The aim of this chapter is to identify the suitable research philosophy, research approach, research design and research strategy for the process of gathering primary data. Additionally, the data collection technique and analysis procedure for the empirical findings is outlined. At the end of the methodology, the means of ensuring that the research is trustworthy is presented.
A research philosophy consists of vital assumptions of how researchers view the world. The research philosophy applied in a study needs to be well-considered, since it establishes how the environment is interpreted by the researchers. Moreover, the chosen philosophy affects the latter steps of the methodology such as the research strategy and method used to conduct the research (Saunders, Lewis & Thornhill, 2009). According to the research ‘onion’ by Saunders, Lewis and Thornhill (2009), there are four focal research philosophies being positivism, realism, interpretivism and pragmatism. Each philosophy differs in terms of ontology, epistemology, and axiology, which essentially refers to characteristics that create a perspective of how knowledge is perceived, and which methodological strategies that are used to obtain certain scientific knowledge (Saunders et al., 2009; Bryman, 2012).
As the aim of this study is to explore consumers’ attitudes and purchase intentions regarding legume-based products, we are concerned with gathering valuable human insights to be able to fulfil the given purpose. Therefore, the philosophy of interpretivism is a recommended approach within this field of study, since this philosophy relates to the study of social phenomena in people’s natural environment. This refers to the necessity for the researchers to understand the differences between humans and their roles as social actors. Interpretivism emphasises the conduct of research among humans rather than objects, adopting an empathic stance as a researcher to comprehend people’s social context and the meaning these individuals contribute with from their point of view (Saunders et al., 2009; Saunders & Tosey, 2012). Yet another advantage of an interpretivist approach lies in its ability to address the complexity and meaning of situations related to consumption (Black, 2006), which this study aspires to investigate. Additionally, an interpretivist research philosophy adopts a subjective approach which allows for investigating issues in greater depth, without the concern of objectivity (Bryman, 2012). Collection and analysis of data using an interpretivist philosophy are likely to involve qualitative data from in-depth investigations with smaller samples (Saunders & Tosey, 2012).
Having decided on an appropriate research philosophy for the study, a suitable research approach needs to be determined. According to the research ‘onion’ by Saunders et al. (2009), there are primarily two different research approaches that are applicable depending on how the process of research is related to theory. These approaches are known as deductive and inductive. Through deductive reasoning, a theory and hypotheses are developed and the research strategy is thereafter intended to test the given hypothesis. With an inductive approach, data is collected and a theory is developed based on the analysis of the data. Since research approaches are connected to the different research philosophies, deduction is more applicable to a positivistic philosophy and quantitative studies, whereas induction is more applicable to an interpretivist research philosophy and qualitative research methods (Saunders et al., 2009). Since the approach of deduction is more applicable to a quantitative study, and thus, this approach will not be adapted in this thesis. Alvesson and Sköldberg (2009) suggest that the approaches of deduction and induction tend to be one-dimensional and have the tendency to limit the research in certain situations. When that is the case, a third research approach of abduction is applicable (Alvesson & Sköldberg, 2009).
An abductive approach shares similarities with both inductive and deductive approaches in the sense that the researcher emphasises gaining understanding and drawing generalizable inferences from empirical findings similar to that of an inductive approach, while not dismissing the use of existing literature to develop theory in line with deductive reasoning (Alvesson & Sköldberg, 2009; Saunders et al., 2009). This approach facilitates moving back and forth between empirical findings while comparing and reinterpreting existing literature (Suddaby, 2006). It is a more flexible research structure that allows for changes of research as the research progresses, and can collect data through qualitative research methods (Saunders et al., 2009; Bryman, 2012). The abductive approach is proficient when there are large amounts of information available to a broader context, as for example food, but is not necessarily concerning the subject associated with the specific study (Bryman, 2012), applicable to this thesis investigating legume-based food. The reason why an abductive research approach is appropriate in this study is because the authors aim to explore factors affecting consumers purchase intentions similar to that of inductive reasoning. The authors have built a model to explore attitude and intention formation of health-conscious consumers’ regarding legumes, based on previous academic findings. However, the empirical findings are intended to qualitatively examine if the factors are equivalent to the proposed modified theory of planned behaviour model. Moreover, the abductive approach aids in answering the given research questions and fulfilling the given purpose since it allows the authors to observe human behaviour as well as identify patterns in existing theory more proficiently (Bryman, 2012).
