Entering the world of NONPARAMETRIC
Sunday, August 28th, 2011Oh my, I’ve been working on Chapter 4 statistics and it looks like my data is nonnormal…so will I have to enter the world of NONPARAMETRIC? What dangers lurk in that galaxy? Will I be taken captive and forced to morph into a NONPARAMETRICON?
AARRGGHHHHHH!
I spent so much time working on understanding MANOVA that in all honesty, I have not investigated nonparametric possibilities. However, I have read a number of articles indicating that often data collected “in the wild” turns up as nonparametric rather than parametric. In some of the analyses, it appears dissertations have been completed using parametric when they would have been better suited as nonparametric. Unfortunately, the scenarios and problems completed in stat classes are mostly parametric sprinkled with a few nonparametric models.
It appears as though the Mann-Whitney U will soon become my best friend. SPSS analysis is revealing the same conclusions as my work with MANOVA…it appears there is little (or no) difference in my two groups. This is surprising as I had supposed there were some similarities, but not to this extent.
So next time you see me, I may look a little different; morphing into a NONPARAMETRICON of house Mann-Whitney U is quite painful!