I need to validate if two independent samples give significantly different results. The sample has a non-normal distribution, so I will use Mann-Whitney U Test

The dependent variable is a five-point Likert scale, with four items. My understanding is that I need to do the next steps:

1-Create a variable that is the median of the four items that compose the scale. Keep this variable as ordinal 2-Do Mann-Whitney U Test with SPSS

It is the first time I do this kind of test, and I am not sure the results are right. It says that the Mean Rank for one group is 46,62 and the for the other 90,36. I am using a five-point Likert scale, so I am not sure how the means could be higher than 5?

I am quite new, so I am sure I am misunderstanding something. in the case that the result is right, how do you write it down for the paper?

  • $\begingroup$ It doesn't sound like you're doing the right thing but your situation could be clearer. How is the scale made from the items? By adding them? How many observations (/subjects) are there? $\endgroup$ – Glen_b May 24 '17 at 1:32
  • $\begingroup$ 46,62 ? is it 46.62 ? $\endgroup$ – Subhash C. Davar Jun 19 '17 at 14:39
  • $\begingroup$ How many subjects did you examine? $\endgroup$ – Subhash C. Davar Jun 19 '17 at 14:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.