In: Statistics and Probability
The following data are from an experiment designed to investigate the perception of corporate ethical values among individuals who are in marketing. Three groups are considered: management, research and advertising (higher scores indicate higher ethical values). Marketing Managers Marketing Research Advertising 10 5 7 9 5 8 8 4 7 9 4 6 10 5 7 8 4 7 Compute the values identified below (to 1 decimal, if necessary). Sum of Squares, Treatment 61 Sum of Squares, Error 7.5 Mean Squares, Treatment 30.5 Mean Squares, Error 0.5 Use = .05 to test for a significant difference in perception among the three groups. Calculate the value of the test statistic (to 2 decimals). 61 The p-value is What is your conclusion? Using = .05, determine where differences between the mean perception scores occur. Calculate Fisher's LSD value (to 2 decimals). 1.00 Test whether there is a significant difference between the means for marketing managers (1), marketing research specialists (2), and advertising specialists (3). Difference Absolute Value Conclusion 1 - 2 4.5 1 - 3 2 2 - 3 2.5
I only need this "Calculate Fisher's LSD value "
Applying one way ANOVA: (use excel: data: data analysis: one way ANOVA: select Array): |
Source | SS | df | MS | F | P value |
Between | 61.00 | 2 | 30.500 | 61.000 | 0.000 |
Within | 7.50 | 15 | 0.5000 | ||
Total | 68.50 | 17 |
a)
sum of sq;treatment= | 61.00 |
sum of sq; error= | 7.50 |
mean sq;treatment= | 30.50 |
mean square; error= | 0.50 |
test statistic = | 61.00 |
p value is less than 0.01 | |
Conclude the treatment mean for the three groups are not all the same |
b)
critical value of t with 0.05 level and N-k=15 degree of freedom= | tN-k= | 2.131 | |||
Fisher's (LSD) for group i and j =(tN-k)*(sp*√(1/ni+1/nj) = | 0.87 |
Difference | Absolute Value | Conclusion |
x1-x2 | 4.50 | significant difference |
x3-x1 | 2.00 | significant difference |
x3-x2 | 2.50 | significant difference |