Limits...
Respondent driven sampling: determinants of recruitment and a method to improve point estimation.

McCreesh N, Copas A, Seeley J, Johnston LG, Sonnenberg P, Hayes RJ, Frost SD, White RG - PLoS ONE (2013)

Bottom Line: Consistent with these findings, weighting for non-random presentation for interview by age and socioeconomic status group greatly improved the estimate of the proportion of men in the lowest socioeconomic group, reducing the root-mean-squared error of RDS estimates of socioeconomic status by 38%, but had little effect on estimates for age.The weighting also improved estimates for tribe and religion (reducing root-mean-squared-errors by 19-29%), but had little effect for sexual activity or HIV status.Further evaluation of this new method is required.

View Article: PubMed Central - PubMed

Affiliation: School of Medicine, Pharmacy and Health, Durham University, Durham, United Kingdom.

ABSTRACT

Introduction: Respondent-driven sampling (RDS) is a variant of a link-tracing design intended for generating unbiased estimates of the composition of hidden populations that typically involves giving participants several coupons to recruit their peers into the study. RDS may generate biased estimates if coupons are distributed non-randomly or if potential recruits present for interview non-randomly. We explore if biases detected in an RDS study were due to either of these mechanisms, and propose and apply weights to reduce bias due to non-random presentation for interview.

Methods: Using data from the total population, and the population to whom recruiters offered their coupons, we explored how age and socioeconomic status were associated with being offered a coupon, and, if offered a coupon, with presenting for interview. Population proportions were estimated by weighting by the assumed inverse probabilities of being offered a coupon (as in existing RDS methods), and also of presentation for interview if offered a coupon by age and socioeconomic status group.

Results: Younger men were under-recruited primarily because they were less likely to be offered coupons. The under-recruitment of higher socioeconomic status men was due in part to them being less likely to present for interview. Consistent with these findings, weighting for non-random presentation for interview by age and socioeconomic status group greatly improved the estimate of the proportion of men in the lowest socioeconomic group, reducing the root-mean-squared error of RDS estimates of socioeconomic status by 38%, but had little effect on estimates for age. The weighting also improved estimates for tribe and religion (reducing root-mean-squared-errors by 19-29%), but had little effect for sexual activity or HIV status.

Conclusions: Data collected from recruiters on the characteristics of men to whom they offered coupons may be used to reduce bias in RDS studies. Further evaluation of this new method is required.

Show MeSH

Related in: MedlinePlus

Diagram of the RDS recruitment process.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3814964&req=5

pone-0078402-g001: Diagram of the RDS recruitment process.

Mentions: Respondent driven sampling (RDS)[3] is a variant of a link-tracing design that is designed to generate unbiased estimates of the prevalence of a disease or risk factors, or other characteristics, in a socially networked population. First, a small number of ‘seeds’ are selected by convenience. The seeds are given coupons, usually three, to recruit others from the target population. Recruits are given incentives both for taking part in the survey and for recruiting others. After their interview, the recruits are invited to become recruiters. If they accept, recruiters are asked to give coupons to other individuals in the target population with whom they have a relationship (Figure 1). To be recruited, individuals to whom coupons are offered need to accept, present for interview, be eligible and consent. They themselves can then become recruiters. This process continues in recruitment ‘waves’ until a target sample size is reached and/or until equilibrium is reached. Estimation methods are then applied to account for the non-random sample selection in an attempt to generate unbiased estimates for the composition of the target population. Two main estimation methods have been used, ‘RDS-1’, which accounts for patterns of recruitment between subgroups and the average number of other members of the target group recruiters know (the ‘network size’) in each subgroup[4], [5], and ‘RDS-2’, which, in its simplest formulation, accounts for network size only[6]. More recently, a more computationally intensive estimation method that relaxes some of the assumptions underlying the RDS-2 approach has been proposed[7].


Respondent driven sampling: determinants of recruitment and a method to improve point estimation.

McCreesh N, Copas A, Seeley J, Johnston LG, Sonnenberg P, Hayes RJ, Frost SD, White RG - PLoS ONE (2013)

Diagram of the RDS recruitment process.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3814964&req=5

pone-0078402-g001: Diagram of the RDS recruitment process.
Mentions: Respondent driven sampling (RDS)[3] is a variant of a link-tracing design that is designed to generate unbiased estimates of the prevalence of a disease or risk factors, or other characteristics, in a socially networked population. First, a small number of ‘seeds’ are selected by convenience. The seeds are given coupons, usually three, to recruit others from the target population. Recruits are given incentives both for taking part in the survey and for recruiting others. After their interview, the recruits are invited to become recruiters. If they accept, recruiters are asked to give coupons to other individuals in the target population with whom they have a relationship (Figure 1). To be recruited, individuals to whom coupons are offered need to accept, present for interview, be eligible and consent. They themselves can then become recruiters. This process continues in recruitment ‘waves’ until a target sample size is reached and/or until equilibrium is reached. Estimation methods are then applied to account for the non-random sample selection in an attempt to generate unbiased estimates for the composition of the target population. Two main estimation methods have been used, ‘RDS-1’, which accounts for patterns of recruitment between subgroups and the average number of other members of the target group recruiters know (the ‘network size’) in each subgroup[4], [5], and ‘RDS-2’, which, in its simplest formulation, accounts for network size only[6]. More recently, a more computationally intensive estimation method that relaxes some of the assumptions underlying the RDS-2 approach has been proposed[7].

Bottom Line: Consistent with these findings, weighting for non-random presentation for interview by age and socioeconomic status group greatly improved the estimate of the proportion of men in the lowest socioeconomic group, reducing the root-mean-squared error of RDS estimates of socioeconomic status by 38%, but had little effect on estimates for age.The weighting also improved estimates for tribe and religion (reducing root-mean-squared-errors by 19-29%), but had little effect for sexual activity or HIV status.Further evaluation of this new method is required.

View Article: PubMed Central - PubMed

Affiliation: School of Medicine, Pharmacy and Health, Durham University, Durham, United Kingdom.

ABSTRACT

Introduction: Respondent-driven sampling (RDS) is a variant of a link-tracing design intended for generating unbiased estimates of the composition of hidden populations that typically involves giving participants several coupons to recruit their peers into the study. RDS may generate biased estimates if coupons are distributed non-randomly or if potential recruits present for interview non-randomly. We explore if biases detected in an RDS study were due to either of these mechanisms, and propose and apply weights to reduce bias due to non-random presentation for interview.

Methods: Using data from the total population, and the population to whom recruiters offered their coupons, we explored how age and socioeconomic status were associated with being offered a coupon, and, if offered a coupon, with presenting for interview. Population proportions were estimated by weighting by the assumed inverse probabilities of being offered a coupon (as in existing RDS methods), and also of presentation for interview if offered a coupon by age and socioeconomic status group.

Results: Younger men were under-recruited primarily because they were less likely to be offered coupons. The under-recruitment of higher socioeconomic status men was due in part to them being less likely to present for interview. Consistent with these findings, weighting for non-random presentation for interview by age and socioeconomic status group greatly improved the estimate of the proportion of men in the lowest socioeconomic group, reducing the root-mean-squared error of RDS estimates of socioeconomic status by 38%, but had little effect on estimates for age. The weighting also improved estimates for tribe and religion (reducing root-mean-squared-errors by 19-29%), but had little effect for sexual activity or HIV status.

Conclusions: Data collected from recruiters on the characteristics of men to whom they offered coupons may be used to reduce bias in RDS studies. Further evaluation of this new method is required.

Show MeSH
Related in: MedlinePlus