UNISTAT - the ultimate Excel statistics add-in

Kruskal-Wallis One-Way ANOVA in Excel with Unistat

The Unistat statistics add-in extends Excel with Kruskal-Wallis One-Way ANOVA capabilities.

Further documentation is available at section 6.5.1 of the manual, Nonparametric Tests > Kruskal-Wallis One-Way ANOVA.

Sample output

Here we provide annotated example report output from the Unistat Excel statistics add-in for data analysis.

Kruskal-Wallis One-Way ANOVA

CasesRank SumMean Rank
North 12 270.5000 22.5417
East 12 197.0000 16.4167
South 12 421.5000 35.1250
West 12 287.0000 23.9167
Total 48 1176.0000 24.5000

Correction for Ties 0.0030
Chi-Square Statistic 11.2012
Degrees of Freedom 3
Right-Tail Probability 0.0107

Multiple Comparisons with Rank Sums (Tukey-HSD)

Method: 95% Tukey-HSD interval.
** denotes significantly different pairs. Vertical bars show homogeneous subsets.
A pairwise test result is significant if its q stat value is greater than the table q.

GroupCasesRank SumEastNorthWestSouth
East 12 197.0000**
|
North 12 270.5000
||
West 12 287.0000
||
South 12 421.5000**
 |

ComparisonDifferenceStandard Errorq StatTable qSignificanceLower 95%Upper 95%Result
South - East 224.5000 48.4974 4.6291 3.6332 0.0059 48.3011 400.6989**
West - East 90.0000 48.4974 1.8558 3.6332 0.5551-86.1989 266.1989
North - East 73.5000 48.4974 1.5155 3.6332 0.7069-102.6989 249.6989
South - North 151.0000 48.4974 3.1136 3.6332 0.1227-25.1989 327.1989
West - North 16.5000 48.4974 0.3402 3.6332 0.9951-159.6989 192.6989
South - West 134.5000 48.4974 2.7733 3.6332 0.2029-41.6989 310.6989

Homogeneous Subsets:
Group 1: East North West
Group 2: North West South

Multiple Comparisons with Mean Ranks (Tukey-HSD)

Method: 95% Tukey-HSD interval.
** denotes significantly different pairs. Vertical bars show homogeneous subsets.
A pairwise test result is significant if its q stat value is greater than the table q.

GroupCasesMean RankEastNorthWestSouth
East 12 16.4167**
|
North 12 22.5417
||
West 12 23.9167
||
South 12 35.1250**
 |

ComparisonDifferenceStandard Errorq StatTable qSignificanceLower 95%Upper 95%Result
South - East 18.7083 4.0415 4.6291 3.6332 0.0059 4.0251 33.3916**
West - East 7.5000 4.0415 1.8558 3.6332 0.5551-7.1832 22.1832
North - East 6.1250 4.0415 1.5155 3.6332 0.7069-8.5582 20.8082
South - North 12.5833 4.0415 3.1136 3.6332 0.1227-2.0999 27.2666
West - North 1.3750 4.0415 0.3402 3.6332 0.9951-13.3082 16.0582
South - West 11.2083 4.0415 2.7733 3.6332 0.2029-3.4749 25.8916

Homogeneous Subsets:
Group 1: East North West
Group 2: North West South

Multiple Comparisons with Rank Sums (S-N-K)

Method: 95% Student-Newman-Keuls interval.
** denotes significantly different pairs. Vertical bars show homogeneous subsets.
A pairwise test result is significant if its q stat value is greater than the table q.

GroupCasesRank SumEastNorthWestSouth
East 12 197.0000**
|
North 12 270.5000
||
West 12 287.0000**
||
South 12 421.5000****
 |

ComparisonDifferenceStandard Errorq StatTable qSignificanceLower 95%Upper 95%Result
South - East 224.5000 48.4974 4.6291 3.6332 0.0059****
West - East 90.0000 48.4974 1.8558 3.3145 0.3883**
North - East 73.5000 48.4974 1.5155 2.7718 0.2839**
South - North 151.0000 48.4974 3.1136 3.3145 0.0709**
West - North 16.5000 48.4974 0.3402 2.7718 0.8099**
South - West 134.5000 48.4974 2.7733 2.7718 0.0499****

Homogeneous Subsets:
Group 1: East North West
Group 2: North West South

Multiple Comparisons with Mean Ranks (S-N-K)

Method: 95% Student-Newman-Keuls interval.
** denotes significantly different pairs. Vertical bars show homogeneous subsets.
A pairwise test result is significant if its q stat value is greater than the table q.

