About

Dr. Catherine (Kate) Calder is a Professor in the Department of Statistics and Data Sciences at the University of Texas at Austin and serves as the Senior Vice Provost for Academic Administration and Funding. She previously spent 16 years at The Ohio State University as a faculty member in the Department of Statistics. From 2018 to 2019, she also co-directed the Mathematical Biosciences Institute, one of eight NSF-funded Mathematical Sciences Research Institutes. Her research focuses on the development of statistical methodology for complex, structured data, including spatial statistics and relational data. She has contributed to dimension reduction techniques for spatio-temporal data, covariate-driven nonstationary spatial models, hierarchical pathways models for exposure assessment, data-augmentation algorithms for spatial generalized linear models, and latent space models for relational data, among others. Her current applied projects aim to better measure routine activity patterns of individuals using mobile-tracking devices to understand social ties and the impact of non-residential activity space exposures on youth health and well-being. Her research has been funded by NIH, NSF, and other federal agencies and foundations. Dr. Calder holds a B.A. in Mathematics from Northwestern University, and both an M.S. and Ph.D. in Statistics from Duke University. She is a Fellow of the American Statistical Association, the American Association for the Advancement of Science, and the Institute of Mathematical Statistics, and has received the Founders Award from the American Statistical Association.

Research topics

• Sociology
• Geography
• Psychology
• Social Science
• Computer Science
• Demography
• Mathematics
• Cartography
• Psychiatry
• Data science
• Demographic economics
• Medicine
• Gerontology
• Economic growth
• Statistics
• Criminology
• Environmental health

Selected publications

• Geographic isolation, compelled mobility, and adolescent risk behavior

  Urban Studies · 2026-03-18

  articleSenior author

  We explore alternative hypotheses regarding the association between activity space racial composition, and risk behavior among Black-identifying urban youth. Racial isolation perspectives argue that exposure to Black-segregated neighborhoods limits access to mainstream institutions and influence, increasing participation in risk behavior (violence, delinquency, and substance/alcohol use). An alternative compelled mobility perspective argues that Black youth spend a substantial amount of time in low-proportion Black, largely white neighborhoods seeking organizational resources typically less available in segregated areas. These exposures may lead to discrimination-related strain, detachment from conventional norms, and elevated physiological stress, increasing the likelihood of risk behavior compared to Black youth who spend more time in same-race-dominated activity spaces. We test these competing hypotheses employing data from the Columbus, Ohio, USA-based Adolescent Health and Development in Context study on the geospatial exposures and both survey and ecological momentary assessment-reported behaviors of 506 Black youth aged 11–17. Contrary to the expectations of the isolation model, we find that greater exposure to residentially low-proportion Black areas is associated with an increased likelihood of risk behavior for Black males. We consider implications of findings for extant theories and data collection approaches in research examining spatial effects on adolescent risk behavior.

  PublisherDOI

• Markov Random Fields: Structural Properties, Phase Transition, and Response Function Analysis

  Open MIND · 2026-02-02

  preprintSenior author

  This paper presents a focused review of Markov random fields (MRFs)--commonly used probabilistic representations of spatial dependence in discrete spatial domains--for categorical data, with an emphasis on models for binary-valued observations or latent variables. We examine core structural properties of these models, including clique factorization, conditional independence, and the role of neighborhood structures. We also discuss the phenomenon of phase transition and its implications for statistical model specification and inference. A central contribution of this review is the use of response functions, a unifying tool we introduce for prior analysis that provides insight into how different formulations of MRFs influence implied marginal and joint distributions. We illustrate these concepts through a case study of direct-data MRF models with covariates, highlighting how different formulations encode dependence. While our focus is on binary fields, the principles outlined here extend naturally to more complex categorical MRFs and we draw connections to these higher-dimensional modeling scenarios. This review provides both theoretical grounding and practical tools for interpreting and extending MRF-based models.

  DOI

• Hyperbolic Latent Space Models for Network Embedding: Model Specification and Bayesian Inference

  arXiv (Cornell University) · 2026-05-11

  preprintOpen accessSenior author

  Many real-world networks exhibit hierarchical, tree-like structure and heavy-tailed degree distributions, phenomena not readily captured by standard statistical models for network data. Extensions of the popular continuous latent space modeling framework have been proposed to accommodate such networks. Drawing on insights from statistical physics, continuous latent space models with underlying hyperbolic geometry have been proposed as a natural framework, probabilistically embedding nodes in a latent Riemannian manifold with constant negative curvature. Most statistical implementations, however, simplify the original physics-based model by omitting the ``temperature parameter," which controls the sharpness of the latent distance-to-probability mapping. We argue this omission is critical. We demonstrate that temperature is the fundamental parameter governing a network's tree-like topology, and that failing to infer it weakens model expressiveness. We formalize a Bayesian hyperbolic continuous latent space model with an unknown, learnable temperature parameter. We then develop two inferential procedures: a Hamiltonian Monte Carlo approach for rigorous posterior characterization and a scalable auto-encoding variational Bayes algorithm for large-scale networks. Through simulation and real data examples, we show that our model outperforms models with fixed temperature and misspecified Euclidean geometries in graph reconstruction tasks in most settings, confirming temperature is a crucial and inferable feature of complex networks.

