Here you will find a comprehensive consumer preference
study for threebrands of energy bars: Pria,
Balance, and Snickers. Various statistical methods
were used for this analysis and each is discussed in
detail in the following pages.
I hope you enjoy my findings.
Factor Analysis
A factor analysis was preformed for each brand of energy bar in order to place related brand attributes into independent factors, for the calculation of a brand attitude score. The ten brand attributes were used as the dependant variables. Next a paired t-test was done to determine the statistical significance for the variations in means between the brand attitude scores of the three energy bar brands.
Pria |
|||||
Pria Brand Attributes |
Communalities |
Factor Matrix |
Verimax Rotated Matrix |
||
I |
II |
I |
II |
||
Good Brand |
0.7 |
0.8 |
-0.1 |
0.6 |
0.6 |
Eat as a Regular Meal |
0.6 |
0.6 |
0.4 |
0.7 |
0.0 |
Over Priced |
0.5 |
0.5 |
0.5 |
0.7 |
-0.1 |
Healthy Snack |
0.8 |
0.9 |
-0.1 |
0.5 |
0.7 |
Provide Energy I need |
0.7 |
0.8 |
0.3 |
0.8 |
0.3 |
Low in Calorie |
0.5 |
0.7 |
-0.1 |
0.4 |
0.6 |
Not a Brand I Trust |
0.8 |
0.6 |
-0.6 |
0.1 |
0.8 |
High in Protein |
0.4 |
0.6 |
0.2 |
0.6 |
0.3 |
Tastes Better Than Other Brands |
0.7 |
0.8 |
0.1 |
0.6 |
0.5 |
Just Like Other Brands |
0.6 |
0.5 |
-0.6 |
-0.0 |
0.7 |
Pria |
|||
Factor |
Eiginvalue |
% of Variance |
Cumulative Variance |
I |
4.8 |
47.9 |
47.9 |
II |
1.3 |
13.5 |
61.4 |
III |
0.9 |
8.6 |
69.9 |
IV |
0.7 |
7.0 |
77.0 |
V |
0.7 |
6.7 |
83.7 |
VI |
0.5 |
5.5 |
89.2 |
VII |
0.3 |
3.4 |
92.6 |
VIII |
0.3 |
3.0 |
95.7 |
IX |
0.3 |
2.7 |
98.4 |
X |
0.2 |
1.6 |
100 |
For the Pria brand 61.4% of the variance can be explained by two factors. These factors (I and II) each have a value greater than 1. For this analysis the Factor Matrix was used instead of the Verimax Rotated Matrix since they both had three ambiguous variables. According to the Factor Matrix seven attributes are evaluative in nature. This is determined by analyzing which attributes are loaded under the same factor as the “Good Brand” attribute. These include the following: Good Brand, Eat as a Regular Meal, Healthy Snack, Provide Energy I need, Low in Calorie, High in Protein, and Tastes Better Than Other Brands. The remaining three attributes were ambiguous and therefore ignored.
Balance |
|||||
Balance Brand Attributes |
Communalities |
Factor Matrix |
Verimax Rotated Matrix |
||
I |
II |
I |
II |
||
Good Brand |
0.7 |
0.9 |
0.0 |
0.8 |
0.4 |
Eat as a Regular Meal |
0.6 |
0.7 |
0.1 |
0.6 |
0.4 |
Over Priced |
0.6 |
0.4 |
0.6 |
0.1 |
0.8 |
Healthy Snack |
0.8 |
0.7 |
-0.4 |
0.9 |
-0.0 |
Provide Energy I need |
0.6 |
0.8 |
-0.0 |
0.7 |
0.3 |
Low in Calorie |
0.7 |
0.7 |
-0.4 |
0.8 |
-0.1 |
Not a Brand I Trust |
0.7 |
0.8 |
0.4 |
0.7 |
0.4 |
High in Protein |
0.6 |
0.5 |
0.6 |
0.1 |
0.8 |
Tastes Better Than Other Brands |
0.5 |
0.7 |
0.1 |
0.6 |
0.3 |
|
0.3 |
0.6 |
-0.3 |
0.5 |
0.3 |
Balance |
|||
Factor |
Eiginvalue |
% of Variance |
Cumulative Variance |
I |
4.9 |
49.0 |
49.0 |
II |
1.2 |
11.7 |
60.7 |
III |
0.9 |
8.7 |
69.3 |
IV |
0.8 |
7.8 |
77.1 |
V |
0.6 |
6.3 |
83.4 |
VI |
0.5 |
5.2 |
88.6 |
VII |
0.4 |
3.8 |
92.5 |
VIII |
0.3 |
3.1 |
95.6 |
IX |
0.3 |
2.5 |
98.1 |
X |
0.2 |
1.9 |
100 |
For the Balance brand 60.7% of the variance can be explained by two factors. These factors (I and II) each have a value greater than 1. For this analysis the Verimax Rotated Matrix was used instead of the Factor Matrix since there was one ambiguous variable in the Factor Matrix. According to the Verimax Rotated Matrix eight attributes are evaluative in nature. These include the following: Good Brand, Eat as a Regular Meal, Healthy Snack, Provide Energy I need, Low in Calorie, Not a Brand I Trust, Tastes Better Than Other Brands, and Just Like Other Brands. The remaining two attributes load on Factor II, indicating that these attributes relate to each other in a way that is independent from the other attributes.
