A simulative comparison of respondent driven sampling with incentivized snowball sampling - The "strudel effect"

V. Anna Gyarmathy, Lisa G. Johnston, Irma Caplinskiene, Saulius Caplinskas, Carl A Latkin

Research output: Contribution to journalArticle

Abstract

Background: Respondent driven sampling (RDS) and incentivized snowball sampling (ISS) are two sampling methods that are commonly used to reach people who inject drugs (PWID). Methods: We generated a set of simulated RDS samples on an actual sociometric ISS sample of PWID in Vilnius, Lithuania ("original sample") to assess if the simulated RDS estimates were statistically significantly different from the original ISS sample prevalences for HIV (9.8%), Hepatitis A (43.6%), Hepatitis B (Anti-HBc 43.9% and HBsAg 3.4%), Hepatitis C (87.5%), syphilis (6.8%) and Chlamydia (8.8%) infections and for selected behavioral risk characteristics. Results: The original sample consisted of a large component of 249 people (83% of the sample) and 13 smaller components with 1-12 individuals. Generally, as long as all seeds were recruited from the large component of the original sample, the simulation samples simply recreated the large component. There were no significant differences between the large component and the entire original sample for the characteristics of interest. Altogether 99.2% of 360 simulation sample point estimates were within the confidence interval of the original prevalence values for the characteristics of interest. Conclusions: When population characteristics are reflected in large network components that dominate the population, RDS and ISS may produce samples that have statistically non-different prevalence values, even though some isolated network components may be under-sampled and/or statistically significantly different from the main groups. This so-called "strudel effect" is discussed in the paper.

Original languageEnglish (US)
Pages (from-to)71-77
Number of pages7
JournalDrug and Alcohol Dependence
Volume135
Issue number1
DOIs
StatePublished - 2014

Fingerprint

Sampling
Lithuania
Chlamydia Infections
Hepatitis A
Network components
Population Characteristics
Syphilis
Hepatitis C
Hepatitis B Surface Antigens
Hepatitis B
Pharmaceutical Preparations
Seeds
HIV
Confidence Intervals
Surveys and Questionnaires
Population
Seed

Keywords

  • Incentivized snowball sampling
  • People who inject drugs
  • Prevalence estimates
  • Respondent driven sampling
  • Sampling methodology
  • Simulations

ASJC Scopus subject areas

  • Psychiatry and Mental health
  • Toxicology
  • Pharmacology
  • Pharmacology (medical)

Cite this

A simulative comparison of respondent driven sampling with incentivized snowball sampling - The "strudel effect". / Gyarmathy, V. Anna; Johnston, Lisa G.; Caplinskiene, Irma; Caplinskas, Saulius; Latkin, Carl A.

In: Drug and Alcohol Dependence, Vol. 135, No. 1, 2014, p. 71-77.

Research output: Contribution to journalArticle

Gyarmathy, V. Anna ; Johnston, Lisa G. ; Caplinskiene, Irma ; Caplinskas, Saulius ; Latkin, Carl A. / A simulative comparison of respondent driven sampling with incentivized snowball sampling - The "strudel effect". In: Drug and Alcohol Dependence. 2014 ; Vol. 135, No. 1. pp. 71-77.
@article{1d7024fd99a74ae9aef3fc07aa48fcdb,
title = "A simulative comparison of respondent driven sampling with incentivized snowball sampling - The {"}strudel effect{"}",
abstract = "Background: Respondent driven sampling (RDS) and incentivized snowball sampling (ISS) are two sampling methods that are commonly used to reach people who inject drugs (PWID). Methods: We generated a set of simulated RDS samples on an actual sociometric ISS sample of PWID in Vilnius, Lithuania ({"}original sample{"}) to assess if the simulated RDS estimates were statistically significantly different from the original ISS sample prevalences for HIV (9.8{\%}), Hepatitis A (43.6{\%}), Hepatitis B (Anti-HBc 43.9{\%} and HBsAg 3.4{\%}), Hepatitis C (87.5{\%}), syphilis (6.8{\%}) and Chlamydia (8.8{\%}) infections and for selected behavioral risk characteristics. Results: The original sample consisted of a large component of 249 people (83{\%} of the sample) and 13 smaller components with 1-12 individuals. Generally, as long as all seeds were recruited from the large component of the original sample, the simulation samples simply recreated the large component. There were no significant differences between the large component and the entire original sample for the characteristics of interest. Altogether 99.2{\%} of 360 simulation sample point estimates were within the confidence interval of the original prevalence values for the characteristics of interest. Conclusions: When population characteristics are reflected in large network components that dominate the population, RDS and ISS may produce samples that have statistically non-different prevalence values, even though some isolated network components may be under-sampled and/or statistically significantly different from the main groups. This so-called {"}strudel effect{"} is discussed in the paper.",
keywords = "Incentivized snowball sampling, People who inject drugs, Prevalence estimates, Respondent driven sampling, Sampling methodology, Simulations",
author = "Gyarmathy, {V. Anna} and Johnston, {Lisa G.} and Irma Caplinskiene and Saulius Caplinskas and Latkin, {Carl A}",
year = "2014",
doi = "10.1016/j.drugalcdep.2013.11.020",
language = "English (US)",
volume = "135",
pages = "71--77",
journal = "Drug and Alcohol Dependence",
issn = "0376-8716",
publisher = "Elsevier Ireland Ltd",
number = "1",

