Effects of sample size and distributional assumptions on competing models of the factor structure of the PANSS and BPRS

Stephen J. Tueller, Kiersten L. Johnson, Kevin Grimm, Sarah L. Desmarais, Brian G. Sellers, Richard A. Van Dorn

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Factor analytic work on the Positive and Negative Syndrome Scale (PANSS) and Brief Psychiatric Rating Scale (BPRS) has yielded varied and conflicting results. The current study explored potential causes of these discrepancies. Prior research has been limited by small sample sizes and an incorrect assumption that the items are normally distributed when in practice responses are highly skewed ordinal variables. Using simulation methodology, we examined the effects of sample size, (in)correctly specifying item distributions, collapsing rarely endorsed response categories, and four factor analytic models. The first is the model of Van Dorn et al., developed using a large integrated data set, specified the item distributions as multinomial, and used cross-validation. The remaining models were developed specifying item distributions as normal: the commonly used pentagonal model of White et al.; the model of Van der Gaag et al. developed using extensive cross-validation methods; and the model of Shafer developed through meta-analysis. Our simulation results indicated that incorrectly assuming normality led to biases in model fit and factor structure, especially for small sample size. Collapsing rarely used response options had negligible effects.

Original languageEnglish (US)
JournalInternational Journal of Methods in Psychiatric Research
DOIs
StateAccepted/In press - 2016

Fingerprint

Brief Psychiatric Rating Scale
Sample Size
Normal Distribution
Meta-Analysis
Research

Keywords

  • BPRS
  • Data integration
  • Factor analysis
  • PANSS
  • Psychopathology

ASJC Scopus subject areas

  • Psychiatry and Mental health

Cite this

Effects of sample size and distributional assumptions on competing models of the factor structure of the PANSS and BPRS. / Tueller, Stephen J.; Johnson, Kiersten L.; Grimm, Kevin; Desmarais, Sarah L.; Sellers, Brian G.; Van Dorn, Richard A.

In: International Journal of Methods in Psychiatric Research, 2016.

Research output: Contribution to journalArticle

@article{f4a305f3106a4b43a12e1c8f9b2ed351,
title = "Effects of sample size and distributional assumptions on competing models of the factor structure of the PANSS and BPRS",
abstract = "Factor analytic work on the Positive and Negative Syndrome Scale (PANSS) and Brief Psychiatric Rating Scale (BPRS) has yielded varied and conflicting results. The current study explored potential causes of these discrepancies. Prior research has been limited by small sample sizes and an incorrect assumption that the items are normally distributed when in practice responses are highly skewed ordinal variables. Using simulation methodology, we examined the effects of sample size, (in)correctly specifying item distributions, collapsing rarely endorsed response categories, and four factor analytic models. The first is the model of Van Dorn et al., developed using a large integrated data set, specified the item distributions as multinomial, and used cross-validation. The remaining models were developed specifying item distributions as normal: the commonly used pentagonal model of White et al.; the model of Van der Gaag et al. developed using extensive cross-validation methods; and the model of Shafer developed through meta-analysis. Our simulation results indicated that incorrectly assuming normality led to biases in model fit and factor structure, especially for small sample size. Collapsing rarely used response options had negligible effects.",
keywords = "BPRS, Data integration, Factor analysis, PANSS, Psychopathology",
author = "Tueller, {Stephen J.} and Johnson, {Kiersten L.} and Kevin Grimm and Desmarais, {Sarah L.} and Sellers, {Brian G.} and {Van Dorn}, {Richard A.}",
year = "2016",
doi = "10.1002/mpr.1549",
language = "English (US)",
journal = "International Journal of Methods in Psychiatric Research",
issn = "1049-8931",
publisher = "John Wiley and Sons Ltd",

}

TY - JOUR

T1 - Effects of sample size and distributional assumptions on competing models of the factor structure of the PANSS and BPRS

AU - Tueller, Stephen J.

AU - Johnson, Kiersten L.

AU - Grimm, Kevin

AU - Desmarais, Sarah L.

AU - Sellers, Brian G.

AU - Van Dorn, Richard A.

PY - 2016

Y1 - 2016

N2 - Factor analytic work on the Positive and Negative Syndrome Scale (PANSS) and Brief Psychiatric Rating Scale (BPRS) has yielded varied and conflicting results. The current study explored potential causes of these discrepancies. Prior research has been limited by small sample sizes and an incorrect assumption that the items are normally distributed when in practice responses are highly skewed ordinal variables. Using simulation methodology, we examined the effects of sample size, (in)correctly specifying item distributions, collapsing rarely endorsed response categories, and four factor analytic models. The first is the model of Van Dorn et al., developed using a large integrated data set, specified the item distributions as multinomial, and used cross-validation. The remaining models were developed specifying item distributions as normal: the commonly used pentagonal model of White et al.; the model of Van der Gaag et al. developed using extensive cross-validation methods; and the model of Shafer developed through meta-analysis. Our simulation results indicated that incorrectly assuming normality led to biases in model fit and factor structure, especially for small sample size. Collapsing rarely used response options had negligible effects.

AB - Factor analytic work on the Positive and Negative Syndrome Scale (PANSS) and Brief Psychiatric Rating Scale (BPRS) has yielded varied and conflicting results. The current study explored potential causes of these discrepancies. Prior research has been limited by small sample sizes and an incorrect assumption that the items are normally distributed when in practice responses are highly skewed ordinal variables. Using simulation methodology, we examined the effects of sample size, (in)correctly specifying item distributions, collapsing rarely endorsed response categories, and four factor analytic models. The first is the model of Van Dorn et al., developed using a large integrated data set, specified the item distributions as multinomial, and used cross-validation. The remaining models were developed specifying item distributions as normal: the commonly used pentagonal model of White et al.; the model of Van der Gaag et al. developed using extensive cross-validation methods; and the model of Shafer developed through meta-analysis. Our simulation results indicated that incorrectly assuming normality led to biases in model fit and factor structure, especially for small sample size. Collapsing rarely used response options had negligible effects.

KW - BPRS

KW - Data integration

KW - Factor analysis

KW - PANSS

KW - Psychopathology

UR - http://www.scopus.com/inward/record.url?scp=85006314748&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85006314748&partnerID=8YFLogxK

U2 - 10.1002/mpr.1549

DO - 10.1002/mpr.1549

M3 - Article

C2 - 27910162

AN - SCOPUS:85006314748

JO - International Journal of Methods in Psychiatric Research

JF - International Journal of Methods in Psychiatric Research

SN - 1049-8931

ER -