Women’s Fragrance: A Brand Preference Report
by Danae Manika
This is a comprehensive consumer preference analysis that includes copy testing research for print ads of three brands of women’s fragrance: Inspiration by Lacoste, Ralph Lauren Romance and Euphoria by Calvin Klein. Several statistical techniques were used for the analysis and will be discussed in detail on the following pages.
This website is a course requirement of Professor Leckenby's Fall 2006 Advertising 380J graduate seminar: Quantitative and Qualitative Research at the University of Texas at Austin.
|
| Executive Summary | Introduction | Conclusions | Summary | Appendix A | Appendix B |
Executive Summary
The purpose of this survey was to determine brand preference for three brands of women’s fragranceusing copy testing. The three brands examined were Inspiration by Lacoste, Ralph Lauren Romance and Euphoria by Calvin. This project determined initial brand preference, analyzed common attributes associated with each brand, and then sought to find the effect of print ads. The original survey can be found in Appendix B. Respondents for the survey were solicited through email, and asked to fill out an online form. The results were collected in database format, and then analyzed using the statistics software program SPSS. Sixty people completed the survey, and there was no control group. Descriptive statistics for each variable measured are included on the Survey in Appendix A. A number of different analyses were conducted in SPSS: basic statistics (independent samples and paired t-tests, frequencies, and correlations), multiple regression analysis, discriminant analysis, analysis of variance (ANOVA/MANOVA) and factor analysis. Statistical significance throughout (indicating that in 85 or more samples drawn from the same population as this sample, the expected values would be the same as those reported here) was set at p < .15. Throughout the analysis, the three brands were significantly different from one another. The tests revealed a significantly more positive attitude towards Ralph Lauren Romance, followed by Calvin Klein; Euphoria and then by Lacoste; Inspiration. Detailed results for all analyses mentioned above can be found in the Statistical Analysis section.
| Executive Summary | Introduction | Conclusions | Summary | Appendix A | Appendix B |
Introduction
This study used copy-testing to understand the effectiveness of print advertisements for three brands of women’s fragrance. Copy-testing is a way to determine the effect an ad has on consumer preferences. The three brands used in this project were Inspiration by Lacoste, Ralph Lauren Romance and Euphoria by Calvin. All three share similar price points and compete for market share at a common point of purchase. Consumer attitudes towards the brands were measured both pre- and post-ad exposure. All statistical analyses used p < .15 as significant. If significance is noted, this means that in 85 or more samples drawn from the same population as this sample, the expected scores would be the same as found in this sample. The survey itself (Appendix B) was taken online after friends, family and colleagues were sent a hyperlink via email. The 60 results were downloaded to a database and then analyzed using the statistical software package SPSS. Average values for all of the measured variables are added to the survey in Appendix A. This report compiles the results from this project and includes the following: Executive Summary, Introduction, Terminology, Methodology, Analysis & Conclusions, Summary, and two Appendices; survey with mean values for each answer (A) and the survey as viewed by respondents (B).
Methodology
Design
This project was designed to measure brand attitudes and responsiveness to advertisements using a copy-test format with pre- and post-advertising exposure data collection. Respondents were asked to score brands prior to and immediately after viewing an advertisement. In addition, other attitudinal responses were measured after viewing the ad. Due to time constraints, there was no control group against which to compare responses. The survey was designed with Dreamweaver.
Sampling
A convenience sample was used for this survey due to both time and resource limitations. This project uses the results from sixty completed surveys for analysis.
On-line Data Collection
The data in this study were collected via the questionnaire contained in Appendix B. Respondents accessed the survey through a link in an e-mail sent to friends and family. The results submitted by each respondent were linked through a Cold fusion program and integrated into a master database file. This information was retrieved through Microsoft Access and all analyses were conducted using SPSS and simple mathematical formulas.
