Skip to page content

NewGen Psychometrics and Data Science Analytics (NPDA) Lab

Introduction

NPDA Word Cloud image

Welcome to the NewGen Psychometrics and Data Science Analytics (NPDA) Lab at the University of Georgia. Our mission is to advance human development by reimagining how learning, performance, and growth are measured, modeled, and understood. By integrating psychometrics, statistics, data science toolkits, and educational measurement, we develop innovative methods and tools that lead to better assessments, deeper insights, and more informed decisions in education and beyond. Through this work, we aim to shape the future of measurement and analytics in ways that better support human potential.

Research Projects

Featured Projects

The Item Bank Calibration and Replenishment for Computerized Adaptive Testing in Small Scale Assessments

This project is funded by the National Science Foundation .

  • We will develop a series of statistical estimation methods, theories, algorithms, and software aimed at accelerating the development of CAT and promoting its applications in personalized educational assessment and learning.
  • The project will address a fundamental challenge in educational research: the lack of efficient statistical methods to cope with limited sample sizes and missing data, especially in low-stakes assessment settings.
    •  
Intelligent, Adaptive Program with Just-in-Time Feedback for Preservice Teachers

This project is funded by the National Science Foundation.

  • We propose to tailor the interactive, intelligent computer-based program with just-in-time feedback specifically for the needs of middle-grade preservice teachers.
  • We will test whether the program increase preservice teachers’ CK and PCK of ratios and proportional reasoning as well as their teaching of these concepts.
  • This project will provide a proof of concept for developing intelligent computerized programs that use AI technologies to interact with and provide just-in-time feedback for teachers that can be completed by anyone with access to the internet.
    •  
Bayesian Inference for Attribute Hierarchy in Cognitive Diagnosis Models

This project is funded by the National Science Foundation .

  • We have developed Bayesian inference on attribute hierarchy for both static and dynamic CDM models.
  • We aim to promote the use of CDMs in conjunction with attribute hierarchy to facilitate learning.
  • We are in the process of exploring attribute hierarchy and cognitive process learning with multimodal data, such as response accuracy, response times, and eye-tracking data.
    •  
Revision and Review Behavior in Large-Scale Computer-Based Assessments: An Analysis of NAEP Mathematics Process Data

This project is funded by NSF-AERA

  • The overall objective of this project is to analyze the recently released NCES restricted-used process, outcome, and survey data from the NAEP 2017 Grade 8 Mathematics Assessment. We plan to attain the overall objective by pursuing four research questions:
  1. How to analyze and interpret unstructured revision and review process data?
  2. Are there latent constructs (i.e. discrete classes or continuous traits) explaining examinees’ patterns of revision and review in large-scale, low-stakes computer-based math assessments?
  3. How do latent constructs for revision/review relate to other important cognitive and noncognitive traits, such as math proficiency and motivation, persistence, and pressure?
  4. How does revision/review behavior relate to demographic and instructional covariates (e.g., student race and ethnicity, disability status, and teacher and school characteristics)?
    5.  

Faculty Member

Shiyu Wang

Associate professor Shiyu Wang serves as the program chair for the Quantitative Methodology Program in the Department of Educational Psychology at the University of Georgia.

Her research is centered around methodological advancements and innovations across three key domains:

  1. The first area of focus involves the creation of cutting-edge adaptive testing designs. These designs are tailored to offer efficient, personalized assessments hin a testing environment that is user-friendly and accommodating to examinees.

  2. The second facet of her research entails establishing the statistical underpinnings for a family of restricted latent class models. These models serve as a foundation, providing essential guidelines for the estimation and selection of appropriate models.

  3. Wang also focuses on pioneering dynamic psychometric models that can effectively model test-taking and learning process and behaviors.

Her work uses complex multimodal data derived from virtual testing and learning environments. This data encompasses various elements such as response times, log data, textual responses, and eye-tracking data. Beyond her academic pursuits, Wang finds joy in engaging in sports and cherishing quality time with her family.

Ph.D. Students

Tamlyn Lahoud

Tamlyn Lahoud is a Ph.D. student at the University of Georgia in the Quantitative Methodology Program. She completed her master’s in statistics and bachelor’s in mathematical statistics and human kinetics and ergonomics at Rhodes University in South Africa.

Her research interests include item response theory, differential item functioning and topic modeling. Outside of academics, she enjoys rock climbing, traveling, and trying out new recipes.

Nohwon Park

Nohwon Park is a Ph.D. student in the Quantitative Methodology program. She holds a bachelor and master’s degree in Education, specializing in Sociology of Education from Ewha Womans University in South Korea. With her educational background and seven years of experience as a middle school teacher teaching moral philosophy, she is passionate about addressing educational inequity. Her aim is to focus on personalized learning to help students unlock their full potential regardless of their family background. Her research interests lie in diagnostic classification models, computerized adaptive testing and machine learning. In her free time, she enjoys listening to music and taking ballet classes.

Cheng Tang

Cheng Tang is a Ph.D. student in the Quantitative Methodology (QM) program at UGA. Originally from Chongqing, China, he holds both bachelor’s and master’s degrees in civil engineering from Chongqing University. Currently, he is also pursuing a master’s degree in computer science at UGA. Cheng has a deep passion for educational data mining and measurement. In his leisure time, he enjoys reading, listening to R&B music, watching Marvel movies, and hiking.

Yaxuan Yang

Yaxuan Yang is a Ph.D. student in the Quantitative Methodology program. She holds a bachelor’s degree from the University of North Carolina at Chapel Hill and a master’s degree from Columbia University. Her research interests include educational data mining, learning analytics, and AI in education. In her free time, she enjoys skiing and traveling.

Research Assistants

Cheng Cong

Cheng Cong, a dedicated Ph.D. student in the Department of Statistics who completed his undergraduate studies at the University of Science and Technology of China in 2020. He possesses a fervent enthusiasm for exploring the synergy between statistical methodologies and educational insights.

Aside from his academic life, he enjoys journeying through the fantasy world of Genshin Impact.

Collaborative Research Projects

A Multi-Phase Development of the Electric Circuit Concepts Diagnostic Tool

This project is funded by the National Science Foundation.

  • The Electric Circuit Concepts Diagnostic (ECCD) project team will create web-based electric circuit concept inventory that: (1.) provides an immediate and multi-purpose feedback system for reporting about students’ circuits and electricity prior knowledge; (2.) differentiates, with a high probability, between a lack of prior knowledge and misconceptions; and (3.) uses a scheme of multidimensional knowledge profiles to report on students’ prior knowledge and misconceptions.
  • The project team will integrate the affordances of cognitive diagnostic modeling, multi-tier testing frameworks, and computer-assisted testing to realize these project objectives.
    •  
The Rational Numbers Playground: Applying and Refining a Model for Dynamic, Discussion-Based Professional Development for Fractions, Ratios, and ProportionsThis project is funded by the National Science Foundation.

Resources

Lab Publications   Lab News

Software

Hmcdm R package

This R package can be used for fitting hidden Markov models of learning under the cognitive diagnosis framework. The estimation of the hidden Markov diagnostic classification model, the first order hidden Markov model, the reduced-reparametrized unified learning model, and the joint learning model for responses and response times

Contact Info

Shiyu Wang, associate professor, Department of Educational Psychology
© University of Georgia, Athens, GA 30602
706‑542‑3000