  PublisherDOI

• Hyperbolic Latent Space Models for Network Embedding: Model Specification and Bayesian Inference

  ArXiv.org · 2026-05-11

  articleOpen accessSenior author

  Many real-world networks exhibit hierarchical, tree-like structure and heavy-tailed degree distributions, phenomena not readily captured by standard statistical models for network data. Extensions of the popular continuous latent space modeling framework have been proposed to accommodate such networks. Drawing on insights from statistical physics, continuous latent space models with underlying hyperbolic geometry have been proposed as a natural framework, probabilistically embedding nodes in a latent Riemannian manifold with constant negative curvature. Most statistical implementations, however, simplify the original physics-based model by omitting the ``temperature parameter," which controls the sharpness of the latent distance-to-probability mapping. We argue this omission is critical. We demonstrate that temperature is the fundamental parameter governing a network's tree-like topology, and that failing to infer it weakens model expressiveness. We formalize a Bayesian hyperbolic continuous latent space model with an unknown, learnable temperature parameter. We then develop two inferential procedures: a Hamiltonian Monte Carlo approach for rigorous posterior characterization and a scalable auto-encoding variational Bayes algorithm for large-scale networks. Through simulation and real data examples, we show that our model outperforms models with fixed temperature and misspecified Euclidean geometries in graph reconstruction tasks in most settings, confirming temperature is a crucial and inferable feature of complex networks.

  PublisherOA PDF

• Markov Random Fields: Structural Properties, Phase Transition, and Response Function Analysis

  ArXiv.org · 2026-02-02

  articleOpen accessSenior author

  This paper presents a focused review of Markov random fields (MRFs)--commonly used probabilistic representations of spatial dependence in discrete spatial domains--for categorical data, with an emphasis on models for binary-valued observations or latent variables. We examine core structural properties of these models, including clique factorization, conditional independence, and the role of neighborhood structures. We also discuss the phenomenon of phase transition and its implications for statistical model specification and inference. A central contribution of this review is the use of response functions, a unifying tool we introduce for prior analysis that provides insight into how different formulations of MRFs influence implied marginal and joint distributions. We illustrate these concepts through a case study of direct-data MRF models with covariates, highlighting how different formulations encode dependence. While our focus is on binary fields, the principles outlined here extend naturally to more complex categorical MRFs and we draw connections to these higher-dimensional modeling scenarios. This review provides both theoretical grounding and practical tools for interpreting and extending MRF-based models.

  PublisherOA PDF

• Mobility Network-based Measurement of Local Collective Efficacy and its Consequences for the Spatial Patterning of Violent Crime

  Journal of Quantitative Criminology · 2025-08-12

  articleOpen access1st author

  Abstract Objectives Estimate the extent to which local variation in collective efficacy, a measure of social cohesion and norms toward intervention among individuals, is associated with the sub-neighborhood spatial patterning of violent crime in Columbus, OH. Methods Using estimates of local collective efficacy derived from survey data on individuals’ perceptions of collective efficacy in the neighborhoods and at their routine activity locations collected as part of the Adolescent Health and Development in Context Study and incident-level, point-referenced crime data from the Ohio Incident-Based Reporting System, we fit inhomogeneous Poisson process models. Results We find that net of neighborhood-level collective efficacy, a one standard deviation increase in deviation in the local collective efficacy score from the neighborhood average local collective efficacy score is associated with a decline in the violent crime intensity by a factor of 0.858. Conclusions Affirming Jane Jacobs’ arguments about informal social control dynamics at fine-grained levels, our study illustrates that collective efficacy operates and can be measured at sub-neighborhood levels across the entirety of a city. Our findings highlight the need to measure social processes at finer spatial scales in order to understand their effects on crime. Future research should prioritize collecting additional fine-grained data on social processes using individuals’ perceptions of the locations they frequent as part of their everyday mobility patterns.

