A Practical Introduction to Mixed Methods for Business and Management

Patricia Bazeley

• Publisher: SAGE Publications Ltd
• Year: 2018
• ISBN:
  9781526462909
• Edition: 1
• Topic: Betriebswirtschaft
• Subtopic: F&D

Mixed Methods Research for Business and Management guides students and researchers through how to use this methodology successfully in a research-based dissertation or project.

The book introduces the concepts and debates associated with combining methods, and illustrates the many benefits, and hazards, of undertaking a mixed methods study. Example studies from across business and management disciplines bring the text to life throughout. The reader is taken step-by-step through the mixed methods process from developing a mixed methods study, through designing and conducting it, and finally, reporting on the results. Suitable for business and management students and researchers undertaking their own mixed methods research.

1 Foundations

Mixed Methods Exemplified in Classic and Recent Studies

A classic study: How do employees and companies respond to the stresses of a merger?

In a landmark article, published in a 1979 special issue of Administrative Science Quarterly focusing on qualitative methods, Todd Jick portrayed the considerable advantages and the hazards of using multiple sources and methods to study employee and organisational responses to a merger between companies. To assess levels and symptoms of employee stress, he used multi-strand surveys, interviews, and observations, he explored archives and more recent company memos, and he used unobtrusive measures including worker visits to the archives (as an indicator of anxiety), absenteeism, and rumours. He found congruence across different measures and observations for most of his findings, but also some ‘surprising discrepancies’ at some sites. He resolved discrepancies between symptomatic stress levels and actual job security, for example, through ‘piecing together’ information about changing contextual factors from the mixed data sources. In so doing, he turned divergences arising from use of multiple methods into beneficial new knowledge. He concluded that comparing and combining data from varied sources functioned as a means to improve the definition and analysis of the organisational research problem, revealed sources of variance that might be missed using a single method, and portrayed what was being studied in a more holistic way. This enriched understanding and instilled confidence in the conclusions drawn. Working with multiple sources together, in his experience, was a process in which methods typically associated with qualitative data and analysis functioned ‘as the glue that cements the interpretation of multimethod results’ (1979: 609).

Mixed methods in recent practice: Why do farmers tolerate high labour turnover, rather than taking steps to control it?

Staelens and Louche (2017) investigated management responses to labour turnover in the emerging cut-flower industry in Ethiopia, in a study designed to add a subjective dimension to explanations based on rational cost accounting. The Ethiopian cut-flower industry, largely located on the urban fringe of Addis Ababa, employs a large, low-skill, low-pay, primarily female workforce in which a high degree of labour turnover generally occurs at two critical periods each year. The questions being addressed by the researchers asked why farm managers tolerated this high turnover, rather than trying to reduce it, and what they did to control it. The phenomenon of tolerance could be partly explained by the lower cost of turnover compared to retention, but Staelens and Louche were not fully satisfied with this explanation. They looked to institutional theory, which recognised the role of regulative, normative, and cultural-cognitive forces, to inform the design and interpretation of their data.

The study was conducted in two phases, with data from both phases contributing to the overall outcomes. In the first phase, they undertook progressive preliminary interviewing of 21 managers and other stakeholders (with adjustment of interview guides as they went), complementing these with field notes. These broadened their understanding of the industry, confirmed labour turnover as an issue in the industry, and assisted in refining their research questions and in selecting a farm cluster to use as their case study for further investigation. In the second phase, they conducted additional interviews with the managers of the eight farms in their selected cluster and a joint interview with two relevant government representatives to deepen their understanding of the turnover issue. They carried out mixed-form surveys with 375 workers on five farms, and also six focus groups with workers from those five farms. These not only enriched their understanding, but also served as a source of validation for conclusions arising from their interviews with managers. Methods employed were usefully summarised in a diagram showing their sequence and content, pointing to the centrality of their interviews with managers (Figure 1.1). The authors applied existing theory as an organising framework for their analysis, but they also reported working in a ‘highly iterative’ manner between theory, data, and literature, and referred to ‘emerging results’ which were then subjected to both local and academic peer checking.

Analysis of their interview and fieldwork data revealed how managers perceived their environment as one where they felt powerless to change the situation. The managers therefore used prediction, containment, and accommodation strategies to control rather than reduce labour turnover and its effects. Analysis of data contributed by workers through their survey and focus group responses confirmed the patterns on which managers based their predictions, and the roles played by regulatory, normative, and cultural-cognitive influences on turnover. Thus, legitimation of managers’ employment strategies was seen to be responsive to local institutions and conditions, rather than economic forces. The article finished with considerations of the practical strategies that might be needed to break the ‘deadlock’ of institutional forces, and of the potential relevance of this study for other industries employing low-skilled workers.

