**Confidence Interval for ***p*

For large *n*, a
confidence interval for *p* is given by

**Determining The Sample Size**

For
error margin for the estimation of *p*, we use the
expression
.
The required sample size is obtained by
equating
,
where *d* is the specified error margin.

If the value of *p* is known to be roughly in the neighborhood of a value
*p*^{*}, then *n* can be determined from

Without prior knowledge of *p*, *pq* can be replaced by its maximum possible
value 1/4 and *n* determined from the relation