ABSTRACT
Suppose a health scientist is interested in estimating the incidence rate of a particular disease in a large city. One needs to specify the margin of error attached to statement, d, and the chance (1 − α) for the validity of the statement. If the scientist estimates the proportion as pˆ, the real proportion is then in the interval (pˆ − d, pˆ + d) and in repeated sampling; the statement is true in (1 − α) 100% samples. Usually, d = 0.03, 0.05, or 0.07 and α = 0.1, 0.05, or 0.01. Suppose, the scientist considers d = 0.03 and α = 0.05, then she needs to have an idea of this proportion and assumes this to the best to her knowledge and experience as p . When this information is not available, the scientist considers p = 0.5, realizing that the sample size would be larger than required. The number of respondents for the study needs to be determined and the following program will give the necessary sample size, n.