Figure 1.1 Data sources used in studying the Ethiopian flower industry

Source: Staelens and Louche, 2017: 1471, Figure 1

Introduction

This book focuses on the application of mixed methods research in business and management. It will take you through the main steps involved in carrying out a mixed methods research project, illustrated by studies conducted by Jick, by Staelens and Louche, and by many others. At the same time, you will be introduced to theoretical issues and debates associated with combining methods. Much of what is written applies also to studies that use a single methodological approach, but the focus here will be on the particular considerations involved when a combination of methods is being employed. Thus, the emphasis will be on effective integration of those methods, rather than on how to carry out research using particular qualitative or quantitative methods independently. For that reason, it is anticipated that students will have some basic knowledge of the steps involved in methods of collecting and analysing quantitative and/or qualitative data, gained through earlier coursework or reading the research literature.

In this first chapter you will learn:

• what distinguishes mixed methods as an approach to research
• the historical development of mixed methods as a distinct methodological approach
• how mixed methods research has been applied in business and management, and the benefits of doing so
• and you will be introduced to the issues and debates that have surrounded and informed mixed methods as a methodological approach.

What is Mixed Methods Research?

Mixed methods research is a methodological approach in which the researcher integrates varied types of data (usually including both numbers and text) and/or different ways of analysing data within a single study or programme of study, in the service of a common theoretical or research goal. Jick used surveys, interviews, observations, and archival data to explore employee responses to change; Staelens and Louche interviewed people in different roles, recorded field observations, and conducted surveys and focus groups in order to understand workforce issues in a particular industry. The roots of combining methods in this way lie in diverse fields across the natural and social sciences; more recent use is strongly associated with the professional disciplines of education and health. Business and management theory and professional practice similarly draw on a diverse range of economic, social, and behavioural disciplines. In using diverse methods of investigation to capture different aspects of a phenomenon, a mixed methods approach is entirely appropriate for those wanting to build depth and/or breadth into their research in business and management.

A defining aspect of mixed methods research is that different approaches and the methods used are integrated within the study (Bazeley, 2009, 2012, 2018b; Fetters & Molina-Azorín, 2017). Information might be exchanged between methods during the course of data collection. Data and preliminary results from different sources and methods of data collection will be brought together during the analysis phase of the research. Final results and conclusions will draw on all methods used.

Integrating methods has been said to offer ‘potential to provide opportunities to see things that have not yet been seen’ (Mertens et al., 2016: 222), to give access to insights and understanding beyond those that might have been provided by use of quantitative or qualitative methods alone – a whole that is more than the sum of its parts (Creswell & Plano Clark, 2018; Fetters & Freshwater, 2015; Teddlie & Tashakkori, 2009). Use of mixed methods by business and management students and researchers, therefore, can also be expected to contribute to advancing business and management research through the synergy gained from integrating both qualitative and quantitative methods (Molina-Azorín, 2011).

Mixed Methods Research as a Third Methodological Movement

Quantitative and qualitative methods, which focus, respectively, on the quantities and qualities of phenomena, have been combined to some extent in both the natural and social sciences throughout their respective histories, without this being seen as doing something distinct and special (Johnson & Gray, 2010; Maxwell, 2016). The term mixed methods began to gain recognition as a way to describe research combining quantitative and qualitative data in the 1990s, first in education and then health, but the mixed methods label was not widely used in any discipline until the 2000s when the term became more established. Prior to that, researchers were more likely to use both qualitative and quantitative as descriptive terms in their study titles and methods, or to describe their studies as using multiple methods.

Quantitative methodologies had dominated the social and behavioural sciences during the twentieth century as researchers, believing that truth was discoverable, pursued the goal of being objective (value-free) in their work in an attempt to emulate the physical sciences. The application of measurement tools and the use of experimental methods were seen to be more scientific in revealing and measuring human behaviour than was interpretive description obtained through more open-ended, smaller-scale qualitative methods. The adequacy of controlled experimental approaches for understanding human behaviour and evaluating social and educational programmes was increasingly questioned, however. Qualitative methodologists rejected the idea of discoverable, absolute truth and emphasised the necessity of understanding participants’ constructions of their experiences in and of the world. Qualitative approaches to research gained recognition as a growing methodological movement during the 1970s and 1980s.

