A comparison of randomized response, computer-assisted self-interview, andface-to-face direct questioning - Eliciting sensitive information in the context of welfare and unemployment benefit

This article assesses the validity of responses to sensitive questions usin g four different methods. In an experimental setting, the authors compared a computer-assisted self-interview (CASI), face-re-face direct questioning, and two different varieties of randomized response. All respondents interv iewed had been identified as having committed welfare and unemployment bene fit fraud. The interviewers did not know that respondents had been caught f or fraud and the respondents did not know that the researchers had this inf ormation. The results are evaluated by comparing the percentage of false ne gatives. The authors also looked for variables that might explain why same respondents admit fraud and others do not. The proportions of respondents a dmitting fraud are relatively low, between 19 percent and 49 percent. The t wo randomized response conditions were superior in eliciting admissions of fraud. A number of background variables, notably gender; age, still receivi ng benefit, and duration and perception of fraud are related to admitting f raud Although the randomized response conditions performed much better than face-to-face direct questioning and CASI, the percentage of respondents ad mitting fraud is only around 50 percent. Some possible reasons for this are discussed.