Calculates the probability of an observed (sample) proportion (in the range 0-1) against a hypothetical proportion. No input data are required in the spreadsheet. The 95% confidence interval for the proportion is calculated using two different methods. The normal approximation CI is more commonly used, but the exact CI (Clopper-Pearson) is more accurate especially for small N. For large N the two methods will give similar results.
The (two-tailed) test against a given proportion (typically 0.5) uses the normal approximation.
Example use of results for publication:
The sample proportion of 0.45 (N=100) is not significantly different from 0.5 (proportion test with normal approximation, p=0.32)
A 95% Clopper-Pearson confidence interval for the proportion is (0.35, 0.55).