Survey
The survey was structured into 10 sections with a final section focused on demographic information. It initially requests that respondents list the brands of body lotion, body spray and perfumes they last purchased, and divide 10 points among each brand name for that category through a constant sum scale. After viewing the ads, they were again asked to divvy up 10 points among the perfume products. Impressions of the ads were then taken followed by 10 Likert items and a series of identification checkboxes concerning attitude towards each brand. Lastly, 10 yes-and-no questions were displayed concerning attitudes of the ad as it related to its brand.
Limitations
Interpretation of the data collected for this project reflects a number of limitations. The sample size of 60 respondents is relatively small, but should be adequate for the statistical procedures employed. A convenience sample was used instead of a preferred random sample and a control group was not available due to constraints of both time and resources. Finally, this consumer preference survey did not use randomization to correct or lessen order and interaction effects. This means that the three fragrances were not randomly presented each time ads were displayed or responses taken through repeat measures of a randomized order. Instead, both the advertisements and questions regarding them always appeared in the same order, with Inspiration by Lacoste first, followed by Ralph Lauren Romance and then by Euphoria by Calvin Klein.
Analysis
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
| Executive Summary | Introduction | Conclusions | Summary | Appendix A | Appendix B |
Basic Statistics
Paired t-test comparing Brand Index Scores for all fragrance brands
Table 1: Mean Brand Index Scores for all brands
BrandMean
Std. Deviation
Lacoste
28.8
4.1
Ralph Lauren
36.3
5.1
Calvin Klein
33.0
5.1
#Respondents = 60
The mean Brand Index Score was higher for Ralph Lauren fragrance, followed by Calvin Klein, and then by Lacoste; which means that the respondents were more favorable towards Ralph Lauren fragrance.
Table 2: Brand Comparison of Brand Index Score
Fragrance Brands Pairst ratio
Lacoste/Ralph Lauren
8.93*
Lacoste/Calvin Klein
4.55*
Ralph Lauren/Calvin Klein
4.56*
#Respondents = 60 *p < .15
Each of the three pairs is significantly different from the others. This means that in 85 or more samples, out of 100 samples drawn from the population, as this sample, it would be expected that the mean Brand Index Scores for each brand would be about what they are in this sample.
To determine respondents’ feelings towards each of the three test brands a brand index score was calculated for each brand by summing up the responses to 10 brand-related Likert items appearing on the survey. A score of 50 was the highest possible positive response indicating favorable feelings, and a score of 10 the least, indicating less favorable feelings towards the brand.
Independent samples t-test
Table 3: Comparison of Ralph Lauren fragrance Brand and Ad Index Scores between those who moved up versus down on the Ad Exposure Score after viewing ads.
Brand Index Score
Ad Index Score
Number of Respondents
Mean/St. Deviation
Mean/St. Deviation
Up movers
39.0 / 4.0
7.3 / 1.6
26
Down movers
35.0 / 4.4
3.0 / 1.9
5
t ratio
1.9*
4.8*
*p < .15
There was a significant difference for both the Brand and Ad Index Score after viewing the ad. This means that in 85 or more samples, out of 100 samples drawn from the population, as this sample, it would be expected that the mean Brand Index Scores would be about what they are in this sample for the up movers and down movers. The same can be said for the Ad Index Scores.
Chi-squared test
Table 4: Chi-squared test for Calvin Klein median scores with Change Score
Calvin Klein Score
Above Median
Below Median
Up movers
Count
Mover Row %
Change Column %
Total %6
60.0
20.7
10.34
40.0
13.8
6.9Same
Count
Mover Row %
Change Column %
Total %6
24.0
20.7
10.319
76.0
65.5
32.8Down movers
Count
Mover Row %
Change Column %
Total %17
73.9
58.6
29.36
26.1
20.7
10.3
Chi-squared = 12.42* Total count = 58 respondents
*p < .15The respondents with a Brand Index Score above median who moved down in the Row percentage are 73.9%. This shows that there is a negative relation between the liking of the Brand and the Ad, which is surprising because it means that many respondents had a negative reaction to the specific ad (post < pre ad exposure). Also it is important to underline the fact that 76% of the respondents with Brand Index Score below median stayed the same and had no change whatsoever (Row %). These two results may indicate that the concept of the ad is not working, towards convincing respondents to buy Calvin Klein fragrance.
