Measuring euthymia within the Neuroticism Scale from the NEO Personality Inventory: A Mokken analysis of the Norwegian general population study for scalability

P Bech; D Carrozzino; S F Austin; S B Møller; O Vassend

doi:10.1016/j.jad.2015.12.039

Measuring euthymia within the Neuroticism Scale from the NEO Personality Inventory: A Mokken analysis of the Norwegian general population study for scalability

J Affect Disord. 2016 Mar 15:193:99-102. doi: 10.1016/j.jad.2015.12.039. Epub 2015 Dec 29.

Authors

P Bech¹, D Carrozzino², S F Austin², S B Møller², O Vassend³

Affiliations

¹ Psychiatric Research Unit, Psychiatric Centre North Zealand, University of Copenhagen, Hillerød, Denmark. Electronic address: per.bech@regionh.dk.
² Psychiatric Research Unit, Psychiatric Centre North Zealand, University of Copenhagen, Hillerød, Denmark.
³ Department of Psychology, Psychological Institute, University of Oslo, Oslo, Norway.

PMID: 26773899
DOI: 10.1016/j.jad.2015.12.039

Abstract

Background: Whereas the Eysenck Neuroticism Scale only contains items covering negative mental health to measure dysthymia, the NEO Personality Inventory (NEO-PI) contains neuroticism items covering both negative mental health and positive mental health (or euthymia). The consequence of wording items both positively and negatively within the NEO-PI has never been psychometrically investigated. The aim of this study was to perform a validation analysis of the NEO-PI neuroticism scale.

Methods: Using a Norwegian general population study we examined the structure of the negatively and positively formulated items by principal component analysis (PCA). The scalability of the identified two groups of euthymia versus dysthymia items was examined by Mokken analysis.

Results: With a response rate of 90%, 1082 individuals with a completed NEO-PI were available. The PCA identified the neuroticism scale as the most distinct where 14 items had acceptable loadings for the euthymia subscale, another 14 items for the dysthymia subscale. However, the Mokken analysis coefficient of homogeneity only found acceptable scalability for the euthymia subscale.

Limitations: A comparison with the Eysenck Neuroticism Scale was not performed.

Conclusion: The NEO-PI neuroticism scale contains two subscales consisting of items worded in an opposite direction where only the positive euthymia items have an acceptable scalability.

Keywords: Dysthymia; Euthymia; Mokken analysis; NEO Personality Inventory.

MeSH terms

Anxiety Disorders / diagnosis
Anxiety Disorders / psychology
Cyclothymic Disorder / diagnosis*
Cyclothymic Disorder / psychology
Dysthymic Disorder / diagnosis
Dysthymic Disorder / psychology
Humans
Neuroticism
Norway
Personality Inventory*
Principal Component Analysis
Psychiatric Status Rating Scales*
Psychometrics
Reproducibility of Results