ANALYSIS
After the data were collected, analysis was conducted using SPSS. Many different analyses were performed. Here you will find the results from several statistical analyses performed on these data. Click the links below to see the various analyses.
Paired t-test
Correlated t-tests were conducted on all brands of whiskey (Knob Creek, Makers Mark and Jack Daniels), comparing the mean brand index scores (BIS). Brand index scores were derived from the summation of input given by subjects in response to Likert items.
Brand |
Mean BIS |
Sample Size |
Standard Deviation |
Knob Creek |
27.4 |
76 |
3.8 |
Maker’s Mark |
29.8 |
76 |
4.5 |
Jack Daniel’s |
29.9 |
76 |
3.7 |
Paired Samples |
t |
Significance |
Knob Creek & Maker’s Mark |
-3.68 |
.000* |
Knob Creek and Jack Daniel’s |
-3.62 |
.001* |
Maker’s Mark and Jack Daniel’s |
-.077 |
.939 |
* p <= .15
From the first chart, we can see that the pool of respondents slightly favored Jack Daniels over Maker’s Mark, with respective brand index scores of 29.9 and 29.7. Fewer respondents had positive feelings towards Knob Creek, which had a brand index score of only 27. 4.
Due to the level of significance found in the Knob Creek vs. Maker’s Mark brands and the Knob Creek vs. Jack Daniel’s brands, it can be concluded that, in 85 or more samples out of every 100 taken from the same population as this one, their respective brand index scores can be projected to the population.
It is impossible to tell whether or not 85 out of 100 samples drawn from the same population of this sample would result in similar mean scores when comparing Maker’s Mark and Jack Daniel’s because their statistical significance is greater than the alpha level of .15. Therefore, we cannot project the results on the population.
Top
Between-groups t-tests ( Aveeno)
The following table shows those respondents whose perceptions of Knob Creek moved up or down after viewing the advertisements.
Brand Index Score Change for Knob Creek
Change Score |
Number of People |
Mean Score |
Std. Deviation |
t |
Moved up |
22 |
30.1 |
3.8 |
4.81* |
Moved down |
39 |
25.7 |
3.2 |
*p<= .15
Ad Index Score Change for Knob Creek
Change Score |
Number of People |
Mean Score |
Std. Deviation |
t |
Moved up |
22 |
14.7 |
3.3 |
3.45* |
Moved down |
39 |
11.9 |
2.7 |
*p<= .15
These numbers are all statistically significant and can there for be projected 85 out of 100 people from this population. This means that more people will have a negative view of Knob Creek after they see the advertising. This could be devastating for a brand, as it means that its advertising is creating a negative brand image that could ultimately result in a lack of sales and brand equity.
Top
Chi-Square Significant test (Jack Daniel's)
A chi-square significance test was conducted to find out if there is a significant relationship between "up", "no change", and "down" pre-to-post movers on Jack Daniel's. The test also worked to observe whether or not these respondents weere above or below the median brand index score for Jack Daniel's. The results were obtained via an online questionnaire. Sixty-five people were used for this analysis.
Change in score, Pre/Post |
Above Median |
Below Median |
Up |
29 |
10 |
% within Change Score |
74.4 |
25.6 |
% above/below median |
76.3 |
37.0 |
% total Percentage |
44.6 |
15.4 |
No Change |
5 |
5 |
50.0 |
50.0 |
13.2 |
18.5 |
7.7 |
7.7 |
Down |
4 |
12 |
25.0 |
75.0 |
10.5 |
44.4 |
6.2 |
18.5 |
Chi Square |
Degrees of Freedom |
Significance |
11.73** |
2 |
.003* |
* p <=0.15
**1 cell has expected count less than 5. The minimum expected count is 4.15.
Due to the level of significance we would, in 85 or more samples of 100 samples taken from the same population as this one, be able to project a cross tab distribution as this sample for Jack Daniel’s.
The median brand index score for Jack Daniel’s is 29. From this sample, there were 5 people or 7.7% of the population whose pre and post test measurements did not change and whose brand index scores were below the median. There were also 5 people or 7.7% of the population whose pre and post test measurements did not change and whose brand index scores were above the median. Those who favored the Jack Daniel’s brand more after viewing the print ad comprised of 29 people, or 44.6% of the population, who rated the brand above the median and 10, or 15.4% of the population, who still rated below the median. Those who favored the brand less after viewing still had 4 (6.2%) who rated the brand above the median and 12 (18.5%) below.
