Exploratory factor analysis (EFA) was done with the PCA method and it revealed several factors that were in line with the factors that showed up in the analysis of Brakus et al (2009). However, there were some small differences in the way the factors loaded for the individual brands, so it was decided that the responses for the three biggest brands would be combined to get a result without a brand bias. When PCA was conducted for the combined responses for the three brands, the result was that all the items relating to eco-friendliness load principally on the first component or factor. The first factor is the one where also nearly all of the positive behavioral, sensory and intellectual statements load the strongest. All the negative statements for all of the dimensions load negatively on the first factor and positively on either factor 2 or both factors 2 and 3.
For all of the dimensions, one factor mainly explained the most of the data variation, and none of the other items had a loading of zero or less on the first dimension, so the development of the scale development proceeded with the presumption that there would be a simple linear combination of the individual items and that no individual item is sufficient on its own, but that the entire scale is required to measure the construct of brand experience (Nunnally and Bernstein, 1994). See Table 9 for the detailed results of the PCA for the combined responses for the three brands. The table shows the unrotated loadings of each item. SPSS uses the Kaiser criterion and retains all of the components with the Eigen value of over 1.
Table 9. PCA on the combined responses for the three brands
Component
1 2 3
ThreeBrand_3-i- This brand stimulates my curiosity and problem solving. .843
ThreeBrand_16-B- This brand makes me behave in an eco-friendly way. .824 .321 ThreeBrand_15-b- This brand results in bodily experiences. .816
ThreeBrand_4-s- I find this brand interesting in a sensory way. (sight, touch, hearing, taste, and smell).
.814
ThreeBrand_14-s- This brand makes a strong impression on my visual sense or other senses. (sight, touch, hearing, taste, and smell).
.808
ThreeBrand_13-A- This brand creates eco-friendly emotions. .807 .375 ThreeBrand_17-b- I engage in physical actions and behaviors when I use
this brand.
.779
ThreeBrand_2-I- This brand makes me think about the state of the environment.
.769
ThreeBrand_11-S- This brand makes an eco-friendly impression. (eco-friendly = not environmentally harmful)
.719 .433
ThreeBrand_5-a- This brand is an emotional brand. .701 -.334 .335 ThreeBrand_8-i- I engage in a lot of thinking when I encounter this brand. .649
ThreeBrand_1-a- This brand induces feelings and sentiments. .566 -.405 .440 ThreeBrand_7-i- This brand does not make me think. -.392 .585
ThreeBrand_12-a- I do not have strong emotions for this brand. -.404 .563
ThreeBrand_9-b- This brand is not action oriented. -.305 .381 .682 ThreeBrand_10-s- This brand does not appeal to my senses. (sight, touch,
hearing, taste, and smell).
-.469 .440 .567
.
As the results for the PCA indicated that the items of eco-friendliness loaded principally on the first factor as the majority of the other items, it was worthwhile moving to the next phase in the study, and test what the results for the eco-friendliness items would be when analyzing the models with CFA.
In this study, the number of scale items in the original 12-item BBX were increased by four items on eco-friendliness. The stability of the scale was already originally tested by Brakus et al. (2009), first by using students and then by a sample from the wider population and the 12 scale items in the original BBX model was then proven to be brand and respondent independent and it demonstrated the general brand experience of the respondents. As original the BBX model had already undergone such a thorough analysis with the various model options to find the four-factor model and it had been verified for model fit with structural equation modelling and confirmatory factor analysis, in this study it was taken as the starting point for testing how the results could be replicated with a different data set.
In the actual survey there were 5 mobile phone brands included that the respondents were asked to respond to, however, only the responses for three of the major brands were analyzed for this study: Samsung, Nokia Lumia and Apple. The reason why the responses for the Sony and hTc brands were not included in the analyses is that the brands were not that familiar to the respondents and the number of ‘Do not know' response options were high for these two brands. The sample includes responses from 506 respondents on three brands evaluating the extent to which the 16 items of the scale describe their experience with each brand using an eight-point Likert scale where all the response options are anchored.
The positioning of the environmental dimension in the BBX scale was tested with four items on eco-friendliness, one designed for each of the four brand experience dimensions used in the BBX model: affective, sensory, behavioral, intellectual. In the following section, the conceptual models are analyzed individually and then compared to each other to identify the option that has the best explanatory power. In the conceptual modelling, the original BBX model has been used as a basis for measuring the brand experience, but for the second and third research question it has also been extended with a set of items on how an environmental aspect is considered to be included in the brand experience. The four-factor BBX model was first analyzed without including the items on eco-friendliness. Then the four-factor model was tested with an eco-friendliness item added in each of the four factors. And lastly, the five-factor model with a separate five-factor for the eco-friendliness items was tested. In the analysis of all of the three models, the same data set was used. The major part of the research analysis has to do with the testing of the three different BBX model constructs that are the focus of this study. Below the results for the model fit of each of the models will be described and discussed in detail as well as the factor loadings and results from the CFA done on the basis of the SEM.