The Chi-squared value is 12.42. This means that in 85 or more samples, out of 100 samples drawn from the population, as this sample, it would be expected that the percentage distribution would be about the same as found in these 6 cells (Table 4).
Change Score for all brands (Frequency)
Table 5: Change Score for all brands
Lacoste
Ralph Lauren
Calvin Klein
Up movers
13
26
10
Same
38
29
25
Down movers
9
5
25
#Respondents = 60
Change in response to the advertisement, as indicated by change in post → pre allocation of “buying points” was similar for Lacoste and Ralph Lauren fragrance. As it was expected the highest number of respondents stayed the same for all brands.
The up movers were a higher number than the down movers for Lacoste and Ralph Lauren fragrance, which means that there is a positive response towards the two ads for those two brands. However, for Calvin Klein fragrance the down movers were more than the up movers, as shown by Table 5. This may indicate, once more, the ineffectiveness of the specific ad in convincing the respondents to buy the product (Calvin Klein fragrance).
Number of respondents scoring Ralph Lauren > Calvin Klein on the Brand Index Scale
Brand Index Score: Ralph Lauren > Calvin Klein = 40
#Respondents = 60The number of respondents with a Ralph Lauren Index Score higher than a Calvin Klein Index Score was 40 out of 60 respondents. This means that those respondents scored higher, on the summed Likert responses, for Ralph Lauren than for Calvin Klein. This result indicates a more favorable attitude towards Ralph Lauren fragrance after viewing the ad.
Correlation between Lacoste and Ralph Lauren Brand Index Scores
Pearson Correlation ( r ) = 0.4
*p < .15The correlation between the Brand Index Scores of Lacoste and Ralph Lauren is r = 0.4, which is a positive (direct) correlation. This means that they are moderately correlated. The two brands tend to move up and down together. However, the results were not significant.
Correlation between Ralph Lauren and Calvin Klein Brand Index Scores for the group of respondents who said that they would try a new fragrance if they show an ad they liked.
Pearson Correlation ( r ) = 0.4*
*p < .15The correlation between Ralph Lauren and Calvin Klein Brand Index Scores for the group of respondents who said that they would try a new fragrance if they saw an ad they liked is r = 0.4, which is a positive (direct) correlation. This means that they are moderately correlated. The two brands tend to move up and down together. The results were significant which means that in 85 or more samples drawn from the same population as this sample that the mean scores would be about what they are in this sample.
Back to the top
| Executive Summary | Introduction | Conclusions | Summary | Appendix A | Appendix B |
Regression Analysis
For each of the three test brands, a multiple linear regression analysis was conducted where the dependent variable is its change score and the ten Likert items are the independent variables. The purpose of the analysis is to uncover the connection, that the sample respondents made between the advertisements and how they rated the brands.
