|RESEARCH SERIES NO. 24
|Year : 2021 | Volume
| Issue : 2 | Page : 143-147
Mixed-methods research: Why, when and how to use
Senior Resident, Department of Community Medicine, Christian Medical College, Vellore Chittoor Campus, Tamil Nadu, India
|Date of Submission||03-Sep-2021|
|Date of Decision||25-Oct-2021|
|Date of Acceptance||05-Nov-2021|
|Date of Web Publication||07-Dec-2021|
Dr. Dorothy Lall
Department of Community Medicine, Christian Medical College, Vellore Chittoor Campus, Vellore, Tamil Nadu
Source of Support: None, Conflict of Interest: None
Mixed-methods research emerged in the late 1970s as a methodological approach that uses both quantitative and qualitative data collection methods in one single study. In this article, I present an overview of what mixed-methods research is and identify its defining features. Key considerations in designing mixed-methods research are priority, sequence, stage of integration and use of theory. Each of these is discussed briefly with pointers that could serve as a starting point in designing a mixed-methods research. There is a brief description of the core mixed-methods research designs, namely convergent, exploratory sequential and explanatory sequential, with some examples relevant to the Indian context. Finally, I present some challenges commonly faced in conducting mixed-methods research.
Keywords: Convergent, explanatory, exploratory, mixed methods, pragmatism, sequential
|How to cite this article:|
Lall D. Mixed-methods research: Why, when and how to use. Indian J Cont Nsg Edn 2021;22:143-7
| Introduction|| |
Health research is often described as complex. There is complexity within the health system, its problems and its solutions. Complexity is described as a dynamic and constantly emerging set of processes and objects that not only interact with each other, but come to be defined by those interactions. As the understanding of health systems and its problems has evolved, the need for methods and tools to answer research questions about the health system has also changed. In this article, I will briefly discuss the emergence of mixed-methods research and present an overview of the three core designs in mixed-methods research, its advantages and some challenges commonly faced in conducting mixed-methods research.
Mixed-methods research is an emergent methodology that systematically integrates quantitative and qualitative approaches to answer research questions. Creswell and Plano Clark define mixed-methods research as studies that include at least one quantitative strand and one qualitative strand. A strand is a part of the study that is defined by the research process of posing a research question, defining methods, collecting and analysing the data. Integration of quantitative and qualitative types of data during the research process is fundamental to mixed-methods research. The 'mixing' of data to answer research questions provides a complete picture and allows researchers to overcome weaknesses inherent in the individual methods and approaches. It provides both an 'in depth' as well as a 'broad' understanding of research problems.
Philosophical underpinnings and the emergence of mixed methods
Mixed-methods research emerged in the social and behavioural sciences during the late 1970s as a third research methodological movement. The two dominant research methodologies till then were the quantitative and qualitative research approaches. Quantitative and qualitative research approaches belong to two distinct research paradigms, the positivist and the constructivist paradigms, respectively [Figure 1]. A positivist paradigm is characterised by an understanding of a single reality that is objectively assessed and measured through the collection and analysis of quantitative data. The constructivist paradigm of research, on the other hand, involves the construction of reality through the experiences of people. The constructivist viewpoint acknowledges multiple realities that are not measurable but discovered through qualitative inquiry of questioning and reasoning. The last two decades of the 20th century were known for the so-called 'paradigm wars' that put these two ways of knowing and studying reality (positivists and constructivists) in opposition. However, in the late 1980s, methodology experts from around the world were simultaneously working on the idea of using both quantitative and qualitative approaches in a single study. Researchers realised that there are strengths in both the approaches that could meaningfully enhance research in which one method alone is used.
A third philosophical position of pragmatism arose from these debates and discussions that allowed researchers to meaningfully use both these research approaches to answer research questions. Pragmatism is based on the proposition that researchers use the philosophical and/or methodological approach that works best for a particular research problem that is being investigated. The research question is king, from a pragmatists viewpoint, and is the driver of the entire research process. Pragmatist scholars are of the particular opinion that there is an objective reality that exists apart from human experience. However, this reality is grounded in the environment and can only be encountered through human experience. Mixed-methods research is rooted in the pragmatic philosophical position that is focused on the consequences of research and on research questions rather than on the methods. Mixed-methods research is particularly well suited to health policy and systems research questions that seek to answer 'why' and 'how' questions in addition to the 'what' questions. For example, the question can be 'Is an intervention package for improving quality of primary care effective and if so, how and why does it improve quality?' The how and why questions are important in health system and implementation research as they allow us to learn lessons for wider application. In the section below, we will look more closely at considerations in the design of mixed-methods research.
