Respondent-driven sampling (RDS) is a network-based method for sampling hidden and hard-to-reach ... more Respondent-driven sampling (RDS) is a network-based method for sampling hidden and hard-to-reach populations that has been shown to produce asymptotically unbiased population estimates when its assumptions are satisfied. This includes resolving a major concern regarding bias in chain-referral samples—that is, produc-ing a population estimate that is independent of the seeds (initial subjects) with which sampling began. However, RDS estimates are limited to nominal variables, and one of the assumptions re-quired for the proof of lack of bias is the absence of differential recruitment. One aim of this paper is to analyze the role of dif-ferential recruitment, quantify the bias it produces, and propose a This research was made possible by grants from the National Endowment
Rest breaks ff the clock work eal breaks vertime inimum wage ight to organize etaliation Unprotec... more Rest breaks ff the clock work eal breaks vertime inimum wage ight to organize etaliation Unprotected orkers’
Background: Hidden populations at elevated risk for HIV, such as illicit drug users and bisexual ... more Background: Hidden populations at elevated risk for HIV, such as illicit drug users and bisexual men of color, can be hard to reach and are not amenable to random sampling. Novel approaches should be developed and tested for studying such populations. Respondent-driven sampling (RDS) seeks to construct samples from populations that lack comprehensive sampling frames needed for random sampling, while reducing biases associated with chain-referral methods. RDS is based on the recognition that, if allowed to continue through enough waves, the sample attains a stable composition (equilibrium) independent of the initial choice of subjects. Methods: Respondent-driven sampling was used to recruit persons who use hard' drugs, men who had sex with men, and sex partners of both groups for a study of the sexual diffusion of HIV. RDS assumes that those best able to access members of hidden populations are their own peers. RDS motivates peers by offering incentives both for being interviewed...
Researchers generally use nonprobability methods such as chain-referral sampling to study populat... more Researchers generally use nonprobability methods such as chain-referral sampling to study populations for which no sampling frame exists. Respondent-driven sampling is a new form of chain-referral sampling that was designed to reduce several sources of bias associated with this method, including those from the choice of initial participants, volunteerism, and masking. This study expands this method by introducing "steering incentives," supplemental rewards for referral of members of a specific group, injection drug users (IDUs) aged 18-25. The results are based on an interrupted time series analysis in which 196 IDUs from Meriden, CT, were interviewed before introduction of the steering incentives, and another 190 were interviewed afterwards. The steering incentives increased the percentage of younger IDUs sampled by 70%. We compared recruitment patterns with institutional data and self-reported personal networks to determine representativeness and whether volunteerism or ...
est breaks ff the clock work eal breaks vertime inimum wage ight Working to organize etaliation o... more est breaks ff the clock work eal breaks vertime inimum wage ight Working to organize etaliation orkers ’ comp est A Survey breaks
est breaks ff the clock work eal breaks vertime inimum wage ight to organize etaliation Unprotect... more est breaks ff the clock work eal breaks vertime inimum wage ight to organize etaliation Unprotected orkers’
Many populations of interest present special challenges for traditional survey methodology when i... more Many populations of interest present special challenges for traditional survey methodology when it is difficult or impossible to obtain a traditional sampling frame. In the case of such “hidden” populations at risk of HIV/AIDS, many researchers have resorted to chain-referral sampling. Recent progress on the theory of chain-referral sampling has led to Respondent Driven Sampling (RDS), a rigorous chain-referral methodwhich allows unbiased estimation of the target population. In this article we present new probability-theoretic methods for making estimates fromRDS data. The new estimators offer improved simplicity, analytical tractability, and allow the estimation of continuous variables. An analytical variance estimator is proposed in the case of estimating categorical variables. The properties of the estimator and the associated variance estimator are explored in a simulation study, and compared to alternative RDS estimators using data from a study of New York City jazz musicians. ...
