we must divide the value of alpha in two.
Please note that calculating two separate /single sided bounds/, each with risk
-level [alpha][space]is not the same thing as calculating a two sided interval.
+level [alpha] is not the same thing as calculating a two sided interval.
Had we calculate two single-sided intervals each with a risk
that the true value is outside the interval of [alpha], then:
quite close to the experimentally calculated value of 0.2. Furthermore
the difference between the two calculation methods is now really quite small.
-[endsect]
+[endsect] [/section:binom_conf Calculating Confidence Limits on the Frequency of Occurrence for a Binomial Distribution]
[section:binom_size_eg Estimating Sample Sizes for a Binomial Distribution.]
[endsect] [/section:binom_size_eg Estimating Sample Sizes for a Binomial Distribution.]
-[endsect][/section:binom_eg Binomial Distribution]
+[endsect] [/section:binom_eg Binomial Distribution]
[/
Copyright 2006 John Maddock and Paul A. Bristow.