Mixture general diagnostic model
AssigneeEducational Testing Service
InventorsMatthias Von Davier, Kentaro Yamamoto, Xueli Xu
Life of this patent
Estimating the term…
Abstract
Disclosed herein is a method of analyzing examinee item response data comprising constructing a diagnosis model for reporting skill profiles of examinees, wherein the diagnosis model comprises at least a variable representing unobserved subpopulations, creating an item design matrix, distributing examinees across the unobserved subpopulations, iteratively estimating values for a plurality of variables within the diagnosis model, and reporting the estimated values to a user.
Claims14
A method of analyzing examinee item response data comprising:
- constructing a diagnosis model for reporting skill profiles of examinees, wherein the diagnosis model comprises at least a variable representing unobserved subpopulations;
- creating an item design matrix;
- distributing examinees across the unobserved subpopulations;
- iteratively estimating values for a plurality of variables within the diagnosis model;
- and
- reporting the estimated values to a user.
A method of analyzing examinee item response data comprising:
- constructing a diagnosis model for reporting skill profiles of examinees, wherein the diagnosis model comprises at least a variable representing unobserved subpopulations and a cluster variable;
- creating an item design matrix;
- distributing examinees across the unobserved subpopulations;
- iteratively estimating values for a plurality of variables within the diagnosis model;
- and
- reporting the estimated values to a user.
Dependencies are inferred from the claim text.