Figure B.1 shows the required effective sample size on the vertical axis and the proportion estimated, p, on the horizontal axis. The graph shows that the required effective sample sizes are very high for proportions that are estimated to be close to 0 or 1. The required effective sample size decreases from 153 at an estimated proportion of 0.01 to a sample size of 51 for an estimated proportion of 0.17. The required sample size remains at 51 for estimated proportions 0.17 to 0.25 and then increases to a local maximum of 68 for an estimated proportion of 0.5. The required sample size decreases to 51 at an estimated proportion of 0.75. The required effective sample size remains at this level for estimated proportions of 0.75 to 0.83. The required effective sample size increases for proportion estimates larger than 0.83. An effective sample size of 153 will be required when the proportion estimated is 0.99.
Return to Figure B.1