Designing mixed-methods research
A research design is a purposeful framework that provides a strategy and method to answer the research question. Specific research designs, therefore, are shaped by the way research questions are formulated. Research questions requiring mixed-methods research designs are mixed in nature, that is, they have questions that need a quantitative answer as well as questions that need qualitative information. Research questions may be framed as an overarching single question requiring both quantitative and qualitative data or as two or more questions. This can give rise to a number of research designs and indeed, Teddlie and Tashakori note 40 mixed-methods research designs reported in the literature
In designing a mixed-methods research, Creswell proposes four important questions that should be answered at the start.
- In what sequence will the qualitative and quantitative data collection be implemented?
- What relative priority will be given to the qualitative and quantitative data collection and analysis?
- At what stage of the project will the qualitative and quantitative data be integrated?
- Will an overall theoretical perspective be used to guide the study?
Sequence of the strands
This is an important consideration in designing mixed-methods research and is aligned to the research question. The framing of the research question decides whether the quantitative strand should precede, follow or be conducted at the same time as the qualitative strand. The two basic research types based on sequence are concurrent and sequential designs. Concurrent designs, as the name suggests, are when both strands are conducted at the same time, in parallel. Sequential arrangements can further be exploratory or explanatory. Exploratory sequential designs are when the qualitative strand precedes the quantitative strand. Explanatory mixed-methods studies, on the other hand, are when the qualitative work is conducted after the quantitative strand and provides explanations for the findings of the quantitative study.
The priority of quantitative or qualitative is also determined by the framing of the research question. Priority refers to the relative weight assigned to the qualitative and quantitative research components. Typically, in exploratory studies, when concepts, variables and relationships are unclear, greater priority is often assigned to qualitative elements and relationships among them are subsequently studied quantitatively. On the other hand, in explanatory research where qualitative research is mostly used to substantiate findings generated in a population-level survey, priority is mostly assigned to the quantitative component. Sequence and priority give rise to various combinations that define the types of mixed-methods research studies [Table 1].
Integration is the defining feature of mixed-methods research studies that distinguishes it from multimethod studies. There is a clear distinction between mixed-methods research and the mere use of multiple methods to answer research questions. The distinction is, therefore, very important to make a clear statement of this in designing and reporting mixed-methods research is an important consideration. True mixed-methods design includes a purposeful integration of qualitative and quantitative methods. Integration can occur at various stages of the research process, during the data collection, data analysis and/or data interpretation phases. There are mainly three ways in which mixing occurs: (1) merging or converging the two data sets, often done by quantifying the qualitative findings, (2) making connections between the two datasets such that the findings of one are further explored in another and (3) embedding one dataset within the other so that one type of data provides a supportive role for the other dataset. In short, it is not enough to simply collect and analyse qualitative and quantitative data; they need to be mixed in some way so that together they form a more complete picture than alone. The decision on when and how to integrate the data is related to the formulation of the research question. Integration ought not to be an afterthought but is very much part of the design and is a pivotal step in mixed-methods research.
Mixed-methods studies may incorporate a theoretical perspective that influences the selection of a particular research design and shapes the research process. A wide range of theories have been used, ranging from formalised empirical theories to epistemological positions, social theories or theoretical propositions with regard to socioeconomic, cultural or lifestyle factors. Mixed-methods designs that are guided by theoretical perspectives are often referred to as transformative designs.
Research designs in mixed methods
Before we look at core research designs in mixed methods, I would like to highlight the notations used to communicate research designs. The strands in a mixed-methods research study are referred to as quant (for quantitative research methods) or qual (for qualitative research methods) [Table 1]. The upper case for quant or qual is used to indicate the priority of the strand. For example, if the study has a large quantitative survey followed by a relatively small qualitative component, the quantitative strand will be denoted by QUANT. Arrows are used to indicate sequence of the strands such as QUANT Qual as in the case mentioned. In case there is no sequence, a plus sign is used to denote the equal emphasis of both the strands.
