Estimation of the Operating Characteristics of Item Response Categories I: Introduction to the Two-Parameter Beta Method.
Abstract
Following the Normal Approximation Method, in this study, the Two-Parameter Beta Method is developed and introduced, as a method of estimating the operating characteristics of a test without assuming any prior model. The maximum likelihood estimation, Theta-bar, and its asymptotic property such that its conditional distribution, given ability Theta, is N(Theta, 1/I(theta), where I(Theta) is the test information function, are fully utilized. An approximation is made for the conditional distribution of Theta, given Theta-bar, by a Beta distribution using the method of moments, with two a priori set parameters and the other two estimated parameters from the first two conditional moments of Theta, given Theta-bar. The least square principle is adopted in estimating two parameters in the normal ogive model for each item.
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 01, 1977
- Accession Number
- ADA049618
Entities
People
- Fumiko Samejima
Organizations
- University of Tennessee