Question

In the Turkey paired comparisons. What treatment shows better outcomes? Support with data of the Tuckey analyses.

inces Mailings Review View Help Acrobat E Title Tide Tele TITLE TIE GI EDIT A Colors Fonts ISO gt F sig Document Formatting Symptoms Improvement Indicator Sum of Squares df Between Groups 251.800 Within Groups 640.700 Total 892.500 Mean Square 62.950 4 4.421 004 45 14.238 49 0.004<0.01 Table 3 Dependent Variable: Symptoms Improvement Indicator Tukey HSD Mean Difference O Treatment (J) Treatment (1-3) Std. Error Escitalopram No Treatment 6.700 1.687 Placebo 2.500 1.687 Seroxat (Paxil 3.500 1.687 Effexor 4.800 1.687 The mean difference is significant at the 0.01 level Sig 002 99% Confidence Interval Lower Bound Upper Bound 86 12.54 -3.34 8.34 -2.34 9.34 - 1.04 10.64 580 249 050

Answer 1

Solution:

Define hypothesis:

H0 :there is not significant difference between the group means. i.e all the group means are same.

H1:there is significant difference between the group means . atleast two of the group means are significantly different .i.e all the group means are not same.

Significance level=0.01 (given)

From the anova table,

We get p-value =0.004

p-value is less than the significance value

0.004<0.01

This is significant . so we reject the null hypothesis (H0) at 0.01 level of significance and conclude that there is significant difference between the group means.

Now we check which pair of groups are significant different using Tukey paired comparisons.

p-value less than 0.01 is regarded as significant different pair

p-value greater than 0.01 is regarded as not significant different pair.

p-value for pair (Escitalopram,no treatment) is 0.002

this is less than 0.01

so this is only significant different pair.

p-value for other pairs is greater than 0.01 so these are not significant different pairs.