University of Wisconsin–Madison

Quantitative Methods Students

Daniel Adams

Daniel Adams is a graduate student in the Quantitative Methods area. Prior to attending the University of Wisconsin-Madison, he earned an MS degree in statistics at Ball State University in Indiana in 2014.   He also earned BS degrees in accountancy and business administration from San Diego State University in 2010.

Daniel currently studying educational measurement and his advisor is Daniel Bolt.   He is interested in latent variable models, including models of response style and their application to self-report rating scale measures.


  • BA., Statistics, Ball State University, Muncie, Indiana, 2014
  • BS., Accountancy, San Diego State University, San Diego, California, 2010
  • BS., Business Administration, San Diego State University, San Diego, California, 2010

Nana Kim

Nana Kim is a graduate student in the Quantitative Methods area.  She completed her B.A. in Education and M.A. in Educational Evaluation at Yonsei University.   She is continuing her study in educational measurement working with Daniel Bolt.

Nana’s primary research interests are item response theory (IRT) and its applications to areas such as the measurement of student growth. She is currently working as a graduate assistant at the National Conference of Bar Examiners (NCBE) in Madison, WI.  Other interests include hierarchical linear modeling, longitudinal data analysis, and test equating.

Yongnam Kim

Yongnam Kim is a graduate student in the Quantitative Methods area. He received his BA. in Education & Psychology and MA in Educational Psychology from Seoul National University in South Korea. He is currently working on a grant funded by IES under Dr. Peter Steiner that seeks to identify suitable matching strategies in multilevel contexts.

Yongnam’s research interest is in causal inference using randomized and non-randomized studies. Currently, he is working on how to obtain evidence of a possible hidden bias in observational studies, which has been considered almost impossible. By clarifying some old ideas on this, he hopes to suggest useful new strategies to detect hidden bias. Also, he is interested in graphical models like directed acyclic graphs (DAGs) and applications of the models to his studies.


  • Steiner, P. M., & Kim, Y. (2014, August). Implications of measurement error on covariate selection for causal inference. Paper presented at the 2014 Joint Statistical Meetings, Boston, Massachusetts
  • Kim, Y., & Steiner, P.. M. (2014, July). Bias-amplification in observational studies. Paper presented at the 2014 International Meeting of the Psychometric Society, Madison, Wisconsin.

Sora Lee

Sora Lee is a graduate student in the Quantitative Methods area. Her interests are in educational measurement and her advisor is Daniel Bolt. She had worked until 2015 as a project assistant in the UW-Madison Office of Testing and Evaluation Services (T&E) in conducting nationally recognized research on measurement-related topics. Following this, she worked in NCBE (National Conference of Bar Examiners) as a research assistant in conducting data manipulation and test-related research on issues related to automated scoring, standard-setting, and test security.   She earned a BA in Education from Seoul National University of Education and an MA in Educational Measurement and Evaluation from Seoul National University prior to coming to Wisconsin.

Sora’s primary research interests are in the theory and application of psychometric methods in education and psychology. Her research focuses in particular on (1) IRT models that yield asymmetric item characteristic curves, and their potential to provide insight into response process, and (2) multidimensional IRT models that accommodate item sensitivity to student’s ability growth/change. These interests stem from her goal of using latent trait models to improve measurement.


  • Lee, S., & Bolt, D. M. (2017). Asymmetric item characteristic curves and item complexity: Insights from simulation and real data analyses, Psychometrika, DOI 10.1007/s11336-017-9586-5.
  • Lee, S., & Bolt, D. M. (2016). Using the asymmetry of item characteristic curves (ICCs) to learn about underlying item response processes. In van der Ark, L.A., Bolt, D.M., Wang, W.-C., Douglas, J.A., & Wiberg, M. (Eds.). Quantitative Psychology Research. The 80th Annual Meeting of the Psychometric Society, Beijing, 2015. New York: Springer.
  • Bolt, D., Deng, S. and Lee, S. (2014). IRT model misspecification and measurement of growth in vertical scaling. Journal of Educational Measurement, 51(2), 141-162.

Yen Lee

Yen Lee is a PhD student in the Quantitative Methods program and works with David Kaplan. Yen majored in psychometrics in the psychology department of the National Chengchi University in Taiwan, where she received a masters degree. Before joining the Wisconsin program, Yen worked as a research assistant at the National Cheng Kung University.