Table 1: Correlations between Brands’ Change Score and Brand Index Items
Brand
r
R squared
Std. Error
of EstimateF
Lacoste
0.7
53.6%
1.9
5.7*
Ralph Lauren
0.7
60%
1.6
7.3*
Calvin Klein
0.6
47.3%
1.8
4.4*
*p< .15 # Respondents = 60
Table 2: Standardized and Unstandardized Coefficients
Lacoste
Ralph Lauren
Calvin Klein
Likert Items
B
Beta
t
B
Beta
t
B
Beta
t
Constant
-8.9
3.7*
-1.9
1.0*
4.5
2.0*
Good
2.2
0.5
3.8*
-2.1
0.5
2.9*
0.5
0.1
0.8
Smell
-1.1
0.0
0.1
-0.2
0.1
0.5
-2.0
0.7
2.4*
Expensive
1.0
0.3
2.5*
-0.6
0.2
1.7*
0.6
0.2
1.7*
Better
0.6
0.1
1.4*
1.1
0.2
2.4*
-0.6
0.2
1.3*
Trust
0.4
0.1
0.5
0.4
0.1
1.0
-1.6
0.5
3.0*
Sexy
-3.0
0.6
3.7*
3.6
1.2
3.8*
0.5
0.2
0.9*
Buy
0.9
0.3
2.1*
-0.1
0.0
0.3
2.3
0.9
2.8*
Confidence
2.1
0.4
3.1*
-1.2
0.4
1.6*
0.3
0.1
0.5
Not likeable
-4.6
0.1
0.1
-0.4
0.1
1.0
-0.7
0.2
1.4*
Wear
-1.4
0.4
2.5*
0.4
0.1
0.9
-0.9
0.3
2.0
*p< .15 # Respondents = 60
Lacoste: According to Table 1, the Brand Index Items (total) had a moderate, positive correlation with the Change Score. This correlation means that 53.6% of how much the respondents liked the ad (as defined by a positive Change Score) is explained by how much they liked the brand. The F value (5.7*) is significant (*p < .15), which means that in 85 or more samples drawn from the same population as this sample would mean that the correlation would be about what it is in this sample for that item. Table 2 shows that “Good”, “Expensive”, “Better”, “Trust”, “Buy”, and “Confidence” are all positively correlated with the Change Score, which means that as these item values increase, so does the Change Score. “Smell”, “Sexy”, “Not Likeable”, and “Wear” are negatively correlated with the Change Score, which means that as the values for these items increase, the Change Score decreases. The items that are important variables in explaining variance in the Change Score in this sample include those described as “Good”, “Sexy”, “Confidence”, and “Wear”. Most of the independent variables (10 Likert Items) were significant (*p < .15), except “Smell”, “Trust”, and “Not Likeable”, which means that in 85 or more samples drawn from the same population as this sample that the correlation would be about what it is in this sample for that item.
Ralph Lauren: According to Table 1, the Brand Index Items (total) had a moderate, positive correlation with the Change Score. This correlation means that 60% of how much the respondents liked the ad (as defined by a positive Change Score) is explained by how much they liked the brand. The F value (7.3*) is significant (*p < .15), which means that in 85 or more samples drawn from the same population as this sample would mean that the correlation would be about what it is in this sample for that item. Table 2 shows that “Better”, “Trust”, “Sexy”, and “Wear” are positively correlated with the Change Score, which means that as these item values increase, so does the Change Score. “Good”, “Smell”, “Expensive”, “Buy”, “Confidence”, and “Not Likeable” are negatively correlated with the Change Score, which means that as the values for these items increase, the Change Score decreases. The items that are important variables in explaining variance in the Change Score in this sample include those described as “Smell”, “Trust”, and “Buy”. Most of the independent variables (10 Likert Items) were significant (*p < .15), except “Smell”, “Trust”, “Buy”, “Not Likeable”, and “Wear”, which means that in 85 or more samples drawn from the same population as this sample that the correlation would be about what it is in this sample for that item.
Calvin Klein: According to Table 1, the Brand Index Items (total) had a moderate, positive correlation with the Change Score. This correlation means that 47.3% of how much the respondents liked the ad (as defined by a positive Change Score) is explained by how much they liked the brand. The F value (4.4*) is significant (*p < .15), which means that in 85 or more samples drawn from the same population as this sample would mean that the correlation would be about what it is in this sample for that item. Table 2 shows that “Good”, “Expensive”, “Sexy”, “Buy”, and “Confidence” are all positively correlated with the Change Score, which means that as these item values increase, so does the Change Score. “Smell”, “Better”, “Trust”, “Not Likeable”, and “Wear” are negatively correlated with the Change Score, which means that as the values for these items increase, the Change Score decreases. The items that are important variables in explaining variance in the Change Score in this sample include those described as “Smell”, “Trust”, and “Buy”. Most of the independent variables (10 Likert Items) were significant (*p < .15), except “Good”, “Confidence”, and “Wear, which means that in 85 or more samples drawn from the same population as this sample that the correlation would be about what it is in this sample for that item.
The equation for the brand Calvin Klein is (Standard Coefficient):
Calvin Chang Score = A + “Good” + “Smell” + “Expensive” + “Better” + “Trust” + “Sexy” + “Buy” + “Confidence” + “Not Likeable” + “Wear” =
4.5 + 0.5 + (-2.0) + 0.6 + (-0.6) + (-1.6) + 0.5 + 2.3 + 0.3 + (-0.7) + (-0.9) =
4.5 + 0.5 - 2.0 + 0.6 - 0.6 - 1.6 + 0.5 + 2.3 + 0.3 - 0.7 - 0.9 = 2.9
Back to the top
| Executive Summary | Introduction | Conclusions | Summary | Appendix A | Appendix B |
Multiple Discriminant Analysis
Lacoste
The purpose of Multiple Discriminant Analysis is to examine how well independent variables (10 likerts items) classify sample members into different groups (up movers/down movers). This analysis determines the connection between how respondents rated the brand Lacoste (10 likert items) and their pre to post exposure (changescore).
Table 1: Group Mean Scores and Standard Deviation
Likert Items
Up Movers
Respondents = 13Down Movers
Respondents = 9
Mean
Standard Deviation
Mean
Standard Deviation
Good
3.6
0.6
3.2
0.8
Smell
3.0
0.3
2.8
0.4
Expensive
2.8
0.8
2.7
0.7
Better
2.7
1.0
2.4
0.5
Trust
3.3
0.5
3.3
0.7
Sexy
3.3
0.5
3.3
0.7
Buy
3.8
0.9
2.9
1.3
Confidence
3.4
0.5
2.4
0.5
NotLikeable
2.8
0.7
3.3
0.9
Wear
3.0
0.9
2.7
0.5
These mean scores represent the averages for all 10 Likert items for the Up and Down Mover categories. These values are used in further analysis to determine discriminant function coefficients.
Table 2: Standardized and Unstandardized discriminant function coefficients
Likert Items
Standardized
Unstandardized
Constant
-29.2
Good
2.9
-4.0
Smell
2.4
6.8
Expensive
1.3
1.7
Better
0.3
0.3
Trust
0.7
1.1
Sexy
0.0
-0.0
Buy
1.4
1.3
Confidence
1.8
3.5
Not Likeable
0.2
-0.3
Wear
0.1
-0.1
From Table 2 we can see that the important variables in discriminating between the Up and Down Movers are “Good”, “Smell”, “Expensive”, “Trust”, “Buy” and “Confidence” out of the 10 Likert Items
Table 3: Summary Statistics
Wilks’ Lambda
Chi-squared
0.11
32.42*
*p < .15 level
Wilks’ Lambda score was found to be significant; therefore in 85 out of every 100 samples drawn from the same sample population as this sample, we could expect similar results. This means that these function coefficient values can be projected to the general population.
Table 4: Group Centroid Function
Grouping
Up Movers
Down Movers
Centroid Function
2.2
-3.2
The group centroid for the Up Movers is 2.2 compared with (-3.2) for the Down Movers.
Comparing the distance between group centroids (Up movers & Down movers) shows that the distance between them is large relative to the variance within groups. The group centroids are not too similar in value and can reach significance, indicating that the Likert Items have good discriminant function between Up and Down Movers.
