The use of questionnaires in child psychiatric research

Measuring their performance and choosing an optimal cut-off

Research output: Contribution to journalArticle

110 Citations (Scopus)

Abstract

This paper introduces the basic principles of Receiver Operating Characteristics (ROC) analysis, a technique to evaluate and compare the discriminant power of questionnaires employed in psychiatric research. ROC curves may assist in the selection of an optimal cut-off point, which has been shown to depend upon the prevalence rate, the consequences of correct and incorrect classifications, and the distributions of scale scores among normal individuals and cases. Then, ROC analysis is applied to the Child Behavior Checklist on a sample of 564 6-11-year old French children. The performances of the social competence scales of the instrument are poorer than those of the behavior problem scales. The findings suggest also that the cut-offs should be varied within this same age group and be higher for screening in community surveys. Further applications of ROC analysis are discussed.

Original languageEnglish (US)
Pages (from-to)677-693
Number of pages17
JournalJournal of Child Psychology and Psychiatry and Allied Disciplines
Volume32
Issue number4
StatePublished - 1991
Externally publishedYes

Fingerprint

Child Psychiatry
ROC Curve
Research
Child Behavior
Checklist
Psychiatry
Age Groups
Surveys and Questionnaires

ASJC Scopus subject areas

  • Psychiatry and Mental health
  • Developmental and Educational Psychology

Cite this

@article{b37a6b59bc8f4dd19fca18e0acc3af37,
title = "The use of questionnaires in child psychiatric research: Measuring their performance and choosing an optimal cut-off",
abstract = "This paper introduces the basic principles of Receiver Operating Characteristics (ROC) analysis, a technique to evaluate and compare the discriminant power of questionnaires employed in psychiatric research. ROC curves may assist in the selection of an optimal cut-off point, which has been shown to depend upon the prevalence rate, the consequences of correct and incorrect classifications, and the distributions of scale scores among normal individuals and cases. Then, ROC analysis is applied to the Child Behavior Checklist on a sample of 564 6-11-year old French children. The performances of the social competence scales of the instrument are poorer than those of the behavior problem scales. The findings suggest also that the cut-offs should be varied within this same age group and be higher for screening in community surveys. Further applications of ROC analysis are discussed.",
author = "Eric Fombonne",
year = "1991",
language = "English (US)",
volume = "32",
pages = "677--693",
journal = "Journal of Child Psychology and Psychiatry and Allied Disciplines",
issn = "0021-9630",
publisher = "Wiley-Blackwell",
number = "4",

}

TY - JOUR

T1 - The use of questionnaires in child psychiatric research

T2 - Measuring their performance and choosing an optimal cut-off

AU - Fombonne, Eric

PY - 1991

Y1 - 1991

N2 - This paper introduces the basic principles of Receiver Operating Characteristics (ROC) analysis, a technique to evaluate and compare the discriminant power of questionnaires employed in psychiatric research. ROC curves may assist in the selection of an optimal cut-off point, which has been shown to depend upon the prevalence rate, the consequences of correct and incorrect classifications, and the distributions of scale scores among normal individuals and cases. Then, ROC analysis is applied to the Child Behavior Checklist on a sample of 564 6-11-year old French children. The performances of the social competence scales of the instrument are poorer than those of the behavior problem scales. The findings suggest also that the cut-offs should be varied within this same age group and be higher for screening in community surveys. Further applications of ROC analysis are discussed.

AB - This paper introduces the basic principles of Receiver Operating Characteristics (ROC) analysis, a technique to evaluate and compare the discriminant power of questionnaires employed in psychiatric research. ROC curves may assist in the selection of an optimal cut-off point, which has been shown to depend upon the prevalence rate, the consequences of correct and incorrect classifications, and the distributions of scale scores among normal individuals and cases. Then, ROC analysis is applied to the Child Behavior Checklist on a sample of 564 6-11-year old French children. The performances of the social competence scales of the instrument are poorer than those of the behavior problem scales. The findings suggest also that the cut-offs should be varied within this same age group and be higher for screening in community surveys. Further applications of ROC analysis are discussed.

UR - http://www.scopus.com/inward/record.url?scp=0025875070&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0025875070&partnerID=8YFLogxK

M3 - Article

VL - 32

SP - 677

EP - 693

JO - Journal of Child Psychology and Psychiatry and Allied Disciplines

JF - Journal of Child Psychology and Psychiatry and Allied Disciplines

SN - 0021-9630

IS - 4

ER -