While there are many mixed-methods research designs possible, Creswell, in the 5th edition of their book on mixed-methods research, identifies three core designs: convergent, explanatory sequential and exploratory sequential that are at the heart of mixed-methods studies [Table 2]. Individuals may engage in a study that uses one or more of the core designs and sometimes apply the core designs within larger frameworks or approaches (such as in experiments or evaluation projects).
The convergent design, also referred to as the concurrent parallel design, intends to bring together the results of quantitative and the qualitative data analysis so that they can be compared and combined. The aim of doing so is to gain a more complete understanding of the research problem. Convergent designs are also used to validate findings. In the convergent design, the two types of data are integrated to answer the research question. An example of a convergent mixed-methods research study from India is a study conducted to look at the implementation of partograph use to monitor labour in facilities providing the (Janani Suraksha Yojana) cash transfer programme for facility births in India. In this study, the authors used both quantitative and qualitative methods concurrently. They conducted (1) obstetric case record review, (2) a vignette-based survey among nurse midwives and (3) interviews with staff.
The explanatory sequential design has two distinct phases. The first phase includes collection and analysis of quantitative data. This first phase is followed by the collection and analysis of qualitative data to explain or expand the quantitative results. An example from India is a study conducted to evaluate the performance of a multi-strategy national health programme to reduce maternal and child health disparities in Haryana, India. The researchers in this study first assessed the degree of implementation of MCH plans by estimating the budget utilisation rates of each mean corpuscular haemoglobin (MCH) plan, and the effectiveness of these plans by comparing demographic health survey data collected at post (2012–2013), during (2007–2008) and pre (2002–2004) NRHM implementation period, in the quantitative study. Then, perceptions and beliefs of stakeholders regarding extent and effectiveness of the National Rural Health Mission (NRHM) in Haryana were explored in the qualitative study during 2013. They conducted a logistic regression analysis for the quantitative data, and inductive applied thematic analysis for qualitative data. The findings of the quantitative and qualitative parts of the study were mixed at the interpretation level.
The exploratory sequential design begins with the collection and analysis of qualitative data in the first phase. This is followed by quantitative methods that build on the qualitative findings. This design has been used to design survey instruments, develop activities for interventions and generate new variables. An example from India is a study conducted to develop and validate a context-specific knowledge, attitude and practice questionnaire on micronutrients for literate mothers of school-age children. The study had two distinct phases that included a qualitative phase, where literature was reviewed to prepare an item pool, and focus group discussions were conducted to get an overview of existing knowledge and practices. In quantitative phase, the item pool was tested for its validity, internal consistency and reliability.
Why use mixed methods
Greene et al., in a review of mixed-methods studies, identified triangulation, complementarity, development, initiation and expansion as the most common reasons for researchers to use mixed-methods designs.
Triangulation is a well-established concept in social science research methods that involves the use of multiple data collection methods, multiple researchers, multiple theories and multiple methodologies to answer the same research question. Triangulation is considered an antecedent to mixed methods as it is known today. Quantitative and qualitative methodologies are, therefore, used together to find convergence in the findings. The use of multiple methods that are integrated provides greater insight than any one of the methods alone. Complementarity is the second reason why mixed methods are used and refers to the use of quantitative and qualitative data to get a complete understanding of the phenomenon. Qualitative data are often used to understand and provide explanations of the quantitative findings. The third reason is development and involves using one method after the other so that the first method guides the second in terms of decisions about sampling, measurement and implementation. The sequential arrangement enables testing theories, establishing generalisability, development of survey tools and identification of variables, especially when little is known about the research problem. Comparison and analysis of discrepancies or discordant findings that lead to initiation of new research is the fourth reason that justifies the use of mixed methods. Initiation occurs in mixed methods when the findings from quantitative and qualitative methods are not consistent or discrepancies are found. These are compared and analysed for new perspectives and insights that yield new questions. Finally, mixed-methods research is used for expanding the scope of the study to answer the research question. Qualitative and quantitative components are included in a study to increase its scope and breadth, respectively.
