This paper proposes an alternative randomized response technique by improving existing works on tripartite randomized response technique (TRRT) using unrelated questions. To improve the technique, this work studies TRRT under sampling with unequal probabilities with or without replacement. The study of percentage relative efficiency shows that the proposed estimator is far better than the Hybrid Tripartite Randomized Response Technique with increase in the sensitive character and that of the unrelated character. In addition, conducting survey using both the proposed technique and direct (traditional) technique shows that the proposed technique is more efficiency than the direct technique of data collection.