I would be grateful for some advice or comments on this output from kwallis2
(Stata 8). On the basis of Kruskal-Wallis I would reject the null hypothesis
of no difference between regions, but the subsequent multiple comparison
indicate no significant differences between any of the regions.
Thanks,
Martyn
. kwallis2 score if q==13,by(region)
One-way analysis of variance by ranks (Kruskal-Wallis Test)
region Obs RankSum RankMean
--------------------------------
1 48 4074.50 84.89
2 17 1064.50 62.62
3 62 4638.00 74.81
4 20 1101.00 55.05
Chi-squared (uncorrected for ties) = 8.336 with 3 d.f. (p = 0.03956)
Chi-squared (corrected for ties) = 9.424 with 3 d.f. (p = 0.02416)
Multiple comparisons between groups
-----------------------------------
(Adjusted p-value for significance is 0.004167)
Ho: score(region==1) = score(region==2)
RankMeans difference = 22.27 Critical value = 31.70
Prob = 0.031944 (NS)
Ho: score(region==1) = score(region==3)
RankMeans difference = 10.08 Critical value = 21.60
Prob = 0.109120 (NS)
Ho: score(region==1) = score(region==4)
RankMeans difference = 29.84 Critical value = 29.90
Prob = 0.004234 (NS)
Ho: score(region==2) = score(region==3)
RankMeans difference = 12.19 Critical value = 30.75
Prob = 0.147871 (NS)
Ho: score(region==2) = score(region==4)
RankMeans difference = 7.57 Critical value = 37.06
Prob = 0.295024 (NS)
Ho: score(region==3) = score(region==4)
RankMeans difference = 19.76 Critical value = 28.89
Prob = 0.035590 (NS)
All the groups are comparable from each other.
"The difference between winners and losers is that winners tell the jokes
and the losers talk about the run of the ball."
I've stopped 3,059 spam messages. You can too!
One month FREE spam protection at http://www.cloudmark.com/spamnetsig/}
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
