A heterogeneity based iterative clustering approach for obtaining samples with reduced bias

Medical and social sciences demand sampling techniques which are robust, reliable, replicable and give samples with the least bias. Majority of the applications of sampling use randomized sampling, albeit with stratification where applicable to lower the bias. The randomized technique is not consistent, and may provide different samples each time, and the different samples themselves may not be similar to each other. In this paper, we introduce a novel sampling technique called Wobbly Center Algorithm, which relies on iterative clustering based on maximizing heterogeneity to achieve samples which are consistent, and with low bias. The algorithm works on the principle of iteratively building clusters by finding the points with the maximal distance from the cluster center. The algorithm consistently gives a better result in lowering the bias by reducing the standard deviations in the means of each feature in a scaled data.