Part I
Critically designing and evaluating diary studies
Introduction
In higher education research, as well as across the wider social sciences, there is increased awareness that combining multiple research methods enables researchers to better understand the complex and often non-linear phenomena that occur in todayâs societies (Creswell & Creswell, 2018; Creswell & Plano Clark, 2017; Greene, Caracelli & Graham, 1989; Johnson & Onwuegbuzie, 2004). In this chapter, we consider solicited diaries as one possible method to incorporate within a mixed methods higher education research design. We start by defining mixed methods research broadly and outlining various approaches, purposes and key considerations for its use. The second half of this chapter highlights the benefits and challenges of mixed methods diary designs through reflection on our own research examining doctoral studentsâ social community building experiences. The chapter closes with six key considerations for higher education researchers for developing rigorous mixed methods diary studies.
Approaches and considerations for mixed methods research
While there is debate about how to label and define mixed methods research, we define it in this chapter in line with Bazeley (2017, p. 7): âAny research that involves multiple sources and types of data and/or multiple approaches of those data, in which integration of data and analyses occurs before drawing final conclusions about the final topic of the investigation.â Some mixed methods researchers argue that mixed methods must involve both quantitative and qualitative elements (see, for example: Johnson, Onwuegbuzie & Turner, 2007), but there is increasing recognition that this strict dichotomy between research paradigms is perhaps superficial (Bazeley, 2017).
Mixed methods diary research offers many benefits for approaching complex research topics, including those present in many higher education studies. One consideration is that researchers can draw upon the strengths and counterbalance the weaknesses of chosen methods, as well as more broadly balance quantitative versus qualitative research paradigms (Symonds & Gorard, 2010). One additional strength of mixed methods approaches is that they allow for data triangulation by combining multiple insights into a phenomenon from different perspectives. As described by Cohen, Manion and Morrison (2018, p. 195), âtriangular techniques in the social sciences attempt to map out, or explain more fully, the richness and complexity of human behaviour by studying it from more than one standpoint.â
Mixed methods designs are inherently flexible and allow researchers to think pragmatically about research approaches, paradigms and methods that best answer their research questions (Johnson & Onwuegbuzie, 2004). In higher education research, the underlying pragmatism of mixed methods research is well suited, given that research topics often combine more subjective perspectives (such as studentsâ experiences or preferences) with more positivistic approaches (such as measuring studentsâ learning behaviours or progression routes) (Scoles, Huxham & McArthur, 2014).
Despite this flexibility, there are many considerations researchers must make in designing mixed methods diary research, including managing the sample across multiple methods, developing complementary research instruments, deciding on approaches to analysis, and outlining procedures for drawing together findings from multiple sources (Teddlie & Tashakkori, 2009). Plano Clark and Ivankova (2016) also note that considerations must be made related to timing (when and in what order to collect data), priority (the relative importance of the various research strands in light of the research questions), and integration (the strategies used to combine multiple sets of data throughout the research design). Taken together, this means there is no single, ârightâ way to design a mixed methods diary study, as approaches are often complex and unique to the aims of the individual study. Indeed, Creswell and Plano Clark (2017) highlight six overall approaches for developing a mixed methods research design, all of which could be incorporated into a solicited diary study:
- Convergent parallel design: data from multiple methods are collected separately, but at the same time, and then drawn together in the analysis.
- Explanatory sequential design: an initial quantitative data collection phase is followed by qualitative data collection, explaining the initial results of the quantitative analysis.
- Exploratory sequential design: an initial qualitative data collection phase is followed by quantitative data collection, allowing for an initial exploration phase before bringing the study to scale.
- Embedded design: data from multiple methods are collected together at the same time in one single research phase.
- Transformative design: a transformative theoretical framework is used to inform all decisions within the mixed methods design.
- Multiphase design: convergent and sequential designs are combined over a larger programme of study.
An additional consideration in mixed methods research is the careful design of when, where and how multiple methods are integrated throughout the research process (Bazeley, 2017). Woolley (2009, p. 7) describes integration in mixed methods research as:
the extent that these components [multiple research methods] are explicitly related to each other within a single study and in such a way as to be mutually illuminating, thereby producing findings that are greater than the sum of the parts.
Decisions related to integration are important to mixed methods diary designs, particularly as there are multiple approaches to integrating data in mixed methods research (see Creswell and Plano Clark 2017).
