3

I have two groups, both male soccer players divided into professional and amateur, with knee proprioception scores for each player scored from 0-50. When analysed together they are not normally distributed, but when analysed as two groups individually they are. Which way should I do it? I can't decide if they should be assessed for normality together or separately, which will decided if I run independent samples t-tests or Mann-Whitney U-test.

gung - Reinstate Monica
  • 132,789
  • 81
  • 357
  • 650
Allan
  • 31
  • 2
  • 2
    To what does "it" refer in "Which way should I do it"? What analysis are you intending to perform? Have you searched our site for similar questions about normality? – whuber May 11 '15 at 15:00

1 Answers1

2

The data only need to be normal 'by group'. More specifically, the residuals need to be normally distributed, not your response (proprioception scores) marginally. To understand this further, it may help to read my answer here: What if residuals are normally distributed, but y is not?

Community
  • 1
gung - Reinstate Monica
  • 132,789
  • 81
  • 357
  • 650