From this analysis we can conclude that those who had a lesser opinion of Jack Daniel’s prior to viewing the advertising would be more likely to change their opinion following ad exposure. This means that Jack Daniel’s advertising is having a positive effect on a greater number of people than those upon whom it is having a negative effect.
Top
Frequencies
A frequency test was conducted on all three brands to show pre-to-post ad exposure on the constant-sum scale. The results were obtained via an online questionnaire. Seventy-six people were used in this analysis.
pre to post exposure measurement
|
Knob Creek |
Maker’s Mark |
Jack Daniel’s |
Increase |
22 |
25 |
45 |
No Change |
15 |
34 |
11 |
Decrease |
39 |
17 |
20 |
Jack Daniel’s had the most individuals think positively of its advertising following exposure (45 individuals or 59.2% of the population) while Maker’s Mark had the fewest respondents moving down in their opinions (17 or 22.4%). Knob Creek had the fewest positive increase in opinions (22 ior 28.9%), and the most individuals whose opinion of the brand decreases after ad exposure (39 or 51.3%). Maker’s Mark had the greatest number of people with unchanging opinions (34, 44.7%).
From this analysis, we can infer that for the majority of the sample, Jack Daniel’s had the strongest advertising, as brand perception increased. However, Maker’s Mark already had a strong brand awareness, as many people also maintained opinions that they had about the brand prior to ad exposure. Knob Creek clearly has the weakest advertising, as brand perception decreased significantly following ad exposure.
Top
Frequency Count
Frequency tests were conducted to compare Jack Daniel’s and Maker’s Mark brand index scores.
Brand Comparison |
Number of respondents |
Percent of respondents |
Jack Daniel’s > Maker’s Mark |
36 |
47.4% |
Jack Daniel’s <= Maker’s Mark |
35 |
46.1% |
The results indicate that 36 respondents (47.4%) rated Jack Daniel’s higher than Maker’s Mark and 35 respondents (46.1%) rated Jack Daniel’s as being less than or equal to Maker’s Mark. From this analysis, we can infer that the Jack Daniel’s advertising had a slightly larger positive impact than that of Maker’s Mark, making it (albeit slightly) more effective.
Top
Simple Correlation Coefficient
A correlation test was conducted to determine if there was any relationship between Jack Daniel’s and Maker’s Mark’s brand index scores.
Brand Index Scores Compared |
Correlation |
Jack Daniel’s and Maker’s Mark |
.364 |
* p <=.15
The correlation between Jack Daniel’s and Maker’s Mark is slightly less than moderate, as it is closer to zero than it is to one. Correlation allows us to determine the possibility of an inverse relationship between the two brands (ie. if a person preferred Jack Daniel’s would they be less likely to prefer Maker’s Mark and vice versa) or if there is some other relationship that is strong. However since the correlation we have found is not quite moderate, we cannot come to the conclusion that there is an inverse relationship, or even a strong relationship, between Jack Daniel’s and Maker’s Mark. We can only assume that, if projected onto 85 out of 100 people, the relationship found through these analyses between the two brands might exist, making any conclusions dubious.
Top
Modified Simple Correlation Coefficient:
A separate correlation test was conducted between Jack Daniel’s and Maker’s Mark’s brand index scores to determine if any relationship exists. This time, only male respondents were selected from the sample.
Brand Index Scores Compared |
Correlation |
Jack Daniel’s and Maker’s Mark (Males) |
.246 |
Sample size = 42
The correlation between Jack Daniel’s and Maker’s Mark if males are selected from the test group is even lower than before. It is considered to be not terribly significant, therefore projecting that there is no discernable relationship between the two brands with regard to respondents. However there still is somewhat of a relationship, as the correlation statistic is not zero.
Top
Regression Analysis
This report contains regression analysis of copy-testing research on print ads for three brands of whiskey: Knob Creek, Maker’s Mark and Jack Daniel’s. The following multiple regression analysis illustrates the relationship between 10 Likert questions that were asked following the respondents’ respective exposure to one advertisement for each brand in their above order, and the ranking change score for each brand. The ranking change score was calculated by taking the final ranking (0-10) of one of the brands following exposure to the three brands and subtracting out the initial ranking (0-10) of one of the brands prior to exposure.
