Dunn’s Test
This article covers meaning & overview of Dunn’s Test from statistical perspective.
What is Dunn’s Test?
Dunn’s test is a statistical test used to do a specific number of comparisons between groups of data and find out which one of them is significant. This was developed by Olive Dunn in 1965 who provided the tables. The test is also commonly known as Bonferroni t.
When dealing with large number of data groups, it is desirable to find out which set of groups have significant difference. This is easier because it is tedious to make all the possible comparisons. To so the comparisons between a specific set of groups, Dunn’s test is useful. It is used as a post-hoc test after Analysis of Variance which concludes that there is a significant difference between the groups.
In this test, the significance level for a given number of comparisons is calculated as Overall significance level (α)/ number of comparisons (c). If the overall significance level is 5% and there are 5 comparisons to be made then the new α = 0.05/5 = 0.01
The usual Student’s t-tables does not provide the critical values of t for values of α between 0.01 and 0.05. In this case, the Dunn’s t tables are useful in finding out the critical value. But these days, the test is usually done on software packages which directly yield the results and therefore hardly the tables are used.
Hence, this concludes the definition of Dunn’s Test along with its overview.
Browse the definition and meaning of more similar terms. The Management Dictionary covers over 1800 business concepts from 5 categories.
Continue Reading:
