CHAPTER 3. RESEARCH DESIGN AND METHODOLOGY
3.2. Mixed methods and triangulation in case study
3.2.1. Mixed methods
3.2.1.1. What is mixed methods?
Mixed methods may be described as a type of research ‘where there is a substantial element of qualitative data collection as well as a substantial element of quantitative data collection in the same research project’ (Robson, 2011: 161). However Robson (2011) stated that he preferred to refer to ‘multi-strategy designs’
rather than mixed methods, because ‘they involve not only combining methods in some way but also use more than one research strategy’ (Robson, 2011: 161). A mixed methods design is not simply understood as the mixed research form with qualitative and quantitative data themselves. Tashakkori and Teddlie (2003:10) stated thus: ‘For several years, the terms multi-method design and mixed methods design have been confused with one another’. They explained multiple method designs as a research design that ‘more than one method or more than one worldview is used’ (Tashakkori & Teddlie, 2003:10). Multiple method designs are divided into three categories; multi-method research, mixed method research, and mixed model research by Tashakkori & Teddlie (2003). They are called mixed methods research as an umbrella term for mixed methods and mixed model research.
Table 3.3. Three categories of multiple method designs Multi-Method
Research
The research questions are answered by using two data collection procedures or two research methods, each of which is from the same QUAL or QUAN tradition Mixed Method
Research
Qualitative and quantitative data collection and analysis techniques are used in either parallel or sequential phases. This mixing occurs in the methods section of a study. Mixed Model
Research
Mixed model research is mixed in many or all stages of the study (questions, research methods, data collection and analysis, and the inference process). It is indeed possible to have two paradigms, or two worldviews, mixed throughout a single research project.
In essence, mixed methods designs are not only the research design to simply mix qualitative quantitative research methods within one research project, but also the research strategy to pursue the harmony of qualitative and quantitative aspects in all stage of the study as a cover term for mixed method research and mixed model research.
3.2.1.2. Pragmatic mixed methods
Punch (2009: 289) summarized the methodological history as below.
The dominance of quantitative methods as wave 1
The emergence of qualitative methods as wave 2
The growth of mixed methods as wave 3.
According to Robson (2011:164) ‘the positivist quantitative paradigm dominant between the 1950s and mid-1970s was followed by one in which the qualitative interpretivist /constructivist research paradigm became established as a viable alternative in the mid-1970s to the 1990s.’ After the quantitative qualitative paradigm war, pragmatic mixed methods emerged as the third alternative research methodology since the 1990s.
When a researcher starts a research project, the question of whether it is qualitative or quantitative research at first appears important. However, this
dichotomy is not accepted in pragmatism. Pragmatists reject the distinction between quantitative and qualitative research, and often adopt a mixed methods approach as a way of avoiding traditional dualism (Denscombe, 2007). It is for this reason that
Maxcy (2003: 79) states that pragmatism ‘seems to have emerged as both a method of inquiry and a device for the settling of battles between research purists and more practical-minded scientists’.
3.2.1.3. The special features of mixed methods
One of the biggest benefits of mixed methods is that it brings the
combination of the specialities of qualitative and quantitative research, and at the same time the compensation of the weak points of each (Punch, 2009). Quantitative research allows for ‘conceptualizing variables, profiling dimensions, tracing trends and relationships, formalizing comparisons and using large and perhaps
representative samples’; meanwhile, qualitative research allows ‘the strengths of sensitivity to meaning and to context, local grounded(-)ness, the in-depth study of smaller samples, and great methodological flexibility which enhances the ability to study process and change’ (Punch, 2009: 290). Mixed methods can bring both strengths. While undertaking an in-depth study of smaller samples and seeking to trace trends and relationships with quantitative data, and while formalizing comparison, qualitative data provide the sensitivity to meaning and to context. Therefore, a mixed methods approach supplies a more complete, broader view of the subject of the research (Robson, 2011 based on Bryman 2006). Mixed methods can help us to understand the complicated nature and phenomenon of the real world research, and to approach a wider range of research questions through qualitative and quantitative methods, to study the same subject. In addition, it helps to obtain
research (Robson, 2011: 167 based on Bryman 2006). However Denscombe (2010: 147) emphasizes that the researcher has to have a clear logic for using distinct research methods, so as to empower a mixed methods design.
3.2.1.4. The types of mixed methods design
Tashakorri and Teddlie (2003) mentioned the difficulty of dealing with this issue because there are around forty different types of mixed methods designs in literature. Punch (2009) introduced Creswell and Plano Clark’s (2007) division of mixed methods; triangulation design, embedded design, explanatory design, and exploratory design. In triangulation design quantitative and qualitative data are collected with equal significance in the same time structure. Normally, they are collected and analysed separately, but are blended for data transformation and interpretation (Creswell and Plano Clark, 2007: 62-4). Embedded design indicates a mixed method design, in which one type of data supports the other type of data at the secondary level, so qualitative and quantitative data are not regarded as equal in embedded design. The basic idea of this design is that data from only the qualitative or quantitative method are not adequate to answer different forms of question within a study (Creswell and Plano Clark, 2007: 67-71). Explanatory design is a two stage mixed methods design, in which qualitative data are used to develop an explanation for the project. Usually, quantitative data are gathered first, and qualitative methods follow (Creswell and Plano Clark, 2007: 71-2). An exploratory design differs from an explanatory design. The first phase is qualitative, while the second phase is
quantitative (Creswell and Plano Clark, 2007: 75). Punch (2009) states that the division of each design is made based on three facets; timing, weighting and mixing.