The loneliness questionnaire-short version: An evaluation of reverse-worded and non-reverse-worded items via item response theory

Chad Ebesutani, Christopher Drescher, Steven P. Reise, Laurie Heiden, Terry L. Hight, John D. Damon, John Young

Research output: Contribution to journalArticle

24 Citations (Scopus)

Abstract

Although reverse-worded items have often been incorporated in scale construction to minimize the effects of acquiescent reporting biases, some researchers have more recently begun questioning this approach and wondering whether the advantages associated with incorporating reverse-worded items is worth the complexities that they bring to measures (e.g., Brown, 2003; Marsh, 1996). In this study, we used item response theory (IRT) to determine whether there is statistical justification to eliminate the reverse-worded items (e.g., I have lots of friends) from the Loneliness Questionnaire (LQ; Asher, Hymel, & Renshaw, 1984) and retain only the non-reverse-worded items (e.g., Im lonely) to inform the provision of a shortened LQ version. Using a large sample of children (Grades 2-7; n = 6,784) and adolescents (Grades 8-12; n = 4,941), we examined the psychometric properties of the 24-item LQ and found support for retaining the 9 non-reverse-worded LQ items to make up a shortened measure of loneliness in youth. We found that the non-reverse-worded items were associated with superior psychometric properties relative to the reverse-worded items with respect to reliability and IRT parameters (e.g., discrimination and item information). A 3-point Likert-type scale was also found to be more suitable for measuring loneliness across both children and adolescents compared to the original 5-point scale. The relative contributions of reverse-worded and non-reverse-worded items in scale development for youth instruments are also discussed.

Original languageEnglish (US)
Pages (from-to)427-437
Number of pages11
JournalJournal of Personality Assessment
Volume94
Issue number4
DOIs
StatePublished - Jul 1 2012
Externally publishedYes

Fingerprint

Loneliness
brotizolam
Psychometrics
Wetlands
Research Personnel
Surveys and Questionnaires
Questionnaire
Evaluation
Item Response Theory

ASJC Scopus subject areas

  • Clinical Psychology
  • Psychiatry and Mental health
  • Health, Toxicology and Mutagenesis

Cite this

The loneliness questionnaire-short version : An evaluation of reverse-worded and non-reverse-worded items via item response theory. / Ebesutani, Chad; Drescher, Christopher; Reise, Steven P.; Heiden, Laurie; Hight, Terry L.; Damon, John D.; Young, John.

In: Journal of Personality Assessment, Vol. 94, No. 4, 01.07.2012, p. 427-437.

Research output: Contribution to journalArticle

Ebesutani, Chad ; Drescher, Christopher ; Reise, Steven P. ; Heiden, Laurie ; Hight, Terry L. ; Damon, John D. ; Young, John. / The loneliness questionnaire-short version : An evaluation of reverse-worded and non-reverse-worded items via item response theory. In: Journal of Personality Assessment. 2012 ; Vol. 94, No. 4. pp. 427-437.
@article{e8833bb30ce545db8d9b7fc099988adb,
title = "The loneliness questionnaire-short version: An evaluation of reverse-worded and non-reverse-worded items via item response theory",
abstract = "Although reverse-worded items have often been incorporated in scale construction to minimize the effects of acquiescent reporting biases, some researchers have more recently begun questioning this approach and wondering whether the advantages associated with incorporating reverse-worded items is worth the complexities that they bring to measures (e.g., Brown, 2003; Marsh, 1996). In this study, we used item response theory (IRT) to determine whether there is statistical justification to eliminate the reverse-worded items (e.g., I have lots of friends) from the Loneliness Questionnaire (LQ; Asher, Hymel, & Renshaw, 1984) and retain only the non-reverse-worded items (e.g., Im lonely) to inform the provision of a shortened LQ version. Using a large sample of children (Grades 2-7; n = 6,784) and adolescents (Grades 8-12; n = 4,941), we examined the psychometric properties of the 24-item LQ and found support for retaining the 9 non-reverse-worded LQ items to make up a shortened measure of loneliness in youth. We found that the non-reverse-worded items were associated with superior psychometric properties relative to the reverse-worded items with respect to reliability and IRT parameters (e.g., discrimination and item information). A 3-point Likert-type scale was also found to be more suitable for measuring loneliness across both children and adolescents compared to the original 5-point scale. The relative contributions of reverse-worded and non-reverse-worded items in scale development for youth instruments are also discussed.",
author = "Chad Ebesutani and Christopher Drescher and Reise, {Steven P.} and Laurie Heiden and Hight, {Terry L.} and Damon, {John D.} and John Young",
year = "2012",
month = "7",
day = "1",
doi = "10.1080/00223891.2012.662188",
language = "English (US)",
volume = "94",
pages = "427--437",
journal = "Journal of Personality Assessment",
issn = "0022-3891",
publisher = "Taylor and Francis Ltd.",
number = "4",