Table 5: Predictive Analysis: Classification Results for Up Movers and Down Movers
Counts
Predicted Up
NumberPredicted Down
NumberActual Up
12
1
Actual Down
0
9
95.5% of original grouped cases correctly classified.
t(observed) = (0.955 – 0.500) / [ √ { (0.955)(0.045) / (22) + (0.5)(0.5) / (22) } ] = 0.325
*p < .15 level95.5% of the original grouped cases were correctly classified as either an Up or Down Movers. This value is non-significant (p < .15 level) which means that these results (the expected percentage of predictive accuracy using these same discriminant functions) apply only to this sample and cannot be projected to the general population.
Back to the top
| Executive Summary | Introduction | Conclusions | Summary | Appendix A | Appendix B |
Analysis of Variance: ANOVA / MANOVA
The ANOVA analysis is conducted in order to uncover the relationship between the Level of Education of the respondents and their Change Scores for the variable “good brand”; for the brand Ralph Rauren Romance. The dependent variable is the Likert Item: “goodbrand” and the two Independent variables are Ralph Lauren move and the level of education of the respondents. The MANOVA analysis is the same; the only difference is that the Dependent variable this time was all 10 Likert Items of the Questionnaire. This analysis is conducted in order to find out about all the Likert Items at once, to see what is happening.
ANOVA: Ralph Rauren Change Score & “good brand” Likert Item
Table 1: Mean scores, by level of education, for respondents who moved up, down or stayed the same on Change Score for “goodbrand” variable
Change ScoreUp movers
Same
Down movers
Basic Edu.
Mean / Std. Dev. #Respondents
4.25 / 0.5
4
3.2 / 0.4
9
4.25 / 0.5
4Higher Edu.
Mean/Std. Dev.
#Respondents
4.23 / 0.6
22
3.7 / 0.7
16
4.2 / 0.4
5Table 2 Group Effects for Change Score and level of education: ANOVA Table
Group Variance
Sum of Squares
Degrees of freedom
Mean Square
F
ratioBetween Groups Effect
Main : Change Score
Main : level of education
Interaction: Change x level of education
6.3
0.2
0.9
2
1
2
3.2
0.2
0.4
9.15*
0.65
1.30Within Groups Effect
Unexplained error
18.7
54
0.3
*p < .15
Of the three between-groups effects, only the main effect for the Change Score was significant (*p < .15). This means that in 85 or more samples drawn from the same population as this sample, the expected mean scores on this Likert item would be the same as those reported here. The other main effect for level of education and the interaction effect were not significant. This means that these mean scores apply only to this sample and do not project to the general population.
MANOVA: Ralph Rauren Change Score & Brand Index Items
Table 3: Mean scores, by level of education, for respondents who moved up, down or stayed the same on Change Score for 10 Likert Items
Change Score
Up movers
Same
Down movers
Basic Edu.
Mean/Std.Dev.
#Respondents
Good brand
Smell
Too expensive
Better brand
Trust
Sexy
Would buy
Confidence
Not like any
Would wear
4.25 / 0.5
3.75 / 0.9
2.50 / 1.7
3.00 / 0.8
4.25 / 0.5
4.50 / 0.6
4.00 / 0.0
4.50 / 0.6
3.00 / 1.2
4.00 / 0.04
3.22 / 0.4
3.22 / 0.6
2.90 / 0.3
3.11 / 0.3
3.11 / 0.3
3.11 / 0.3
3.11 / 0.3
3.11 / 0.3
3.11 / 0.3
3.11 / 0.39
4.25 / 0.5
4.25 / 0.5
3.25 / 1.0
3.50 / 0.6
4.25 / 0.5
4.00 / 0.8
4.25 / 0.5
4.00 / 0.8
3.75 / 0.5
3.75 / 1.04
Higher Edu.
Mean/Std.Dev.
