- Multilevel Modeling
- Growth Modeling / Longitudinal Data Analysis
- Reliability and Meta-analysis Model for Reliability Coefficients
- Multilevel Measurement Model
- Multilevel Factor Model
- Editorial Board: Behaviormetrika
- Ad-hoc Reviewer for Refereed Journals: Psychological Methods, Applied Psychological Measurement, Journal of Applied Measurement, Behaviometrika, Journal of Educational and Behavioral Statistics
- Session discussant, chair, and moderator in professional organization conferences: the Annual Meeting of for the National Council on Measurement in Education, 2010; the Annual Meeting of the American Educational Research Association, 2006; the International Meeting of the Psychometric Society, 2009, 2011.
- Reviewer of Annual Meeting paper proposals: American Educational Research Association, National Council on Measurement in Education.
- Ph.D. Michigan State University
Awards and Honors
2014 Excellence in Access and Inclusion Award, Virginia Tech Students with Disabilities Office and the Office for Diversity and Inclusion, April 22, 2014
Virginia Tech School of Education Teaching Excellence Award, April 2012
Miyazaki, Y., & Sugisawa, T. (in press). A simulation study on appropriate transformations of reliability coefficient in mixed-effects meta-analysis models. Japanese Journal for Research on Testing.
Miyazaki, Y., and Stack, M. (2015). Examining individual and school characteristics associated with child obesity using a multilevel growth model (in press). Social Science & Medicine, 128, 57-66.
Agnich, L. E., & Miyazaki, Y. (2013). A multilevel cross-national analysis of direct and indirect forms of school violence. Journal of School Violence, 12(4), 319-339.
Kamata, A., Bauer, D. J., & Miyazaki, Y. (2008). Multilevel measurement modeling. In A. A. O’Connell, & D. B. McCoach (Eds.). Multilevel Modeling of Educational Data (pp. 345-388). Charlotte, NC: Information Age Publishing.
Lawson, Gerard (Principal Investigator), Patrizio, Kami M (Co-Principal Investigator), Miyazaki, Yasuo(Co-Investigator), Skaggs, Gary (Co-Investigator), "A Multiple Perspectives Analysis of the Influences on the School to Prison Pipeline in Virginia," Sponsored by National Institute of Justice, $914,241.00. (January 1, 2016 – December 31, 2017).
White, S. (PI), Asselin, S. (Co-PI), Miyazaki, Y. (Co-PI), "STEPS: Stepped Transition in Education Program for Students with ASD," Sponsored by NIMH, $666,498. (September 1, 2014 – August 31, 2017).
Miyazaki, Y. (PI) & Creamer, E. (Co-PI). Investigating the accuracy of standard error of ability and necessary number of items using HLM to study K-12 and health and education practices (2015). 2015 - 2016 Niles Research Grant Award from the College of Liberal Arts and Human Sciences at Virginia Tech, $3,796.
I’m interested in developing a new statistical and measurement modeling methodology and/or improving current methodology for the most pressing issues that plague behavioral and social scientists to answer the important questions. I consider one of the most pressing problems in behavioral and social sciences lies in the measurement of key variables in research since many of them are phrased in terms of abstract construct such as ability, anxiety, social skills, and school climate, because they cannot be directly observed.
Since human development through formal education occurs in school contexts, it is also important to consider the validity and reliability of measurements at teacher and school level as well as the individual level constructs. Because of the hierarchical nature of educational settings, I have been focusing on the methodology and applications surrounding Hierarchical Linear/Nonlinear Modeling (also known as Multilevel Modeling), which is a useful methodology for studying the phenomena that occurs in multilevel contexts. Hierarchical Linerar/Nonlinear Modeling shares a common theme with other major methodology used in behavioral and social sciences such as structural equation modeling and Item Response Theory. The common theme I found is the use of latent variables. Thus I’m making efforts to combine and integrate Hierarchical Linear/Nonlinear Modeling with the structural equation modeling and Item Response Theory to develop new and improved methodology and applications.
Specific topics that I’m currently tackling are: A longitudinal growth model from item level data; measurement model to capture the organizational climate; contextual/compositional effects in organizational research, and impacts of nested data structure on IRT model parameters.