In: Statistics and Probability
1. A researcher wants to know what kind of exercising is most likely to lead to improve people’s moods. He has participants engage in one month of regular workouts (cardio, weight training, or yoga) and then has them complete a mood measure (1-10, higher scores = better mood). Is there a significant difference in groups? Hand calculate the ANOVA using a significance level of .05 and then check your work using SPSS.
Cardio | Weights | Yoga |
8 | 9 | 7 |
7 | 5 | 9 |
8 | 7 | 9 |
4 | 6 | 10 |
Hypothesis:
Fcrit /threshold:
Source of Variations | SS | df | MS | Fobt |
Between groups | ||||
within groups | ||||
Total |
Conclusion:
From the question, the data values are
Cardio | Weights | Yoga |
8 | 9 | 7 |
7 | 5 | 9 |
8 | 7 | 9 |
4 | 6 | 10 |
Cardio | Weights | Yoga | Total | |
Sum X | 27 | 36 | 13 | 76 |
Sum X^2 | 193 | 191 | 311 | 695 |
N | 4 | 4 | 4 | 12 |
Mean | 6.75 | 6.75 | 8.75 |
The grand mean is obtained by taking mean of the treatments,
Sum of Squares
Between Groups
The sum of square of treatment is obtained by using the formula,
Where,
j is the column (j=1,2,3,4)
i is the row, (i=1,2,3)
is the average of jth column
is the grand mean
Total Sum of Square
Cardio | 8 | 7.416667 | 0.583333 | 0.340278 |
7 | 7.416667 | -0.41667 | 0.173611 | |
8 | 7.416667 | 0.583333 | 0.340278 | |
4 | 7.416667 | -3.41667 | 11.67361 | |
Weights | 9 | 7.416667 | 1.583333 | 2.506944 |
5 | 7.416667 | -2.41667 | 5.840278 | |
7 | 7.416667 | -0.41667 | 0.173611 | |
6 | 7.416667 | -1.41667 | 2.006944 | |
Yoga | 7 | 7.416667 | -0.41667 | 0.173611 |
9 | 7.416667 | 1.583333 | 2.506944 | |
9 | 7.416667 | 1.583333 | 2.506944 | |
10 | 7.416667 | 2.583333 | 6.673611 | |
Sum | 34.91667 |
Within Groups Sum of Square
Degree of freedoms
Mean Squares
F-statistic
F- critical value
The F-critical value is obtained from F-distribution table for significance level = 0.05, numerator degree of freedom = 2 and denominator degree of freedom = 9
The complete ANOVA table is,
ANOVA | |||||
Source of Variation | SS | df | MS | F | F crit |
Between Groups | 10.66667 | 2 | 5.333333 | 1.979381 | 4.256495 |
Within Groups | 24.25 | 9 | 2.694444 | ||
Total | 34.91667 | 11 |
The values are matched with the results are statistical software.
Conclusion:
Since F-statistic value is less than F-critical value, the null hypothesis is not rejected. There is no significant difference in means.