A well-formulated research design is the blueprint of a marketing research project, as it specifies the procedures necessary to obtain the information desired in order to fulfil the purpose of the study (Malhotra & Birks, 2007). A research design can broadly be classified as either exploratory or conclusive. An exploratory research design seeks to understand, explore and provide insights on a subject that is difficult to measure, whereas a conclusive design aims to describe a specific phenomenon in a structured manner where existing information is clearly defined by, for example, testing hypotheses (Malhotra & Birks, 2007).
This study will adopt an exploratory design to obtain the needed information and data in order to achieve the intended aim of the investigation, which is to explore which key components that affect HCCs intentions to purchase legume-based products. The motivation of the choice of an exploratory research design lies in the fact that this approach does not require a structured and defined understanding of the problem, but rather intends at exploring and understanding the phenomena. Moreover, an exploratory design is considered favourable when the researcher is attempting to interpret attitudes and opinions from the investigation (Malhotra & Birks, 2007), which this study intends to uncover.
Determining an appropriate research strategy is the bottom line of the methodology. A well-defined research strategy enables the researchers to achieve the intended research objectives (Saunders et al., 2009). The research strategy is the general orientation to the conduction of the research (Bryman, 2012). Bryman (2012) considers predominant strategies in social research to be either quantitative or qualitative, and each strategy differs significantly in the pattern of obtaining information.
A quantitative research strategy emphasizes quantification during the data collection and analysis phase. It usually adapts a deductive approach to the relationship between research and theory, where the emphasis is placed on the testing of various theories. The norms and practices of a quantitative strategy are related to a positivism philosophy where the view of social reality is perceived as external (Bryman, 2012). A quantitative method is mostly used for data collection that desires numerical measures to perform statistical analyses, which this study does not aspire. The research questions of this study are non-numerical in nature, which suggests a qualitative method to be more appropriate (Saunders et al., 2009). Based on these statements, a quantitative study will not be adopted in this study.
Contrary, a qualitative research strategy puts emphasis on words rather than numerical measures. It is making use of non-numerical data that has not been quantified (Bryman, 2012; Saunders et al., 2009). Qualitative research is linked with the philosophy of interpretivism, where focus lies on gaining insight and understanding underlying factors such as actions, values and opinions of individuals (Saunders et al., 2009). Qualitative research is furthermore closely related to an abductive or inductive approach to the relationship between theory and research, predominantly emphasising the generation of theory from verbal interactions between individuals (Bryman, 2012). Malhotra & Birks (2007, p. 152) define qualitative research as “An unstructured, primarily exploratory design based on small samples, intended to provide depth, insight and understanding”. Based on these motivations, the use of a qualitative research strategy is deemed more suitable in this study. The aim of this thesis is to understand and describe rather than to measure consumers’ attitudes and purchase intentions, further affirming the choice of research strategy. Recommended primary data collection techniques for a qualitative research are in-depth interviews, focus group discussions and participant observations (Malhotra & Birks, 2007), and the choice of empirical investigation method will be discussed next.
Primary Data through In-Depth Interviews
The authors have chosen to conduct in-depth interviews as the primary empirical investigation of this study. An in-depth interview is an unstructured personal conversation, between an interviewee and interviewer, with the intent to uncover underlying motivations, beliefs, attitudes and feelings about a specific topic (Malhotra & Birks, 2007). The purpose of qualitative interviews is to derive meaning through interpretations from conversing with participants, rather than from clearly stated facts. Therefore, emphasis is based upon a full interaction from the researcher’s perspective to understand the meaning of the participants’ experiences and context (Malhotra & Birks, 2007).
