Two Proportions

Test for Two Proportions - Fisher's Exact Test

  

When you specify a test difference of zero – in other words, when the null hypothesis states that the two proportions are equal, Minitab performs Fisher's exact test in addition to a normal approximation test. This test is valid for all samples, regardless of the number of events or trials in each sample.

Example Output

Sample   X    N  Sample p 

1      725  802  0.903990

2      573  712  0.804775

 

Difference = p (1) - p (2)

Estimate for difference: 0.0992147

95% CI for difference: (0.0636706, 0.134759)

Test for difference = 0 (vs ≠ 0):  Z = 5.47  P-Value = 0.000

 

Fisher's exact test: P-Value = 0.000

Interpretation

For the student employment data, Fisher's exact test calculated a p-value of 0.000. Minitab rounds p-values to three decimal points, so you conclude that the actual p-value is less than 0.0005. This p-value is lower than common a-levels, so you can reject the null hypothesis that the proportions of male and female students with summer employment are equal.

 

Related Charting Tools

Free online chart generators for statistical analysis:

Useful tools: