ANOVA, Tukey and finding the difference

#1
So basically I used an ANOVA on a set of scores (One-way unstacked) and got these values:

Source DF/ SS / MS / F / P
Factor 3 / 300.55 / 100.18 / 14.79 / 0.000
Error 16 / 108.40 / 6.78/
Total 19 / 408.95/

according to the "percentage points of the F-distribution" table:

F= 14.79 > 3.24 a significant difference does exist at the 5% level of significance

A Tukey-Kramer test gives these results:

Tukey 95% Simultaneous Confidence Intervals
All Pairwise Comparisons

Individual confidence level = 98.87%


200 subtracted from:

Lower Center Upper --------+---------+---------+---------+-
300 -6.114 -1.400 3.314 (-----*-----)
400 0.886 5.600 10.314 (-----*-----)
500 3.286 8.000 12.714 (-----*-----)
--------+---------+---------+---------+-
-8.0 0.0 8.0 16.0


300 subtracted from:

Lower Center Upper --------+---------+---------+---------+-
400 2.286 7.000 11.714 (-----*-----)
500 4.686 9.400 14.114 (-----*-----)
--------+---------+---------+---------+-
-8.0 0.0 8.0 16.0


400 subtracted from:

Lower Center Upper --------+---------+---------+---------+-
500 -2.314 2.400 7.114 (-----*-----)
--------+---------+---------+---------+-
-8.0 0.0 8.0 16.0


According to these results where is the difference?


**Sorry I've stuffed up in posting the results of Tukey. In short, none of the intervals contain 0 (they all overlap )
 
Last edited: