Chapter 3 Methodology
3.5 Procedure
Hensel, & James, 2001; Zarantonello & Schmitt, 2010). The brand attitude measurements include three items of evaluation, the Bad/Good, Unpleasant/Pleasant, and Unattractive/Attractive aspects of participants’ brand attitude (See the three items in section 2 of Appendix A). Participants were asked to measure their brand attitude according to the extent of agreement with these three aspects of brand attitude. Brand attitude was measured with a 7-point semantic differential scale. The reliability of the three items in brand attitude measurements was also tested. The Cronbach’s Alpha value for Nike is .89; for Starbucks, it was .84; for Hollister, it was .91; and for BMW, it was .81.
Brand satisfaction was measured with one item. The measure was taken from a three item measures for customer satisfaction (Babakus, Bienstock, & Scotter, 2004). 1 The most relevant one is chosen as our measure for brand satisfaction, which is “Your overall satisfaction”. Participants would be asked to evaluate their overall satisfaction towards the brands (See section 3 of Appendix A). Brand satisfaction was measured with a 7-point Likert scale, from very satisfied to very dissatisfied.
Purchase intention was measured with one item. Participants would be asked to evaluate their extent of agreement with the description of “If budget allowed, I would like to buy a product of this brand” (Yeh, 2011). Purchase intention was measured with a 7-point Likert scale, from strongly disagree (point 1) to strongly agree (point 7).
3.5 Procedure
The experiment was conducted in classrooms. Before the experiment, important information was written on the blackboard, including participants’ class numbers, their assigned groups, and the access information of wifi and Instagram. Participants were separated into four groups according to random assignment. They could check their assigned group from the blackboard. Next, they were instructed to connect to the Internet provided by the researcher.
Three portable wifi routers were rented to provide Internet access in class. Participants could access Internet through the wifi IDs and keys written on the blackboard. After connecting to the Internet, those who did not have Instagram would be asked to download Instagram app.
They could use the Instagram ID and passwords provided by the researcher to access Instagram without having to register for one. Smart phones were required for this research.
Those who did not have smart phones could not participate in this research.
1 The original question measuring customer satisfaction is “How satisfied have you been with (insert brand name)?” With 3 evaluating items following the question: “Your overall satisfaction; products you have purchased; services you have received.” (Babakus, Bienstock, & Scotter, 2004).
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
After technical issues had been dealt with, the researcher called the group name and summoned the participants to collect the paper for brand measurements. The order of distributing the paper was from group I to group IV. The first paper was double sided. Each side has one brand measurement. (See page 52, 53 of Appendix A). Each group would receive different brand measurements. This paper was measuring participants’ brand experience in controlled condition. After participants completed the measurements, they were asked to pass their paper to the front for collection.
The second paper contained only one brand measurement (See page 54 of Appendix A). This paper was distributed with the same manner as the previous procedure. Group name were called, and students were summoned to collect the brand measurements. Before answering the measurements, participants were asked to view Instagram stimulation. When they received their respective brand measurements, they were asked to access the Instagram account that was stated on the paper of measurements. They were instructed to view the Instagram contents for five minutes. Discussion or answering the measurements within the five minutes was not allowed. When five minutes ended, they were instructed to start answering the third brand measurement. Upon completing the measurement, the papers were collected with the same manner.
Finally, the last piece of paper was distributed. This paper was double sided (See page 55, 56 of Appendix A). The front of the paper was the fourth brand measurements, and the back of the paper was basic information. The basic information inquired for participants’
Instagram usage, past purchase experience with brands, and their interaction when viewing Instagram stimulation. After completion, the last paper was collected and students were thanked for their patience and contribution to this study.
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
From participants’ answer of basic information (Table 8), it showed that all of the participants had viewed brand images when viewing the Instagram stimulation. Among these participants, 28 of them gave hearts to the brand contents, two participants had left comments, 39 participants had checked the comments, 23 participants had checked hashtags, and 31 participants had checked tagged brands or users.
Table 8
Participants’ Activities on Instagram during the Experiment
n %
Viewing pictures 112 100
Give Hearts 28 25
Comment 2 2
Check Comments 39 35
Check hashtag 23 21
Check tagged brands or users 31 28
This data showed that participants were fully immersed in the Instagram stimulation.
This greatly helped enhance the effect of Instagram on participants’ brand experience and advertisement effectiveness.
‧
The data collected from the brand measurements were analyzed using SPSS. This study first analyzed the influence of Instagram on the overall brand experience of the four brands. Then each of the four dimensions of brand experience were also analyzed. Lastly, the effect of Instagram on each of the three dimensions of advertisement effectiveness was analyzed.
In the analysis, there was one independent variable with two levels. It was the experimental treatment, with experiment condition (With Instagram stimulation) and controlled condition (Without Instagram stimulation) as levels of independent variable. There were several dependent variables in this analysis, including the four dimensions of brand experience (sensory experience, affective experience, intellectual experience, and behavioral experience), and the three dimensions of advertisement effectiveness (brand attitude, brand satisfaction, and purchase intention). In the brand experience analysis, a one-factor, between subjects Multivariate Analysis of Variance (MANOVA) was used to compare the difference of overall brand experience before and after Instagram stimulation. Four dimensions of brand experience were also analyzed using a follow-up Analysis of Variance (ANOVA). For advertisement effectiveness, a MANOVA analysis and a follow-up Analysis of Variance (ANOVA) were conducted to see how Instagram influenced participants’ brand attitude, brand satisfaction, and purchase intention.
4.1 Brand Experience 4.1.1 Overall Brand Experience
Instagram’s influence on participants’ overall brand experience toward the four brands was analyzed using a one-factor, between subjects Multivariate Analysis of Variance (MANOVA). The dependent variable was the four dimensions of brand experience. The results showed a significant difference between experimental condition and controlled condition in Nike (Wilks’ Λ = .852, F (107,107) = 4.65, p = .002, partial η2 = .15), and Starbucks (Wilks’ Λ = .861, F (107,107) = 4.30, p = .003, partial η2 = .14). However, significance was not found in Hollister and BMW.