Data Analysis with Microsoft Excel: Updated for Office 2007

Chapter 6 Statistical Inference 247

Figure 6-13

t Test
analysis
of the
labor data

t test statistic 95% confidence

interval

parameter values of the

hypothesis test

sample average pvalue

On the basis of our analysis, there is an average increase in the percentage

of women in the labor force of 3.37 percentage points between 1968 and

1972. This is statistically signifi cant with a p value of 0.024, so we reject
      the null hypothesis and accept the alternative. There has been a signifi cant
      change in women’s participation in the workforce in those four years. The
      95% confi dence interval for this estimate ranges from 0.49 percentage point
      up to 6.25 percentage points. Notice the interval does not include 0 and the
      hypothesis test rejects 0, so the interval and the test agree. It is not hard to
      show that they will always agree.
      An economist, viewing other data on this topic, claims that the per centage
      of women in the workforce had actually increased by 5 points from 1968 to


1973. Do your data confl ict with this statement? Let’s test the hypothesis.
      H 0 : m 5 0.05
      Ha: m 2 0.05
      Rather than rerunning the StatPlus command, we can simply enter the
      new hypothesis directly into the t test worksheet, if we have StatPlus set to
      create dynamic output. Otherwise, we have to rerun the command.

To test the new hypothesis:

1 Click cell D2, change the value from 0 to 0.05, and press Enter.