| Conclusions
Basic statistics
Paired t-tests
Based on the Brand Index scores, respondents perceive Corona and Dos XX most positively (each having virtually the same Brand Index Score). Negra Modelo’s was perceived slightly less positively than the leading two brands. This would indicate that while Corona and Dos XX are the favorites, the respondents had fairly similar judgments about each brand.
Independent t-tests
In this sample, the brand perception of Corona was the same for both up movers and down movers. The favorable perception of the Corona advertisement however was greater for up movers than down movers. It was also found that the higher perception of Corona's advertising can be projected to the population from which the sample was drawn, but the brand perception of Corona cannot.
Chi-squared test
The Chi-Square show that these results could not projected in 85 or more samples of every 100 samples drawn from the same population.
Frequency count
The respondents purchase intention after exposure to the ads, either increased, stayed the same, or decreased. For the case of Corona and Dos XX the respondents’ change in purchase intention was a fairly evenly distributed between up movers, down movers, and non movers. In contrast, Negra Modelo, had a very low number of up movers (compared to Corona's 22 and Dos XX's 24).
Comparison of brand perceptions
Out of 10 judgment statements about each brand, 39 respondents had scores, which ranked higher for Dos XX than for Corona. Therefore it can be inferred that 39 people prefer Dos XX over Corona.
Correlation
The Brand Index Scores for Corona and Dos XX were shown to have a statistically significant low positive correlation. Thus In 85 samples out of 100 samples taken from the same population as this sample, it would be expected that the correlation between the brand index scores of Corona and Dos XX would be about what it is in this sample. The results of this sample can be projected to the population from which the sample was drawn.
Paired t-tests for female respondents
A statistically significant medium to low positive correlation was found between the brand index scores of the female respondents for Corona and Dos XX. In 85 samples out of 100 samples taken from the same population as this sample, it would be expected that the correlation between the brand index scores of Corona and Dos XX would be about what it is in this sample. The results of this sample can be projected to the population from which the sample was drawn. Compared to the unisex data in table 8 of question 6, there is a slightly stronger correlation between Corona’s and Dos XX’s brand index scores for the female respondents.
Back To Top
Multiple regression analysis
The multiple regression analysis for the Corona brand reveals a very low multiple determination coefficient (R2=12.0%), while Dos XX and Negra Modelo had relatively higher scores (23.9% and 27.6% respectively). This percentage indicates the level of relationship between the Brand Index Score and the amount of change pre-to-post ad exposure. In other words this percentage illustrates the connection between how the respondents view these brands and how they view the brands’ respective ads. . The results for Dos XX and Negra Modelo appear to be statistically significant thus in 85 samples of 100 Samples drawn from the same population as this sample it would be expected that the multiple correlation coefficients for these brands (0.5 and 0.5) would be about what they are in this sample. The results for these two brands can be projected to the population from which this sample was drawn.
This analysis shows that Corona's brand perceptions played a very small role in the variations between respondents’ changed purchase intentions after exposure to the advertisements whereas Dos XX and Negra Modelo had a relatively larger impact. In addition, for corona eight of the ten brand attributes were found to be ‘important’ compared to six of ten for both Dos XX and Negra Modelo
Back To Top
Discriminant analysis
The analysis of the mean scores of each brand attributes show little difference between the “Up-movers” and the “Down-movers”. Although the largest difference between mean scores was .5, the difference was not great. Thus there appear to be few if any great difference between the two groups’ perceptions of the Corona brand. Looking at the Standardized Discriminant Coefficients, the two brad attributes “Corona is refreshing” and “Corona has a distinctive taste” appear to be the most important in determining whether respondents were up-movers or down movers.
A Wilks Lambda test was converted into Chi-squared and was then shown to be not statistically significant. Thus in 85 or more samples out of every 100 samples drawn from the same population as this sample, we would not expect to find the same magnitude of difference between the two groups (up and down-movers) based on the 10 brand attributes. The group centroids for Corona cannot be projected to the population from which the sample was drawn.
Only 28 out of 43 respondents (65.1%) fall in their predicted groups (up movers or down movers). A 65.1% classification accuracy is not much better than chance, and thus prediction accuracy of which group a respondent will fall in is relatively low. The results were however found to be statistically significant and thus can be projected to the population from which the sample was drawn.
Back To Top
Analysis of variance
In this sample there is no notable relationship between gender and how the groups (up movers and down movers) reaction to the statement “Corona is refreshing”. There is however a mild difference between the means for up-movers and down-movers suggesting that there might be a slight relationship between these mover groups and their response to the statement “Corona is refreshing”. However, it is important to note, none of these responses are statistically significant and thus cannot be projected to the sample from which they were drawn.
Back To Top
Factor Analysis
For Negra Modelo, “Negra Modelo is a good imported beer”, “Negra Modelo is refreshing”, “Negra Modelo has a good taste”, “Negra Modelo is a cool/trendy beer”, “I prefer Negra Modelo over other imported beers”, and “Negra Modelo has a distinctive taste” all loaded on factor I. “Negra Modelo is too expensive” loaded on factor II and “Negra Modelo is a masculine beer” and “Negra Modelo is a well established brand loaded on factor III. “Negra Modelo is a premium beer” loaded ambiguously.
For Dos XX, “Dos XX is a good beer”, “Dos XX is refreshing”, “Dos XX is a premium beer”, “I prefer Dos XX over other imported beers”, and “Dos XX has a distinctive taste” all load on factor I. “Dos XX is a masculine beer” and “Dos XX is a well established brand” loaded on factor II, and “Dos XX is too expensive” loaded on factor III. “Dos XX is a cool/trendy beer” and “Dos XX has a good taste” loaded ambiguously.
For Corona, “Corona is a good beer”, “Corona is refreshing”, “Corona has a good taste”, “Corona is a Premium beer”, , “I prefer Corona over other imported beers”, and “Corona has a distinctive taste” all load on factor I. “Corona is too expensive” loaded on factor II and “Corona is a well established brand” loaded on factor III. “Corona is a cool/trendy beer” and “Corona is a masculine beer” loaded ambiguously.
The attitude score reveals that respondents evaluated Dos XX highest with a score for 3.8 followed by Corona and then Negra Modelo with scores 3.7 and 3.1 respectively. Over all there was very little difference between Corona and Dos XX and thus would suggest that people like these brands about the same. The difference in scores between Corona and Negra Modelo, and Dos XX and Negra Modelo are statistically significant and thus can be projected to the population from which the sample was drawn.
Back To Top
Cluster Analysis
When compared to Cluster 2, Cluster 1 exhibits a higher mean score for every brand attribute. This demonstrates that respondents in Cluster 1 rated and thus liked Negra Modelo better than the respondents in Cluster 2. The F ratios indicate that all the mean scores for the 10 Likert items, with the exception of “Negra Modelo is refreshing” and “Negra Modelo”, is a masculine beer”, are significant meaning that in 85 or more samples out of every 100 samples drawn from the same population as this sample of 64 people, it would be expected that the mean scores would be the same magnitude as in this sample.
Of the two clusters, cluster 2 was found to be female-oriented (by double) and cluster 1 male-oriented. These results are statistically significant, thus can be used to describe a larger population. In 85 (or more) of 100 samples drawn from the same population as this sample, we would expect to find the same results.
Back To Top
|