Analysis
Factor Analysis
A factor analysis was conducted for each of the three brands of cereal bar (Nature Valley, Quaker, and Soyjoy) to place related brand attributes (10 Likert items) into independent groups, or factors, in order to calculate a brand attitude score. A paired t-test was then conducted to determine the significance of the results.
Table 1. Communities (h²) (Sample Size: 63)
|
Nature Valley |
Quaker |
Soyjoy |
Good brand |
.5 |
.7 |
.7 |
Energy |
.5 |
.6 |
.7 |
Healthy |
.5 |
.6 |
.7 |
Substitute |
.4 |
.7 |
.6 |
Expensive |
.4 |
.8 |
.9 |
Appetite |
.5 |
.6 |
.7 |
Good Taste |
.6 |
.8 |
.6 |
Purchase |
.7 |
.8 |
.8 |
Recommendation |
.5 |
.8 |
.7 |
Attractive |
.3 |
.5 |
.7 |
Table 1 shows the communality scores, which is the total of the squared factor loadings divided by all the factors.
Table 2: Total Variance Explained
|
Nature Valley |
Quaker |
Soyjoy |
Factor |
Eigenvalues |
% of variance |
Cumulative
% |
Eigenvalues |
% of variance |
Cumulative
% |
Eigenvalues |
% of variance |
Cumulative
% |
I |
3.2 |
32.2 |
32.2 |
4.5 |
44.8 |
44.8 |
4.6 |
46.3 |
46.3 |
II |
1.7 |
16.9 |
49.1 |
1.4 |
13.5 |
58.3 |
1.3 |
12.8 |
59.1 |
III |
.9 |
9.6 |
58.8 |
1.1 |
10.8 |
69.1 |
1.0 |
10.4 |
69.6 |
IV |
.9 |
8.8 |
67.6 |
.7 |
7.4 |
76.6 |
.9 |
8.6 |
78.3 |
V |
.7 |
7.4 |
75.1 |
.6 |
6.4 |
83.0 |
.6 |
5.9 |
84.2 |
VI |
.7 |
6.8 |
82.0 |
.5 |
5.2 |
88.3 |
.6 |
5.6 |
89.8 |
VII |
.6 |
5.5 |
87.5 |
.4 |
3.8 |
92.1 |
.4 |
3.7 |
93.6 |
VIII |
.5 |
4.7 |
92.3 |
.4 |
3.6 |
95.7 |
.3 |
2.6 |
96.3 |
IX |
4 |
4.0 |
96.3 |
.3 |
2.5 |
98.3 |
.2 |
2.2 |
98.5 |
X |
.4 |
3.6 |
100.0 |
.2 |
1.6 |
100.0 |
.1 |
1.4 |
100.0 |
For the three brands, each had several factors with Eigenvalues > 1.0, indicating that the factors for each brand could explain variance in the respective Likert items. For Nature Valley, two of the ten factors had Eigenvalues greater than 1.0. These two factors combined explained 49.1% of the variance in the original ten Likert items. For Quaker, three of the ten factors had Eigenvalues greater than 1.0. These three factors combined explained 69.1% of the variance in the original ten Likert items. And for Soyjoy, three of the ten factors again had Eigenvalues greater than 1.0. These three factors combined explained 69.6% of the variance in the original ten Likert items. It goes to say that we would want less factors contributing to a higher percentage of variation. Most of the brands had upwards of 60-70% variation, which is good, but not great in terms of being precise on which attribute items caused the greatest variation.
Table 3. Factor Varimax Rotated Matrix (Nature Valley)
|
Factor Matrix |
Varimax rotated matrix |
Likert Items |
I |
II |
I |
II |
Good brand |
.7 |
.1 |
.7 |
.3 |
Energy |
.7 |
.1 |
.7 |
.3 |
Healthy |
.6 |
-.4 |
.7 |
-.2 |
Substitute |
.6 |
-.4 |
.6 |
-.2 |
Expensive |
.2 |
.6 |
-.1 |
.7 |
Appetite |
.7 |
.0 |
.6 |
.2 |
Good Taste |
.2 |
.7 |
-.1 |
.7 |
Purchase |
.8 |
-.1 |
.7 |
.2 |
Recommendation |
.4 |
.6 |
.2 |
.7 |
Attractive |
.5 |
-.2 |
.5 |
-.1 |
As mentioned before, in NatureValley, 49.1% of the variance can be explained by two factors. With multiple factors extracted, the matrix could be rotated to search for greater importance. The item ‘NatureValley is a good brand’ loaded on factor 1 and all other items except item 5 (NatureValley is Expensive), 7(NatureValley has a good taste), and 9 (I would recommend Nature Valley) also loaded on factor 1. These scores would be used to compute the attitude scores presented later. There were no ambiguous items.
