# HL Mann-Whitney U test

## Recommended Posts

I am really confused on what the Mann whitney u test is trying to find, and what kind of answer i need to look for, ive looked at heaps of sites and I still don't really get it and cannot find any sample psych IA's using it so im not sure what to include

im pretty sure i have to use this test as my data is the number of correct words for each mnemonic

I keep getting different answer for U

my R1=147.5 and R2=62.5 and n1 and n2 are =10 crit value for U is= 23

I keep getting two different answers for the equations, either my order of operations is wrong or my calculator is U1= 10 x 10 +10 (10+1)/2 - 147.5

= 7.5 or 52.5 (i know completely different!) (which one?)

U2= 10 x 10 +10(10+1)- 62.5

= 92.7 or 137.5 (which one? ) thank you!

##### Share on other sites

I am really confused on what the Mann whitney u test is trying to find, and what kind of answer i need to look for, ive looked at heaps of sites and I still don't really get it and cannot find any sample psych IA's using it so im not sure what to include

im pretty sure i have to use this test as my data is the number of correct words for each mnemonic

You can verify your calculations here, doing stats by hand is a waste of time in my opinion as long as you understand what the test is doing (comparing medians) so just have software do the calculations for you.

You will need to show us exactly how you input the equation into your calculator for both sets of answers that you get if you want to know what's going on with your order of operations.

Here's what I got

U1=(10x10)+(10x(10+1))/2 - 147.5

100+110/2 - 147.5

100+55 - 147.5

155-147.5=7.5

U2= (10x10)+(10x(10+1))/2 - 62.5

100+110/2 - 62.5

100+55 - 62.5

155-62.5=92.5

According to the significance table I'm using the critical value is 19 for p<.02 for a two-tailed test. The smaller U value needs to be smaller than or equal to the critical value so your results are significant since 7.5 is less than 19 .

Definitely double check by using that online resource that does the test for you.

You can also use a t-test instead since your data are interval/ratio. Mann Whitney is traditionally used for ordinal data but it is also recommended for interval/ratio data when the sample size is small (10 qualifies as small).

## Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account. ×   Pasted as rich text.   Paste as plain text instead

Only 75 emoji are allowed.