Mann-Whitney U Test: Theory and Tutorial in Excel

The video tutorial explains the differences between parametric and nonparametric tests, focusing on when to use nonparametric tests like the Mann Whitney U test. It provides a detailed guide on performing the Mann Whitney U test using Excel, including calculating ranks and U statistics, and interpreting the results to determine statistical significance.

What is a key assumption of parametric tests?

• Data must be binary.

• Data must be nominal.

• Data must be ordinal.

• Data must be normally distributed.

When should a nonparametric test be used?

• When data is binary.

• When data is nominal.

• When data is not normally distributed.

• When data is normally distributed.

Which type of data can nonparametric tests handle that parametric tests cannot?

• Interval data

• Nominal data

• Ordinal data

• Binary data

What is the nonparametric equivalent of the paired T-test?

• Mann Whitney U test

• Chi-square test

• Wilcoxon rank sum test

• Kruskal-Wallis test

What is the Mann Whitney U test equivalent to?

• ANOVA

• Chi-square test

• Independent T-test

• Paired T-test

In the Mann Whitney U test, what is the significance of the U statistic?

• It helps determine statistical significance.

• It measures correlation.

• It indicates the mean difference.

• It determines the sample size.

What is the first step in preparing data for the Mann Whitney U test?

• Calculate the mean.

• Rank the data.

• Determine the median.

• Combine data into one group.

How do you determine if the difference between two groups is significant in the Mann Whitney U test?

• Calculate the mean difference.

• Compare the U statistic to the critical value.

• Perform a regression analysis.

• Use a correlation coefficient.

What happens if the U statistic is less than the critical value?

• Fail to reject the null hypothesis.

• Accept the null hypothesis.

• Reject the null hypothesis.

• Perform a different test.

What is the main challenge of calculating the Mann Whitney U test by hand?

• It is only applicable to nominal data.

• It needs special software.

• It is error-prone and tedious.

• It requires a large sample size.