This report contains analysis of an online survey to copy-test three print advertisements for body wash. The online survey was distributed via e-mail and 58 usable responses were collected. In this report, the results of the survey are subjected to multiple linear regression analyses for each brand. The independent variables are ten Likert questions about attributes of each brand under investigation. The dependent variable is the change in the respondent’s intention to purchase the brand after exposure to the three advertisements on a constant-sum scale.
Table 1: Multiple linear regression analysis of Axe, Old Spice and Zest
Brand |
Multiple correlation coefficient (R) |
Coefficient of multiple determination (R²) |
Standard error of the estimate (Se) |
F |
Axe |
0.4 |
18.1% |
2.2 |
1.04 |
Old Spice |
0.4 |
13.8% |
2 |
0.75 |
Zest |
0.3 |
10.8% |
3.7 |
0.57 |
|
|
|
|
* p ≤ 0.15 |
Table 2: Axe brand attributes
Axe... |
Unstandardized coefficient (b) |
Standardized coefficient (β) |
t |
(Constant: -1.8) |
|
-1 |
|
... is a good brand of body wash. |
0.9 |
0.3 |
1.46* |
... is preferred over other body washes. |
0.5 |
0.2 |
0.95 |
... gets me cleaner than other body washes. |
-1 |
-0.3 |
-1.42 |
... smells nicer than other body washes. |
0.2 |
0.1 |
0.39 |
...isn’t recommended |
-0.3 |
-0.1 |
-0.77 |
... is a good value. |
0.7 |
0.2 |
1.54* |
...is healthy. |
-0.1 |
-0.5 |
-0.23 |
... makes me feel good. |
-0.4 |
-0.2 |
-1.08 |
...is bad for my skin. |
0.1 |
0 |
0.17 |
...is a brand I would purchase |
0 |
0 |
-.03 |
|
|
|
* p ≤ 0.15 |
In 85 samples out of 100 samples drawn from the same population as this sample, it would be expected that the ten Axe brand attitude items would account for a small portion, about 18.1%, variance in pre/post change scores, but this is not statistically significant and cannot be projected to the population. Only two of the brand attributes (Axe is a good brand, Axe is a good value) had a statistically significant impact; in 85 samples out of 100 samples drawn from the same population as this sample, it would be expected that these coefficients would be about what they are here.
Change score = -1.8 +- 0.9(good)+ 0.5(preferred) -1(cleaner)+ 0.2(smell) -0.3(recommended)+ 0.7(value) -0.1(healthy) -0.4(feel)+ 0.1(skin)+ 0(purchase)
Table 3: Old Spice brand attributes
Old Spice... |
Unstandardized coefficient (b) |
Standardized coefficient (β) |
t |
(Constant: 1.5) |
|
|
0.78 |
... is a good brand of body wash. |
-0.4 |
-0.2 |
-0.86 |
... is preferred over other body washes. |
0.4 |
0.2 |
0.7 |
... gets me cleaner than other body washes. |
-0.5 |
-0.2 |
-1 |
... smells nicer than other body washes. |
0.6 |
0.3 |
0.93 |
...isn’t recommended |
-0.1 |
-0.1 |
-0.4 |
... is a good value. |
-0.4 |
-0.2 |
-0.92 |
...is healthy. |
0.5 |
0.2 |
1.13 |
... makes me feel good. |
-0.3 |
-0.1 |
-0.92 |
...is bad for my skin. |
0.2 |
0.1 |
0.54 |
...is a brand I would purchase |
-0.5 |
-0.3 |
-1.26 |
|
|
|
* p ≤ 0.15 |
In this sample, the brand attributes for Old Spice accounted for a small portion (13.8%) of the variance in pre/post change scores, but this is not statistically significant and cannot be projected to the population. None brand had a statistically significant impact.
Table 4: Zest brand attributes
Zest... |
Unstandardized coefficient (b) |
Standardized coefficient (β) |
t |
(Constant: -4.2) |
|
-1.3 |
|
... is a good brand of body wash. |
0.7 |
0.1 |
0.6 |
... is preferred over other body washes. |
0.1 |
0 |
0.2 |
... gets me cleaner than other body washes. |
-0.9 |
-0.2 |
-1.3 |
... smells nicer than other body washes. |
0.6 |
0.1 |
0.9 |
...isn’t recommended |
0.4 |
0.1 |
0.3 |
... is a good value. |
0.9 |
0.2 |
0.9 |
...is healthy. |
0.4 |
0.1 |
0.4 |
... makes me feel good. |
-0.1 |
0 |
-0.2 |
...is bad for my skin. |
-0.8 |
-0.2 |
-0.8 |
...is a brand I would purchase |
0.1 |
0 |
0.1 |
|
|
|
* p ≤ 0.15 |
In this sample, the brand attributes for Zest accounted for a small portion (10.8%) of the variance in pre/post change scores, but this is not statistically significant and cannot be projected to the population. None of the brand attributes had a statistically significant impact.