Seminar in Tests & Measurements

Catalog description

Educational and psychological tests are commonly used to measure latent traits (e.g., knowledge, aptitude, achievement, attitude, and personality), often with the goal of making inferences on individuals or groups of individuals. Item response theory (IRT) measurement models are useful tools in examining the quality of the data that arises from the administration of such tests, which in turn allows us to make judgements about the validity of the test scores (or latent trait estimates) and the quality of the tests (are the items confusing and/or are they biased). This course will introduce students to commonly used IRT models for analyzing dichotomous (e.g., 0=incorrect and 1=correct) and polytomous (0=never, 1=sometimes, 2=always) test data. While much of the course will be application oriented, some technical details that facilitate the understanding of the IRT and highlight the advantages of it over classical test theory (CTT) will be provided. Knowledge of basic algebra, linear modeling, and CTT is required.