Mixture Quantitative Optional Randomized Response Technique (RRT) Models

In this work, we introduce a mixture optional RRT model that helps better mitigate lack of trust in quantitative RRT models. We provide the mathematical framework for the model and validate its properties using extensive simulations. We discuss how to use auxiliary information in this mixture model to find more efficient mean estimators. We also discuss how auxiliary information can lead to loss of privacy for the respondents. We also discuss how to account for potential measurement errors associate with this model.