The literature on jazz is vast, focusing on the biographies of influential figures, history and e... more The literature on jazz is vast, focusing on the biographies of influential figures, history and emergence of performing styles, and the influence of jazz on other areas of culture and society. However, only two studies, and no studies in the United States, have examined jazz musician communities, including not only the famous but also the obscure, the leaders as well as the followers, and both were limited by the use of nonprobability sampling methods. This study reports on studies of jazz musician communities in the greater New York metropolitan area and the San Francisco Bay area, using a sampling method, respondent-driven sampling, which permits representative samples to be drawn even from hard-to-reach populations. The analysis also introduces a new method for studying the structure of very large social networks, termed “breakpoint analysis.” The principle focus of the study is on identifying patterns of affiliation, that is, determining who associates with whom as a means for a...
The primary responsibility of the chair of our section is to organize an exciting series of sessi... more The primary responsibility of the chair of our section is to organize an exciting series of sessions for the Annual Meeting. It has become a tradition that the theory section organizes a “mini-conference,” a tradition to which I will adhere. Once again the section, having over six hundred members, will have four sessions (plus time for a business meeting and a set of roundtables). I have set aside three of our sessions for a mini-conference on the theme of “Sociological Theory and Empirical Research.”
Good estimates of key population sizes are critical for appropriating resources to prevent HIV in... more Good estimates of key population sizes are critical for appropriating resources to prevent HIV infection. We conducted two capture/recapture studies to estimate the number of PWID currently in Hai Phong, Vietnam. A 2014 respondent-driven sampling (RDS) survey served as one capture, and distribution of cigarette lighters at drug use "hotspots" in 2016 served as another "capture." A 2016 survey using RDS, conducted 1 week after lighter distribution, served as "recapture" for both captures. Recaptured participants in the two surveys were identified with a computerized fingerprint reader. Recaptured participants from the lighter distribution were asked to show their lighters. 1385 participants were included in the "recapture" survey. They were 94% male and had a median age of 39. All (100%) injected heroin, and HIV prevalence was 30%. 144 of the 603 participants in the 2014 survey and 152 of the 600 PWID who had received lighters were "recapt...
Respondent-driven sampling (RDS), a link-tracing sampling and inference method for studying hard-... more Respondent-driven sampling (RDS), a link-tracing sampling and inference method for studying hard-to-reach populations, has been shown to produce asymptotically unbiased population estimates when its assumptions are satisfied. However, some of the assumptions are prohibitively difficult to reach in the field, and the violation of a crucial assumption can produce biased estimates. We compare two different inference approaches: design-based inference, which relies on the known probability of selection in sampling, and model-based inference, which is based on models of human recruitment behavior and the social context within which sampling is conducted. The advantage of the latter approach is that when the violation of an assumption has been shown to produce biased population estimates, the model can be adjusted to more accurately reflect actual recruitment behavior, and thereby control for the source of bias. To illustrate this process, we focus on three sources of bias, differential e...
Violent drug markets are not as prominent as they once were in the United States, but they still ... more Violent drug markets are not as prominent as they once were in the United States, but they still exist and are associated with significant crime and lower quality of life. The drug market intervention (DMI) is an innovative strategy that uses focused deterrence, community engagement, and incapacitation to reduce crime and disorder associated with these markets. Although studies show that DMI can reduce crime and overt drug activity, one perspective is prominently missing from these evaluations: those who purchase drugs. This study explores the use of respondent-driven sampling (RDS)-a statistical sampling method-to approximate a representative sample of drug users who purchased drugs in a targeted DMI market to gain insight into the effect of a DMI on market dynamics. Using RDS, we recruited individuals who reported hard drug use (crack or powder cocaine, heroin, methamphetamine, or illicit use of prescriptions opioids) in the last month to participate in a survey. The main survey a...
Classical Respondent-Driven Sampling (RDS) estimators are based on a Markov Process model in whic... more Classical Respondent-Driven Sampling (RDS) estimators are based on a Markov Process model in which sampling occurs with replacement. Given that respondents generally cannot be interviewed more than once, this assumption is counterfactual. We join recent work by Gile and Handcock in exploring the implications of the sampling-with-replacement assumption for bias of RDS estimators. We differ from previous studies in examining a wider range of sampling fractions and in using not only simulations but also formal proofs. One key finding is that RDS estimates are surprisingly stable even in the presence of substantial sampling fractions. Our analyses show that the sampling-with-replacement assumption is a minor contributor to bias for sampling fractions under 40%, and bias is negligible for the 20% or smaller sampling fractions typical of field applications of RDS.
Combined prevention for HIV among persons who inject drugs (PWID) has led to greatly reduced HIV ... more Combined prevention for HIV among persons who inject drugs (PWID) has led to greatly reduced HIV transmission among PWID in many high-income settings, but these successes have not yet been replicated in resource-limited settings. Haiphong, Vietnam experienced a large HIV epidemic among PWID, with 68% prevalence in 2006. Haiphong has implemented needle/syringe programs, methadone maintenance treatment (MMT), and anti-retroviral treatment (ART), but there is an urgent need to identify high-risk PWID and link them to services. We examined integration of respondent-driven sampling (RDS) and strong peer support groups as a mechanism for identifying high-risk PWID and linking them to services. The peer support staff performed the key tasks that required building and maintaining trust with the participants, including recruiting the RDS seeds, greeting and registering participants at the research site, taking electronic copies of participant fingerprints (to prevent multiple participation i...
Respondent-driven sampling (RDS) is a network-based method for sampling hidden and hard-to-reach ... more Respondent-driven sampling (RDS) is a network-based method for sampling hidden and hard-to-reach populations that has been shown to produce asymptotically unbiased population estimates when its assumptions are satisfied. This includes resolving a major concern regarding bias in chain-referral samples—that is, produc-ing a population estimate that is independent of the seeds (initial subjects) with which sampling began. However, RDS estimates are limited to nominal variables, and one of the assumptions re-quired for the proof of lack of bias is the absence of differential recruitment. One aim of this paper is to analyze the role of dif-ferential recruitment, quantify the bias it produces, and propose a This research was made possible by grants from the National Endowment
Rest breaks ff the clock work eal breaks vertime inimum wage ight to organize etaliation Unprotec... more Rest breaks ff the clock work eal breaks vertime inimum wage ight to organize etaliation Unprotected orkers’
Background: Hidden populations at elevated risk for HIV, such as illicit drug users and bisexual ... more Background: Hidden populations at elevated risk for HIV, such as illicit drug users and bisexual men of color, can be hard to reach and are not amenable to random sampling. Novel approaches should be developed and tested for studying such populations. Respondent-driven sampling (RDS) seeks to construct samples from populations that lack comprehensive sampling frames needed for random sampling, while reducing biases associated with chain-referral methods. RDS is based on the recognition that, if allowed to continue through enough waves, the sample attains a stable composition (equilibrium) independent of the initial choice of subjects. Methods: Respondent-driven sampling was used to recruit persons who use hard' drugs, men who had sex with men, and sex partners of both groups for a study of the sexual diffusion of HIV. RDS assumes that those best able to access members of hidden populations are their own peers. RDS motivates peers by offering incentives both for being interviewed...
Researchers generally use nonprobability methods such as chain-referral sampling to study populat... more Researchers generally use nonprobability methods such as chain-referral sampling to study populations for which no sampling frame exists. Respondent-driven sampling is a new form of chain-referral sampling that was designed to reduce several sources of bias associated with this method, including those from the choice of initial participants, volunteerism, and masking. This study expands this method by introducing "steering incentives," supplemental rewards for referral of members of a specific group, injection drug users (IDUs) aged 18-25. The results are based on an interrupted time series analysis in which 196 IDUs from Meriden, CT, were interviewed before introduction of the steering incentives, and another 190 were interviewed afterwards. The steering incentives increased the percentage of younger IDUs sampled by 70%. We compared recruitment patterns with institutional data and self-reported personal networks to determine representativeness and whether volunteerism or ...
