In: Statistics and Probability
The data below are from an independent-measures experiment comparing three different treatment conditions on number of panic attacks. (10 POINTS)
CBT |
Psychotherapy |
Placebo |
|
1 |
2 |
4 |
|
1 |
3 |
2 |
G = 42 |
2 |
1 |
4 |
SX2 = 122 |
3 |
3 |
3 |
|
1 |
2 |
4 |
|
1 |
1 |
4 |
|
T = 9 |
T = 12 |
T = 21 |
|
SS = 3.5 |
SS = 4 |
SS = 3.5 |
1) State the null hypothesis
2) calculate necessary statistics using an alpha of .05
3) make a decision about your null hypothesis and explain what that decision means
4) Make a conclusion about your IV and DV that includes an APA format summary of your statistical analysis.
5) Indicate what you would do next
1) Ho: µ1=µ2=µ3
2)
one way anova
treatment | CBT | psychotherepy | Placebo | |||||
count, ni = | 6 | 6 | 6 | |||||
mean , x̅ i = | 1.500 | 2.00 | 3.50 | |||||
std. dev., si = | 0.8 | 0.9 | 0.8 | |||||
sample variances, si^2 = | 0.700 | 0.800 | 0.700 | |||||
total sum | 9 | 12 | 21 | 42 | (grand sum) | |||
grand mean , x̅̅ = | Σni*x̅i/Σni = | 2.33 | ||||||
( x̅ - x̅̅ )² | 0.694 | 0.111 | 1.361 | |||||
TOTAL | ||||||||
SS(between)= SSB = Σn( x̅ - x̅̅)² = | 4.167 | 0.667 | 8.167 | 13 | ||||
SS(within ) = SSW = Σ(n-1)s² = | 3.500 | 4.000 | 3.500 | 11.0000 |
no. of treatment , k = 3
df between = k-1 = 2
N = Σn = 18
df within = N-k = 15
mean square between groups , MSB = SSB/k-1 =
13/2= 6.5000
mean square within groups , MSW = SSW/N-k =
11/15= 0.7333
F-stat = MSB/MSW = 6.5/0.7333= 8.86
ANOVA | ||||||
SS | df | MS | F | p-value | F-critical | |
Between: | 13.00 | 2 | 6.50 | 8.864 | 0.0029 | 3.6823 |
Within: | 11.00 | 15 | 0.73 | |||
Total: | 24.00 | 17 | ||||
α = | 0.05 |
3)
Decision: p-value<α , reject null
hypothesis
Test is statiscally significant
4) conclusion : there is enough evidence of significant mean difference among three treatments
This test was found to be statistically significant, F(2,15) = 8.864, p < .05;
5)
since, null hypothesis is rejected, we will do post hoc test ,tukey Kramer test to test which pair of means are significantly different