Table 4. Factor Varimax Rotated Matrix (Quaker)
|
Factor Matrix |
Varimax rotated matrix |
Likert Items |
I |
II |
III |
I |
II |
III |
Good brand |
.8 |
.2 |
.2 |
.8 |
.2 |
.1 |
Energy |
.8 |
-.1 |
.1 |
.8 |
.1 |
-.1 |
Healthy |
.7 |
-.2 |
.2 |
.8 |
.0 |
-.1 |
Substitute |
.7 |
-.4 |
-.1 |
.7 |
.0 |
-.4 |
Expensive |
-.1 |
.6 |
.4 |
-.1 |
.0 |
.9 |
Appetite |
.8 |
-.1 |
.1 |
.8 |
.1 |
-.1 |
Good Taste |
.4 |
.7 |
-.2 |
.2 |
.8 |
.3 |
Purchase |
.9 |
.1 |
.1 |
.9 |
.2 |
.0 |
Recommendation |
.4 |
.3 |
-.8 |
.1 |
.8 |
-.4 |
Attractive |
.7 |
.2 |
.0 |
.6 |
.4 |
.1 |
In Quaker, 69.1% of the variance can be explained by three factors. With multiple factors extracted, the matrix could be rotated to search for greater importance. The item ‘Quaker is a good brand’ loaded on factor 1 and all other items except item 5 (Quaker is Expensive), 7 (Quaker has a good taste), and 9 (I would recommend Quaker) also loaded on factor 1. These scores would be used to compute the attitude scores presented later. There were no ambiguous items.
Table 5. Factor Varimax Rotated Matrix (Soyjoy)
|
Factor Matrix |
Varimax rotated matrix |
Likert Items |
I |
II |
III |
I |
II |
III |
Good brand |
.8 |
-.1 |
.1 |
.8 |
.0 |
-.1 |
Energy |
.8 |
.1 |
.0 |
.7 |
.3 |
-.2 |
Healthy |
.8 |
-.1 |
.1 |
.8 |
.0 |
-.1 |
Substitute |
.7 |
.1 |
.0 |
.7 |
.2 |
-.1 |
Expensive |
-.3 |
.3 |
.9 |
-.2 |
.0 |
.9 |
Appetite |
.8 |
-.1 |
-.2 |
.7 |
.1 |
-.4 |
Good Taste |
.2 |
.7 |
.1 |
.1 |
.7 |
.2 |
Purchase |
.8 |
.0 |
.3 |
.9 |
.1 |
.1 |
Recommendation |
.1 |
.8 |
-.3 |
.0 |
.8 |
-.2 |
Attractive |
.8 |
-.2 |
.2 |
.8 |
.-1 |
-.1 |
In Soyjoy, 69.6% of the variance can be explained by three factors. With multiple factors extracted, the matrix could be rotated to search for greater importance. The item ‘Soyjoy is a good brand’ loaded on factor 1 and all other items except item 5 (Soyjoy is Expensive), 7 (Soyjoy has a good taste), and 9 (I would recommend Soyjoy) also loaded on factor 1. These scores would be used to compute the attitude scores presented later. There were no ambiguous items.
Table 6. Attitude Scores
|
Mean |
Standard Deviation |
Nature Valley |
3.4 |
.5 |
Quaker |
3.3 |
.7 |
Soyjoy |
3.3 |
.7 |
(n=63)
The attitude score reveals that respondents evaluated Nature Valley(attitude score of 3.4) the most favorably followed by Quaker and Soyjoy (both attitude scores of 3.3). In this study, Nature Valley is liked the best but not by much. Although Nature Valley has higher brand perception than Quaker and Soyjoy, the magnitude of this difference is not big enough to be considered important.
Table 7. Paired t-test for Attitude Scores
|
Mean Difference |
Standard Deviation |
t-ratio |
Nature Valley & Quaker |
.1 |
.9 |
.56 |
Quaker & Soyjoy |
.1 |
.9 |
.47 |
Soyjoy & Nature Valley |
.1 |
.8 |
1.10* |
(n=63) *p ≤ .15
The paired t-test shows that in 85 or more samples out of every 100 samples drawn from the same population as this sample of 63 people, it is expected that the difference between brand attitude scores for Soyjoy and Nature Valleywould be about what they are in this sample. However, the magnitude of differences between brand attitude scores for Nature Valley & Quaker and Quaker & Soyjoy cannot be projected to the population.