Scribd
Upload
1K views

Mann Whitney U Test

The document provides instructions for conducting and interpreting a Mann-Whitney U test, a non-parametric test used to compare two independent groups when the assumptions for a t-test are violated. It explains that the Mann-Whitney U test compares the mean ranks of scores between two groups to determine if there is a statistically significant difference. An example is given testing for differences in algebra scores between male and female students. The output shows the mean ranks for each group, the U value, Z value, and reports that the female students' scores were significantly higher than the males.

Uploaded by

kinhai_see
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
1K views

Mann Whitney U Test

The document provides instructions for conducting and interpreting a Mann-Whitney U test, a non-parametric test used to compare two independent groups when the assumptions for a t-test are violated. It explains that the Mann-Whitney U test compares the mean ranks of scores between two groups to determine if there is a statistically significant difference. An example is given testing for differences in algebra scores between male and female students. The output shows the mean ranks for each group, the U value, Z value, and reports that the female students' scores were significantly higher than the males.

Uploaded by

kinhai_see
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
You are on page 1/ 4

1

Non-Parametric Test (DR SEE KIN HAI) Mann-Whitney U Test (Mann-Whitney Rank-Sum Test )
(For two independent samples- You use this test if the data violated the assumptions for using the independent sample t - test) Before you can perform this test, the data must fulfill the assumptions below (although not so stringent as parametric technique). 1. Data randomly collected from the population. 2. Variance for the data distribution is equal. 3. Independence of samples. 4. Test H o : there is no significant difference between the 2 independent samples.

Example:

A test on Algebra was given to 30 students after a lesson in this topic of You are to

Mathematics. Student gender was recorded as 1 = male and 2 = female. has violated the assumption of the use of independent sample t-test)

determine if there are gender differences in their scores for Algebra. (Assuming that the data

2 How to run Mann-Whitney U Test 1. Select [Analyze] then [Nonparametric Tests] then [Legacy Dialogs] and [2 Independent Samples] to open the dialogue box below.

2. Move [Algebra] into [Test Variable List:] box and [Gender] into [Grouping Variable] box and select [Define Groups..] Type 1 for male, and 2 for female, then [Continue] and [OK] The groups are defined as 1, 2 or male and female

3 Interpreting the output for the Mann-Whitney U-test The Mean Rank given to [Algebra] for the first group (1=male) is 8.07 and the average rank given to the second group (2=female) is 22.93. This means that the mean scores for female is higher than the male

The basic Mann-Whitney statistic, the Uvalue is 1.000, which is statistically significant at the 0.000 level.

The Z value of - 4.658 is significant at 0.000 level. This is the value of MannWhitney test when a correction for tied ranks has been applied. This has not altered the significance level from 0.000b to 0.000.

Reporting the output for the Mann-Whitney U-test The Mann-Whitney U-test found that the Algebra test scores for female is significantly higher than the male (U = 1.000, N1 =15, N 2 =15, two-tailed p = 0.000). COURSEWORK An English language teacher wanted to know whether there is any significant difference in male and female students English Test. She has administered an English language test to a group of 60 students. 10 male and 10 female students were randomly selected from the 60 students. (Assume that the data has violated the assumption for using independent sample

t - test).

You might also like