Methodology and Method
This section starts with the methodology, which will discuss the chosen research philosophy and approach. This chapter will also go through the method used to define, collect, and analyze the data. The first part comprises the philosophical basis for the research and is followed by an explanation of the method chosen to study the phenomena in question.
Within the methodology, the philosophical foundation of the research and the research de-sign will be presented.
Philosophical Foundation of the Research
In accordance with the qualitative approach chosen to explore the Pinterest usage of Swedes, the philosophical foundation of this thesis is based on an interpretivistic point of view. Interpretivism differs from positivism, which has its roots in the natural science, re-garding their scientific aims (Schwandt, 2000; Malholtra & Birks, 2007). An interpretivist is trying to understand the research subject, whereas the positivist is searching for a universal explanation or law that can be used to predict future incidence (Schwandt, 2000). As “hu-man actions are meaningful” one needs to understand the meaning behind an action in order to draw conclusions from it. These meanings can differ depending on the context (Schwandt, 2000: 191). The interpretivistic research philosophy is based on the idea that there exists no ‘context-free’ theory and that every interpretation of a text or information has to be viewed within its individual framework (Bender, 2014). With regards to the research subject, the in-terpretivistic research philosophy best reflects the goal of this thesis, which is, in depth, to understand the Pinterest usage of selected Swedes, rather than delivering generic explana-tions why people are using this platform. Therefore, the results of this study need to be understood within the context of Swedish users and Pinterest and they cannot automatical-ly be applied to other countries or social media platform
In the context of user behavior, a qualitative research approach has certain advantages over quantitative research. Qualitative research has for example a low degree of abstraction, which results in proximity to the research object that is often lost in quantitative research (Heinze, 2001). Also, as there has neither been sufficient research done on the usage of Pinterest in Europe nor Sweden, qualitative research is the most suitable approach for ex-ploring an unknown subject, since it does not require the formulation of specific hypothe-sizes, that would need to be tested in quantitative research (Heinze, 2001; Malhotra & Birks, 2007; Dworkin, 2012). Further, qualitative research allows conducting a study with a relatively small group of participants compared to the amount of primary data that is need-ed to draw conclusions from quantitative research (Malhotra & Birks, 2007; Dworkin, 2012). Since the usage of Pinterest in Sweden is not very common yet, finding an adequate number of participants is in any case difficult. However, the main advantage of qualitative research and an important task of this study is the ability to truly understand the partici-pants’ behaviors, feelings and motivations when using Pinterest (Dworkin, 2012; Malhotra Birks, 2007). Qualitative studies enable the researcher to uncover these often-subconscious thoughts through bonding with the participants and probing (Malhotra & Birks, 2007). Finally, it can be concluded that only a qualitative approach can picture the holistic dimension of the research subject: Swedes usage of Pinterest.
With regards to the fairly limited theory and research in the field of Pinterest usage, conclu-sions from this study will be drawn by abductive reasoning. Unlike deduction, where the researcher uses general and sound theory to explain specific empirical findings (Gravetter Forzano, 2012) or induction, where the observation of many single empirical events lead to general theory, abduction is not seeking for a universal explanation (Malholtra & Birks, 2007; Thagard & Shelley, 1997). Abductive reasoning tries to find the most likely explana-tion and therefore also a set of incomplete information can be used to draw conclusion from (Stanford Encyclopedia of Philosophy, 2011). This is important, since only a limited amount of selected participants will be interviewed within this study, which, to some ex-tent, can be seen as incomplete information. In the process of abductive reasoning, existing theory or a set of hypotheses are used to explain empirical findings (Thagard & Shelley, 1997). In some cases though, existing theories need to be adapted to new scientific insights. This advantage of abductive reasoning is crucial, since the questionnaire of this study, used for the interviews, is based on four distinct consumer behavior theories, which might need to be extended or reduced depending on the outcome of the interviews. Further, in a quali-tative and exploratory study, where, as it has been explained, no generalizations to a large population can be made, abduction is the appropriate method of inference.
