Audrey Leroux

Associate Professor of Psychology

Education

Ph.D. in Quantitative Methods, The University of Texas at Austin, 2014

B.S. in Mathematics, The University of Texas at Austin, 2008

Research Interests

Multilevel modeling, cross-classified and multiple membership data structures, longitudinal modeling, computerized adaptive testing, item response theory, simulations

About

My methodological research program focuses in two areas: multilevel modeling and computerized adaptive testing, both of which aim to solve methodological dilemmas encountered by applied researchers in the field of psychology as well as more generally in the social and behavioral sciences. Specifically, my multilevel modeling research examines the impacts of different models and approaches for handling complex non-hierarchical data structures where individuals do not remain in their sampled clustering unit over time or when there are two (or more) higher-level units of individuals but the clustering units themselves are not nested within one another. These types of complex multilevel data structures, commonly referred to as cross-classified or multiple membership, arise often in social and behavioral sciences research. My computerized adaptive testing research compares novel and commonly used stopping rules and item exposure control procedures through simulation studies. The use of computerized adaptive testing can generally lessen test length while increasing measurement precision depending on choices made in the different components of the test, such as the item pool, item selection procedure, trait estimation method, and stopping criteria.

Along with my own independent research program, I also engage in research collaborations with colleagues interested in applying quantitative methods. Many of these collaborations have been federally funded research grants from the National Science Foundation, Institute of Education Sciences, and Department of Health and Human Services.

For more details on my research, visit my Google Scholar page.

Selected Publications

Cappelli, C. J., Leroux, A. J., & Masyn, K. E. (2024). Individual mobility across clusters: The impact of ignoring cross-classified data structures in discrete-time survival analysis. Multivariate Behavioral Research, 59(1), 171-186. https://doi.org/10.1080/00273171.2023.2230481

Lee, Y., Leite, W. L., & Leroux, A. J. (2024). Propensity score matching with cross-classified data structures: A comparison of methods. Journal of Experimental Education, 92(2), 359-376. https://doi.org/10.1080/00220973.2023.2164843

Leroux, A. J., Cappelli, C. J., & Fikis, D. R. J. (2021). The impacts of ignoring individual mobility across clusters in estimating a piecewise growth model. British Journal of Mathematical and Statistical Psychology, 74(3), 404-426. https://doi.org/10.1111/bmsp.12229

Leroux, A. J., Waid-Ebbs, J. K., Wen, P., Helmer, D. A., Graham, D. P., O’Connor, M. K., & Ray, K. (2019). An investigation of exposure control methods with variable-length CAT using the partial credit model. Applied Psychological Measurement, 43(8), 624-638. https://doi.org/10.1177/0146621618824856

Leroux, A. J. (2019). Student mobility in multilevel growth modeling: A multiple membership piecewise growth model. Journal of Experimental Education, 87(3), 430-448. https://doi.org/10.1080/00220973.2018.1465384

Chen, J., & Leroux, A. J. (2018). Residual normality assumption and the estimation of multiple membership random effects models. Multivariate Behavioral Research, 53(6), 898-913. https://doi.org/10.1080/00273171.2018.1533445

Leroux, A. J., & Beretvas, S. N. (2018). Estimation of a latent variable regression growth curve model for individuals cross-classified by clusters. Multivariate Behavioral Research, 53(2), 231-246. https://doi.org/10.1080/00273171.2017.1418654

Leroux, A. J., Lopez, M., Hembry, I., & Dodd, B. G. (2013). A comparison of exposure control procedures in CATs using the 3PL model. Educational and Psychological Measurement, 73(5), 857-874. https://doi.org/10.1177/0013164413486802