The prevalence of compulsive buying: A meta-analysis

Aniko Maraz, Mark D. Griffiths, Z. Demetrovics

Research output: Contribution to journalArticle

46 Citations (Scopus)

Abstract

Aims: To estimate the pooled prevalence of compulsive buying behaviour (CBB) in different populations and to determine the effect of age, gender, location and screening instrument on the reported heterogeneity in estimates of CBB and whether publication bias could be identified. Methods: Three databases were searched (Medline, PsychInfo, Web of Science) using the terms 'compulsive buying', 'pathological buying' and 'compulsive shopping' to estimate the pooled prevalence of CBB in different populations. Forty studies reporting 49 prevalence estimates from 16 countries were located (n = 32 000). To conduct the meta-analysis, data from non-clinical studies regarding mean age and gender proportion, geographical study location and screening instrument used to assess CBB were extracted by multiple independent observers and evaluated using a random-effects model. Four a priori subgroups were analysed using pooled estimation (Cohen's Q) and covariate testing (moderator and meta-regression analysis). Results: The CBB pooled prevalence of adult representative studies was 4.9% (3.4-6.9%, eight estimates, 10 102 participants), although estimates were higher among university students: 8.3% (5.9-11.5%, 19 estimates, 14 947 participants) in adult non-representative samples: 12.3% (7.6-19.1%, 11 estimates, 3929 participants) and in shopping-specific samples: 16.2% (8.8-27.8%, 11 estimates, 4686 participants). Being young and female were associated with increased tendency, but not location (United States versus non-United States). Meta-regression revealed large heterogeneity within subgroups, due mainly to diverse measures and time-frames (current versus life-time) used to assess CBB. Conclusions: A pooled estimate of compulsive buying behaviour in the populations studied is approximately 5%, but there is large variation between samples accounted for largely by use of different time-frames and measures.

Original languageEnglish
Pages (from-to)408-419
Number of pages12
JournalAddiction
Volume111
Issue number3
DOIs
Publication statusPublished - Mar 1 2016

Fingerprint

Compulsive Behavior
Meta-Analysis
Population
Publication Bias
Regression Analysis
Databases
Students

Keywords

  • Addictive behaviour
  • Consumer behaviour
  • Cross-cultural comparison
  • Epidemiology
  • Publication bias
  • Shopping addiction

ASJC Scopus subject areas

  • Medicine (miscellaneous)
  • Psychiatry and Mental health

Cite this

The prevalence of compulsive buying : A meta-analysis. / Maraz, Aniko; Griffiths, Mark D.; Demetrovics, Z.

In: Addiction, Vol. 111, No. 3, 01.03.2016, p. 408-419.

Research output: Contribution to journalArticle

Maraz, Aniko ; Griffiths, Mark D. ; Demetrovics, Z. / The prevalence of compulsive buying : A meta-analysis. In: Addiction. 2016 ; Vol. 111, No. 3. pp. 408-419.
@article{496f2323a9a24b64a85097627a13921e,
title = "The prevalence of compulsive buying: A meta-analysis",
abstract = "Aims: To estimate the pooled prevalence of compulsive buying behaviour (CBB) in different populations and to determine the effect of age, gender, location and screening instrument on the reported heterogeneity in estimates of CBB and whether publication bias could be identified. Methods: Three databases were searched (Medline, PsychInfo, Web of Science) using the terms 'compulsive buying', 'pathological buying' and 'compulsive shopping' to estimate the pooled prevalence of CBB in different populations. Forty studies reporting 49 prevalence estimates from 16 countries were located (n = 32 000). To conduct the meta-analysis, data from non-clinical studies regarding mean age and gender proportion, geographical study location and screening instrument used to assess CBB were extracted by multiple independent observers and evaluated using a random-effects model. Four a priori subgroups were analysed using pooled estimation (Cohen's Q) and covariate testing (moderator and meta-regression analysis). Results: The CBB pooled prevalence of adult representative studies was 4.9{\%} (3.4-6.9{\%}, eight estimates, 10 102 participants), although estimates were higher among university students: 8.3{\%} (5.9-11.5{\%}, 19 estimates, 14 947 participants) in adult non-representative samples: 12.3{\%} (7.6-19.1{\%}, 11 estimates, 3929 participants) and in shopping-specific samples: 16.2{\%} (8.8-27.8{\%}, 11 estimates, 4686 participants). Being young and female were associated with increased tendency, but not location (United States versus non-United States). Meta-regression revealed large heterogeneity within subgroups, due mainly to diverse measures and time-frames (current versus life-time) used to assess CBB. Conclusions: A pooled estimate of compulsive buying behaviour in the populations studied is approximately 5{\%}, but there is large variation between samples accounted for largely by use of different time-frames and measures.",
keywords = "Addictive behaviour, Consumer behaviour, Cross-cultural comparison, Epidemiology, Publication bias, Shopping addiction",
author = "Aniko Maraz and Griffiths, {Mark D.} and Z. Demetrovics",
year = "2016",
month = "3",
day = "1",
doi = "10.1111/add.13223",
language = "English",
volume = "111",
pages = "408--419",
journal = "Addiction",
issn = "0965-2140",
publisher = "Wiley-Blackwell",
number = "3",

