Determining Sample Size for Estimating poportions
It is necessary to determining the sample size for a problem involving proportions. This deals with the percentage of frequency of occurrence. The standard deviation is computed considering the proportions of the happening of the particular event. If the frequency of occurrence is known as p and the frequency of non occurrence is known as 1 , the standard deviation of the proportion maybe
p = √p q/n
The samples size is determine as discussed earlier. The precision and population size have been the deciding factors for determining the sample size. The precision can again be expressed absolutely or relatively. Absolute precision involves that the estimate will be within plus or minus of the true value i ,e,± 5 percentage points of the true value. If the sample elements are selected independently and if the sample size is small the correct distribution of the sample proportion will be binomial. It is convenient to employ the normal approximation for estimating the sample size. After the sample is drawn and the sample promotion determined the researcher determines the confidence interval.