Rule Space model

Research Method · All research methods
Evidence · 4 claims (3 for, 1 mixed) · 3 studies (2 causal, 1 associational), q2 · 1 of 3 report an effect size · 3 claims rest on one study

Description

The Rule Space model (Tatsuoka, 1983) is an IRT-based diagnostic classification framework in which "Items are decomposed into attributes, which represent the latent traits that the items assess." From an examinee's response vector the model "infers the most likely combination of attributes the examinee has mastered." Classification proceeds in three steps: determining classification groups via an incidence matrix Q, formulating a two-dimensional classification space of IRT proficiency and a pattern-unusualness index, and classifying responses using Bayes' decision rules over ideal knowledge states. This paper extends it to a semantically rich domain by defining attributes as item characteris

Accounts

• Rule Space Model of cognitive diagnosis: The Rule Space model (Tatsuoka, 1983) is an IRT-based diagnostic classification framework in which "Items are decomposed into attributes, which represent the latent traits that the items assess." From an examinee's response vector the model "infers the most likely combination of attributes the examinee has mastered." Classification proceeds in three steps: determining classification groups via an incidence matrix Q, formulating a two-dimensional classification space of IRT proficiency and a pattern-unusualness index, and classifying responses using Bayes' decision rules over ideal knowledge states. This paper extends it to a semantically rich domain by defining attributes as item characteris (Katz et al. (1993))

Claims

• Rule Space successfully classified most examinees in architecture, with solve attributes the most powerful and check attributes the least powerful [+M]
• Attribute mastery showed significant proficiency-group, attribute, and attribute-by-proficiency effects, but no significant status-group main effect [+M]
• Solve attributes differ in discrimination: the environment attribute strongly separated proficiency groups while the learned-procedure attribute did not [+W]
• The prerequisite-violating knowledge-state pattern is primarily attributable to Q-matrix misspecification, and the original hierarchy shows superior predictive efficiency [~W]

Related Research Methods

• Two-dimensional classification of diagnostic situations: information base and collaboration requirement

Key Sources

• Katz, I. R., Martinez, M. E., Sheehan, K. M., & Tatsuoka, K. K. (1993). Extending the Rule Space Model to a Semantically-Rich Domain: Diagnostic Assessment in Architecture. Educational Testing Service. https://eric.ed.gov/?id=ED366652. doi:10.21236/ada273789

Page metadata

Field	Value
type	research-method
id	rule-space-model
title	Rule Space model
description	The Rule Space model (Tatsuoka, 1983) is an IRT-based diagnostic classification framework in which "Items are decomposed into attributes, which represent the latent traits that the items assess." From an examinee's response vector the model "infers the most likely combination of attributes the exami
status	review
moved_from	theory/rule-space-model-cognitive-diagnosis
generated	by process:sweep-kinds on 2026-10-10
trust	unverified — no one has confirmed this page's content
sources	1 — katz-1993

Frontmatter is the page's machine-readable record — see the Schema & Guide.