Question 849:

Asked on March 5, 2010

Tags: F-test

1

Answer:

No answer provided yet.We want to compare two standard deviations here. To do so we'll actually compare the variances, which are just the square of the standard deviation using the F-test or F-ratio. The Null Hypothesis for this test is that there is no difference in the variances. The alternative hypothesis is that the variance is higher for oceanfront homes.

The two variances are 2,079,360,00/454,968,900 to get a ratio of 4.57. We lookup the significance of this ratio using an F-table with 20 and 17 degrees of freedom, or use the excel function =FDIST(4.57,20,17). We get a 1-tailed p-value of .0013006--doubling this gets us a 2-tailed p-value of .0026. Since both 1-tailed and 2-tailed p-values are less than the alpha significance level of .01 we reject the Null Hypothesis and conclude that the variance of the selling prices of the oceanfront homes is higher than the inland homes.

If you want to compare the two standard deviations in MegaStat you select the Hypothesis Tests > Compare Two Independent Groups and select the box that says "Test for equality of Variances."