Assessment and Test Construction 

This course addresses the major issues of educational assessment, with emphasis on the characteristics, administration, scoring, and interpretation of both formal and informal assessments, including but not limited to tests of achievement. All forms of assessment are examined including observation, portfolios, performance tasks, and paper-and-pencil tests, including standardized tests. Basic techniques of test construction, item writing, and analysis are included. Standardized norm-referenced tests and statewide testing programs are also examined.

Introduction to Mathematical Statistics

Quantitative methods in educational and psychological research have become increasingly complex over time, employing more sophisticated models and estimation strategies. This course helps students to develop a deeper understanding of the strengths and limitations of different approaches to inference and to appreciate some of the ongoing arguments among the adherents of the different philosophies regarding statistical inference.

General Linear Models 

Addresses the construction, interpretation, and application of linear statistical models. Specifically, lectures and computer exercises cover ordinary least squares regression models; matrix algebra operations; parameter estimation techniques; missing data options; power transformations; exploratory versus confirmatory model building; linear-model diagnostics, sources of multicollinearity; diagnostic residual analysis techniques; variance partitioning procedures; dummy, effect, and orthogonal coding procedures; and an introduction to structural equation modeling.

Intermediate Statistics 

Topics and computer exercises address tests of means, partial and part correlations, multiple regression, analysis of variance with planned and post hoc comparisons, analysis of covariance, repeated measures analysis, elements of experimental design, and power analysis.

Multivariate Statistical Analysis

This course provides lectures, examples, and lab analyses that address multinomial and ordinal logistic regression models, multiple group discriminant analysis, cluster analysis, multivariate analysis of variance, principal component analysis, factor analysis, and structural equation modeling. We cover various issues related to research design, model building, and the interpretation of the output from SPSS, R, Lisrel, and SAS software programs.

Psychometric Theory: Classical Test Theory and Rasch Models

Presents a study of theoretical concepts, statistical models, and practical applications in educational and psychological measurement. General topics include the history of measurement, Thurstone and Guttman scales, classical true-score theory, and item response theory. Specific topics include principles of Rasch measurement, parameter estimation procedures, fit statistics, item banking, and computer adaptive testing.

Psychometric Theory II: Item Response Theory

This course presents an advanced study of theoretical concepts, statistical models, and practical applications in educational and psychological measurement. Topics include item response theory, two-parameter model, three-parameter model, methods for estimating latent trait and item parameters, models for polytomously scored items, differential Item Functioning(DIF), test equating , vertical scaling, computerized adaptive testing, standard setting, and multidimensional item response theory models. The IRT software or tools used in this course include R, IRTPRO, BILOG-MG, PARSCALE, NOHARM, DIF related software, STUIRT, ST., EQUATE, etc.

Multilevel Regression Modeling

This course introduces students to multilevel regression modeling (aka hierarchical models or mixed effects models) for analyzing data with a nesting or hierarchical structure. We discuss the appropriate uses of multilevel regression modeling, the statistical models that underpin the approach, and how to construct models to address substantive issues. We consider a variety of types of models, including random intercept, and random slope and intercept models; models for longitudinal data; and models for discrete outcomes. We cover various issues related to the design of multilevel studies, model building and the interpretation of the output from HLM and SPSS software programs.

Master's Comprehensive Exam

In order to ensure that all students graduating from the master's program have a fundamental understanding of the field which they are about to enter, they are required to complete a written comprehensive examination covering the broad areas of the core courses.