Randomized responses: A better way to obtain sensitive information

Companies have an ongoing need to obtain employee information on sensitive issues, such as drug and alcohol use and sexual harassment. A relatively unknown approach to obtaining such sensitive information is the randomized response technique (RRT). RRTs are various methods that use probabilities in some way to completely conceal the true response of the person answering a given question. The procedures guarantee that a true answer can never be matched with a particular individual. Some of the more commonly used designs include: 1. Warner's design, 2. the unrelated-question design, 3. Morton's design, 4. the contamination design, and 5. the responder-set probability design. RRT is judged to be superior to direct questioning because of the generally higher percentage of "yes" answers to the sensitive question. In applying RRT in the corporate environment, the decision maker must be aware of several considerations, including: 1. research methodology, 2. sampling, 3. instrumentation, and 4. administration.