est breaks ff the clock work eal breaks vertime inimum wage ight Working to organize etaliation o... more est breaks ff the clock work eal breaks vertime inimum wage ight Working to organize etaliation orkers ’ comp est A Survey breaks
est breaks ff the clock work eal breaks vertime inimum wage ight to organize etaliation Unprotect... more est breaks ff the clock work eal breaks vertime inimum wage ight to organize etaliation Unprotected orkers’
Many populations of interest present special challenges for traditional survey methodology when i... more Many populations of interest present special challenges for traditional survey methodology when it is difficult or impossible to obtain a traditional sampling frame. In the case of such “hidden” populations at risk of HIV/AIDS, many researchers have resorted to chain-referral sampling. Recent progress on the theory of chain-referral sampling has led to Respondent Driven Sampling (RDS), a rigorous chain-referral methodwhich allows unbiased estimation of the target population. In this article we present new probability-theoretic methods for making estimates fromRDS data. The new estimators offer improved simplicity, analytical tractability, and allow the estimation of continuous variables. An analytical variance estimator is proposed in the case of estimating categorical variables. The properties of the estimator and the associated variance estimator are explored in a simulation study, and compared to alternative RDS estimators using data from a study of New York City jazz musicians. ...
The literature on jazz is vast, focusing on the biographies of influential figures, history and e... more The literature on jazz is vast, focusing on the biographies of influential figures, history and emergence of performing styles, and the influence of jazz on other areas of culture and society. However, only two studies, and no studies in the United States, have examined jazz musician communities, including not only the famous but also the obscure, the leaders as well as the followers, and both were limited by the use of nonprobability sampling methods. This study reports on studies of jazz musician communities in the greater New York metropolitan area and the San Francisco Bay area, using a sampling method, respondent-driven sampling, which permits representative samples to be drawn even from hard-to-reach populations. The analysis also introduces a new method for studying the structure of very large social networks, termed “breakpoint analysis.” The principle focus of the study is on identifying patterns of affiliation, that is, determining who associates with whom as a means for a...
The primary responsibility of the chair of our section is to organize an exciting series of sessi... more The primary responsibility of the chair of our section is to organize an exciting series of sessions for the Annual Meeting. It has become a tradition that the theory section organizes a “mini-conference,” a tradition to which I will adhere. Once again the section, having over six hundred members, will have four sessions (plus time for a business meeting and a set of roundtables). I have set aside three of our sessions for a mini-conference on the theme of “Sociological Theory and Empirical Research.”
Good estimates of key population sizes are critical for appropriating resources to prevent HIV in... more Good estimates of key population sizes are critical for appropriating resources to prevent HIV infection. We conducted two capture/recapture studies to estimate the number of PWID currently in Hai Phong, Vietnam. A 2014 respondent-driven sampling (RDS) survey served as one capture, and distribution of cigarette lighters at drug use "hotspots" in 2016 served as another "capture." A 2016 survey using RDS, conducted 1 week after lighter distribution, served as "recapture" for both captures. Recaptured participants in the two surveys were identified with a computerized fingerprint reader. Recaptured participants from the lighter distribution were asked to show their lighters. 1385 participants were included in the "recapture" survey. They were 94% male and had a median age of 39. All (100%) injected heroin, and HIV prevalence was 30%. 144 of the 603 participants in the 2014 survey and 152 of the 600 PWID who had received lighters were "recapt...