Brand |
R |
R 2 |
Se (Standard Error of the Estimate) |
F ratio |
Knob Creek |
0.5 |
25.3% |
2.8 |
2.20* |
Maker’s Mark |
0.3 |
9.6% |
2.1 |
0.69 |
Jack Daniel’s |
0.6 |
31.0% |
2.6 |
2.92* |
*p <= .15
Variables |
|
Knob Creek |
|
|
Maker’s Mark |
|
|
Jack Daniel’s |
|
|
b |
Beta |
t |
b |
Beta |
t |
b |
Beta |
t |
Constant (a) |
-9.3 |
|
2.27 |
-1.6 |
|
0.69 |
-5.0 |
|
1.14 |
1. This is a good whiskey. |
0.2 |
0.1 |
0.90 |
0.0 |
0.0 |
0.04 |
-0.4 |
-0.2 |
1.66* |
2. More time is spent making this whiskey than any other. |
0.0 |
0.0 |
0.70 |
-0.2 |
0.0 |
0.35 |
0.2 |
0.0 |
0.33 |
3. This whiskey tastes better than other brands of whiskey. |
1.0 |
0.2 |
1.40 |
0.2 |
0.0 |
0.22 |
-0.1 |
0.0 |
0.15 |
4. This whiskey is less expensive than other brands of whiskey. |
0.0 |
0.0 |
0.11 |
0.3 |
0.1 |
0.71 |
-0.2 |
-0.1 |
0.50 |
5. I would want to serve this whiskey to my friends and family. |
0.5 |
0.2 |
1.07 |
0.5 |
0.2 |
1.30 |
0.5 |
0.1 |
0.97 |
6. This whiskey goes down smoother than other whiskey. |
-0.2 |
0.0 |
0.32 |
-0.2 |
-0.1 |
0.48 |
-0.3 |
-0.1 |
0.50 |
7. This whiskey’s packaging is unattractive. |
0.6 |
0.2 |
1.53* |
0.3 |
0.1 |
0.95 |
0.1 |
0.0 |
0.22 |
8. This whiskey does not suit my lifestyle. |
0.0 |
0.0 |
0.05 |
-0.3 |
-0.1 |
0.88 |
-0.1 |
0.0 |
0.20 |
9. This whiskey is more “manly” than other brands of whiskey. |
0.4 |
0.1 |
0.77 |
0.5 |
0.1 |
0.98 |
0.0 |
0.0 |
0.13 |
10. I prefer this whiskey to other brands of whiskey. |
0.5 |
0.1 |
0.85 |
-0.3 |
-0.2 |
0.58 |
1.5 |
0.5 |
2.46* |
*p <= .15, b is the unstandardized coefficient and Beta is the standardized coefficient
Multiple Regression Analysis: Knob Creek
The multiple regression analysis for Knob Creek reveals a low coefficient of multiple determinants (r 2), 25.3%, indicating that there is little to no relationship between how respondents rated the Knob Creek brand on the ten Likert items with regard to brand attributes and their favorability of the Knob Creek ad as indicated by their pre-post ad exposure change score. In layman terms, the Likert items account for only 25.3%, of the variance in the pre-post ad exposure change scores. The standard error of estimate (Se) is fairly high, at 2.8, indicating respondents were, on average, 2.8 units away from the regression line. With the statistically significant F ratio of 2.20, in 85 or more samples of every 100 samples drawn from the same population as this particular sample, we would expect to find a multiple regression coefficient of the same magnitude.
Only one of the Knob Creek brand attributes can be considered important, meaning that this particular attribute is the only one that can somewhat explain the variance in pre-post ad exposure change scores: (7) This whiskey’s packaging is unattractive. In 85 or more of 100 samples drawn from the same population as this sample, we would expect this particular brand attribute, as well as the constant term (a), to be able to be projected and we could expect to find results of the same magnitude.
Multiple Regression Analysis: Maker’s Mark
The multiple regression analysis for Maker’s Mark reveals a low coefficient of multiple determinants (r 2), 9.6%, indicating that there is little to no relationship between how respondents rated Maker’s Mark on the ten Likert items regarding brand attributes and their favorability of the Maker’s Mark advertisement as indicated by their pre-post ad exposure change score. In layman terms, the Likert items account for only 9.6 %, of the variance in the pre-post ad exposure change scores. The standard error of estimate (Se) is fairly high, at 2.1, indicating respondents were, on average, 2.1 units away from the regression line. These findings are not statistically significant and therefore cannot be projected onto the entire population.
None of the ten Maker’s Mark brand attributes can be considered important, and so they cannot explain the variance between pre-post ad exposure change scores relative to other brand attributes. None of these Likert questions’ responses can be projected onto the entire population.