Mixed methods diary research
Mixed methods diary research offers great potential, particularly as many diary research features are well-suited for mixed methods designs. The most striking is the flexibility of diary approaches, which can be easily moulded across the broad range of mixed methods research design options outlined by Creswell and Plano Clark (2017). Diary instruments can be developed in many ways: as highly structured and questionnaire-like (Mullan, 2019), unstructured and more open-ended (Martinez-Vargas, Walker & Mkwananzi, 2020), or a hybrid (Swim, Hyers, Cohen, Fitzgerald & Bylsma, 2003). This means diariesâ role and function can vary greatly between different studies. For example, some mixed methods studies in higher education have used diaries at the studyâs start to provide a baseline understanding of participantsâ experiences in an initial exploratory phase (Chen, Yarnal, Hustad & Sims, 2016), while others have used diaries towards the studyâs end to explain or complement initial quantitative findings (Swim et al., 2003).
Triangulating findings from diary research with other methods provides many benefits to researchers. For example, one common approach is the diary-interview method (Harvey, 2011), whereby preliminary findings from a diary study are used to develop interview questions. This allows researchers to expand upon preliminary diary findings with the interviewee, shedding greater light on findings that seem unclear or providing more depth into the sentiments expressed. Other research designs have collected diary data after a questionnaire (for example: Swim et al., 2003), which can help explain preliminary quantitative findings or provide deeper understandings about why certain patterns were exhibited. In higher education research, triangulating diaries with other methods has previously provided researchers access into more in-depth and well-rounded understandings of highly sensitive or personal research topics, such as studentsâ experiences with racism (Swim et al., 2003) and sexism (Swim, Hyers, Cohen & Ferguson, 2001) or international studentsâ experience (Heng, 2017; Chapters 3 and 12, this volume).
Incorporating other methods alongside diaries can additionally help overcome perceived drawbacks or challenges of the method. For example, one consideration is self-reflection bias, as participants select themselves what to include in their diaries, which may not fully represent their experiences or may demonstrate only extreme examples (Cohen et al., 2018). Responses might also be influenced by social desirability bias; in recognition that their writing will be read by another person for research purposes, participants may wish to portray their lives and experiences in a more positive light (Cohen et al., 2018). However, using mixed methods diary approaches allows for triangulation and member checking, meaning researchers can confirm or contradict the findings from multiple sources to create a more complete picture. An additional consideration is that some participants may not provide much depth in their diary writing or there might be a waning response depth over time, particularly in more unstructured diaries. Mixed methods diary approaches can help overcome such problems by allowing researchers to follow up on interesting ideas that were perhaps initially only explored superficially in the diaries.
Altogether, mixed methods diary approaches have the potential to offer unique insights into participantsâ worlds in higher education research through flexible engagement with multiple facets of their experiences. In the next sections, we critically reflect on the benefits and challenges associated with mixed methods diary research in the higher education field through an in-depth exploration of our own research.
Research methodology adopted
The remainder of this chapter reflects on our experiences conducting a longitudinal mixed methods diary study entitled Social Transition Research into International Doctoral Experiences (STRIDE). This project, situated within doctoral education research, focused on the social transitions and peer community development experiences of 53 doctoral students studying in three higher education departments in institutions based in England, Scotland and China. The studyâs underlying goal was to understand how doctoral students develop social support from peers within their departments and the roles doctoral communities play in supporting studentsâ well-being. In recognition that social relationships and social transitions in higher education are complex experiences, we opted for a mixed methods approach using an explanatory sequential design (Creswell & Plano Clark, 2017). Altogether three research methods were used in this study:
- Social network analysis surveys: Social network analysis is a research method that measures and maps existing social structures and networks (Wasserman & Faust, 1994). In this study, we used a closed network questionnaire, whereby we provided doctoral students with a list of peers in their higher education departments and asked them to mark those from whom they had received social support. These data were then used to visualise the existing network structure and identify quantitative patterns of social support between doctoral students.
- Reflective diaries: The second research method was a semi-structured reflective diary study that lasted for six weeks. Participants were asked to reflect in their diaries about the social community in their higher education departments, including how they had interacted socially within their doctoral communities, what social opportunities occurred, and what factors they felt affected the social support they received. The diary data provided an understanding of the types of social behaviours participants exhibited, as well as a reflective account of why they did or did not feel socially supported in their academic departments.
- Semi-structured interviews: The final research method used was semi-structured interviews with doctoral students. In the interviews, participants were given the social network analysis visualisation as a mediating artefact and asked to reflect on the social network structure and their own role in the doctoral community. Data from the diaries were also used as prompts to explore participantsâ behaviours or reflections in greater depth. Altogether, the interviews provided an opportunity to member check (i.e. check understandings of the findings with the population being studied) and develop a greater understanding of initial findings.
The mixed methods research design investigated doctoral studentsâ social support networks on multiple levels by considering what were studentsâ existing social networks in their higher education departmen...