James Uanhoro

Bio

I am an assistant professor in the Research Methodology, Measurement, and Statistics program within the Department of Educational Psychology at the University of North Texas. I have a background in Information Systems, and in my downtime, I enjoy using my computing skills to develop statistical tools for applied researchers.

Research interests:

Applied Bayesian data analysis
Multilevel, psychometric, and covariance structure models and their interconnections
Communicating statistical findings & effect sizes
Statistical software development

Education:

PhD Educational Studies, Ohio State University (2021)
MS Information Systems, National Tsing Hua University (2016)
BS Information Systems, American University of Nigeria (2012)

I am also a member of the Texas State Employees Union (Communication Workers of America, Local 6186). TSEU is a union for everyone who works for the state of Texas.

Email: james.uanhoro@unt.edu

Teaching

I usually teach regression (fall), structural equation modeling (spring), and simulation (some Falls). And I should teach an applied Bayesian course in the near future.

Additional links

Blog

2024/04/14 Hierarchical ordinal regression for analysis of single subject data OR Bayesian estimation of overlap and other effect sizes
2023/05/04 AERA 2023 Presentation: Modeling misspecification as a parameter in Bayesian SEMs
2023/04/02 The minorbsem package in R
2022/12/14 Hierarchical Covariance Estimation Approach to Meta-Analytic Structural Equation Modeling
2021/01/18 Traditional SEM overly focuses on LISREL equations to the detriment of dealing with measurement error
2019/11/25 Problems with using odds ratios as effect sizes in binary logistic regression and alternative approaches
2019/09/26 How do you communicate the evidence for the practical significance of an intervention using Bayesian methods?
2019/08/02 When data are not so informative, it pays to choose the sum score over the factor score
2018/11/28 Multidimensional CFA with RStan
2018/10/20 Possibility of heteroskedasticity is a good reason not to dichotomize a continuous variable for use as outcome in logistic regression.
2018/10/07 Should you perform logistic regression on a dichotomized continuous variable when you have access to the continuous variable? I’m not sure.
2018/10/06 Two group mean and variance comparisons
2018/05/07 Modeling the error variance to account for heteroskedasticity
2018/05/07 Simulating data from regression models
2018/02/12 Using binary regression software to model ordinal data as a multivariate GLM
2018/01/02 Using glmer() to perform Rasch analysis
2017/11/16 A Chi-Square test of close fit in covariance-based SEM
2017/10/28 Misspecification and fit indices in covariance-based SEM
2017/09/21 Little’s MCAR test at different sample sizes
2017/09/21 Theil-Sen regression in R
2017/09/19 Linear regression with violation of heteroskedasticity with small samples
2017/08/11 On the interpretation of regression coefficients