Multiple Regression Analysis: Jack Daniel’s
The multiple regression analysis for Jack Daniel’s reveals a fairly low coefficient of multiple determinants (r 2), 31.0%, indicating that there is little to no relationship between how respondents rated Jack Daniel’s on the ten Likert items with regard to brand attributes and their favorability of the Jack Daniel’s advertisement as indicated by their pre-post ad exposure change score. In layman terms, the Likert items account for only 31.0% of the variance in the pre-post ad exposure change scores. The standard error of estimate (Se) is high, at 2.6, indicating respondents were, on average, 2.6 units away from the regression line. With the statistically significant F ratio of 2.92, in 85 or more samples of every 100 samples drawn from the same population as this particular sample, we would expect to find a multiple regression coefficient of the same magnitude.
Two of the Jack Daniel’s brand attributes can be considered important, meaning that these particular attributes can somewhat explain the variance in pre-post ad exposure change scores: (1) This is a good whiskey and (10) I prefer this whiskey to other brands of whiskey. In 85 or more of 100 samples drawn from the same population as this sample, we would expect these particular brand attributes to be able to be projected and we could expect to find results of the same magnitude.
Equation illustrating relationship:
The unstandardized coefficients (b) suggest the relationship between each Jack Daniel’s Likert item and how respondents favored the Jack Daniel’s print ad. The following formula explains how to interpret the unstandardized coefficients (b).
Regression Equation:
Pre-Post Ad Exposure Change Score for Jack Daniel’s= -5.0 - 0.4 (good) + 0.2 (time) - 0.1 (taste) - 0.2 (expensive) + 0.5 (serve) – 0.3 (smooth) + 0.1 (unattractive) – 0.1 (lifestyle) + 0.0 (manly) + 1.5 (prefer)
Suggested Relationship between Brand Attributes and Respondents Like/Dislike of Ad:
The unstandardized regression coefficient suggests the relationship between each brand attribute (Likert item) and how much each respondent likes the Jack Daniel’s ad. Below is listed how to interpret the unstandardized coefficients (b):
- The more respondents think that Jack Daniel’s is a good whiskey, the less they like the ad.
- The more respondents think Jack Daniel’s spends more time making the whiskey, the more they like the ad.
- The more respondents think Jack Daniel’s tastes better than other whiskeys, the less they like the ad.
- The more respondents think Jack Daniel’s is less expensive than other whiskeys, the less they like the ad.
- The more respondents would want to serve Jack Daniel’s to their friends and family, the more they like the ad.
- The more respondents think Jack Daniel’s goes down smoother than other whiskeys, the less they like the ad.
- The more respondents think Jack Daniel’s packaging is unattractive, the more they like the ad.
- The more respondents think Jack Daniel’s is suits his or her lifestyle, the less they like the ad.
- There is no connection between how “manly” respondents thought Jack Daniel’s was and how they liked the ad.
- The more respondents prefer Jack Daniel’s to other brands of whiskey, the more they like the ad.
Two of the ten brand attributes can be considered significant, meaning they are the attributes that best explain the variance in pre-post ad exposure change scores relative to the other brand attributes: (1) This is a good whiskey and (10) I prefer this whiskey to other brands of whiskey. In 85 or more of 100 samples drawn from the same population as this sample, we would expect these particular brand attributes to be able to be projected and we could expect to find results of the same magnitude.
Top
DISCRIMINANT ANALYSIS
Discriminant Analysis (Maker’s Mark)
A multiple discriminant analysis was conducted for two groups, those who had positive and negative pre-post ad exposure change scores for Maker’s Mark whiskey. Respondents whose pre-post ad exposure scores did not change were not included in this analysis. The categorical dependent variable in this analysis is group membership and the independent variables are the brand attributes as indicated by the Likert items in the questionnaire. This analysis was conducted to determine whether or not we can accurately predict, with a better than chance accuracy, the up-movers (those with positive change scores) from the down-movers (those with negative change scores) according to their brand attribute (Likert item) responses. While 76 participants completed the survey, only 42 will be used in this analysis.
Brand Attributes (Likert Items) |
Moved UP |
Moved DOWN |
Standardized
Discriminatnt
Coefficient |
Unstandardized
Discriminat
Coefficient |
Mean |
Standard
Deviation |
Mean |
Standard
Deviation |
1. Maker’s Mark is a good whiskey. |
0.8 |
1.6 |
0.5 |
1.3 |
0.1 |
0.1 |
2. More time is spent making Maker’s Mark than any other. |
2.9 |
0.8 |
3.3 |
0.7 |
-1.0 |
-1.4 |
3. Maker’s Mark tastes better than other brands of whiskey. |
3.5 |
0.7 |
3.4 |
0.9 |
1.3 |
1.7 |
4. Maker’s Mark is less expensive than other brands of whiskey. |
2.7 |
0.9 |
2.5 |
0.9 |
0.1 |
0.1 |
5. I would want to serve Maker’s Mark to my friends and family. |
3.7 |
0.8 |
3.7 |
0.9 |
0.1 |
0.2 |
6. Maker’s Mark goes down smoother than other whiskey. |
3.1 |
0.7 |
3.3 |
0.7 |
0.4 |
0.6 |
7. Maker’s Mark packaging is unattractive. |
3.9 |
0.9 |
3.7 |
1.3 |
0.1 |
0.1 |
8. Maker’s Mark does not suit my lifestyle. |
3.2 |
0.7 |
3.2 |
1.1 |
-0.3 |
-0.3 |
9. Maker’s Mark is more “manly” than other brands of whiskey. |
3.0 |
0.6 |
3.0 |
0.8 |
0.5 |
0.7 |
10. I prefer Maker’s Mark to other brands of whiskey. |
3.3 |
0.8 |
3.5 |
0.9 |
-1.3 |
-1.5 |
Based on total sample size n=42 |
n=25 |
n=17 |
|
When looking at the mean scores of each Likert item and comparing these scores between up movers and down movers, there are no items that have dramatically different mean scores. Thus, there does not seem to be much discrimination between those whose opinions moved up and down, respectively, after viewing the Maker’s Mark ad.
The largest differences between mean scores for up-mover and down-mover groups was on item 2(“More time is spent…”). The magnitude of change for this Likert item is not great (0.4), but it reveals the largest difference between group mean scores.
The unstandardized discriminant coefficients suggest the relationship between each Likert item and how much the respondents like the Maker’s Mark ad. They can be interpreted as follows:
- The more respondents think Maker’s Mark tastes better than other brands, the more they like the ad.
- The more respondents think is more “manly” than other brands, the more they like the ad.
Group Centroids
Wilk's Lambda |
Chi-Squared |
Df |
0.79 |
8.11 |
10 |
p<=.15
Change Score Movement for Maker’s Mark |
Average z Score |
Up |
0.4 |
Down |
-0.6 |
The group centroids (average z score for each of the two groups) are not significant (0.4 for the 'up' movers and -0.6 for the 'down' movers) due to the lack of significance found for Wilks' Lambda at 0.79. As Wilks Lambda is statistically insignificant, there is not a significant difference between the Group Centroids of respondents who like the ad and who do not like the ad. Therefore, in 85 or more samples out of every 100 drawn from the same population as this sample of 44, we would not expect to find group centroids of the same magnitude. The Chi-Square is also not significant and shows a lack of a relationship between the 'up' and 'down' movers and the brand attributes. In 85 or more samples of 100 samples taken from the same population as this sample of 42, we would not expect to find a relationship and Chi-Squared of the same magnitude.
|
Predicted Up Movers |
Predicted Down Movers |
Total |
Actual Up Movers |
21 |
4 |
25 |
Actual Down Movers |
8 |
9 |
17 |
Classification Matrix 71.4% of original grouped cases correctly classified
to = (0.714 – 0.5)/√[(0.714*(1-0.714)/42) + (0.5*0.5)/42]
t o = 2.06*
* p ≤ 0.15
The classification matrix shows that 71.4% of the original grouped cases are predicted correctly. Twenty one of the 25 predicted ‘up’ movers actually moved up, and 8 out of the 17 'down' movers were correctly predicted to move down. The observed t-ratio of 2.06 is statistically significant, as it is greater than or equal to 1.04. As it is significant, in 85 out of 100 people of the similar population as the 42 respondents, these numbers could be projected and we could expect to find a relationship of a similar magnitude.
Conclusions
The t-ratio to test the significance of classification was the only number that was of significance and could be projected upon a population of 85 out of 100 people similar to the 42 respondents tested. The relationship, if projected, would be of a similar magnitude.
The rest of the discriminant analysis results, based on this sample of 42 respondents, cannot be projected to the entire population. In other words, we can only project with 71.4% accuracy whether or not people exposed to the ads will be in favor of it or not, using positive pre-post ad exposure change scores.