#Respondents
Good brand
Smell
Too expensive
Better brand
Trust
Sexy
Would buy
Confidence
Not like any
Would wear
4.23 / 0.5
4.32 / 0.8
2.45 / 0.8
3.95 / 0.2
4.18 / 0.4
4.27 / 0.7
3.50 / 1.3
4.05 / 0.6
4.14 / 0.9
4.09 / 0.922
3.75 / 0.7
3.63 / 1.1
2.90 / 0.7
3.38 / 0.6
3.38 / 1.0
3.50 / 0.6
3.75 / 0.7
3.38 / 0.8
3.38 / 0.7
3.69 / 0.616
4.20 / 0.4
3.80 / 0.4
3.20 / 0.8
2.80 / 0.8
3.80 / 0.8
3.20 / 0.4
3.80 / 0.4
3.00 / 0.7
2.80 / 0.8
3.60 / 0.95
Table 4: Group Effects for Change Score and level of education: MANOVA Table
Group Variance
Wilks’ lambda
F
ratioDegrees of freedom
Between groupsDegrees of freedom
Within groupsBetween Groups Effect
Main : Change Score
Main : level of education
Interaction: Change x level of education
0.31
0.75
0.47
3.58*
1.50
2.07*
20.0
10.0
20.0
90.0
45.0
90.0*p < .15
The main effect for the Change Score and the interaction between the change score and the level of education were found significant (*p < .15) in the between-groups effect. This means that in 85 or more samples drawn from the same population as this sample, the expected mean scores on the ten Likert items and their interaction with the level of education would be the same as those reported here. The other main effect for level of education was not significant. This means that the specific mean scores apply only to this sample and do not project to the general population.
Back to the top
| Executive Summary | Introduction | Conclusions | Summary | Appendix A | Appendix B |
Factor Analysis
A factor analysis was conducted for all three brands of women’s fragrance in order to then calculate a brand attitude score for each brand. Paired t-tests were then conducted to determine the statistical significance for the differences of means between brand attitude scores, and thus used to determine which brand survey respondents like best or have the most favorable attitudes towards.
Table 1: Communalities and Factor Loadings for Lacoste
Likert
Items
Communalities
Factor Matrix
Varimax Rotated
Matrix
Good brand
0.6
0.4
0.6
0.4
0.8
-0.2
0.1
Smell
0.7
0.7
-0.4
-0.2
0.1
0.8
0.3
Expensive
0.7
0.6
-0.3
-0.4
0.1
0.8
0.1
Better
0.5
0.5
-0.3
0.4
0.2
0.2
0.6
Trust
0.8
0.8
-0.3
-0.1
0.2
0.7
0.4
Sexy
0.7
0.8
0.2
-0.0
0.6
0.5
0.1
Buy
0.8
0.7
0.3
0.4
0.8
0.1
0.3
Confidence
0.7
0.5
0.5
-0.5
0.5
0.4
-0.4
Not like any other
0.6
0.4
-0.6
0.4
-0.0
0.3
0.7
Wear
0.6
0.6
0.5
-0.1
0.7
0.2
-0.1
Table 2: Total Explained Variance for Lacoste
Factors1
2
3
4
5
6
7
8
9
10
Eigenvalues
3.8
1.8
1.0
0.9
0.7
0.6
0.4
0.3
0.2
0.1
% of
Variance38.4
18.0
10.1
9.4
7.2
6.4
4.3
3.2
2.0
1.0
Cumulative
%38.4
56.4
66.5
75.9
83.1
89.6
93.9
97.1
99.2
100.000
Three factors had Eigenvalues > 1.0 and were extracted from the 10 Likert items. All three factors combined explained 66.5% of the variance in the original 10 Likert items. The Factor Matrix shows seven Likert items loading positively into Factor one, two items into Factor two and one item (‘Confidence”) was removed for ambiguity as it loads equally into all three Factors. That’s why, we selected the Varimax Rotated Matrix which has no ambiguous items.
