ANOVA/MANOVA
This write-up reports the results of a two-way ANOVA analysis that was conducted to determine the relationship between one dependent variable and two independent variables. In this case the dependent variable is the Likert item “Dasani flavored water fits my lifestyle” and the independent variables are gender (male or female) and the move score (up-, same and down-movers). In addition, MANOVA, a multivariate analysis of variance, was conducted to assess the relationship between the same two independent variables gender and move score and all ten Likert items as dependent variables. The data was compiled from an online survey that generated 82 respondents.
Table 1: Descriptive Statistics (ANOVA analysis)
Gender |
Move Score Direction |
||
|
Up |
Same |
Down |
Male |
Mean: 3.3 Std. Deviation: 0.5 Sample size: 4 |
Mean: 3.4 Std. Deviation: 0.5 Sample size: 7 |
Mean: 3.1 Std. Deviation: 0.7 Sample size: 17 |
Female |
Mean: 3.5 Std. Deviation: 0.8 Sample size: 13 |
Mean: 3.4 Std. Deviation: 1.1 Sample size: 15 |
Mean: 3.4 Std. Deviation: 0.8 Sample size: 26 |
As shown in table 1, women have slightly higher mean move scores than men in the up- and down-mover groups. In the same-group, men and women have the same mean move scores. This conveys that, after viewing the ad, women are more likely than men to either agree (up-movers) or disagree (down-movers) with the statement “Dasani flavored water fits my lifestyle.” However, as the table 1 shows, the vast majority of both males and females are Down-Movers (17 males and 26 females) suggesting that most respondents did not agree with the statement " Dasani flavored water fits my lifestyle” after seeing the ad.
Table 2: ANOVA
|
Sum of Squares |
Degrees of Freedom |
Mean Square |
F-ratio |
Move Score (Between Groups) |
0.29 |
2 |
0.14 |
0.23 |
Gender (Between Groups) |
0.38 |
1 |
0.38 |
0.60 |
Move x Gender (Between Groups) |
0.36 |
2 |
0.18 |
0.29 |
* p≤.15
As table 2 indicates, the F-ratios are not significant and therefore results cannot be projected to a larger population. In 85 samples out of 100 samples drawn from the same population as this sample, we cannot expect the same results.
Table 3: Descriptive Statistics (MANOVA analysis)
|
Male |
Female |
||||||||||
|
n=28 |
n=54 |
||||||||||
Brand Attribute (Likert item) |
Up |
Same |
Down |
Up |
Same |
Down |
||||||
|
n=4 |
n=7 |
n=17 |
n=13 |
n=15 |
n=26 |
||||||
|
Mean |
Std. Dev. |
Mean |
Std. Dev. |
Mean |
Std. Dev. |
Mean |
Std. Dev. |
Mean |
Std. Dev. |
Mean |
Std. Dev. |
…good brand … |
3.3 |
0.5 |
3.6 |
0.8 |
3.8 |
0.5 |
3.6 |
0.9 |
3.6 |
1.1 |
3.9 |
0.7 |
…makes me feel healthy. |
3.0 |
0.8 |
3.3 |
0.5 |
3.2 |
0.8 |
3.5 |
0.9 |
2.9 |
0.5 |
3.2 |
0.8 |
…tastes great. |
3.0 |
0.8 |
2.7 |
1.3 |
3.3 |
0.7 |
3.5 |
0.7 |
3.4 |
1.0 |
3.6 |
0.7 |
…brand I trust. |
3.5 |
0.6 |
3.9 |
0.7 |
3.7 |
0.6 |
3.5 |
1.2 |
3.4 |
1.1 |
3.6 |
0.8 |
…fits my lifestyle. |
3.3 |
0.5 |
3.4 |
0.5 |
3.1 |
0.7 |
3.5 |
0.8 |
3.4 |
1.1 |
3.4 |
0.8 |
…is too expensive. |
2.3 |
1.0 |
2.6 |
1.0 |
2.2 |
0.9 |
2.9 |
0.5 |
2.7 |
0.6 |
2.8 |
0.8 |
…is just like any other… |
3.3 |
1.0 |
3.7 |
0.8 |
3.4 |
0.5 |
2.6 |
0.9 |
2.9 |
1.2 |
3.2 |
0.9 |
…would recommend to my friends. |
2.8 |
1.0 |
3.1 |
0.7 |
3.1 |
0.6 |
3.1 |
1.0 |
3.1 |
0.9 |
3.1 |
0.7 |
…prefer over regular water. |
2.5 |
1.3 |
3.3 |
0.8 |
2.7 |
0.8 |
2.8 |
1.1 |
2.8 |
1.3 |
2.6 |
1.0 |
…doesn't taste artificial. |
2.8 |
0.5 |
3.0 |
0.8 |
3.1 |
0.7 |
3.0 |
0.7 |
2.9 |
1.2 |
2.9 |
0.8 |
Table 3 reflects descriptive statistics for pre-post exposure move scores for males and females in relation to all ten Likert items. Among Up-Movers, there were notable differences between the mean scores of males and females on three of the ten brand attributes. Post ad exposure, more women believed Dasani flavored water makes them feel healthy and tastes great. On the other hand, considerably more men stated that Dasani flavored water is just like any other flavored water after viewing the ad. Among those respondents whose change scores remained the same, there were two notable difference between males and females; females felt more strongly in terms of Dasani flavored water tasting great, whereas men again felt strongly about Dasani flavored water being just like any other flavored water.
The only notable difference in mean scores among male and female Down-Movers was that women felt more strongly about Dasani flavored water being too expensive.
Table 4: MANOVA
|
Wilks' Lambda |
F-ratio |
Degrees of Freedom |
Move Score (Between Groups) |
0.74 |
1.08 |
20 |
Gender (Between Groups) |
0.80 |
1.68* |
10 |
Move x Gender (Between Groups) |
0.83 |
0.67 |
20 |
* p≤.15
When comparing the ten brand attributes to the independent variables gender and move score, the F-ratio for gender was found to be significant and can be projected to the population. In 85 (or more) of 100 samples drawn from the same population as this sample, we can expect the same results as in this case.