Top
ANOVA/MANOVA
Using the Likert item "Maker’s Mark is less expensive than other brands of whiskey" as a dependent variable, the two-way Factorial ANOVA test will try to determine the relationship between the dependent variable and two independent variables: 1) the estimated annual income of the respective respondents, and 2) the change score of constant sum scale (up, same, and down movement in Pre-Post Ad Exposure). When using a factorial design, we can draw conclusions regarding effects of the independent variables separately, as well as the combined effect of the independent variables. The MANOVA test will then take all ten Likert items as dependent variables for the same test. These data were obtained via an online questionnaire, and 76 people participated in the study.
ANOVA: Pre-Post Ad Exposure Change Score / Income - Cost
Pre-Post Ad Exposure Change Score |
$30,000 per year and below |
Above $30,000 per year |
TOTAL |
Positive (Up-Movers) |
mean=2.8
standard deviant (s.d.)=1.0
n=9 |
mean=2.6
s.d.=0.8
n=16 |
mean=2.7
s.d.=0.9
n=25 |
Same (No Change) |
mean=2.3
s.d.=0.5
n=21 |
mean=2.5
s.d.=0.7
n=13 |
mean=2.4
s.d.=0.6
n=34 |
Negative (Down-Movers) |
mean=2.3
s.d.=1.3
n=4 |
mean=2.5
s.d.=0.8
n=13 |
mean=2.5
s.d.=0.9
n=17 |
TOTAL |
mean=2.4
s.d.=0.8
n=34 |
mean=2.6
s.d.=0.7
n=42 |
mean=2.5
s.d.=0.8
n=76 |
|
Sum of Squares |
Degree of Freedom |
Mean Square |
F-ratio |
Change Score Movement(Up/Down/Same) |
1.17 |
2 |
0.59 |
1.00 |
Income Level(Above/Below $30000 per year) |
0.19 |
1 |
0.19 |
0.32 |
Change Score Movement x Income Level |
0.57 |
2 |
0.28 |
0.48 |
Error |
41.18 |
70 |
0.59 |
|
*p<=0.15
ANOVA: Pre-Post Ad Exposure Change Score / Income - Cost
There is little difference between the mean scores on the cost of Maker’s attribute (Likert item) between those who make less than $30,000 a year (mean score of 2.4) and those who make more (mean score of 2.6). There was also very little difference among the pre-post ad exposure change score groups, as the down-movers mean score was 2.5, the up-movers had a mean score of 2.7, and the same (no change) group had a mean score of 2.4. The total mean for the entire sample was 2.5. All groups seem to be similarly sensitive to the cost of Maker’s Mark, as their mean scores are respectively similar regardless of income level. Perhaps this means that people drink Maker’s Mark because of the whiskey itself, regardless of any other circumstances.
Based on a sum of squares to test for the significance of pre-post ad exposure change score, income level, and the relationship between change score and income level, we can come to the following conclusion: We cannot project the mean scores among income level (above/below $30,000) in this sample to the entire population, as none of these numbers were found to be significant.
Two-way Factorial MANOVA
MANOVA: Pre-Post Ad Exposure Change Score / Income Level—All Brand Attributes
Brand Attribute
(Likert Item) |
$30,000 a year or below |
Above $30,000 a year |
(n=42) |
(n=34) |
Up |
Same |
Down |
Up |
Same |
Down |
(n=16) |
(n=13) |
(n=13) |
(n=9) |
(n=21) |
(n=4) |
mean |
s.d. |
mean |
s.d. |
mean |
s.d. |
mean |
s.d. |
mean |
s.d. |
mean |
s.d. |
1. Maker’s Mark is a good whiskey. |
0.5 |
1.4 |
0.5 |
1.3 |
0.4 |
1.3 |
1.2 |
1.9 |
0.0 |
0.0 |
1.0 |
2.0 |
