248 Fundamentals of Statistics
The p value changes from 0.024 to 0.249. A p value of 0.249 is not small
enough to reject the null hypothesis. We conclude that our data do not confl ict
with the claim made by this economist in a signifi cant way.
You can also change other values in the hypothesis test. You can switch
to a one-sided test by changing the value of cell D3 to either 2 1 or 1. You
can also change the size of the confi dence interval by changing the value in
cell D4.EXCEL TIPS
You can also perform a paired t test of your data using the Anal-
ysis ToolPak, supplied by Excel. To perform a paired t test, load
the Analysis ToolPak and click the Data Analysis button from
the Analysis group on the Data tab. In the Data Analysis dialog
box, click t Test: Paired Two Sample for Means and specify the
two columns containing the paired data. This command does not
calculate the confi dence interval for you, so you have to calculate
that using the formulas supplied in this chapter.You should not accept your analysis at face value without further inves-
tigating the assumptions of the t test. One of these assumptions is that the
data follow a normal distribution. The t test is robust, but that doesn’t mean
you shouldn’t explore the possibility that the data are seriously nonnormal.
Do this now by creating a histogram and normal probability plot of the data
in the Diff column.To create a histogram of the difference data:1 Click Single Variable Charts from the StatPlus menu and then click
Histograms.
2 Click the Data Values button and select Diff from the list of range
names in the workbook.
3 Click the Normal Curve checkbox to add a normal curve to the
histogram.
4 Click the Output button and save your histogram to the Histogram
chart sheet.
5 Click the OK button twice to create the histogram. See Figure 6-14.