Refined Randomized Response Model for Suspicious Answers: Illicit Drug Users in U.S.A. are illustrated

Abstract

Monte In half a century ago, an ingenious method was proposed by Warner’s (1965) to improve the truthful responses with anonymity in a sample survey on sensitive matter. This method is known as randomized response technique (RRT). Still, the respondents are suspicious in spite of the anonymity and the Warner’s RRT model is insufficient. Therefore, this article amends the RRT model just to quantify and resolve the issues due to the respondent’s suspicion in RRT. The likelihood ratio based hypothesis testing procedure to check the statistical significance of the estimated suspicion level is constructed in this article. The refined RRT model of this article is illustrated using the data on the number of illegal drug users in the United States of America.Carlo simulation method was used to study the effects of the data structure on the quality of the predictions in linear multiple regression. Five hundred forty (540) data files were generated of which the number of variables, R-square, the collinearity between the explanatory variables and the index of coefficient, that measures the importance of the explanatory variables in the model, were controlled. Predictions were influenced by the theoretical value of R-square, the method used to establish the model and, to a lesser extent, the collinearity between the explanatory variables. The determination of the minimal sample size which leads to predicted values better than those obtained by the mean of the dependent variable indicated that this size depends on the number of the explanatory variables, the theoretical value of the R-square and the method used to establish the model.