2. More time is spent making Maker’s Mark than any other. |
2.7 |
0.5 |
3.0 |
0.6 |
3.3 |
0.8 |
3.3 |
1.0 |
3.1 |
0.5 |
3.2 |
0.5 |
3. Maker’s Mark tastes better than other brands of whiskey. |
3.4 |
0.6 |
3.4 |
0.8 |
3.5 |
0.7 |
3.8 |
0.8 |
3.6 |
0.8 |
3.0 |
1.4 |
4. Maker’s Mark is less expensive than other brands of whiskey. |
2.6 |
0.8 |
2.5 |
0.7 |
2.5 |
0.8 |
2.8 |
1.0 |
2.3 |
0.6 |
2.2 |
1.3 |
5. I would want to serve Maker’s Mark to my friends and family. |
3.6 |
0.7 |
3.5 |
0.9 |
3.8 |
0.8 |
4.0 |
0.9 |
3.5 |
1.2 |
3.2 |
1.0 |
6. Maker’s Mark goes down smoother than other whiskey. |
3.0 |
0.4 |
3.2 |
0.9 |
3.2 |
0.7 |
3.3 |
1.0 |
3.2 |
1.0 |
3.5 |
0.6 |
7. Maker’s Mark packaging is unattractive. |
4.1 |
0.6 |
3.7 |
0.6 |
3.9 |
1.1 |
3.5 |
1.2 |
3.6 |
0.9 |
3.0 |
1.8 |
8. Maker’s Mark does not suit my lifestyle. |
3.1 |
0.8 |
3.0 |
1.0 |
3.4 |
1.2 |
3.3 |
0.5 |
3.7 |
0.8 |
2.8 |
1.0 |
9. Maker’s Mark is more “manly” than other brands of whiskey. |
3.0 |
0.4 |
2.9 |
0.5 |
3.0 |
0.9 |
2.9 |
0.8 |
2.9 |
0.7 |
2.8 |
0.5 |
10. I prefer Maker’s Mark to other brands of whiskey. |
3.3 |
0.7 |
3.2 |
1.1 |
3.6 |
0.8 |
3.4 |
1.0 |
3.8 |
0.9 |
3.3 |
1.5 |
|
Wilks’ Lambda |
F-Ratio |
Change Score Movement(Up/Down/Same) |
0.74 |
0.97 |
Income Level(Above/Below $30,000 per year) |
0.86 |
1.02 |
Change Score Movement x Income Level |
0.71 |
1.12 |
* p<=0.15
Based on the Wilks’ Lambda to test for the significance of pre-post ad exposure change score, income level, and the relationship/interaction between change score and income level, we can determine that we cannot project the mean scores of the same magnitude among pre-post ad exposure change score groups (up/same/down) in this sample to the entire population. We can also not project the mean scores among income level in this sample to the entire population. Lastly, we cannot project the mean scores of the same magnitude among pre-post ad exposure change score / income levels in this sample to the entire population.
When looking at each brand attribute with each independent variable (income level and pre-post ad exposure change score, respectively) and the interaction between the two, there are some significant relationships worth noting. In each of the stated relationships below, in 85 or more samples out of every 100 drawn from the same population, we would expect to find results of the same magnitude as this study:
For Pre-Post Ad Exposure Change Score:
None.
For Income Level:
(7) Maker’s Mark packaging is unattractive.
For Interaction/Relationship Between Change Score and Income Level: (1) Maker’s Mark is a good whiskey, and (8) Maker’s Mark does not suit my lifestyle.
Top
FACTOR ANALYSIS
A factor analysis was performed for Knob Creek, Maker’s Mark and Jack Daniel’s whiskeys to place related brand attributes into independent groups (or factors) for the calculation of a brand attitude score. The ten brand attributes (Likert items) variables are analyzed for relationships and new composite factors are created to identify related characteristics of the variables. A paired t-test was then performed to find any potential statistical significance for the differences of means between brand attitude scores. The results were obtained via an online questionnaire, 76 people participated in the study.
