Focus group

66

Respondent audience:

As the purpose of this thesis is to identify a general phenomenon the respondents utilized hereto will not be delimited to any specific target audience per se. The only general factor that will be used to delimit the chosen respondents will be that they must stem from the Western European culture, as the entire thesis has been delimited to have such focus, as stated in section 1.1.1. It could be argued that one might get stronger results and that it would be of more practical interest to Rolex, if the thesis was delimited  to  only  include  Rolex’s  actual  target  audience. This as one might argue that the target audience would tend to be more affected by the celebrity endorsement as well as more relevant to Rolex, as they would be more inclined to purchase. However using such approach would make the findings very case specific and greatly limit the possibility to generalize the results to other cases. Had such approach been utilized it would thus first be possible to say something general after the same phenomenon had been observed by several other similar case studies.

67 celebrity endorsement is viewed and to (2) identify what celebrities would fit well/not so well with the Rolex brand.

For the interview, six respondents that either owned or had owned a luxury watch was recruited from the social network of the authors (predominantly by use of facebook.com or direct emails as intermediate). These consisted of four males and two females, which all were entry level employees from the ages of 25 to 30. The interview was conducted at the home of one of the authors and the respondents were offered refreshments, chips and other treats. The interview was digitally recorded by use of a microphone, so that it could later be transcribed and studied. For the full transcription please see appendix B1.

To encourage the respondents to contribute freely, the interview was kept semi-structured and the questions formed as open and non-leading as possible (for an overview of the research guide please see appendix B2. At the start of the interview, the respondents were given a small exercise.

All respondents were provided a piece of paper showing 12 male celebrities, as seen in Figure 10 (and appendix B3). The respondents had to rate according to how well they believed the celebrities fit with the Rolex brand.

FIGURE 10 - CELEBRITY RATING SCHEMA

Source: Own work

68 The main point of this exercise was to get the respondents thinking of the celebrities, celebrity endorsement and the factors that come into play. Furthermore the aggregated results of this test would be useful as information into what celebrities should be used in the online survey. For the results of this exercise please see appendix B4.

3.1.1.1 Considerations

In developing the interview guide, recruiting the respondents and conducting the focus group interview several noteworthy considerations were made and dealt with accordingly in order to heighten the credibility of the interview.

Developing the interview guide:

Questions were carefully chosen to ensure that the overall purpose of the interview was achieved and in such a way, that the respondents would be able to contribute freely and have their opinions heard. Particular attention was paid to the specific order the questions should be asked, in order to keep the respondents from guessing the purpose of the study beforehand. For an overview of the interview guide please see appendix B2.

Further much thought was put into choosing the 12 celebrities to include in the test. This as it was recognized, that other factors than the personality of the celebrity could come into play, when the respondents were asked to rate the celebrities. As such it was strived to find celebrities of similar gender and age, which were equally physically attractive, credible and had the same amount of expertise in regards to luxury watches. The point of this was to try to neutralize such factors as most as possible, so the respondents would be basing their answers, as most as possible, on the personalities of the celebrities. It is however recognized that such factors may indirectly come into play as they are very interrelated and such factors may influence the perception that the respondents have in regards to the personality of the celebrity. It was further strived to find pictures of each celebrity that were as similar as possible, to avoid biasing the respondents. In this it was chosen to include pictures with good lighting that showed the celebrities in dark suits while they were smiling and looking away from the camera. Followingly the pictures were edited so they would be of similar size and have the torso of the celebrity in equal focus, in order to not put unnecessary focus on any of the celebrities (for the result see Figure 10).

69 Recruiting the respondents:

The potential bias inherent in recruiting respondents from amongst the authors own social network was considered. The concern was that such respondents would tend to not make their true opinions known, as they might simply be trying to answer what they thought the interviewer would be looking for. To avoid such bias, it was initially attempted to recruit respondents through an online watch community (Urforum.dk). This however   proved   not   to   be   feasible.   As   the   author’s   social   network thus had to be used, respondents with limited knowledge of the purpose of the interview was chosen and clearly asked to voice their true opinion.

Furthermore the homogeneity of the respondents was considered. The fear was that respondents of similar age and at a similar life stage (just starting their professional career after university) would tend to have very similar opinions and thus not provide a truthful picture. However, as it was recognized most of the respondents it would be possible for the authors to recruit would also be of a similar age and at a similar life stage, this was not believed to cause a direct problem. It would however possibly cause a generalizability problem (for further see section 6.1).

Conducting the interview:

When conducting a focus group interview there is always the chance that one or two of the respondents may come to dominate the group causing others not to voice their true opinions. To avoid this and to encourage the respondents to contribute freely it was, in the start of the interview, made clear that everyone should feel free to voice their opinions and give room for others to voice theirs. Further it was attempted to control the speaking time of the more active respondents and to involve the quieter respondents by asking them direct questions, so as to involve all participants and avoid ‘Groupthink’⁴.