In-depth interviews can be classified as either structured, semi-structured or unstructured. Structured interviews can be described as verbally conducted questionnaires, where a list of predetermined questions, with little or no variation and ability for follow-up questions to further elaborate upon, are asked to the participants. Contrary, unstructured interviews are performed with low organisation, starting with an open question and will progress based upon the initial response. Unstructured interviews can be difficult to both manage and participate in since the lack of predetermined questions provide difficult guidance in the communication between the interviewer and the interviewee, and should only be considered when virtually no prior information about the subject area is present. Lastly, there exist semi-structured interviews. These interviews consist of several key questions that help to define the areas that are to be explored, but they allow the interview to deviate from the pre-determined questions in order to uncover more depth and detail by ecologically generating follow-up questions. A semi-structured interview approach is deemed more flexible in comparison to structured interviews, while also allowing for discovery of information that might previously not been thought of by the researchers (Gill, Stewart, Treasure & Chadwick, 2008), and will therefore be the deployed interview technique in this study.
Interviews, compared to other qualitative research methods such as focus groups, are structured to be personal and direct and are based around the setting of questioning one participant at a time (Malhotra & Birks, 2007). This allows in-depth interviews to uncover a greater amount of depth and insight than focus groups, by concentrating and developing an issue directly with an individual, which cannot occur as proficiently in a group setting. In-depth interviews further excel in a free exchange of information between the two parties, without the possible obstacle of social pressure that can occur in a group scenario (Malhotra & Birks, 2007). Furthermore, in depth-interviews provide the researcher with the opportunity to further probe participants responses, which adds further significance and depth to the obtained qualitative data (Saunders et al., 2009). Based on this argumentation, this particular qualitative method will be conducted to answer the posed research questions and gather a deep insight about which factors that influence health-conscious consumers’ attitudes and purchase intentions regarding legume-based products.
When deciding upon which method that should be used to collect primary data, researchers should be aware of its drawbacks, and how to act according to these. One of the most prominent limitations of interviews involves the risk of bias, since the direction of the questioning as well as the interpretation of the final results are both dependent on the interviewer (Malhotra & Birks, 2007). In-depth interviews, and qualitative research techniques in general, are dependent on underlying reasons that can be difficult to detect by the researcher carrying out the interview. For instance, data can be difficult to analyse and interpret due to hidden meanings and interpretations in how participants express themselves (Malhotra & Birks, 2007). Moreover, the results that are derived from the in-depth interviews corresponds to a specific target population, meaning that attempting to generalize to a wider population by the same means as with a quantitative study could end up with misleading results (Malhotra & Birks, 2007).
In order to arrive at more dependable results for this study, the authors have made use of data that has been obtained by other researchers as a complement to the primary findings. Saunders et al. (2009) compiled three main sub-groups that compiles the various types of existing secondary data: documentary data, survey-based data and data compiled from multiple sources. For this thesis, documentary data was the most relevant secondary data as it consists of books, journal articles, newspaper articles and organizations’ websites (Saunders et al., 2009), all of whom have been used throughout this study. The documentary secondary data for this investigation has predominantly been collected through the digital library database of Jönköping University, Primo, which provides a large number of full text documents consisting of academic journals, e-books, research studies and other publications (Jönköping University Library, 2016). As a complement to the digital library of Jönköping University, Google Scholar was also used to a lesser extent as a mean to further obtain documentary secondary data for this study (Google, n.d.). Examples of search terms that have been used to acquire secondary data for this investigation are; ‘legumes’, ‘organic food’, ‘consumer attitudes’, ‘purchase intentions’, ‘consumer behaviour’ and ‘theory of planned behaviour’.
Selection of Sample and Sample Composition
After having decided on an appropriate data collection method, the task is to specify the target population and decide upon which sampling technique to employ when sampling the participants for the in-depth interviews (Saunders et al., 2009). In the delimitations of this thesis, it is argued that the focus of the empirical study will be on individuals considered to be health-conscious. Health-conscious consumers refer to individuals that actively care about their diet and lifestyle, striving for both physiological and psychological wellness. That entails to a person’s activities, interests and opinions deemed to be beneficial to his or hers physical, emotional or mental state (Kraft & Goodell, 1993; Healthy, n.d.). The reason for using this particular sample is motivated by the fact that these individuals’ interest in living a healthy lifestyle is in line with the beneficial aspects of consuming plant based-viands (Konsumentföreningen Stockholm, 2015). Additionally, there is a proven substantial growth of this subpopulation (Folkhälsomyndigheten, 2016), which is considered to be of high importance for development of the Swedish market by the four-year transdisciplinary research project that this thesis is a part of. Since the intention of the four-year research project is to develop the Swedish legume-based product market, the sample will solely consist of Swedish consumers. Concerning the aspect of age and gender, the researchers have decided to aim for an equal representation of male and female participants with a representation of ages ranging from age 20 to 58. The motivation of this lies in the ability to cover a larger span of consumers and obtain a more realistic representation of the sample. Further, this will make the authors’ able to explore if there exist potential differences between different sexes and age cohorts. To further clarify, the target population of this empirical study will consist of an equal representation of Swedish health-conscious men and women, aged between 20 to 58.
