Randomized response techniques are useful for promoting respondent cooperation and reducing the inflated response bias experienced in direct response surveys with respect to potentially sensitive issues. But the primary problem is that nobody knows what an issue of survey inquiry should be regarded as sensitive. This study suggests a new technique, which is more useful due to its advance in measuring the sensitivity of survey issues. And it may be utilized to check whether or not a randomized response technique is better in practice. Circumstances in which the proposed technique can be applied is studied and illustrated using a numerical example.
