Paper Summary
Share...

Direct link:

A General Cognitive Diagnosis Model for Partial-Credit Responses

Sun, April 19, 4:05 to 5:35pm, Marriott, Floor: Sixth Level, Northwestern/Ohio State

Abstract

Cognitive diagnosis models (CDMs) have been developed recently for identifying fine-grained skills of interest. Although partial credit items have been widely used in practice, to analyze the polytomous data, most existing CDMs have to dichotomize the responses, ignoring some potential diagnostic information. This study introduces a general CDM for partial credit items based on the generalized deterministic inputs, noisy, "and" gate model (de la Torre, 2011), allowing to specify different attributes and cognitive processes for each category of each item. The article discusses the specification of the proposed model and estimation methods of its parameters. Additionally, a simulation study is conducted to examine its viability and advantages compared with other ways of analyzing the graded data.

Authors