Hi, I’m

Fatih.

I build measurement models that hold up.

Postdoctoral Research Fellow in Behavioral & Social Sciences at the University of Houston. PhD in Quantitative Methods, Baylor University. I work on psychometrics, item response theory, measurement invariance, and network psychometrics.

About

I’m Fatih Ozkan, a postdoctoral research fellow in Behavioral & Social Sciences at the University of Houston, with a PhD in Educational Psychology (quantitative methods) from Baylor University.

My interests include psychometric methods for fair testing (bias-aware network psychometrics, differential item functioning, measurement invariance), item response theory, Bayesian modeling, and longitudinal health studies.

I use this space to share code, course notes, and small research projects as I work on methods for large educational and behavioral datasets.

Tools I use a lot:
  • R
  • Stan/JAGS
  • Quarto
  • Git/GitHub
  • Mplus
  • SAS

Experience

Postdoctoral Research Fellow, Behavioral & Social Sciences - University of Houston
Aug 2026 - Present
  • Design and evaluate measurement models for behavioral and health science data, extending the bias-aware network psychometrics methods from my dissertation to applied research questions.
  • Provide psychometric support across projects, including scale development and validation, measurement invariance testing, and structural equation and multilevel models for nested designs.
  • Maintain reproducible R and Quarto pipelines so analyses, tables, and figures regenerate from source and hand off cleanly to collaborators.
Graduate Research Assistant - Baylor University
2021 - 2026
  • Exceptional Performance Lab (Dir. Todd Kettler), Center for Gifted Education and Talent Development, Department of Educational Psychology.
  • Built reproducible analysis workflows in R, including simulation studies, model checking, and reporting that non-technical collaborators can actually use.
  • Applied psychometrics and measurement work, including IRT, measurement invariance, and reliability and validity evidence.
  • Analyzed large-scale assessment data, including plausible values and replicate weights when the design demands it.
Graduate Student, Educational Psychology
2019 - 2021
  • M.A. in Educational Psychology (Quantitative Methods), with a thesis on the effects of STEM education on gifted students.
2016 - 2019
  • Oversaw STEM programs.
  • Conducted teacher training.
  • Led curriculum development.
  • Managed community partnerships.
Science and Math Teacher - Ministry of National Education
2009 - 2016
  • Classroom instruction.
  • Lesson and assessment design.
  • Lab supervision.
  • Extracurricular mentorship.

Education

August 2026
Ph.D., Educational Psychology (Quantitative Methods)
Baylor University
Dissertation: Detecting Local Item Dependencies in STEM-Related Assessments via Bias-Aware Network Psychometrics: Improving Measurement in Gifted Education.
2021
M.A., Educational Psychology (Quantitative Methods)
Baylor University
Thesis: Analyzing the Effect of STEM Education on the Academic Success and Social-Emotional Development of Gifted Students.
2017
M.A., Science and Mathematics Education
Erciyes University (Kayseri, Türkiye)
Thesis: Determination of Misconceptions by Developing the Two-Stage Multiple-Choice Test for 7th Grade Digestive System.
2009
B.Ed., Math and Science Education
Erciyes University (Kayseri, Türkiye)
Concentration: Math and Science Education.

Projects

biasawareLID
R package Network psychometrics
biasawareLID
R package for bias-aware detection of group-specific local item dependence: residual networks, permutation inference with FDR control, and a group-disparity penalty.
BLNM
Bayesian Latent networks
BLNM
Bayes factors for conditional independence in latent network modeling: a CFA measurement model with a Gaussian graphical model on the latent variables.
Bayesian-Statistics
Bayesian R
Bayesian-Statistics
Course notes and code related to Bayesian statistics.
PIRLS 2021 Türkiye
Psychometrics R
PIRLS 2021 Türkiye
Home and parental predictors of reading achievement in PIRLS 2021, including the Bayesian LASSO analysis behind the PARE paper.
Partial-Measurement-Invariance
Measurement invariance Quarto
Partial-Measurement-Invariance
Materials and assets for working with measurement invariance.

Publications

Get in Touch

My inbox is always open. Whether you have a question or just want to say hi, I’ll try my best to get back to you!