Respondent-driven sampling bias induced by community structure and response rates in social networks

Luis E C Rocha, Anna E. Thorson, Renaud Lambiotte, Fredrik Liljeros

Résultats de recherche: Contribution à un journal/une revueArticle

Résumé

Sampling hidden populations is particularly challenging by using standard sampling methods mainly because of the lack of a sampling frame. Respondent-driven sampling is an alternative methodology that exploits the social contacts between peers to reach and weight individuals in these hard-to-reach populations. It is a snowball sampling procedure where the weight of the respondents is adjusted for the likelihood of being sampled due to differences in the number of contacts. The structure of the social contacts thus regulates the process by constraining the sampling within subregions of the network. We study the bias induced by network communities, which are groups of individuals more connected between themselves than with individuals in other groups, in the respondent-driven sampling estimator. We simulate different structures and response rates to reproduce real settings. We find that the prevalence of the estimated variable is associated with the size of the network community to which the individual belongs and observe that low degree nodes may be undersampled if the sample and the network are of similar size. We also find that respondent-driven sampling estimators perform well if response rates are relatively large and the community structure is weak, whereas low response rates typically generate strong biases irrespectively of the community structure.

langueAnglais
Pages99-118
Nombre de pages20
journalJournal of the Royal Statistical Society. Series A: Statistics in Society
Volume180
Numéro1
Les DOIs
étatPublié - 1 janv. 2017

Empreinte digitale

Community Structure
Social Networks
social network
social relations
trend
community
Contact
Group
Estimator
contact
Sampling Methods
Social networks
Response rate
Community structure
Sampling
lack
methodology
Likelihood
Methodology
Alternatives

mots-clés

    Citer ceci

    @article{2ca58dbba27941bf8f0cf8f961ff47df,
    title = "Respondent-driven sampling bias induced by community structure and response rates in social networks",
    abstract = "Sampling hidden populations is particularly challenging by using standard sampling methods mainly because of the lack of a sampling frame. Respondent-driven sampling is an alternative methodology that exploits the social contacts between peers to reach and weight individuals in these hard-to-reach populations. It is a snowball sampling procedure where the weight of the respondents is adjusted for the likelihood of being sampled due to differences in the number of contacts. The structure of the social contacts thus regulates the process by constraining the sampling within subregions of the network. We study the bias induced by network communities, which are groups of individuals more connected between themselves than with individuals in other groups, in the respondent-driven sampling estimator. We simulate different structures and response rates to reproduce real settings. We find that the prevalence of the estimated variable is associated with the size of the network community to which the individual belongs and observe that low degree nodes may be undersampled if the sample and the network are of similar size. We also find that respondent-driven sampling estimators perform well if response rates are relatively large and the community structure is weak, whereas low response rates typically generate strong biases irrespectively of the community structure.",
    keywords = "Complex networks, Network sampling, Public health, Respondent-driven sampling bias",
    author = "Rocha, {Luis E C} and Thorson, {Anna E.} and Renaud Lambiotte and Fredrik Liljeros",
    year = "2017",
    month = "1",
    day = "1",
    doi = "10.1111/rssa.12180",
    language = "English",
    volume = "180",
    pages = "99--118",
    journal = "Journal of the Royal Statistical Society. Series A: Statistics in Society",
    issn = "0964-1998",
    publisher = "Wiley-Blackwell Publishing",
    number = "1",

    }

    Respondent-driven sampling bias induced by community structure and response rates in social networks. / Rocha, Luis E C; Thorson, Anna E.; Lambiotte, Renaud; Liljeros, Fredrik.

    Dans: Journal of the Royal Statistical Society. Series A: Statistics in Society, Vol 180, Numéro 1, 01.01.2017, p. 99-118.

    Résultats de recherche: Contribution à un journal/une revueArticle

    TY - JOUR

    T1 - Respondent-driven sampling bias induced by community structure and response rates in social networks

    AU - Rocha, Luis E C

    AU - Thorson, Anna E.

    AU - Lambiotte, Renaud

    AU - Liljeros, Fredrik

    PY - 2017/1/1

    Y1 - 2017/1/1

    N2 - Sampling hidden populations is particularly challenging by using standard sampling methods mainly because of the lack of a sampling frame. Respondent-driven sampling is an alternative methodology that exploits the social contacts between peers to reach and weight individuals in these hard-to-reach populations. It is a snowball sampling procedure where the weight of the respondents is adjusted for the likelihood of being sampled due to differences in the number of contacts. The structure of the social contacts thus regulates the process by constraining the sampling within subregions of the network. We study the bias induced by network communities, which are groups of individuals more connected between themselves than with individuals in other groups, in the respondent-driven sampling estimator. We simulate different structures and response rates to reproduce real settings. We find that the prevalence of the estimated variable is associated with the size of the network community to which the individual belongs and observe that low degree nodes may be undersampled if the sample and the network are of similar size. We also find that respondent-driven sampling estimators perform well if response rates are relatively large and the community structure is weak, whereas low response rates typically generate strong biases irrespectively of the community structure.

    AB - Sampling hidden populations is particularly challenging by using standard sampling methods mainly because of the lack of a sampling frame. Respondent-driven sampling is an alternative methodology that exploits the social contacts between peers to reach and weight individuals in these hard-to-reach populations. It is a snowball sampling procedure where the weight of the respondents is adjusted for the likelihood of being sampled due to differences in the number of contacts. The structure of the social contacts thus regulates the process by constraining the sampling within subregions of the network. We study the bias induced by network communities, which are groups of individuals more connected between themselves than with individuals in other groups, in the respondent-driven sampling estimator. We simulate different structures and response rates to reproduce real settings. We find that the prevalence of the estimated variable is associated with the size of the network community to which the individual belongs and observe that low degree nodes may be undersampled if the sample and the network are of similar size. We also find that respondent-driven sampling estimators perform well if response rates are relatively large and the community structure is weak, whereas low response rates typically generate strong biases irrespectively of the community structure.

    KW - Complex networks

    KW - Network sampling

    KW - Public health

    KW - Respondent-driven sampling bias

    UR - http://www.scopus.com/inward/record.url?scp=85000786107&partnerID=8YFLogxK

    U2 - 10.1111/rssa.12180

    DO - 10.1111/rssa.12180

    M3 - Article

    VL - 180

    SP - 99

    EP - 118

    JO - Journal of the Royal Statistical Society. Series A: Statistics in Society

    T2 - Journal of the Royal Statistical Society. Series A: Statistics in Society

    JF - Journal of the Royal Statistical Society. Series A: Statistics in Society

    SN - 0964-1998

    IS - 1

    ER -