| Abstract: | Used Monte Carlo techniques to generate sampling distributions of the correlation coefficient between 2 variables, each measured by 2 quantitative models of the randomized response technique (RRT)—the additive constants model and the unrelated question model. As expected, the RRT models yielded less statistically efficient estimates of the population correlation coefficient, and inefficiency increased as the population correlation increased. Comparisons of efficiency between the 2 models favored the additive constants model for constants less than certain limiting values. (13 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved) |
