ABSTRACT
Respondent-driven sampling (RDS) is a method for sampling from a target population by leveraging social connections. RDS is invaluable to the study of hard-to-reach populations. However, RDS is costly and can be infeasible. RDS is infeasible when RDS point estimators have small effective sample sizes (large design effects) or when RDS interval estimators have large confidence intervals relative to estimates obtained in previous studies or poor coverage. As a result, researchers need tools to assess whether or not estimation of certain characteristics of interest for specific populations is feasible in advance. In this paper, we develop a simulation-based framework for using pilot data-in the form of a convenience sample of aggregated, egocentric data and estimates of subpopulation sizes within the target population-to assess whether or not RDS is feasible for estimating characteristics of a target population. in doing so, we assume that more is known about egos than alters in the pilot data, which is often the case with aggregated, egocentric data in practice. We build on existing methods for estimating the structure of social networks from aggregated, egocentric sample data and estimates of subpopulation sizes within the target population. We apply this framework to assess the feasibility of estimating the proportion male, proportion bisexual, proportion depressed and proportion infected with HIV/AIDS within three spatially distinct target populations of older lesbian, gay and bisexual adults using pilot data from the caring and Aging with Pride Study and the Gallup Daily Tracking Survey. We conclude that using an RDS sample of 300 subjects is infeasible for estimating the proportion male, but feasible for estimating the proportion bisexual, proportion depressed and proportion infected with HIV/AIDS in all three target populations.
ABSTRACT
The current bioassay development literature lacks the use of statistically robust methods for calculating the limit of detection of a given assay. Instead, researchers often employ simple methods that provide a rough estimate of the limit of detection, often without a measure of the confidence in the estimate. This scarcity of robust methods is likely due to a realistic preference for simple and accessible methods and to a lack of such methods that have reduced the concepts of limit of detection theory to practice for the specific application of bioassays. Here, we have developed a method for determining limits of detection for bioassays that is statistically robust and reduced to practice in a clear and accessible manner geared at researchers, not statisticians. This method utilizes a four-parameter logistic curve fit to translate signal intensity to analyte concentration, which is a curve that is commonly employed in quantitative bioassays. This method generates a 95% confidence interval of the limit of detection estimate to provide a measure of uncertainty and a means by which to compare the analytical sensitivities of different assays statistically. We have demonstrated this method using real data from the development of a paper-based influenza assay in our laboratory to illustrate the steps and features of the method. Using this method, assay developers can calculate statistically valid limits of detection and compare these values for different assays to determine when a change to the assay design results in a statistically significant improvement in analytical sensitivity.
