# Step 8

# Data Analysis 3

# Non- Parametric Inferential Test

# You can only use a non-parametric inferential test if you have ordinal or nominal data!!

# â€‹

# You have three choices of parametric inferential tests: A Wilcoxon Signed Rank Test, A Mann Whitney U Test, or A Spearman's Rank-Order Correlation Test

## WILCOXON SIGNED RANK TEST

You use a Wilcoxon Signed Rank Test when you want to prove or disprove there is a difference between two related scores. By related scores it could mean comparing the same subjects' scores before and after an intervention. For example, is there a significant difference between subjects' effort rating (1-10) before taking an energy drink and after taking an energy drink. Another example could be is there a significant difference between your participants' satisfaction scores before a warm up and after a warm up. The similarities between the two examples given are that they have a before and after score from the same participants. If you have this then you can use a Wilcoxon signed rank test!

## MANN WHITNEY U TEST

You use a Mann Whitney U Test when you want to prove or disprove there is a difference between two different group's scores. For example males compared to females, Man Utd compared to Fulham FC, Creatine compare to protein. For example, is there a significant difference between the group taking creatine and the group taking protein? Another example could be is there a significant different between the female group's satisfaction scores and the male group's satisfaction scores. The similarities between the two examples given are that they are comparing different groups!. If you have this then you can use a Mann Whitney U Test!

## SPEARMAN'S RANK ORDER CORRELATION TEST

Used if you are trying to identify a relationship (correlation) between two variables. For example, is there a correlation between players' effort in training rating and their like for the coach? Is there a relationship between taking protein shakes and an increase in gym training satisfaction? Is there a relationship/correlation between the type of trainers warn by a runner and their comfort rating?