I had ranked data showing preferences from 1 to 7 where 1 was the most preferred and 7 the least, but on the Kruskal-Wallis I get mean ranks of 368, 168, and 155 etc.
How do I relate this to my actual ranks?
Hi Marvelous
Give the rank of 1 to the least preferred, 2 to the next least preferred, and so on. I mean you have to reverse your coding, then it will be clear. The out put is in 'Rank Sum' not a mean rank. So, the groups which takes the max value will be the most preferred in that case.
Those are not mean ranks (an oxymoron?) they look like totals. Why not just plot counts of ranks ( number of 1s, 2s, 3s, etc.........)?
Hi Marvelous, the problem is that when a Kruskal-Wallis test is performed by statistical software, the data are ranked automatically, but if your data are already ranked (in an ordinal scale) this gives a high values in Sum Ranks or Mean Ranks, depending of your sample size of each group.
To avoid this and if you want to make inferences about the original ordinal scale of your data, you can perform the K-W test directly without the ranking procedure. The problem is that the most popular statistical software don’t make this (I don’t know but maybe on R this could be possible), but instead you must do that using Microsoft Excel by “hand made” formulas.
Good luck.
- Badges
- Science topic
- Similar topics
- Medicine
- Physiology
- Biosignals
- Biomedical Signal Processing
- Statistical Analysis