Dr. Shear's primary research interests address topics in psychometrics and applied statistics aimed at improving the fairness of test development and use in educational research. His areas of expertise include validity theory, differential item functioning (DIF), and categorical data analysis techniques. His work aims to inform our perspectives about test score meaning and use for research and accountability purposes, and to help answer questions such as, “what do test scores measure, and how do we know?” His work addresses these questions by developing and applying statistical models to generate and evaluate test scores and to make better use of existing large-scale assessment data. An example of the former is the use of DIF analyses to detect biased test items; an example of the latter is the novel application of ordered probit models to summarize large-scale state assessment data.
False Positives in Multiple Regression.
Educational and Psychological Measurement: devoted to the development and application of measures of individual differences.
733-756.
2013
Teaching
courses taught
EDUC 7396 - Latent Variable and Structural Equation Modeling
Primary Instructor
-
Spring 2019 / Spring 2020
Introduces contemporary advanced multivariate techniques and their application in social science research. Focus on factor analysis and variance, structural equation modeling. Prior experience with multiple regression is assumed.
EDUC 7456 - Multilevel Modeling
Primary Instructor
-
Fall 2022 / Fall 2024
Covers in depth two advanced multivariate models common to social science research: latent variable (structural equation) models and multi-level (hierarchical) models. Topics may be taught with a particular analytic context, such as measurement of change (longitudinal analysis) or experimental design. Recommended prerequisite of one year of graduate-level stats course.
EDUC 8230 - An Introduction to Quantitative Methods in Educational Research
Primary Instructor
-
Fall 2018 / Fall 2019
Explores the use of statistics to formalize study designs in educational research contexts. Introduces causal inference, experimental design, descriptive statistics, linear regression, probability, and the basics of statistical inference. Includes lab-based instruction in the use of statistical software (e.g., R, Excel) to conduct data analysis.
EDUC 8240 - Applied Regression Analysis
Primary Instructor
-
Spring 2021 / Spring 2022 / Spring 2023
Statistical analysis can be a powerful tool for understanding social, educational, psychological, and developmental processes. In this course, we will learn to answer such questions using multiple regression analysis, to develop an understanding of the strengths and limitations of this approach, and practice communicating results clearly and accurately. By the end of the semester, students in this course should be able to critically examine published research using regression and carefully perform their own regression analyses using empirical data. Recommended prerequisites of EDUC 8230 or another course in basic statistical methods.
EDUC 8710 - Measurement in Survey Research
Primary Instructor
-
Fall 2020 / Fall 2021 / Fall 2023
Introduces students to classical test theory and item response theory. Emphasizes the process of developing, analyzing and validating a survey instrument. Focuses on developing a survey instrument with items that derive from a clearly delineated theory for the construct to be measured. Analyzes item responses and put together a validity argument to support the proposed uses of the survey.
EDUC 8720 - Psychometric Modeling: Item Response Theory
Primary Instructor
-
Spring 2018
Focuses on psychometric models for measurement and their applications in educational and psychological research. Emphasizes understanding and evaluating the utility of models from item response theory (IRT). Applies and compares measurement models in the context of simulated or empirical data sets. Recommended prerequisite: EDUC 8710.
