevel and latent variable models, along with a publicly available software package called gllamm, which has been utilized in over 1000 peer-reviewed papers since 2002. Her work has significantly impacted research in education, sociology, political science, economics, medicine, and statistics. Rabe-Hesketh has held prominent roles such as President of the Psychometric Society and elected member of the National Academy of Education. She is also a member of the Interdepartmental Group in Biostatistics at the University of California, Berkeley, and has previously served as Professor of Social Statistics at the Institute of Education, University of London.

Selected publications

• Bayesian Comparison of Latent Variable Models: Conditional Versus Marginal Likelihoods

  Psychometrika · 2019-07-11 · 106 citations

  articleOpen accessSenior author

  Typical Bayesian methods for models with latent variables (or random effects) involve directly sampling the latent variables along with the model parameters. In high-level software code for model definitions (using, e.g., BUGS, JAGS, Stan), the likelihood is therefore specified as conditional on the latent variables. This can lead researchers to perform model comparisons via conditional likelihoods, where the latent variables are considered model parameters. In other settings, however, typical m…

  PublisherOA PDFDOI

• Trifactor Models for Multiple-Ratings Data

  Multivariate Behavioral Research · 2019-03-28 · 53 citations

  article

  In this study we extend and assess the trifactor model for multiple-ratings data in which two different raters give independent scores for the same responses (e.g., in the GRE essay or to subset of PISA constructed-responses). The trifactor model was extended to incorporate a cross-classified data structure (e.g., items and raters) instead of a strictly hierarchical structure. we present a set of simulations to reflect the incompleteness and imbalance in real-world assessments. The effects of th…

  PublisherDOI

• Computational Strategies and Estimation Performance With Bayesian Semiparametric Item Response Theory Models

  Journal of Educational and Behavioral Statistics · 2022 · 15 citations

  Item response theory (IRT) models typically rely on a normality assumption for subject-specific latent traits, which is often unrealistic in practice. Semiparametric extensions based on Dirichlet process mixtures (DPMs) offer a more flexible representation of the unknown distribution of the latent trait. However, the use of such models in the IRT literature has been extremely limited, in good part because of the lack of comprehensive studies and accessible software tools. This article provides g…

  PublisherDOI

• Role of marriage, motherhood, son preference on adolescent girls’ and young women’s empowerment: Evidence from a panel study in India

  PLoS ONE · 2023-09-28 · 11 citations

  articleOpen accessSenior author

  BACKGROUND: Marriage is a key determinant of health and well-being of adolescent girls and young women (AGYW) in India. It is a key life event in which girls move to their marital households, often co-residing with their in-laws and begin childbearing. The change in the normative environment in conjunction with cultural norms surrounding son preference influences women's overall life course. However, there is scant research about the association between these life transitions and changes in empo…

  PublisherDOI

• Ignoring Non-ignorable Missingness

  Psychometrika · 2022-12-20 · 11 citations

  articleOpen access1st authorCorresponding

  The classical missing at random (MAR) assumption, as defined by Rubin (Biometrika 63:581-592, 1976), is often not required for valid inference ignoring the missingness process. Neither are other assumptions sometimes believed to be necessary that result from misunderstandings of MAR. We discuss three strategies that allow us to use standard estimators (i.e., ignore missingness) in cases where missingness is usually considered to be non-ignorable: (1) conditioning on variables, (2) discarding mor…

  PublisherOA PDFDOI

Awards & honors

• Elected to the National Academy of Education (2015)
• President of the Psychometric Society (2014-2015)

Similar researchers at University of California, Berkeley

• Marcia C. Linnh-75
• Alan H. Schoenfeldh-54
• Elliot Turielh-54