ABSTRACT
Viewing the world of statistics through the medium of confidence intervals, in this chapter, the author sees that a smaller error level just generates a larger critical value and a wider interval. He discusses the relationship between intervals and tests in more detail. Notice that any interval can be employed to perform a significance test. The fundamental assumption of the tests can be stated in simple terms as follows: An observed dataset is a random sample drawn from a much larger population. The central principle of Binomial tests is quite difficult to grasp on first reading. Intervals and tests have a straightforward relationship. Test calculations can be simpler, but intervals are more flexible. The author needs an interval to plot upper and lower bounds on data. The Binomial formula may be converted into an ‘exact’ Binomial interval using a computer search procedure. This interval is known as the ‘Clopper-Pearson’ interval.
