Consumer Preference Analysis

 

Analysis

Regression Analysis

ANOVA MANOVA Factor Analysis Cluster Analysis

 

Basic Statistics

 

1. Paired t-test comparing Brand Index Scores for Nature Valley, Quaker and Soyjoy

Table 1-A. Brand Index Score

Brand

Brand Index Score Mean

Standard Deviation

Sample Size

Nature Valley

33.2

4.1

63

Quaker

33.0

5.4

Soyjoy

32.3

5.1

There are small differences of brand index score mean among three brands. Even though Nature Valley has slightly higher brand index score mean than other two brands, sixty-three respondents reported that the brand index score mean values of three brands are not significantly different.

Table 1-B. Paired Sample t-test

Pairs

t-ratio

Nature Valley and Quaker

.28

Quaker and Soyjoy

.80

Soyjoy and Nature Valley

1.12*

*p ≤ .15

Paired sample t-tests are conducted in order to determine whether the results can be appropriately projected to a large population. One pair, Soyjoy and Nature Valley, is statistically significant, it means that in 85 or more samples out of every 100 samples drawn from the same population as this sample, it would be expected that the mean brand index score for Soyjoy and Nature Valley would be about what they are in this sample.

 

2. Between group t-tests comparing Soyjoy up and down movers

Table 2-A. Brand Index Score and Ad Index Score Change for Soyjoy

Group

Sample
(n)

Brand Index Score

Ad Index Score

Mean

Standard Deviation

t-ratio

Mean

Standard Deviation

t-ratio

Move Up

33

34.8

4.7

3.89*

6.3

1.8

4.66*

Move Down

11

29.4

3.8

2.9

2.2

*p ≤ .15

 Table 2-A shows that the mean brand index score and ad index score for Soyjoy for the up and down movers on the constant sum score are statistically significant and can be projected to the population. It would be expected for the mean brand index score and ad index score for Soyjoy for the up and down movers on the constant sum score would be what they are in this sample.

 

3. Chi-squared significance test for Soyjoy

Table 3-A. Median Change for Soyjoy

Pre-Post Change

Above Median

Below Median

Up

Count

20

13

%Row

60.6%

39.4%

% Column

69.0%

38.2%

Total

31.7%

20.6%

Same

Count

6

13

%Row

31.6%

68.4%

% Column

20.7%

38.2%

Total

9.5%

20.6%

Down

Count

3

8

%Row

27.3%

72.7%

% Column

10.3%

23.5%

Total

46.0%

54.0%

Median: 32, Sample Size: 63

Table 3-B. Chi-Squared Test

 

Value

Chi-squared

5.98*

*p ≤ .15

This chart indicates that 29 of the respondents had above median Brand Index Score in comparison with 34 below median. To demonstrate, the reading of the first crosstab would show that of these 63 respondents, 20 respondents were: both up-movers and above median in their Brand Index Scores, accounted for 69.0% of all of those above median, accounted for 60.6% of all post-ad up-movers, and accounted for 31.7% of the total sample. The other cells read the same. The Chi-squared test indicates that these results are significant at the 0.15 level and that in 85 or more samples out of every 100 samples drawn from the same population as this sample, the same results for the Soyjoy ad could be expected.

4. Pre to Post Exposure Brand Index Score Change

4-A. Pre-Post Ad Exposure Movement

Brand

Sample Size

Up

Same

Down

Nature Valley

63

12

15

36

Quaker

34

13

16

Soyjoy

33

19

11

This frequency test was conducted in order to determine whether the ads used in the survey increased positive feelings towards any of the three brands. The two most positively increasing ads were Quaker and Soyjoy; where as, the Nature Valley ad had a majority of down-movers and those that stayed the same.

 

5. Brand Preference – Frequencies

Table 5-A. Higher Brand Index Score Frequency

 

Frequency

Percent (%)

Nature Valley > Quaker

29

46.0%

In this analysis, respondents’ brand index scores for Nature Valley were compared to those for Quaker. Out of 63 total respondents, 29 had a higher brand index score for Nature Valley than Quaker.

 

6. Simple Correlation Coefficients

Table 6-A. Correlation Coefficients for Compared Brands

Compared Brand

Correlation Coefficient

Sample Size

Quaker and Soyjoy

.2*

63

*p ≤ .15

There is a weakly positive correlation between respondents who like Quaker and Soyjoy. In 85 or more samples out of 100 samples drawn from the same population of this sample, it would be expected that there to be a weak direct relationship between two brands as there are in this sample. In terms of brand index scores, when the Quaker brand’s score increases or decreases in a certain direction, Soyjoy’s brand index score tends to follow the same pattern. This test was drawn from a sample size of 63 respondents.

 

7. Simple Correlation Coefficients for Selected Cases

Table 7-A. Correlation Coefficients for Compared Brands (Amongst male respondents)

Compared Brand

Correlation Coefficient

Sample Size

Quaker and Soyjoy

.2*

28

*p ≤ .15

In 85 or more out of 100 samples drawn from the same population as this sample, the same results for the correlation between brands could be expected for Quaker and Soyjoy. A weak correlation of 0.2 was found to exist between the Quaker and Soyjoy brands of snack bars, which suggests that males who like the Quaker brand might slightly tend to like the Soyjoy brand too and vice versa. The sample was taken from 28 respondents.

 

Top

Next

 

ADV 380J Qualitative and Quantitative Research, Fall 2007