non-parametric tests: mann-whitney and wilcoxon

Question 1

when should you use mann-whitney?

Answer 1

• independent groups (unrelated)
• looking for a difference
• ordinal data

Question 2

how do you calculate the table of ranks?

Answer 2

• rank data from groups A and B at the same time
• lowest number on scale has rank of 1
• when two data items are the same you add the rank and give the mean of those ranks
• if there are multiple ranks it may help to use a frequency table
• include the number of participants (N =) for each group

Question 3

how do you work out the value of U? (mann-whitney)

Answer 3

U = Uₐ = Rₐ - ( [Nₐ (Nₐ + 1)] ) / 2

Question 4

mann-whitney and wilcoxon: calculated and critical values

Answer 4

if the calculated value of U is less than the critical value, the result is significant (p ≤ 0.05). we reject the null hypothesis and accept the null hypothesis

Question 5

when should you use wilcoxon?

Answer 5

• looking for a difference
• repeated measures design (related)
• ordinal data

Question 6

(W) step 1: calculating a difference and rank the difference

Answer 6

• rank difference between scores in the two sets of data
• if the difference is zero the data is not included in the ranking and is deducted from the N value
• when ranking, sign is ignored

Question 7

(W) step 2: working out the value of T

Answer 7

calculated value of T is the sum of the ranks of the less frequent sign