Running a Mood test to compare medians with XLSTAT
An Excel sheet with both the data and the results can be downloaded by clicking here.
The data are from a study where four different methods of growing corn are analyzed. This aims at assessing the yield per acre. These data have already being used in the book of Conover [Conover W.J. (1999). Practical Nonparametric Statistics, 3rd edition,Wiley].
Our goal is to test if the medians are the same whatever the method of growing.
Setting up a Mood test on k independant samples
After opening XLSTAT, select the XLSTAT / Non parametric tests / Comparison of 2 or more medians (Mood test) command, or click on the corresponding button of the Non parametric tests toolbar.
Once you've clicked on the button, the dialog box appears.
The format of the dataset is such that you have one variable per sample. The option One column per variable can be used when data are in one column and the labels in another one.
In the Options tab, three approaches are proposed to calculate/estimate the p-value. The significance level α can also be modified. Here, we leave the default options, i.e. a significance level α=5% and the asymptotic approach to estimate the p-value.
The computations begin once you have clicked on OK. The results will then be displayed in a new sheet.
Interpreting the results of a Mood test
The first results displayed are the statistics for the various samples. Next, the results of the Mood test and the multiple pairwise comparisons are displayed.
Hereafter are the Mood test results for the available dataset. Since the p-value is under the significance level (5%), the null hypothesis can be rejected.
Other options such as multiple comparisons using Mood test are also available in XLSTAT.