Table 3: Communalities and Factor Loadings for Ralph Lauren
Likert
ItemsCommunalities
Factor
MatrixVarimax
Rotated
Matrix
Good brand
0.7
0.8
0.1
0.8
0.1
Smell
0.6
0.8
0.1
0.8
0.1
Expensive
0.6
-0.3
0.7
-0.2
0.8
Better
0.4
0.6
-0.1
0.6
-0.2
Trust
0.6
0.7
0.3
0.7
0.2
Sexy
0.8
0.9
-0.1
0.9
-0.2
Buy
0.6
0.2
0.7
0.2
0.7
Confidence
0.7
0.8
0.0
0.8
-0.0
Not like any other
0.8
0.8
-0.3
0.8
-0.4
Wear
0.6
0.8
0.2
0.7
0.1
Table 4: Total Explained Variance for Ralph Lauren
Factors1
2
3
4
5
6
7
8
9
10
Eigenvalues
5.1
1.4
0.9
0.8
0.6
0.4
0.3
0.2
0.1
0.0
% of Variance
50.7
14.5
9.1
7.8
6.4
4.3
3.0
2.2
1.5
0.5
Cumulative
%50.7
65.2
74.4
82.1
88.5
92.8
96.0
98.0
99.5
100.0
Two factors had Eigenvalues > 1.0 and were extracted from the 10 Likert items. All three factors combined explained 65.2% of the variance in the original 10 Likert items. The Factor Matrix shows eight Likert items loading positively into Factor one, and two items into Factor two. Three were no ambiguous items.
Table 5: Communalities and Factor Loadings for Calvin Klein
Likert ItemsCommunalities
Factor Matrix
Varimax Rotated Matrix
Good brand
0.8
0.6
-0.2
-0.6
-0.2
0.7
0.4
-0.2
-0.0
Smell
0.9
0.7
0.5
0.0
-0.2
0.1
0.9
0.1
0.1
Expensive
0.9
0.2
0.5
0.1
0.7
0.1
0.2
-0.0
0.9
Better
0.7
0.6
0.0
0.6
-0.1
0.2
0.4
0.7
0.1
Trust
0.6
0.7
-0.1
-0.2
0.2
0.7
0.3
0.0
0.2
Sexy
0.8
0.8
-0.3
-0.0
0.3
0.8
0.2
0.3
0.1
Buy
0.9
0.8
0.5
-0.0
-0.2
0.2
0.9
0.1
0.1
Confidence
0.8
0.7
-0.5
-0.1
0.3
0.8
0.1
0.2
0.0
Not like any other
0.8
0.4
-0.4
0.7
-0.1
0.2
0.0
0.9
-0.2
Wear
0.7
0.8
0.1
-0.0
-0.2
0.4
0.7
0.2
-0.1
Table 6: Total Explained Variance for Calvin Klein
Factors1
2
3
4
5
6
7
8
9
10
Eigenvalues
4.4
1.4
1.2
1.0
0.6
0.5
0.3
0.3
0.1
0.0
% of Variance
44.1
14.2
12.6
10.1
5.8
4.7
3.5
3.2
1.3
0.5
Cumulative %
44.1
58.3
70.9
81.0
86.8
91.5
95.0
98.2
99.5
100.0
Four factors had Eigenvalues > 1.0 and were extracted from the 10 Likert items. All three factors combined explained 81.0% of the variance in the original 10 Likert items; which makes this factor analysis very good. The Factor Matrix shows six Likert items loading positively into Factor one, one item into Factor three, one item into Factor four and two items (“Good”) & (“Better Brand”) were removed for ambiguity as it loads equally into two Factors. That’s why, we selected the Varimax Rotated Matrix which has no ambiguous items.
Table 7: Mean Attitude Score
BrandMean
Std. Deviation
Lacoste
3.1
0.5
Ralph Lauren
3.7
0.6
Calvin Klein
3.5
0.6
#Respondents = 60
Table 8: Paired t-tests
