Ordinal Measures in Multiple Indicator Models: A Simulation Study of Categorization Error

Q: What does this research mean for the field?

Categorization error introduced by using ordinal measures systematically affects the estimation and accuracy of multiple indicator models. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

David R. Johnson; James C. Creech

doi:10.2307/2095231

June 1, 1983American Sociological Review

Ordinal Measures in Multiple Indicator Models: A Simulation Study of Categorization Error

View at Publisher

Ask AI

Bookmark

Share

Key Points

Assess how transforming continuous concepts into discrete ordinal categories influences parameter estimates and model fit in multiple-indicator structural equation models.
Simulated multiple-indicator models across varying conditions of categorization error, varying the number of response categories and degrees of distributional skewness.
Evaluated structural coefficients, factor loadings, and goodness-of-fit measures by comparing categorized ordinal data against underlying continuous variables.
Categorization error caused minimal attenuation of structural coefficients when indicators contained four or more categories with moderate skewness.
Severe skewness combined with few categories (such as dichotomous indicators) produced the greatest parameter bias and model distortion.

Ask about this paper

Answers are grounded in this paper and cite their sources.

Implication

Simulation study reveals the effects of categorization error in multiple indicator models, indicating that ordinal variables with four or more categories maintain parameter accuracy.

Abstract

David Richard Johnson, James C. Creech, Ordinal Measures in Multiple Indicator Models: A Simulation Study of Categorization Error, American Sociological Review, Vol. 48, No. 3 (Jun., 1983), pp. 398-407

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Johnson et al. (1983) studied this question.

synapsesocial.com/papers/6a0a8714c0fb98abde37993d https://doi.org/10.2307/2095231

View at Publisher

Ask AI

Bookmark

Share

Ordinal Measures in Multiple Indicator Models: A Simulation Study of Categorization Error

Key Points

Ask about this paper

Implication

Abstract

Discussion

Cite This Study

Ordinal Measures in Multiple Indicator Models: A Simulation Study of Categorization Error

Key Points

Ask about this paper

Implication

Abstract

Discussion

Cite This Study

Keep reading

Keep reading