}

TY - JOUR

T1 - The loneliness questionnaire-short version

T2 - An evaluation of reverse-worded and non-reverse-worded items via item response theory

AU - Ebesutani, Chad

AU - Drescher, Christopher

AU - Reise, Steven P.

AU - Heiden, Laurie

AU - Hight, Terry L.

AU - Damon, John D.

AU - Young, John

PY - 2012/7/1

Y1 - 2012/7/1

N2 - Although reverse-worded items have often been incorporated in scale construction to minimize the effects of acquiescent reporting biases, some researchers have more recently begun questioning this approach and wondering whether the advantages associated with incorporating reverse-worded items is worth the complexities that they bring to measures (e.g., Brown, 2003; Marsh, 1996). In this study, we used item response theory (IRT) to determine whether there is statistical justification to eliminate the reverse-worded items (e.g., I have lots of friends) from the Loneliness Questionnaire (LQ; Asher, Hymel, & Renshaw, 1984) and retain only the non-reverse-worded items (e.g., Im lonely) to inform the provision of a shortened LQ version. Using a large sample of children (Grades 2-7; n = 6,784) and adolescents (Grades 8-12; n = 4,941), we examined the psychometric properties of the 24-item LQ and found support for retaining the 9 non-reverse-worded LQ items to make up a shortened measure of loneliness in youth. We found that the non-reverse-worded items were associated with superior psychometric properties relative to the reverse-worded items with respect to reliability and IRT parameters (e.g., discrimination and item information). A 3-point Likert-type scale was also found to be more suitable for measuring loneliness across both children and adolescents compared to the original 5-point scale. The relative contributions of reverse-worded and non-reverse-worded items in scale development for youth instruments are also discussed.

AB - Although reverse-worded items have often been incorporated in scale construction to minimize the effects of acquiescent reporting biases, some researchers have more recently begun questioning this approach and wondering whether the advantages associated with incorporating reverse-worded items is worth the complexities that they bring to measures (e.g., Brown, 2003; Marsh, 1996). In this study, we used item response theory (IRT) to determine whether there is statistical justification to eliminate the reverse-worded items (e.g., I have lots of friends) from the Loneliness Questionnaire (LQ; Asher, Hymel, & Renshaw, 1984) and retain only the non-reverse-worded items (e.g., Im lonely) to inform the provision of a shortened LQ version. Using a large sample of children (Grades 2-7; n = 6,784) and adolescents (Grades 8-12; n = 4,941), we examined the psychometric properties of the 24-item LQ and found support for retaining the 9 non-reverse-worded LQ items to make up a shortened measure of loneliness in youth. We found that the non-reverse-worded items were associated with superior psychometric properties relative to the reverse-worded items with respect to reliability and IRT parameters (e.g., discrimination and item information). A 3-point Likert-type scale was also found to be more suitable for measuring loneliness across both children and adolescents compared to the original 5-point scale. The relative contributions of reverse-worded and non-reverse-worded items in scale development for youth instruments are also discussed.

UR - http://www.scopus.com/inward/record.url?scp=84862853653&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84862853653&partnerID=8YFLogxK

U2 - 10.1080/00223891.2012.662188

DO - 10.1080/00223891.2012.662188

M3 - Article

C2 - 22404209

AN - SCOPUS:84862853653

VL - 94

SP - 427

EP - 437

JO - Journal of Personality Assessment

JF - Journal of Personality Assessment

SN - 0022-3891

IS - 4

ER -