Multi-stage Optional Unrelated Question RRT Model

oleh: Anu Chhabra, B.K. Dass, Samridhi Mehta

Format: Article
Diterbitkan: Springer 2016-02-01

Deskripsi

Sihm et al. (2014) introduced modified optional unrelated question RRT model in both binary and quantitative response situations wherein the prevalence of the sensitive variable and the sensitivity level of the underlying sensitive question could be estimated simultaneously without using a split sample approach. In this study, we propose a three-stage optional unrelated question RRT model for both binary and quantitative response situations which combines the essence of Sihm et al. (2014) model and the three-stage optional additive RRT model proposed by Mehta et al. (2012). The efficiencies of Sihm et al. (2014) model and proposed three-stage optional unrelated question RRT model are compared using simulations. The privacy measures of the two models in question are also discussed. Comparisons for the binary models are based on Lanke (1976) measure while Yan et al. (2009) measure is used to compare the privacy measures for the quantitative models.