Microsoft Office Tutorials and References

In Depth Information

Exhibit 6.7
ANOVA: Single factor dialog box

Table 6.11 shows the result of the analysis. Note that the F-statistic, 8.63465822

(cell L15) is larger than the critical F-value, 3.13591793 (cell N15), so we can reject

the null that all means come from the same population of expected reefer losses.

Also, as before, if the p-value, 0.0004666 (cell M15) is less than our designated

α

(0.05), which is the case, we reject the null hypothesis. Thus, we have rejected the

notion that the average monthly losses at the various ports are similar. At least one

of the averages seems to come from a different distribution of monthly losses and is

not similar to the averages of the other ports. We are not surprised to see this result

given the much lower average at the Port of Singapore
—
about 10.5 versus 18.8 and

18.1 for the Port of NY/NJ and Amsterdam, respectively.

6.7 Experimental Design

There are many possible methods by which we conduct a data collection effort.

Researchers are interested in carefully controlling and designing experimental

studies, not only the analysis of data, but also its collection. The term used for

Search JabSto ::

Custom Search