Maybe you were doing something wrong? It's pretty hard to tell why KS would have blank values and SW would be fine if we don't know what program you're using or what your data looks like. Did you have any missing values?
Hello Fello Stats Talks users
I ran a Kruskal-Wallis test on my data set and also selected the Monte Carlo method test of random association. How do I report the Monte Carlo results? Neither the H test or Monte Carlo were significant (p>0.05).
Occasionally, when I explore the dataset, the Kolmogorov-Smirnov comes up with blank values whilst the Shapiro-Wilk does not and computes the values. Why is this?
Thanking you in advance for your patience and assistance!
A complete novice,
Auds
Maybe you were doing something wrong? It's pretty hard to tell why KS would have blank values and SW would be fine if we don't know what program you're using or what your data looks like. Did you have any missing values?
"His programming is malfunctioning. It begins! Get your weapons, he's going to become a killbot!!!" - bryangoodrich
Hi Dason
Thanks for your reply. I am using SPSS 19. There were no missing values. The associations were arcsin sqrt transformed and the transformed values are:
1.00
.90
.98
1.05
.88
.93
.93
.88
.93
.85
.82
.88
.90
.77
.49
.77
The Normality table is as follows:
Tests of Normality
Kolmogorov-Smirnova Shapiro-Wilk
YSTI Statistic df Sig. Statistic df Sig.
BullAssTrans YSTI 0 .216 4 . .977 4 .887
YSTI 3 .307 4 . .729 4 .024
YSTI 7 .186 4 . .975 4 .875
YSTI 10 .329 4 . .894 4 .402
a. Lilliefors Significance Correction
Any ideas as to what I am doing wrong?
I appreciate your feedback!
Kind regards
Auds
Oops.... here is the Normality table output.
|
|