Item Response Theory

Item response theory (IRT) is a measurement framework for the development of tests and the scoring of item responses on tests. Key aspects of the IRT framework include the focus on items as the units of observed measurement, the fitting of parametric statistical models to categorical item response data, the estimation of a latent trait variable, and the conditional nature of reliability and the standard error of measurement. IRT is relevant to the field of educational measurement in that it is widely used by measurement practitioners for developing and scoring standardized tests, such as the SAT, ACT, and statewide K–12 achievement tests in mathematics, reading, and other academic domains. IRT is also widely used for the scoring of scale data, such as attitude scales consisting ...