If you mean that the dependent variable is a single 7-point item (as opposed to the sum or mean of several such items), then you are bound to have lots of tied scores. And in that case, the KW test does not work so well. E.g., here's a comment from the GraphPad website (
http://graphpad.com/help/prism5/prism5help.html?how_the_kruskal_wallis_test_works.htm):"If you have large sample sizes and a few ties, no problem. But with small data sets or lots of ties, we're not sure how meaningful the P values are."
How large are your 3 groups? And how are the data distributed across the 7 categories?
johnyo99 wrote
Hello all,
Having read the thread on Required Variable Levels in Nonparametric Tests I am no wiser as to if it is possible to use Ordinal data for the 'Test Fields' in nonparametric tests, specifically Kruskal Wallis. I am using Version 19 of SPSS. IS it OK to change the field to scale data even thought the data is ordinal just to get an output.
Being a relative novice I understand that my data is not Scale, but all of my research suggests that the Kruskal Wallis test is the appropriate method of analysis. I am comparing the responses of three unrelated groups who have answered a seven-point Likert Item and I am trying to find if there is a difference between the three groups.
Any help would be much appreciated,
John
--
Bruce Weaver
bweaver@lakeheadu.ca
http://sites.google.com/a/lakeheadu.ca/bweaver/"When all else fails, RTFM."
PLEASE NOTE THE FOLLOWING:
1. My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above.
2. The SPSSX Discussion forum on Nabble is no longer linked to the SPSSX-L listserv administered by UGA (
https://listserv.uga.edu/).
Locked