Yen’s current research interests concern applications of the maximum entropy principle to data simulations and Bayesian analysis. She is currently working on two primary topics: (1) non-normal multivariate data generation and (2) the impact of different objective prior distributions on posterior distributions in Bayesian analysis. These research topics are part of her overall goal of applying known information obtained from previous empirical research to future studies for more precise estimation and evaluation.

Stan Lubanski

Stan Lubanski is a graduate student in the Quantitative Methods area. In 2013, he earned a BS in Psychology at Arizona State University.  Now an advisee of Dr. Peter Steiner, he is funded through an IES project on evaluating the effectiveness of Iowa’s “Authentic Intellectual Work” (AIW) high school program.

Stan is interested in identifying the conditions where causal claims can be made (i.e., causal inference), especially randomized experiments and quasi-experimental designs. This has led to his current research, which concerns the impact of adding propensity score estimation to a quasi-experimental design, specifically, the effect it might have on the sampling variance of the treatment effect. His other interests include graphical models (directed acyclic graphs or “DAGs”) and philosophy of science.

Yiqin Pan

Yiqin Pan is a Ph.D. student in Quantitative Methods Program. She earned a BS in Psychology and an MS in Psychometrics at Beijing Normal University. She is currently studying educational measurement and her advisor is James Wollack.

Yiqin’s primary research interests are item response theory (IRT) and its applications to test security, specifically, detection of cheating on tests. Other interests include latent variable models and hierarchical linear modeling.

Dan Su

Dan Su is a PhD student in Quantitative Methods Program. She received her Master’s degree in Quantitative Methods and a Minor in Statistics in 2014. She worked on the design and analysis of vignette experiments (factorial surveys) in her Master’s Thesis. For her dissertation, she received an AERA dissertation grant award for work on planned missing data designs for causal inference in large-scale assessments (e.g., PISA). Dan’s general research interests include planned missing data designs in large surveys, imputation methods for missing data, experimental designs for factorial surveys, and causal inference with the Rubin causal model and graphic models. Besides studying quantitative methodology, Dan is also a contemporary dancer and choreographer.


  • Su, D. & Steiner, P. M. (forthcoming). An evaluation of experimental designs for constructing vignette sets in factorial surveys. Sociological Methods & Research.
  • Kaplan, D. & Su, D. (2016). On matrix sampling and imputation of context questionnaires with implications for the generation of plausible values in large-scale assessments. Journal of Educational and Behavioral Statistics. doi: 10.3102/1076998615622221
  • Steiner, P. M., Atzmüller, C., & Su, D. (2016). Designing valid and reliable vignette experiments for survey research: A case study on the fair gender income gap. Journal of Methods and Measurement in the Social Sciences, 7(2), 52-94.
  • Steiner, P. M., Kim, Y., Hall, C. E. & Su, D. (2015). Graphical models for quasi-experimental designs. Sociological Methods & Research, doi: 0049124115582272

Youmi Suk

Youmi Suk is a graduate student in Quantitative Methods in Educational Psychology. She is studying educational statistics under Dr. Jee-Seon Kim. Youmi currently works at the UW-Madison Office of Testing and Evaluation Services (T&E) as a project assistant, and as a statistical consultant in the Laboratory for Experimental Design (LED).  Previously, she received a BS in Earth Science Education in 2014 and an MA in Educational Measurement and Evaluation in 2016, both from Seoul National University.

Youmi’s research interests center on multilevel modeling and causal inference. She is particularly interested in measuring treatment effects with multilevel observational data and finding and understanding the heterogeneity of treatment effects. More generally, she enjoys applying and connecting quantitative models to real-life situations to address practical and important problems in the educational, social and behavioral sciences.



Sinan Yavuz is currently a PhD student in the Quantitative Methods program. He is a project assistant in the Interactive Learning and Design Lab. He received his combined bachelors and masters degrees in the Biology Teaching Department from Gazi University in Turkey. He has worked for seven years as a research assistant in three different universities and meanwhile, he taught Educational Measurement and Evaluation and Learning and Developmental Psychology courses to university graduates for three years in private institutions in Turkey. He was a PhD candidate in the Educational Measurement and Evaluation Department at Hacettepe University. During his previous PhD studies in Turkey, he participated in national projects supported by UNICEF and the Turkish National Ministry of Education.

His primary interests are in Bayesian approaches for educational research designs, Bayesian growth models in education, Bayesian approaches to missing data analyses, and extensions of Bayesian model averaging to multilevel models.