An alternative technique of sensitive questions survey

The paper proposes a simple and obvious randomized response procedure for estimating an unknown proportion of people bearing a sensitive characteristic in a given community. The proposed method does not require direct answers on sensitive topics from the respondents. Instead, each respondent is required to report an outcome of ordered pair according to the colors of the cards drawn from two packs and whether he(or she) has sensitive characteristic or not. The proposed strategy turns out to be more efficient than previous strategies by choosing appropriate design parameters. In addition, this method is easy to carry out and protect the privacy of the respondents to a better degree.