  PublisherOA PDFDOI

• Toward a Data Processing Pipeline for Mobile-Phone Tracking Data

  ArXiv.org · 2025-07-01

  preprintOpen access

  As mobile phones become ubiquitous, high-frequency smartphone positioning data are increasingly being used by researchers studying the mobility patterns of individuals as they go about their daily routines and the consequences of these patterns for health, behavioral, and other outcomes. A complex data pipeline underlies empirical research leveraging mobile phone tracking data. A key component of this pipeline is transforming raw, time-stamped positions into analysis-ready data objects, typically space-time "trajectories." In this paper, we break down a key portion of the data analysis pipeline underlying the Adolescent Health and Development in Context (AHDC) Study, a large-scale, longitudinal study of youth residing in the Columbus, OH metropolitan area. Recognizing that the bespoke "binning algorithm" used by AHDC researchers resembles a time-series filtering algorithm, we propose a statistical framework - a formal probability model and computational approach to inference - inspired by the binning algorithm for transforming noisy, time-stamped geographic positioning observations into mobility trajectories that capture periods of travel and stability. Our framework, unlike the binning algorithm, allows for formal smoothing via a particle Gibbs algorithm, improving estimation of trajectories as compared to the original binning algorithm. We argue that our framework can be used as a default data processing tool for future mobile-phone tracking studies.

  PublisherOA PDFDOI

• Racial Differences in Activity Space Exposures and Everyday Perceptions of Safety Among Urban Youth

  Journal of Adolescent Health · 2024-03-12 · 8 citations

  articleOpen access
  PublisherOA PDFDOI

• Leveraging Experience Sampling/Ecological Momentary Assessment for Sociological Investigations of Everyday Life

  Annual Review of Sociology · 2024-04-17 · 12 citations

  articleOpen access

  Experience sampling (ES) - also referred to as ecological momentary assessment (EMA) - is a data collection method that involves asking study participants to report on their thoughts, feelings, behaviors, activities, and environments in (or near) real time. ES/EMA is typically administered using an intensive longitudinal design (repeated assessments within and across days). Although use of ES/EMA is widespread in psychology and health sciences, uptake of the method among sociologists has been limited. We argue that ES/EMA offers key advantages for the investigation of sociologically relevant phenomena, particularly in light of recent disciplinary emphasis on investigating the everyday mechanisms through which social structures and micro (individual and relational) processes are mutually constitutive. We describe extant and potential research applications illustrating advantages of ES/EMA regarding enhanced validity, disentangling short-term dynamics, and the potential for linkage with spatially and temporally referenced data sources. We also consider methodological challenges facing sociological research using ES/EMA.

  PublisherDOI

• Land-use filtering for nonstationary spatial prediction of collective efficacy in an urban environment

  The Annals of Applied Statistics · 2024-01-31 · 5 citations

  articleOpen accessSenior author

  Collective efficacy-the capacity of communities to exert social control toward the realization of their shared goals-is a foundational concept in the urban sociology and neighborhood effects literature. Traditionally, empirical studies of collective efficacy use large sample surveys to estimate collective efficacy of different neighborhoods within an urban setting. Such studies have demonstrated an association between collective efficacy and local variation in community violence, educational achievement, and health. Unlike traditional collective efficacy measurement strategies, the Adolescent Health and Development in Context (AHDC) Study implemented a new approach, obtaining spatially-referenced, place-based ratings of collective efficacy from a representative sample of individuals residing in Columbus, OH. In this paper we introduce a novel nonstationary spatial model for interpolation of the AHDC collective efficacy ratings across the study area, which leverages administrative data on land use. Our constructive model specification strategy involves dimension expansion of a latent spatial process and the use of a filter defined by the land-use partition of the study region to connect the latent multivariate spatial process to the observed ordinal ratings of collective efficacy. Careful consideration is given to the issues of parameter identifiability, computational efficiency of an MCMC algorithm for model fitting, and fine-scale spatial prediction of collective efficacy.

  PublisherOA PDFDOI

Recent grants

• Bayesian Methods for Socio-Spatial Point Patterns and Networks

  NSF · $180k · 2012–2016

• Adolescent Health in an Urban Environment

  NIH · $1.1M · 2017–2024

• Adolescent Health in an Urban Environment

  NIH · $310k · 2017–2022

Frequent coauthors

• Christopher R. Browning

  The Ohio State University

  29 shared

• Mark D. Risser

  24 shared

• Bethany Boettner

  The Ohio State University

  17 shared

• Candace Berrett

  17 shared

• Anna L. Smith

  University of Kentucky

  15 shared

• Noel Cressie

  University of Wollongong

  12 shared

• David C. Wheeler

  University College London

  12 shared

• Dena Marie Asta

  11 shared

Education

• PhD, Statistics

  Duke University

  2003

• MS, Statistics

  Duke University

  2001

• BA, Mathematics

  Northwestern University

  1999

Awards & honors

• Fellow, American Statistical Association
• Fellow, American Association for the Advancement of Science
• Fellow, Institute of Mathematical Statistics
• Founders Award, American Statistical Association