Kruskal Wallis

Home | StatisticsShow all | About | Parameters | Filters | Buttons | Sample output

About

This one-way analysis of variance by ranks is used to decide whether n independent samples are from different populations. It is assumed that the variable in question has essentially a continuous distribution, and at least an ordinal measurement is required. The null hypothesis is tested (i.e. that the n samples come from the same population or from identical populations with respect to averages).

Requires

Variables to be compared, specified as the Y variables.

Parameters

Kruskal Wallis dialog

Variables

Select two or more Y variables (See Selecting multiple variables for details).

Options

Subtitle

Allows you to specify an optional subtitle which will be included in the statistical output.

Filters

Kruskal Wallis dialog

Allows you to select rows for analysis based upon conditions in other variables (See Filtering datasets for details)

Buttons

?

Opens the help system for this dialog.

Dataset…

Sends the output from this analysis to a new OxonStat Dataset. See Dataset output for more details.

Print…

Sends the output from this analysis to a printer.

Cancel

Cancels the analysis and closes the dialog.

Screen…

Displays the output from this analysis in the Results window.

Sample output

OxonStat20:45, 12 Aug 2023

Variables:Weight_1 and Height
Subtitle:<none>
Statistic:Kruskal Wallis
Dataset:OxonStat Demo
Range:Row 1 [Green] to Row 38 [Hall]
Filter:<no filter>
Count:38 rows selected (20 groups of ties)

VariableRank SumMean RankCount
Weight_1762.520.138
Height2163.556.938

Hdfp
52.9591<0.0001
Corrected for ties53.0101<0.0001