Challenges in using mixed-methods research designs
It is clear that mixed methods provide a more complete understanding of research problems compared to the use of only a single research approach. A mixed-methods design offers the best chance of answering research questions by combining two sets of strengths while compensating at the same time for the weaknesses of each method. However, the use of mixed methods is not without challenges. The use of two different methodological approaches requires methodological expertise in multiple areas which is often a challenge. Mixed-methods research studies are also time consuming requiring relatively more resources such as training, logistical support and finances. Another concern is that mixed-methods research should be well justified. Just because we can combine two methodological approaches does not mean it is always required or justified. Researchers need to carefully consider and undertake mixed-methods research only when combining quantitative and qualitative approaches contributes to a more complete answer of the research question. Mixed-methods research typically generates a large volume of data and managing the data, analysing it to make sense in meaningful ways can pose a challenge. Presentation and dissemination of the findings can also be a challenge.
| Conclusion|| |
In this article, I presented an introduction to mixed-methods research, considerations in design, core research designs, advantages and some challenges faced. Mixed-methods research should be considered when the research question demands both quantitative and qualitative approaches to be used in the same study. Integration or mixing of data is the defining feature of a mixed-methods study and should be clearly planned at the start of the study. This should also be stated in the report of a mixed-methods study. The philosophical position that enables mixing of the different types of data is one of pragmatism. When designing a mixed-methods research, due consideration to sequence, priority, integration and use of theory should be given. The three core designs in mixed-methods research are the convergent, exploratory sequential and explanatory sequential designs. Mixed-methods research needs due attention to ethical and availability of methodological expertise. There is an increasing trend towards using mixed-methods research and despite its challenges in terms of logistical resources required, the advantage it provides is in the ability to holistically answer research questions.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Cohn S, Clinch M, Bunn C, Stronge P. Entangled complexity: Why complex interventions are just not complicated enough. J Health Serv Res Policy 2013;18:40-3.
Tashakkori A, Newman I. Mixed methods: integrating quantitative and qualitative approaches to research. In B. McGaw, E. Baker & P. P. Peterson (Eds) International Encyclopaedia of Education. 3rd
ed. Oxford, U.K.: Elsevier. 2010.
Creswell JW, Plano Clark VL. Designing and Conducting Mixed Methods Research. 2nd
ed. Thousand Oaks, CA: Sage Publications. 2011.
Tashakkori A, Johnson B, Teddlie C. Foundations of Mixed Methods Research: Integrating Quantitative and Qualitative Approaches in the Social and Behavioral Sciences . CA: Sage Publications. 2020.
Guba EG, Lincoln YS. Competing paradigms in qualitative research. In N. K. Denzin & Y. S. Lincoln (Eds.), Handbook of qualitative research. CA: Sage Publications. 1994.
Abbas T, Teddlie C. Handbook of Mixed Methods in Social & Behavioral Research. 6th
ed. Thousand Oaks, Calif: SAGE Publications; 2003.
Creswell JW, Creswell JD. Research design: Qualitative, quantitative, and mixed methods approaches. CA: Sage publications; 2017.
Kroll T, Neri M. Designs for mixed methods research. Mix Methods Res Nurs Heal Sci 2009;31:31-49.
Zacharias FJ. Letter: Cross-sensitivity between practolol and other beta-blockers? Br Med J 1976;1:1213.
Chaturvedi S, Upadhyay S, De Costa A, Raven J. Implementation of the partograph in India's JSY cash transfer programme for facility births: a mixed methods study in Madhya Pradesh province. BMJ Open 2015;5:e006211.
Gupta M, Bosma H, Angeli F, Kaur M, Chakrapani V, Rana M, et al
. A mixed methods study on evaluating the performance of a multi-strategy national health program to reduce maternal and child health disparities in Haryana, India. BMC Public Health 2017;17:698.
Konapur A, Krishnapillai MN, Nagalla B, Gavaravarapu SM. A sequential, exploratory, mixed-methods approach for development and validation of a context-specific knowledge, attitude and practice questionnaire on micronutrients for literate mothers of school-age children. Public Health Nutr 2019;22:2120-31.
Lune H, Berg B. Qualitative Research Methods for the Social Sciences. 9th
ed. Harlow: Pearson. 2004.
Johnson RB, Onwuegbuzie AJ. Mixed methods research: A research paradigm whose time has come. Educ Res 2004;33:14-26.
[Table 1], [Table 2]