The qualitative ‘revolution’ also contributed to the development of mixed methods as a differentiated methodological approach to research. Mixed methodologists joined their qualitative colleagues in their critique of the so-called scientific approaches to the evaluation of educational and social programmes (e.g., Howe, 2004; Maxwell, 2004). Rather than eschewing quantitative approaches entirely, however, these researchers argued that combining these with qualitative approaches provided added value to the evaluation of social and educational interventions – and to social research more generally – by contributing to understanding of both process and outcome.

A problem with the growing recognition of qualitative research as a methodological approach was that the distinction between qualitative and quantitative approaches to research was sharpened, and battle lines were drawn. Acrimonious debates across the qualitative and mixed methods communities focused on conflicts in the philosophical assumptions underpinning these different approaches to methodology. Some saw quantitative and qualitative methods as being based on fundamentally different assumptions about the nature of existence (ontology) and of what constitutes justifiable knowledge (epistemology) in a way that prevented their combination. Referred to as the ‘paradigm wars’, these debates were especially prevalent in the United States. Resolutions to the argument that quantitative and qualitative methods were incompatible were found eventually, for some in the rediscovery of pragmatism as a philosophical basis for research, for others in alternative philosophical or theoretical positions, while others ignored it as a problem and worked on, simply recognising that there is an interpretive quality to all research. A resolution was also found in the recognition that there is not necessarily a direct correspondence between epistemologies, methodologies, and the actual methods used. Thus Bryman (1988) drew a distinction between combining epistemological positions based on different views about how social reality should be studied, which might pose some problems, and the technical nature of combining methods in practice, which he saw as being generally unproblematic. By the later 1990s, after two decades of discord, the arguments were sufficiently resolved (or worn out!) to allow researchers to get on with the business of further developing mixed methods. (Further clarification of some of the language used in these arguments is provided below. More detail on philosophical foundations, the paradigm wars, and the debates associated with these is provided later in this chapter for those who are interested.)

Rapid growth in the adoption of mixed methods research across the social and behavioural sciences following the turn of the twenty-first century led to its becoming increasingly articulated as a distinct approach to doing research, and to being recognised as a ‘third methodological movement’ (Johnson & Onwuegbuzie, 2004; Tashakkori & Teddlie, 2003). The publication since 2007 of a specific interdisciplinary methodological journal focused on mixed methods (Journal of Mixed Methods Research) and the establishment of an international professional association in 2013 (Mixed Methods International Research Association) have further contributed to this recognition. The decision about which methods to choose to best investigate a research problem is no longer a question of whether one should use quantitative or qualitative methods. Rather, when use of either or both is appropriate to the purpose of the study, the question becomes how to integrate the strengths of each in a combined approach.

A clarification of terms

The widespread adoption of mixed methods research has been described as a third methodological movement, following on from quantitative and qualitative movements in research. The term methodological movement has been used to describe these three very broad approaches to research, but each of these, in turn, embraces a range of more specific methodologies – systematic sets of principles and strategies for investigating particular kinds of research problems. Quantitative approaches to research include descriptive, correlational, and experimental methodologies, with methods that generally employ data in the form of numbers and use mathematics or statistics for analysis. Qualitative approaches generally employ text, audio or visual data, and a variety of methods of analysis that seek to interpret meaning and patterns in those data. Methodologies usually associated with qualitative approaches include grounded theory, phenomenology, and ethnography, among others. Methods, which are the actual tools and techniques used for data gathering and analysis, are often shared by different methodologies, and sometimes span the (overlapping) boundaries of the broader methodological approaches. A measure of an object or an experience and a description of that object or experience are simply different ways of representing that same phenomenon.

Mixed Methods in Business and Management Research

Business and management researchers appear, at first sight, to have been relatively slow to adopt mixed methods approaches. Systematic reviews of articles published in a variety of business and management fields, using manual methods of identification, report proportions of mixed methods studies as ranging from 22 per cent in vocational training and 17 per cent in international business to 3 per cent in information systems journals (Molina-Azorín & Cameron, 2015). Reviews using electronic search methods report even lower rates (Ivankova and Kawamura, 2010). What is apparent from the reviews undertaken is that when business and management researchers combined qualitative and quantitative sources and analyses they rarely described it as mixed methods, nor did they refer to methodological literature that supports the use of mixed methods (Bazeley, 2015a; Molina-Azorín, 2011). This is, perhaps, a reflection of there being limited resources and training for business and management researchers in using mixed methods.

More positively, Molina-Azorín and Cameron (2015) pointed to the increasing incidence of mixed methods tracks in business conferences and of j...