Research design

3.1.1 Research aim

The goal of this research was to provide empirical evidence that offering the elements digitalization, socialization and personalization (together: an in-store social experience) would indeed increase the store visit intention of consumers and thus the necessity of physical stores in luxury fashion brands.

The elements have been covered in existing literature. However, the effects of those elements on store visit intention has not been researched yet. This research answered the question if these three elements have a positive effect on millennials and Generation Z, and would in fact attract them to physical luxury fashion stores and thus increase their store visit intention. This is relevant, because if so, it will mean that luxury fashion brands can attract people that make up almost 70% of the luxury market in five years to their physical stores, therefore securing store necessity in the future. Thus, the elements of an in-store social experience could play a significant role in the luxury fashion industry and in the necessity of the physical stores of the luxury fashion brands. Therefore, the research question this thesis aimed to answer was: How do various elements of an in-store social experience affect the store visit intention in luxury fashion brands?

To properly answer this research question, several sub-questions were formulated:

1. What is an ‘in-store social experience’ in luxury fashion stores and what elements does it consist of?

2. What effect do digital elements have on the store visit intention of millennials and Generation Z when it comes to luxury fashion stores?

3. What effect do social elements have on the store visit intention of millennials and Generation Z when it comes to luxury fashion stores?

4. What effect does personalization have on the store visit intention of millennials and Generation Z when it comes to luxury fashion stores?

3.1.2 Method

The research was explanatory. In order to scientifically test the hypotheses and answer the proposed research question, quantitative research was conducted by means of a vignette experiment in survey research (Atzmüller et al., 2010). Therefore, the research was correlational, to see if the elements of the in-store social experience were paired with an increased store visit intention of consumers in the luxury fashion industry. The researcher thus collected primary data to understand the intentions and opinions of multiple groups of people in certain situations. A survey was used as the data collection instrument to gather cross-sectional data, because this is useful to generalize the findings and for using statistics to test hypotheses. It is also the most common in this research field. One survey with all of the vignettes was administered digitally, to create a within-subjects vignette research design with five conditions (digitalization, socialization, personalization, all three elements, no element).

While research about the effects of these three elements on store visit intention in the luxury fashion industry has not yet been conducted, there are a few related subjects that have been researched by means of a survey. One of them is about one of the elements, digitalization, in fashion retail stores (Kim et al., 2020). This research investigates how the digital atmosphere affects consumers’

purchase behavior patterns. The method that was used was a survey. Another example is a research paper about the effect of smart, innovative technologies in physical store retail settings, which was also investigated by means of a survey (Vojvodić, 2019). A third example is a research paper about personalization and its effect on customers while in a physical store, which is also researched by means of a survey (Wetzlinger et al. 2017).

Since these research papers are slightly related to this research, which contains multiple elements and their effect on the increased store visit intention, this thesis also used a survey as the method for the research.

Increasingly more experiments in survey research have gained attention, since the experiment’s internal validity is complemented by the external validity of surveys (Schlüter & Schmidt, 2010;

Gaines et al., 2007; Sniderman & Grob, 1996). Specifically, vignette experiments embedded in surveys—or factorial surveys— are getting more attention (Atzmüller & Steiner, 2010). A vignette experiment within survey research was the best choice for this particular research, because this research consisted of measuring the intentions of consumers in multiple scenarios concerning the in-score social experience in physical luxury fashion stores. The scenarios entailed the situation that the luxury fashion store offered one of the three elements of the in-store social experience

(digitalization, socialization or personalization). The fourth scenario was the situation that the luxury fashion store offered none of the elements. An extra scenario was that the luxury fashion store offered the full in-store social experience, containing all three elements. Every respondent got to see all scenarios, after which they needed to indicate their store visit intentions. A vignette experiment through survey research was the ideal way to research these five situations, because a vignette experiment consists of presenting participants with realistic and carefully constructed scenarios (vignettes) to research dependent variables like behaviors, intentions and attitudes, thus enhancing experimental realism and also enabling researchers to control and manipulate independent variables (Aguinis and Bradley, 2014). This is why conducting a vignette experiment was a good way to measure multiple situations regarding the in-store social experience in physical luxury fashion stores presented to the same group of millennials and Generation Z.

3.1.3 Internal and external validity

Internal validity refers to the validity of inferences about the cause effect relationship between the presented vignette stimuli and respondents’ reaction to the stimuli (Shadish, Cook & Campbell, 2002). Internal validity was established by experimental control which enables unique assessment of the vignette factors’ causal effect on the outcome variable. This means that the effect estimates are free of bias. External validity is the extent to which the results of a study can be generalized to and across people, stimuli, other situations, and times (Aronsen et al. 2007). Reliability of a vignette experiment means a balanced and blocked experimental design, reliable vignette measurements, and

a stratified respondent sample of sufficient size, resulting in sufficiently powered hypothesis tests and precize effect estimates (Hinkelmann & Kempthorne, 1994).

An advantage of vignette experiments in survey research is that vignette questions are less abstract and more realistic than regular surveys, the multivalent character of vignettes made a simultaneous investigation of the factors varied in the vignette experiment possible, interaction effects among vignettes could be tested and estimated. Vignettes are flexible and can be used for different purposes and different formats. Another reason why a vignette experiment was the ideal method of research here, is because it was the best way to solve the dilemma of experimental research that results in high levels of internal validity but also in threats to the generalizability and external validity versus conducting nonexperimental research (like surveys) that usually has outstanding external validity but whose conclusions are ambiguous regarding causal relationships (Aguinis and Bradley, 2014). And the flexibility of the vignettes was a good technique to avoid socially desirable or politically correct answers. Finally, many respondents preferred vignettes over monotonous survey questions. All these advantages contributed to the internal validity, reliability and construct validity. Highly contextualized vignettes increase construct validity, which is the degree to which the vignettes measure what the researcher intends to measure. And embedding the vignette experiment in a survey (with random sampling) extends a vignette experiment’s external validity to the survey’s target population, and thus made sure of generalizable results. This is because for example a high level of realism presented in the stimulus presentation, so that the scenarios simulate more approximately experiences in the real world (Heslin, Vandewalle, & Latham, 2006;

Hughes & Huby, 2002).

Thus, the vignette experiment in survey research had both very good internal and external validity and good reliability, which made this a strong way of testing this relatively new research subject.