Randomized response model
Randomized response model is a statistical technique used to gather sensitive information from survey participants without compromising their privacy. It was first introduced by S.L. Warner in 1965 as a way to estimate the prevalence of sensitive behaviors, such as illegal drug use or tax evasion, in a population.
The basic idea behind the randomized response model is to ask participants a randomized question that may or may not be related to the sensitive behavior of interest. By adding randomness to the question, participants can respond truthfully without revealing their true answer. This helps to reduce response bias and increase the accuracy of the survey results.
For example, instead of asking “Have you ever used illegal drugs?” directly, a researcher might ask a question like “Have you ever flipped a coin and answered ‘yes’ regardless of the outcome?” Participants who answer “yes” to this question are then asked if they have used illegal drugs. By analyzing the responses to both questions, researchers can estimate the true prevalence of drug use in the population.
Overall, randomized response model is a useful tool for collecting sensitive information in surveys while maintaining the privacy and anonymity of participants.
References: