Suppose that you can use a scale to measure the weight of a ball, or the total weight of two balls, or the difference in weight between two balls. Each measurement has no bias but is subject to a standard deviation of . Consider two schemes in estimating the weights of ball A and ball B as follows. Scheme 1: take one measurement for each ball, obtaining the weights as WA and WB, respectively. Scheme 2: take one measurement for the total weight of the two balls to get W(A+B), and then one measurement for the difference in weight between the two balls to get W(AB). Which scheme would you prefer in estimating the weights of the two individual balls? Explore