Snickers |
|||||||
Snickers Brand Attributes |
Communalities |
Factor Matrix |
Verimax Rotated Matrix |
||||
I |
II |
III |
I |
II |
III |
||
Good Brand |
0.8 |
0.9 |
-0.1 |
-0.1 |
SPSS was unable to perform the Verimax Rotated Matrix for Snickers. |
||
Eat as a Regular Meal |
0.7 |
0.7 |
-0.0 |
-0.4 |
|||
Over Priced |
0.9 |
0.3 |
0.4 |
0.8 |
|||
Healthy Snack |
0.8 |
0.8 |
-0.2 |
0.1 |
|||
Provide Energy I need |
0.7 |
0.8 |
-0.1 |
0.2 |
|||
Low in Calorie |
0.7 |
0.7 |
-0.3 |
0.3 |
|||
Not a Brand I Trust |
0.5 |
0.7 |
0.1 |
-0.3 |
|||
High in Protein |
0.6 |
0.7 |
0.3 |
-0.1 |
|||
Tastes Better Than Other Brands |
0.5 |
0.6 |
0.4 |
-0.1 |
|||
Just Like Other Brands |
0.8 |
-0.0 |
0.9 |
-0.1 |
|||
Snickers |
|||
Factor |
Eiginvalue |
% of Variance |
Cumulative Variance |
I |
4.4 |
44.3 |
44.3 |
II |
1.4 |
14.1 |
58.4 |
III |
1.1 |
10.5 |
68.9 |
IV |
0.7 |
7.2 |
76.1 |
V |
0.6 |
6.1 |
82.2 |
VI |
0.5 |
5.3 |
87.5 |
VII |
0.4 |
4.4 |
91.9 |
VIII |
0.4 |
3.5 |
95.4 |
IX |
0.3 |
3.1 |
98.5 |
X |
0.1 |
1.5 |
100 |
For the Snickers brand 68.9% of the variance can be explained by three factors. These factors (I, II, and III) each have a value greater than 1. For this analysis the Factor Matrix was used instead of the Verimax Rotated Matrix because SPSS was unable to generate the Verimax Rotated Matrix for this brand. According to the Factor Matrix eight attributes are evaluative in nature. These include the following: Good Brand, Eat as a Regular Meal, Healthy Snack, Provide Energy I need, Low in Calorie, Not a Brand I Trust, High in Protein, and Tastes Better Than Other Brands.
Attitude Scale (N = 66)
|
Mean |
Standard Deviation |
Pria |
3.3 |
0.7 |
Balance |
3.3 |
0.8 |
Snickers |
2.9 |
0.7 |
Paired t-Test for Attitude Scale
|
t-Ratio |
Pria vs. Balance |
0.19 |
Pria vs. Snickers |
3.69* |
Balance vs. Snickers |
3.58* |
*p £ 0.15
The Mean attitude scales for Pria and Balance are equivalent at 3.3, which is higher than the 2.9 scale for Snickers. Therefore we can infer that respondents preferred Pria and Balance over Snickers. According to the paired t-test the correlations between the attitude scales for Pria and Snickers, and Balance and Snickers are statistically significant at an alpha level of 0.15. Therefore it can be expected that in 85 or more samples out of every 100 samples drawn from the same population as the sample we would expect to find nearly the same results for these two pairs and can project these results to the population. However the results for Pria vs. Balance are not statistically significant and therefore cannot be projected to the population.