GroupCasesMean RankEastNorthWestSouth
East 12 16.4167**
|
North 12 22.5417
||
West 12 23.9167**
||
South 12 35.1250****
 |

ComparisonDifferenceStandard Errorq StatTable qSignificanceLower 95%Upper 95%Result
South - East 18.7083 4.0415 4.6291 3.6332 0.0059****
West - East 7.5000 4.0415 1.8558 3.3145 0.3883**
North - East 6.1250 4.0415 1.5155 2.7718 0.2839**
South - North 12.5833 4.0415 3.1136 3.3145 0.0709**
West - North 1.3750 4.0415 0.3402 2.7718 0.8099**
South - West 11.2083 4.0415 2.7733 2.7718 0.0499****

Homogeneous Subsets:
Group 1: East North West
Group 2: North West South

Multiple Comparisons with t Distribution

Method: 95% t interval.
** denotes significantly different pairs. Vertical bars show homogeneous subsets.
A pairwise test result is significant if its q stat value is greater than the table q.

GroupCasesMeanEastNorthWestSouth
East 12 16.4167**
|
North 12 22.5417**
|
West 12 23.9167**
|
South 12 35.1250******
 |

ComparisonDifferenceStandard Errorq StatTable qSignificanceLower 95%Upper 95%Result
South - East 18.7083 5.1477 3.6343 2.0154 0.0007 8.3339 29.0828**
West - East 7.5000 5.1477 1.4570 2.0154 0.1522-2.8745 17.8745
North - East 6.1250 5.1477 1.1899 2.0154 0.2405-4.2495 16.4995
South - North 12.5833 5.1477 2.4445 2.0154 0.0186 2.2089 22.9578**
West - North 1.3750 5.1477 0.2671 2.0154 0.7906-8.9995 11.7495
South - West 11.2083 5.1477 2.1774 2.0154 0.0349 0.8339 21.5828**

Homogeneous Subsets:
Group 1: East North West
Group 2: South

Multiple Comparisons (Dunn)

Method: 95% Dunn interval.
** denotes significantly different pairs. Vertical bars show homogeneous subsets.
A pairwise test result is significant if its q stat value is greater than the table q.

GroupCasesMean RankEastNorthWestSouth
East 12 16.4167**
|
North 12 22.5417
||
West 12 23.9167
||
South 12 35.1250**
 |

ComparisonDifferenceStandard Errorq StatTable qSignificanceLower 95%Upper 95%Result
South - East 18.7083 5.7069 3.2782 2.6383 0.0063 3.6520 33.7647**
West - East 7.5000 5.7069 1.3142 2.6383 1.0000-7.5564 22.5564
North - East 6.1250 5.7069 1.0733 2.6383 1.0000-8.9314 21.1814
South - North 12.5833 5.7069 2.2049 2.6383 0.1648-2.4730 27.6397
West - North 1.3750 5.7069 0.2409 2.6383 1.0000-13.6814 16.4314
South - West 11.2083 5.7069 1.9640 2.6383 0.2972-3.8480 26.2647

Homogeneous Subsets:
Group 1: East North West
Group 2: North West South

Comparisons against a Control Group (Dunnett)

Method: 95% Dunnett interval.
Control Group: North, Two-Tailed Test
** denotes significantly different pairs. Vertical bars show homogeneous subsets.
A pairwise test result is significant if its q stat value is greater than the table q.

GroupCasesRank SumNorth
East 12 197.0000
|
North 12 270.5000
||
West 12 287.0000
 |
South 12 421.5000
 |

ComparisonDifferenceStandard Errorq StatTable qSignificanceLower 95%Upper 95%Result
South - North 151.0000 68.5857 2.2016 2.3491 0.0722-10.1137 312.1137
West - North 16.5000 68.5857 0.2406 2.3491 0.9904-144.6137 177.6137
East - North-73.5000 68.5857 1.0717 2.3491 0.5739-234.6137 87.6137

Homogeneous Subsets:
Group 1: East North
Group 2: North West South

Comparisons against a Control Group (Dunn)

Method: 95% Dunn interval.
Control Group: North, Two-Tailed Test
** denotes significantly different pairs. Vertical bars show homogeneous subsets.
A pairwise test result is significant if its q stat value is greater than the table q.

GroupCasesMean RankNorth
East 12 16.4167
|
North 12 22.5417
||
West 12 23.9167
 |
South 12 35.1250
 |

ComparisonDifferenceStandard Errorq StatTable qSignificanceLower 95%Upper 95%Result
South - North 12.5833 5.7069 2.2049 2.3940 0.0824-1.0790 26.2456
West - North 1.3750 5.7069 0.2409 2.3940 1.0000-12.2873 15.0373
East - North-6.1250 5.7069 1.0733 2.3940 0.8495-19.7873 7.5373

Homogeneous Subsets:
Group 1: East North
Group 2: North West South