Research Design

The following subsections are devoted to articulating the designs of the online survey study and focus group interviews, respectively. Here, we will specify the use of each method for our stated purposes and provide preparative and procedural information.

Online Survey

Saunders et al. (2007) argued for the validity and reliability of the data collected to be to a large degree dependent on the design, layout, and administration of the survey questionnaire. This is applied to our design as introduced in this section.

Appendix 9.4 introduces the data requirements table, which is created for the particular purpose of ensuring all relevant aspects of inquiry are included in the survey design, while simultaneously ensuring that the layout of our self-administered questionnaire appears attractive to encourage the respondents (Krosnick, 1999).

We have chosen to use the online survey tool SurveyMonkey.com for designing a visually and technically pleasing layout. Hence, the questions are limited to a total of 20 items agreed as the most central to our inquiry.

33

The ordering and formulation of the questions are considered to ensure the validity of the findings. We agree upon to avoid leading questions, or questions proposing a risk to the psychological effect of priming (Krosnick, 1999). For these purposes, we order the questions to first include items related to skincare product category and preferences in general, then moving to items with a specific focus on natural and organic skincare, and lastly, items including the country-of-origin on both supernational and national levels (Appendix 9.3).

The question types in the design include categorical questions for items with the purpose of acquiring attribute variables of respondents´ characteristics, including gender, age, and current use of skincare products. Particularly, we perceive that by separating the respondents on the basis of their use of natural skincare products, we enable to derive potentially relevant information. This is shown in that question 4 is attributed to those who use natural skincare products, and question 5 and its sub-questions to those who do not (Appendix 9.5). Furthermore, list questions are introduced for the particular items with the need to be sure that the respondent has considered all possible response alternatives. However, for the purpose of acquiring perceptions and preferences, items including opinion variables are introduced as rating questions.

These include both independent rating items and matrixes, the latter which enables to record the responses to two or more similar questions at the same time (Saunders et al., 2007). Whilst less consuming for the respondents, the rating scales provide better validity compared to categorical ´yes´ and ´no´ questions (ibit).

The type of rating questions used to consist of the Likert-style rating scale in which the respondent is asked how strongly she agrees or disagrees with a statement or series of statements, organized on five-point rating scales. By labeling the scale points with words, we thus provide clarity to the meanings assigned to the scale points for better response validity and reliability (Saunders et al., 2007). For the purpose of analysis, we use coding schemes to account for the scales agreeableness according to following:

Value Code Code

5 Completely agree Very good choice

4 Agree Good choice

3 Neither Agree or disagree Nor good nor bad choice

2 Disagree Bad choice

1 Completely disagree Very bad choice

By assigning these values, we establish a median of 3 to each rating question. These will enable calculations on distributions among the values, thus easier comparison among the two data sets.

34

To maximize the benefits of the survey to reach the stated objectives and minimize weaknesses and adjust the design accordingly, a pilot test is conducted both for the English and Korean versions (OECD, 2012).

Therefore, the iterations facilitated by this feedback mechanism allow perfecting the question items also from a translation perspective: That is, focus points in relation to the lexical, idiomatic, experiential, grammar and syntax levels of compatibility among the two surveys using different languages (Saunders et al., 2007).

After the data collection was completed, the data is transferred from SurveyMonkey to Microsoft Excel spreadsheet for descriptive analysis. Due to the nature of the data items as categorical of nature, namely nominal and ordinal, the descriptive analysis covers the following aspects: Distribution and deviation for ordinal items, and percentages and proportional measures for nominal data items. Finally, based on the analysis for both two samples, major findings are summarized into the charts for the comparisons.

Focus Group

The purpose of the two focus groups is to acquire consumers’ perception of natural skincare and the particular aspects affecting their perceptions, preferences, as well as the influence of the Finnish origin appeal. This is facilitated through everyday ways of talking about natural skincare with both Danish and Korean natives. Two focus groups, one for each Koreans and Danish participants, are conducted to compare consumers’ similarities and differences. The interviews were held on separate days, 21^st for Koreans and 22^nd of March 2018 for Danish. The site of the interviews was at a meeting room in Dalgas Have of Copenhagen Business School. The Danish focus group interview was conducted in English and the Korean focus group interview was conducted in Korean with a native speaking researcher. The researcher performs the role of the moderator in all the focus groups, which includes asking the questions, taking notes and managing the discussion flow.

For the overall aim of employing two methods for primary inquiry in triangulation, the semi-structured interview guide is developed based on central topics similar to the survey design (Appendix 9.7). However, as for inquiring a deeper understanding of cross-cultural similarities and differences, hence to fully employ the unstructured and intuitive nature of the method, the focus group topics vary for the two groups. Common questions for both groups are chosen from the survey analysis for comparison, while country-specific questions regarding the English slogan of the case company Supermood for Korea are included for translation issue.

35

For the purpose of moving from merely product category associations and perceptions to the level of usage benefits and values assigned to natural skincare, a laddering model by Reynolds & Gutman (1988) is proposed.

By drawing on Gutman's Means-end Theory (1982), the laddering model ultimately acts as a marketing research technique that provides a chain of product features, benefits, and emotions and further brings out consumers’ emotional reasons and nuances towards the purchase decision (Reynolds & Gutman, 1988).

Therefore, a set of laddering questions from the broad question of features to specific probing questions regarding personal benefits are used to account for the emotional motivators for using natural skincare usage, which contribute to understanding the meaning of the product category in the lives of the consumer. The results using the laddering model will be illustrated as a figure for each country in the following sections;

5.2.1 and 5.2.2.

The participants are informed to speak freely that there are no right or wrong answers, and both positive or negative opinions are equally valued in the research. Their anonymity and the sole research purpose of use of the data is assured. Important rules are stated in that participants wait until each other finishes on their speech, and mobiles phones are requested to be switched off. As a practicality, refreshments of coffee and cookies are provided. The discussions are between 1 and 1.5 hours long and are voice-recorded. A complete Danish group interview transcription and a summary table of Korean group result in English are included in Appendix 9.9.