MannWhitneyTestT Class

public sealed class MannWhitneyTest<T> : TwoSampleTest<T>

Public NotInheritable Class MannWhitneyTest(Of T)
	Inherits TwoSampleTest(Of T)

generic<typename T>
public ref class MannWhitneyTest sealed : public TwoSampleTest<T>

[<SealedAttribute>]
type MannWhitneyTest<'T> =  
    class
        inherit TwoSampleTest<'T>
    end

Type Parameters

T

The element type of the samples.

The MannWhitneyTestT type exposes the following members.

	Name	Description
	MannWhitneyTestT	Constructs a new Mann-Whitney test.
	MannWhitneyTestT(VectorT, ICategoricalVector)	Constructs a new Mann-Whitney test for the specified samples.
	MannWhitneyTestT(VectorT, VectorT)	Constructs a new Mann-Whitney test for the specified samples.

	Name	Description
	Distribution	Gets the probability distribution used in the hypothesis test. (Inherited from HypothesisTest.)
	Exactness	Gets or sets whether an exact test should be performed.
	HypothesisType	Gets or sets whether the test is one or two-tailed. (Inherited from HypothesisTest.)
	Name	Gets the name of the hypothesis test. (Overrides HypothesisTestName.)
	PValue	Gets the probability that the test statistic would take on the calculated value under the alternate hypothesis. (Inherited from HypothesisTest.)
	Sample1	Gets or sets the first sample this test is being applied to. (Inherited from TwoSampleTestT.)
	Sample2	Gets or sets the second sample this test is being applied to. (Inherited from TwoSampleTestT.)
	SignificanceLevel	Gets the significance level used to test the null hypothesis. (Inherited from HypothesisTest.)
	Statistic	Gets the value of the test statistic. (Inherited from HypothesisTest.)

	Name	Description
	Equals	Determines whether the specified object is equal to the current object. (Inherited from Object.)
	GetConfidenceInterval	Returns the confidence interval for the test parameter for the default confidence level. (Inherited from HypothesisTest.)
	GetConfidenceInterval(Double)	Returns the confidence interval for the test parameter for the specified confidence level. (Inherited from HypothesisTest.)
	GetCount	Returns the number of observations in the specified group.
	GetHashCode	Serves as the default hash function. (Inherited from Object.)
	GetLowerCriticalValue	Gets the lower critical value for the hypothesis test's current significance level. (Inherited from HypothesisTest.)
	GetLowerCriticalValue(Double)	Gets the lower critical value for the hypothesis test at the specified significance level. (Inherited from HypothesisTest.)
	GetMeanRank	Returns the mean of the ranks of the observations from the specified group.
	GetPValue	Gets the probability that the test statistic would take on the calculated value under the specified alternate hypothesis. (Inherited from HypothesisTest.)
	GetSumOfRanks	Returns the sum of the ranks of the observations from the specified group.
	GetType	Gets the Type of the current instance. (Inherited from Object.)
	GetUpperCriticalValue	Gets the upper critical value for the test statistic at the hypothesis test's current significance level. (Inherited from HypothesisTest.)
	GetUpperCriticalValue(Double)	Gets the upper critical value for the test statistic at the specified significance level. (Inherited from HypothesisTest.)
	Reject	Returns whether the null hypothesis is rejected using the default significance level. (Inherited from HypothesisTest.)
	Reject(Double)	Returns whether the null hypothesis is rejected using the specified significance level. (Inherited from HypothesisTest.)
	Summarize	Returns a string containing a human-readable summary of the object. (Inherited from HypothesisTest.)
	Summarize(SummaryOptions)	Returns a string containing a human-readable summary of the object using the specified options. (Inherited from HypothesisTest.)
	ToString	Returns a string that represents the current object. (Inherited from HypothesisTest.)

Use the MannWhitneyTestT class to test the hypothesis that two samples are drawn from the same distribution. The test uses the ranks in the combined sample. No assumptions are made about the distribution.

The test is also known as the Mann-Whitney-Wilcoxon test or the Wilcoxon rank-sum test.

For smaller sample sizes, an exact test is used under the assumption that there are no ties. For larger sample sizes (greater than 80), or when ties are present, the sampling distribution of the Mann-Whitney statistic is approximated by a normal distribution with a correction for tied values. This behavior can be overriden through the Exactness property.

Reference