The present study introduces an unbiased estimator of population mean of a sensitive quantitative variable based on multiple selections of numbers from a scrambling distribution to confound the actual response on sensitive variable with some unrelated variable. The proposed method may be viewed as more protective against the reprisal or stigma. A simulation study is done to observe the performance of the proposed method relative to Ryu et al. (2005) Randomized Response (RR) method. The relative efficiency of proposed estimator with respect to Ryu et al. (2005) RR method is calculated and found appreciable.