Knob Creek |
Maker’s Mark |
Jack Daniel’s |
Factor |
Eigenvalue Total |
% of Variance |
Cum. Var. |
Eigenvalue Total |
% of Variance |
Cum. Var. |
Eigenvalue Total |
% of Variance |
Cum. Var. |
I |
3.5 |
34.7 |
34.7 |
4.2 |
41.7 |
41.7 |
3.9 |
39.0 |
39.0 |
II |
1.3 |
13.5 |
48.1 |
1.4 |
13.6 |
55.3 |
1.5 |
14.5 |
53.5 |
III |
1.2 |
12.0 |
60.1 |
1.0 |
10.1 |
65.4 |
1.2 |
11.8 |
65.3 |
IV |
0.8 |
8.1 |
68.3 |
0.8 |
8.3 |
73.7 |
1.0 |
10.3 |
75.6 |
V |
0.7 |
7.0 |
75.2 |
0.7 |
7.4 |
81.1 |
0.7 |
7.1 |
82.7 |
VI |
0.6 |
6.5 |
81.7 |
0.6 |
6.2 |
87.3 |
0.6 |
6.3 |
88.9 |
VII |
0.6 |
6.0 |
87.7 |
0.5 |
5.2 |
92.4 |
0.4 |
3.8 |
92.7 |
VIII |
0.5 |
4.8 |
92.5 |
0.4 |
3.7 |
96.2 |
0.3 |
3.3 |
96.0 |
IX |
0.4 |
4.2 |
96.7 |
0.3 |
2.5 |
98.7 |
0.2 |
2.3 |
98.3 |
X |
0.3 |
3.3 |
100.0% |
0.1 |
1.3 |
100.0% |
0.2 |
1.7 |
100.0% |
Total Variance Explained: Knob Creek, Maker’s Mark, Jack Daniel’s
There are three factors out of ten (Factors I, II, and III) for each whiskey that had Eigenvalues greater than one. For all three brands, the ‘good,’ ‘time’ and ‘taste’ factors (These are good whiskeys respectively, more time is spent making these whiskeys respectively and these whiskeys taste better than the other to which they are compared, respectively) are loaded on Factors I-III. Thus, results loaded in each respective Factor are in some way related to one another independent of all other factors. These ten items were also used in computing the attitude scores for each brand of whiskey.
For the Knob Creek, 34.7% of the variance can be explained by Factor I (“Knob Creek is a good whiskey”), while 65.3% is unexplained or lost variance. 13.5% of the variance can be explained by Factor II (“More time is spent making Knob Creek than other whiskeys”), while 86.5% is unexplained or lost variance. 12.0% of the variance can be explained by Factor III (“Knob Creek tastes better than other whiskeys”), while 88.0% is unexplained or lost variance.
For the Maker’s Mark, 41.7% of the variance can be explained by Factor I (“Maker’s Mark is a good whiskey”), while 58.3% is unexplained or lost variance. 13.6% of the variance can be explained by Factor II (“More time is spent making Maker’s Mark than other whiskeys”), while 86.4% is unexplained or lost variance. 10.1% of the variance can be explained by Factor III (“Maker’s Mark tastes better than other whiskeys”), while 89.9% is unexplained or lost variance.
For the Jack Daniel’s, 39.0% of the variance can be explained by Factor I (“Jack Daniel’s is a good whiskey”), while 61.0% is unexplained or lost variance. 14.5% of the variance can be explained by Factor II (“More time is spent making Jack Daniel’s than other whiskeys”), while 85.5% is unexplained or lost variance. 10.3% of the variance can be explained by Factor III (“Jack Daniel’s tastes better than other whiskeys”), while 89.7 % is unexplained or lost variance.
Attitude Scale & Paired t-tests
Attitude Scale |
|
Paired T-Tests for Attitude Scores |
|
Mean Attitude Score |
Sample Size |
Standard Deviation |
|
|
Mean |
Standard Deviation |
t |
Knob Creek |
2.9 |
76 |
0.4 |
|
Knob-Maker’s |
0.0 |
0.1 |
-1.34 |
Maker’s Mark |
2.9 |
76 |
0.4 |
|
Knob-Jack Daniel’s |
-0.4 |
0.8 |
-4.09* |
Jack Daniel’s |
3.3 |
76 |
0.6 |
|
Maker’s-Jack Daniel’s |
-0.4 |
0.8 |
-3.93* |
n=71 |
|
* p ≤ 0.15 |
The mean attitude score for Jack Daniel’s was 3.3, while the mean attitude scores for Knob Creek and Maker’s Mark were lower, at 2.9. From this, we can infer that Jack Daniel’s has the higher brand perception than the other two whiskeys. It is important to note, however, that this difference in mean attitude scores is still slight, and so the perception that Jack Daniel’s is the better brand is only of a semi-important magnitude.
Based on the results from a paired t-test to test for the significance of the mean attitude scores, we can make the following statements:
1) In 85 or more out of 100 samples drawn from the same population as this one, we would expect to find the same magnitude of difference between the Knob Creek and Jack Daniel’s mean attitude scores
2) In 85 or more out of 100 samples drawn from the same population as this one, we would expect to find the same magnitude of difference between the Maker’s Mark and Jack Daniel’s mean attitude scores.
3) From the results of this analysis, we cannot predict with confidence or accuracy that we would find the same magnitude of difference between the Knob Creek and Maker’s Mark attitude scores.
Top |
