Creative Research Methods 2e. Kara, Helen
can be challenging (Hemmings et al 2013: 261–2). Joseph Teye, from his doctoral analysis of the formulation and implementation of forest policy in Ghana, identified a number of challenges. These include the following.
•Choice of sample size: quantitative researchers prefer large samples, while qualitative researchers are happy to work with a few participants in detail; this may be why the iterative approach of gathering and analysing quantitative data, then using the findings to inform qualitative investigation, is so popular.
•Mission creep: the scope of the research can end up being wider than originally planned, as new information is brought to light that is difficult or impossible to ignore; this can lead to more data being gathered than was at first envisaged, which then leads to the data analysis being more time consuming than expected.
•Resource constraints: using more than one method, for example, for data gathering and/or data analysis, takes more time and expertise than using a single method.
•Difficulty prioritising research methods: prioritisation can be dictated by mission creep or resource constraints or both, rather than by the researcher’s plans; it is hard to give equal priority to both or all methods used.
•Conflicts in data interpretation: these can happen particularly between qualitative and quantitative analyses.
•Difficulty integrating findings: this applies to analysis and reporting, and there are a range of views about the benefits or disadvantages of integrating findings at either stage or both, which means that no researcher’s solution is likely to please all their readers.
•Managing power relations: gathering data from different groups of participants can mean the researcher has to adapt quickly to changing power relations (Teye 2012: 385–8).
Box 2.12: Researching the researchers by Lunde et al
Åshild Lunde and her colleagues in Norway set out to investigate an interdisciplinary research project on knee injuries in athletes. The original research team involved researchers from different academic disciplines, including physiology and phenomenology, and physiotherapy practitioners. They gathered quantitative data about the nature and extent of knee injuries in athletes, and qualitative data about athletes’ experiences of knee injuries. The intention was to integrate these datasets, with the overall aim of predicting the outcome of rehabilitation. However, despite being highly skilled and motivated, making careful plans and trying hard, the original researchers found it impossible to integrate the datasets. The research funders decided to commission some more interpretive work from external researchers to investigate the barriers to integration. Lunde and her colleagues took on this investigation and identified a number of barriers, including different views about what constitutes good-quality research, and lack of strong project leadership to help reconcile these views. Perhaps more importantly, the findings from the qualitative and quantitative datasets contradicted each other. These contradictions could have been used as a resource for the research, in the form of a springboard for further exploration, but instead they were seen as an obstacle. Lunde et al (2013: 206) stress that this is not ‘the typical narrative of expressed prejudice and hostility between quantitative and qualitative researchers’ and that considerable collaborative efforts were made. However, the desired middle ground was not reached, perhaps because the process of reaching that middle ground would have compromised the professional identity of all researchers. There was not enough ‘external force’ or ‘internal drive’ to make this happen (Lunde et al 2013: 209), so the disciplinary status quo was maintained.
It does appear that most difficulties in multi-modal research occur when disciplines collide (Hemmings et al 2013: 262). Some researchers advocate a ‘qualitatively driven mixed-methods approach’ that is not intended to privilege qualitative over quantitative research, but to ensure a good level of interpretation (Creswell 2006, cited in Hall and Ryan 2011: 106–7). Given the findings of Lunde et al, combined with evidence that some quantitative researchers are not highly skilled in interpretation (for example, Laux and Pont 2012: 3), this would seem worthy of consideration. Either way, for good-quality and consistent multi-modal research it is important to plan which methods to use from the start, within an appropriate theoretical and methodological framework for clarity about why and how you will use those methods, rather than adding methods or devising a framework as you go along (Franz et al 2013: 386).
There are a huge number of examples of multi-modal research in the literature, and only a tiny fraction can be represented here.
Research using technology in practice
Technology can be used to support and enhance all stages of the research process. It is most commonly invoked for data gathering, transcription and analysis. For example, data can be speedily and effectively gathered online using a dedicated program such as SurveyMonkey® or by trawling social media platforms such as Twitter or Pinterest. Audio recorders can be used to record data from interviews and focus groups, and to play it back for transcription, which can be done straight into a computer by typing on a keyboard. There are a number of software packages to help with data analysis, such as SPSS for quantitative data, and NVivo for qualitative data. However, technology can also support other parts of the research process, such as dialogue, literature reviews, collaboration, co-authorship, representation of findings, ethics and reflexivity (Paulus et al 2013: 639; Barnes and Netolicky 2019: 382).
Many ethnographers have embraced the possibilities offered by technology, both for use within conventional ethnographic studies and to shift the boundaries of ethnography itself (van Doorn 2013: 392) (for example, Box 2.13).
Box 2.13: Compressed ethnography by Pope
Technology enabled New Zealand researcher Clive Pope to create a ‘compressed ethnography’ (Jeffrey and Troman 2004, cited in Pope 2010: 134) of the Maadi Regatta. This is New Zealand’s primary rowing competition, a seven-day regatta with hundreds of races. Due to the time-bounded nature of this event, Pope could not conduct a conventional ethnography, using participant observation over an extended period of time. Instead, he ‘spent 10 days and nights at the regatta site, living the everyday life of rower and rowing’ (Pope 2010: 133). This was an intense experience that did not allow for full understanding at the time, so Pope used digital photography and video to record parts of the regatta for later consideration. These enabled him to ‘rewind, revisit and reframe the setting, repeatedly seeking new learnings and understandings’ that ‘replaced the inductive and emerging discoveries that often evolve in situ during prolonged conventional ethnographies’ (Pope 2010: 135).
Doing research online can seem like a great idea in certain circumstances. For example, some geographically dispersed communities, such as distance learners and expatriates, come together in online environments. This can make it seem very appealing to study members of those communities in virtual locations (Lewis and McNaughton Nicholls 2014: 60), whether through observing them at their usual locations or by consensual interaction at a dedicated location such as a chat room or forum set up specifically for research. There are logistical advantages for the researcher: for example, you don’t have to go anywhere, and your data can simply be copied and pasted from the web. This is economical in time and cost, and can make the prospect of doing research online almost too tempting to resist.
However, it is also important to identify and address the limitations of doing research online, and the challenges it may present (Ignacio 2012: 239; Lewis and McNaughton Nicholls 2014: 58; Quinton and Reynolds 2018: 211). The following are a few of those challenges and limitations.
•Technical skills: the researcher may need a certain level of technical skill, or help from someone who has that level of skill, for example, to create a forum, or to make a web page of information about the research to use in seeking informed consent from potential participants.
•Sampling: research online throws up all sorts of problems with sampling, for various reasons – for example, that not everyone has access to online environments, or that the identity of online participants may be wholly