This investigation will implement a non-probability judgemental sampling technique for the in-depth interviews. This technique enables researchers to select participants who are suitable for the purpose of the study, based on the judgement of the authors, rather than by an equal chance of selecting participants with a probability sampling method (Malhotra & Birks, 2007). This method is particularly suitable for exploratory studies and is often used when a study is conducted with a relatively small sample (Malhotra & Birks, 2007; Saunders et al., 2009). However, since the technique depends on the researcher’s judgement, the authors need to bear in mind the risk of being subjective and biased when selecting the participants to be included in the final sample (Neuman, 2005). The positive aspects of judgemental sampling are that it is a relatively convenient and inexpensive method of defining the target population, particularly advantageous when broad population assumptions are not required (Malhotra & Birks, 2007).
In the non-probability judgemental sampling, participants will be selected based the following conditions. First and foremost, the participants need to be considered as health-conscious. It is, however, difficult to sample participants based on the nutritional and physiological entitlements that health-conscious individuals possess, and therefore the judgemental sampling procedure will solely focus on the physical exercising aspect of being considered health-conscious. The Swedish National Institute of Public Health recommends a daily low-intensity physical activity of 30 minutes for individuals to enhance their health and physical well-being. Additionally, it is shown that 20 minutes to an hour of high-intensive physical activity three times per week dramatically enhances human health (Ståhle, 2008). Based on these recommendations, the researchers will assure that each participant meet the aforementioned criterias before conducting the empirical investigation. Additionally, the participants should be consumers within the geographical boundaries of Sweden, to further ensure that the intention of the four-year research project this thesis is a part of stays relevant.
The optimal sample size when conducting in-depth interviews is contradicting and largely dependent on each specific research (Mason, 2010; Baker & Edwards, 2012). However, a consensus to achieve saturation for a qualitative research project is to conduct at least ten to fifteen interviews (Mason, 2010). To achieve optimal results for this investigation, 16 in-depth interviews will be conducted and will contain an appropriate mixture of gender, age variation and personal characteristics amongst the respondents.
Table of Contents
1.2 Problem Definition
1.3 Purpose and Research Questions
1.6 Definitions of Key Terms
2 Frame of Reference
2.1 Environmental Impact of Current Food Consumption
2.3 Health-Conscious Consumers (HCCs): Understanding their Behaviour
2.4 Theory of Planned Behaviour
2.5 Modified TPB Model by Tarkiainen and Sundqvist (2005)
2.6 The Proposed Research Framework
3.1 Research Philosophy
3.2 Research Approach
3.3 Research Design
3.4 Research Strategy
3.5 Data Collection
3.6 Data Analysis and Interpretation
3.7 Assessing the Quality
4 Empirical Findings
4.1 Sample Display
4.2 Factors Influencing Attitudes
4.3 Attitude Leading to Purchase Intention
4.4 Factors Influencing Purchase Intentions
5.1 Factors Affecting Health-Conscious Consumers Attitudes towards Legumes
5.2 Summary: Factors Affecting Health-Conscious Consumers Attitudes towards Legumes
5.3 Does a Positive Attitude Lead to a Purchase Intention?
5.4 Factors Affecting Health-Conscious Consumers Purchase Intentions towards Legume-Based Products
5.5 Summary: Factors Affecting Health-Conscious Consumers Purchase Intentions towards Legume-Based Products
5.6 Revised Research Framework
6 Conclusion and Discussion
6.1 Purpose and Research Questions
6.4 Suggestions for Future Research
7 List of References
GET THE COMPLETE PROJECT