}

TY - JOUR

T1 - A simulative comparison of respondent driven sampling with incentivized snowball sampling - The "strudel effect"

AU - Gyarmathy, V. Anna

AU - Johnston, Lisa G.

AU - Caplinskiene, Irma

AU - Caplinskas, Saulius

AU - Latkin, Carl A

PY - 2014

Y1 - 2014

N2 - Background: Respondent driven sampling (RDS) and incentivized snowball sampling (ISS) are two sampling methods that are commonly used to reach people who inject drugs (PWID). Methods: We generated a set of simulated RDS samples on an actual sociometric ISS sample of PWID in Vilnius, Lithuania ("original sample") to assess if the simulated RDS estimates were statistically significantly different from the original ISS sample prevalences for HIV (9.8%), Hepatitis A (43.6%), Hepatitis B (Anti-HBc 43.9% and HBsAg 3.4%), Hepatitis C (87.5%), syphilis (6.8%) and Chlamydia (8.8%) infections and for selected behavioral risk characteristics. Results: The original sample consisted of a large component of 249 people (83% of the sample) and 13 smaller components with 1-12 individuals. Generally, as long as all seeds were recruited from the large component of the original sample, the simulation samples simply recreated the large component. There were no significant differences between the large component and the entire original sample for the characteristics of interest. Altogether 99.2% of 360 simulation sample point estimates were within the confidence interval of the original prevalence values for the characteristics of interest. Conclusions: When population characteristics are reflected in large network components that dominate the population, RDS and ISS may produce samples that have statistically non-different prevalence values, even though some isolated network components may be under-sampled and/or statistically significantly different from the main groups. This so-called "strudel effect" is discussed in the paper.

AB - Background: Respondent driven sampling (RDS) and incentivized snowball sampling (ISS) are two sampling methods that are commonly used to reach people who inject drugs (PWID). Methods: We generated a set of simulated RDS samples on an actual sociometric ISS sample of PWID in Vilnius, Lithuania ("original sample") to assess if the simulated RDS estimates were statistically significantly different from the original ISS sample prevalences for HIV (9.8%), Hepatitis A (43.6%), Hepatitis B (Anti-HBc 43.9% and HBsAg 3.4%), Hepatitis C (87.5%), syphilis (6.8%) and Chlamydia (8.8%) infections and for selected behavioral risk characteristics. Results: The original sample consisted of a large component of 249 people (83% of the sample) and 13 smaller components with 1-12 individuals. Generally, as long as all seeds were recruited from the large component of the original sample, the simulation samples simply recreated the large component. There were no significant differences between the large component and the entire original sample for the characteristics of interest. Altogether 99.2% of 360 simulation sample point estimates were within the confidence interval of the original prevalence values for the characteristics of interest. Conclusions: When population characteristics are reflected in large network components that dominate the population, RDS and ISS may produce samples that have statistically non-different prevalence values, even though some isolated network components may be under-sampled and/or statistically significantly different from the main groups. This so-called "strudel effect" is discussed in the paper.

KW - Incentivized snowball sampling

KW - People who inject drugs

KW - Prevalence estimates

KW - Respondent driven sampling

KW - Sampling methodology

KW - Simulations

UR - http://www.scopus.com/inward/record.url?scp=84891831672&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84891831672&partnerID=8YFLogxK

U2 - 10.1016/j.drugalcdep.2013.11.020

DO - 10.1016/j.drugalcdep.2013.11.020

M3 - Article

C2 - 24360650

AN - SCOPUS:84891831672

VL - 135

SP - 71

EP - 77

JO - Drug and Alcohol Dependence

JF - Drug and Alcohol Dependence

SN - 0376-8716

IS - 1

ER -