Generally, there are three distinctive research designs known as descriptive, causal and ex-ploratory research. As it can be derived from its name, the objective of descriptive research is to describe and measure individual variables (Malholtra & Birks, 2007). The research de-sign is very structured and based on specific hypothesizes that need to be tested. However, compared to causal research design, descriptive research is not investigating the relation-ships between different variables (Gravetter & Forzano, 2012). Exploratory research differs from the other two research approaches through its flexibility (Malholtra & Birks, 2007). The focus of exploratory research lays on understanding the research subject and gaining new insights without the necessity of quantitative measurement (Malholtra & Birks, 2007). In the light of the limited research that has been done regarding the use of Pinterest, there is no background of previous studies to form hypothesizes from, which makes neither de-scriptive nor causal research an appropriate research design at this point. Instead, an ex-ploratory research designs is seen as the adequate method, to explore the research subject from different angles. Especially with regards to the research questions of this study, which, among others, want to explore the underlying motivations people have for using Pinterest. Further, the question of how users perceive brand activity requires a deep under-standing, independent of quantitative measurements and hypothesizes, of how users inter-act with this new social media platform and what they expect from it.
The method consists of information about the data collection, in-depth interviews and how the analysis of the data was conducted.
Secondary data is, other than primary data, not specifically collected for the current re-search problem (Malholtra & Birks, 2007). Despite that, secondary data is important for the theoretical background in order to reach a broad understanding of the research problem. In this case previous developed theories were used to develop a question guide for the in-depth interviews. Furthermore, secondary data can aid the interpretation and verification of findings from primary data collection (Malholtra & Birks, 2007).
Primary data collection enables the researcher to gather information, which directly ad-dresses the research question (Malholtra & Birks, 2007). For this thesis, primary data was collected through qualitative semi-structured in-depth interviews with Pinterest users from Sweden. As the advantages of qualitative research over quantitative research with regards to the research problem have already been explained earlier, the next step is to understand how the quality and accuracy of qualitative data can be assessed.
Trustworthiness in Qualitative Research
Every study, whether it is of qualitative or quantitative nature, needs to be “open to critique and evaluation” (Long & Johnson, 2000: 30). Quantitative studies are evaluated based on their reliability and validity. Reliability refers to the extent to which findings from a study can be repeated and the method and same measurement techniques will lead to the same results (Malholtra & Birks, 2007). Validity questions the appropriateness of the measure-ment tool and asks whether the chosen tool is actually measuring what it claims to do (Malholtra & Birks, 2007; Long & Johnson, 2000). The use of the terms validity and relia-bility for qualitative studies is a highly controversial topic among researchers and the con-cept trustworthiness is often found to be more appropriate (Shenton, 2004). To reach trustworthiness, four criteria have been identified by Lincoln and Guba (1985) and are ac-cepted by most qualitative researchers that need to be fulfilled: credibility, transferability, dependability and conformability.
Credibility in this case can be compared to internal validity and refers to the accuracy of the research data in representing reality (Lincoln & Guba, 1985). There are many different pro-cedures that can be undertaken to ensure credibility; the following was realized within this study. First, the adoption of well established research models and method for data analysis. In this case well-known models such a as the Technology Acceptance Model (TAM) and the Theory of Reasoned Action were used to develop a question guide, further the method of content analysis was used to analyze the data (Shenton, 2004). Second, through pro-longed engagement in the usage of Pinterest, the researchers themselves have gathered in-formation and knowledge on how to approach this topic (Lincoln & Guba, 1985; Shenton, 2004). Third, equivalence, a technique that uses alternative wordings for questions with the same meaning and projective questioning, a technique which “requires the subjects to give opinions of other people’s actions, feelings or attitudes” (Donoghue, 2000: 49), were used to increase honesty during the interviews (Long & Johnson, 2000; Shenton, 2004). As a fourth procedure, peer scrutiny was used to receive feedback from colleagues regarding the approaches and methods used to explore the topic of Pinterest usage in Sweden (Shenton, 2004; Long & Johnson, 2000). Fifth, the researchers kept a reflective journal throughout the interview process in order to constantly evaluate e.g. the effectiveness of the chosen methods (Long & Johnson, 2000; Shenton, 2004). At last, each interviewee was asked to validate his or her transcripts, in order to improve the credibility of the data (Long & John-son, 2000; Lincoln & Guba, 1985; Shenton, 2004).
The second criterion for establishing trustworthiness in qualitative research is transferabil-ity (Lincoln & Guba, 1985). Transferability is supposed to represent external validity or generalizability, which is not a strength of qualitative studies, since they can hardly be ap-plied to a context other than the one investigated in the study (Long & Johnson, 2000; Shenton, 2004). However, a “thick description” (Shenton, 2004: 70) of how the study was conducted, as it is presented in the methodology and method part, facilitates transfers (Lin-coln & Guba, 1985).
Dependability asks whether the study’s results are consistent and if the same results can be repeated, it therefore is similar to reliability used to evaluate quantitative studies (Lincoln & Guba, 1985). Again, a repetition of qualitative research may be difficult due to its strong contextual character, but a detailed description of the research design, how information were gathered and a reflection regarding the effectiveness of the chosen methods increase the study’s dependability (Shenton, 2004).
The last criterion that ensures the trustworthiness of a qualitative study is conformability, which evaluates a study’s neutrality (Lincoln & Guba, 1985). Especially in qualitative stud-ies, a researcher’s motivation and interest can bias the outcome of the study. Therefore, de-tailed reasoning with regards to the strength and weaknesses of the chosen methods and a strong focus on the researchers own reflectivity maximize conformability (Lincoln & Guba, 1985; Shenton, 2004).
In-depth interviews were chosen as the appropriate qualitative method for conducting this study. In-depth interviews are personal and direct interviews, where only one participant at the time will be questioned through an experienced interviewer (Malhotra & Birks, 2007). This method is used to gain deep and rich knowledge and understandings from individuals on a specific topic (Nagy Hesse-Biber & Leavy, 2011; Chirban, 1996). Especially, the inter-viewees’ personal experience with Pinterest, their emotions and motivations behind the us-age and the everyday context in which they are using this social media platform is im-portant with regards to the research questions (Chirban, 1996; Malhotra & Birks, 2007).
The question guide of the in-depth interviews is semi-structured, which means that there is a predetermined set of open ended questions, but depending on the process of the inter-view, the interviewer will ask follow-up questions, probe and through active engagement motivate the participant to give rich information (Nagy Hesse-Biber & Leavy, 2011). The interviews took place in private surroundings, where the participants were able to relax and feel comfortable answering to the questions; the participants were further informed that their identities would be kept anonymous (Chirban, 1996). In this study some interviews were online in-depth interviews, which means they were conducted over Skype or another online video call software. This is due to the fact, that some interviewees are not living within the reach of the interviewers. In total fourteen interviews were conducted. In gen-eral, there exist no universal rule about the sample size of in-depth interviews and research-ers have been trying to quantify this issue throughout the years with recommendations var-ying between 5 and 50 interview participants (Dworkin, 2012). Nevertheless, most re-searchers agree, that the point of saturation, when any further data collection is not giving any new insights, is an important indicator for when sufficient data has been found (Dworkin, 2012). The researchers believe, the point of saturation was reached with four-teen participants; otherwise the number of interviews would have been adapted.
The interviews were on average 45 minutes long. In the beginning of the interview, the in-terviewer clarified the goals and objectives to the participant. Through this procedure, the participants better understood what is expected from them and were additionally given the possibility to clarify questions for their part (Chirban, 2012). Further, throughout the inter-view, the interviewers were attentive and empathetic with the participant in order to suc-cessfully probe and receive the desired depth of answers (Chirban, 2012). Despite that, the interviewers needed to be self-aware of their individual impact on the participant (Chirban, 2012). Any conspicuousness was part of the notes the interviewers took during the inter-view session and were further written down in the reflection journals that both researchers kept. The notes from the interview were additionally supported by audio-records and the journals were discussed between the interviewers in order to improve later interviews. The question guide for the interviews can be found in appendix 2.
1.2 Problem Definition
1.4 Research Questions
1.5 Delimitation of the Study
1.6 Key Terms
2 Theoretical Framework
2.1 Social Media Usage and the Place of Pinterest
2.2 Need Fulfillment
2.3 The Theory of Reasoned Action
2.4 The Technology Acceptance Model
2.5 Social Media Marketing and the Awareness of Brands on Pinterest
3 Methodology and Method
4 Presentation of Empirical Findings
4.1 Need Fulfillment Findings
4.2 Theory of Reasoned Action Findings
4.3 Technology Acceptance Model Findings
4.4 Awareness of Brand Activity Findings
5 Analysis and Discussion
5.1 Need Fulfillment Analysis
5.2 Theory of Reasoned Action Analysis
5.3 Technology Acceptance Model Analysis
5.4 Awareness of Brand Activity Analysis
6.1 Answers to Research Questions
GET THE COMPLETE PROJECT