Application of a randomized response technique for the sensitive question survey in female sex workers in Xichang, China

BACKGROUND The technique of sampling is a particularly important aspect surveys, and the method determining sample size key link in technique. However, determination for complex surveys sensitive issues using randomized response model not yet available. METHODS In this work, we drew from theory probability mathematical statistics, developed series formulas that are required to estimate overall proportion mean on dichotomous questions 3-stage Simmons models. RESULTS Shortly after generating formulas, successfully applied them pre-survey regarding among female sex workers (FSWs) Xichang, found their practical application achieve satisfactory effects. By large number Monte Carlo simulation tests, reliability survey methods question were be high, but (RRT) cannot completely eliminate possibility social acceptability bias. CONCLUSIONS RRT an acceptable within populations, new liable used populations. We believe results obtained field public health can provide some data support clinical practice.