Test of Proportions


On occasion statisticans are given data that cannot be analyzed in the quantitative fashion we have become accustomed to - finding the population mean or standard deviation for instance.  For example, imagine being given a barrel of apples and oranges.  It wouldn't make sense to find the standard deviation, or mean for either of those set population quantities.  However, we could find the ratio of apples to oranges in the barrel, and that information would have its own utility.

Tests of Proportions involve a binomial distribuiton, and require familiarity with statistical terms unused in quantitative tests.  The equation for a test of proportions is as follows:

= x/n (the sample population)

p  = population proportion

q  = 1-p

As you can see, a test of proportions involves the use of the Standard distribution. The principle which allows us to use this distribution is that if np 5 and nq 5, then the binomial distribution is approximated by the standard distribution.

Case Example: The Ratio of Males to Females
In Graduate School


  

4:24