How to ask sensitive questions using statistics: a case study on academic dishonesty
Our project focus was to determine the proportion of students who have cheated on a test at least once in the past year. Out of the students that cheated we were then to determine if a student was more likely to be a freshman, sophomore, junior or senior. The first thing we had to do, was to define cheating. The following is the definition of cheating, which we used for our project: you have cheated if you (1) copied answers from someone else on a test; (2) turned in a paper that you did not write; (3) used an unauthorized “cheat sheet”; or (4) discussed the answers on take home test with somebody else . Once we had decided on what was meant by cheating, we had to determine how to collect the data for our project. Since cheating is a “sensitive” sub- ject, we were afraid some students would be unwilling to respond truthfully. Therefore, we had to use a type of survey that would respect students’ privacy. The survey technique which we chose, allowing students to answer truthfully without having their privacy invaded, was that of the Randomized Response Survey (RRS).