Abstract
In this article, we propose randomized response-group testing (RR-GT) models. Randomized response technique and the group testing method have long been recognized as sampling schemes that can provide substantial benefits. In order to obtain both confidentiality and economic benefits from a sensitive issue survey sampling, we incorporate the group testing method into the most popular two randomized response models, which are Warner's RR model and unrelated question RR model. Empirical comparisons are provided and discussed.
Original language | English (US) |
---|---|
Pages (from-to) | 33-48 |
Number of pages | 16 |
Journal | Journal of Statistical Theory and Practice |
Volume | 7 |
Issue number | 1 |
DOIs | |
State | Published - Jan 1 2013 |
Keywords
- Randomized response
- confidentiality
- group testing
- maximum likelihood
- sensitive questions