Respondent-driven sampling (RDS), a link-tracing sampling and inference method for studying hard-... more Respondent-driven sampling (RDS), a link-tracing sampling and inference method for studying hard-to-reach populations, has been shown to produce asymptotically unbiased population estimates when its assumptions are satisfied. However, some of the assumptions are prohibitively difficult to reach in the field, and the violation of a crucial assumption can produce biased estimates. We compare two different inference approaches: design-based inference, which relies on the known probability of selection in sampling, and model-based inference, which is based on models of human recruitment behavior and the social context within which sampling is conducted. The advantage of the latter approach is that when the violation of an assumption has been shown to produce biased population estimates, the model can be adjusted to more accurately reflect actual recruitment behavior, and thereby control for the source of bias. To illustrate this process, we focus on three sources of bias, differential e...
Violent drug markets are not as prominent as they once were in the United States, but they still ... more Violent drug markets are not as prominent as they once were in the United States, but they still exist and are associated with significant crime and lower quality of life. The drug market intervention (DMI) is an innovative strategy that uses focused deterrence, community engagement, and incapacitation to reduce crime and disorder associated with these markets. Although studies show that DMI can reduce crime and overt drug activity, one perspective is prominently missing from these evaluations: those who purchase drugs. This study explores the use of respondent-driven sampling (RDS)-a statistical sampling method-to approximate a representative sample of drug users who purchased drugs in a targeted DMI market to gain insight into the effect of a DMI on market dynamics. Using RDS, we recruited individuals who reported hard drug use (crack or powder cocaine, heroin, methamphetamine, or illicit use of prescriptions opioids) in the last month to participate in a survey. The main survey a...
Classical Respondent-Driven Sampling (RDS) estimators are based on a Markov Process model in whic... more Classical Respondent-Driven Sampling (RDS) estimators are based on a Markov Process model in which sampling occurs with replacement. Given that respondents generally cannot be interviewed more than once, this assumption is counterfactual. We join recent work by Gile and Handcock in exploring the implications of the sampling-with-replacement assumption for bias of RDS estimators. We differ from previous studies in examining a wider range of sampling fractions and in using not only simulations but also formal proofs. One key finding is that RDS estimates are surprisingly stable even in the presence of substantial sampling fractions. Our analyses show that the sampling-with-replacement assumption is a minor contributor to bias for sampling fractions under 40%, and bias is negligible for the 20% or smaller sampling fractions typical of field applications of RDS.
Combined prevention for HIV among persons who inject drugs (PWID) has led to greatly reduced HIV ... more Combined prevention for HIV among persons who inject drugs (PWID) has led to greatly reduced HIV transmission among PWID in many high-income settings, but these successes have not yet been replicated in resource-limited settings. Haiphong, Vietnam experienced a large HIV epidemic among PWID, with 68% prevalence in 2006. Haiphong has implemented needle/syringe programs, methadone maintenance treatment (MMT), and anti-retroviral treatment (ART), but there is an urgent need to identify high-risk PWID and link them to services. We examined integration of respondent-driven sampling (RDS) and strong peer support groups as a mechanism for identifying high-risk PWID and linking them to services. The peer support staff performed the key tasks that required building and maintaining trust with the participants, including recruiting the RDS seeds, greeting and registering participants at the research site, taking electronic copies of participant fingerprints (to prevent multiple participation i...
Background: Hidden populations at elevated risk for HIV, such as illicit drug users and bisexual ... more Background: Hidden populations at elevated risk for HIV, such as illicit drug users and bisexual men of color, can be hard to reach and are not amenable to random sampling. Novel approaches should be developed and tested for studying such populations. Respondent-driven sampling (RDS) seeks to construct samples from populations that lack comprehensive sampling frames needed for random sampling, while reducing biases associated with chain-referral methods. RDS is based on the recognition that, if allowed to continue through enough waves, the sample attains a stable composition (equilibrium) independent of the initial choice of subjects. Methods: Respondent-driven sampling was used to recruit persons who use hard' drugs, men who had sex with men, and sex partners of both groups for a study of the sexual diffusion of HIV. RDS assumes that those best able to access members of hidden populations are their own peers. RDS motivates peers by offering incentives both for being interviewed...
Uploads
Papers by Douglas Heckathorn