}

TY - JOUR

T1 - The prevalence of compulsive buying

T2 - A meta-analysis

AU - Maraz, Aniko

AU - Griffiths, Mark D.

AU - Demetrovics, Z.

PY - 2016/3/1

Y1 - 2016/3/1

N2 - Aims: To estimate the pooled prevalence of compulsive buying behaviour (CBB) in different populations and to determine the effect of age, gender, location and screening instrument on the reported heterogeneity in estimates of CBB and whether publication bias could be identified. Methods: Three databases were searched (Medline, PsychInfo, Web of Science) using the terms 'compulsive buying', 'pathological buying' and 'compulsive shopping' to estimate the pooled prevalence of CBB in different populations. Forty studies reporting 49 prevalence estimates from 16 countries were located (n = 32 000). To conduct the meta-analysis, data from non-clinical studies regarding mean age and gender proportion, geographical study location and screening instrument used to assess CBB were extracted by multiple independent observers and evaluated using a random-effects model. Four a priori subgroups were analysed using pooled estimation (Cohen's Q) and covariate testing (moderator and meta-regression analysis). Results: The CBB pooled prevalence of adult representative studies was 4.9% (3.4-6.9%, eight estimates, 10 102 participants), although estimates were higher among university students: 8.3% (5.9-11.5%, 19 estimates, 14 947 participants) in adult non-representative samples: 12.3% (7.6-19.1%, 11 estimates, 3929 participants) and in shopping-specific samples: 16.2% (8.8-27.8%, 11 estimates, 4686 participants). Being young and female were associated with increased tendency, but not location (United States versus non-United States). Meta-regression revealed large heterogeneity within subgroups, due mainly to diverse measures and time-frames (current versus life-time) used to assess CBB. Conclusions: A pooled estimate of compulsive buying behaviour in the populations studied is approximately 5%, but there is large variation between samples accounted for largely by use of different time-frames and measures.

AB - Aims: To estimate the pooled prevalence of compulsive buying behaviour (CBB) in different populations and to determine the effect of age, gender, location and screening instrument on the reported heterogeneity in estimates of CBB and whether publication bias could be identified. Methods: Three databases were searched (Medline, PsychInfo, Web of Science) using the terms 'compulsive buying', 'pathological buying' and 'compulsive shopping' to estimate the pooled prevalence of CBB in different populations. Forty studies reporting 49 prevalence estimates from 16 countries were located (n = 32 000). To conduct the meta-analysis, data from non-clinical studies regarding mean age and gender proportion, geographical study location and screening instrument used to assess CBB were extracted by multiple independent observers and evaluated using a random-effects model. Four a priori subgroups were analysed using pooled estimation (Cohen's Q) and covariate testing (moderator and meta-regression analysis). Results: The CBB pooled prevalence of adult representative studies was 4.9% (3.4-6.9%, eight estimates, 10 102 participants), although estimates were higher among university students: 8.3% (5.9-11.5%, 19 estimates, 14 947 participants) in adult non-representative samples: 12.3% (7.6-19.1%, 11 estimates, 3929 participants) and in shopping-specific samples: 16.2% (8.8-27.8%, 11 estimates, 4686 participants). Being young and female were associated with increased tendency, but not location (United States versus non-United States). Meta-regression revealed large heterogeneity within subgroups, due mainly to diverse measures and time-frames (current versus life-time) used to assess CBB. Conclusions: A pooled estimate of compulsive buying behaviour in the populations studied is approximately 5%, but there is large variation between samples accounted for largely by use of different time-frames and measures.

KW - Addictive behaviour

KW - Consumer behaviour

KW - Cross-cultural comparison

KW - Epidemiology

KW - Publication bias

KW - Shopping addiction

UR - http://www.scopus.com/inward/record.url?scp=84957435501&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84957435501&partnerID=8YFLogxK

U2 - 10.1111/add.13223

DO - 10.1111/add.13223

M3 - Article

C2 - 26517309

AN - SCOPUS:84957435501

VL - 111

SP - 408

EP - 419

JO - Addiction

JF - Addiction

SN - 0965-2140

IS - 3

ER -