Report on 8th BAAL SIG Seminar: "Focus on the Researcher: Dealing with Distressing Data"
(by Kate Barber, Cardiff University):
The 8th annual BAAL Language and New Media SIG event, hosted by Kate Barber (Cardiff University), brought together researchers who analyse distressing linguistic data with the aim of discussing the psychological challenges inherent in these studies. In reflecting on the mental and emotional impacts of difficult data and sharing strategies to mitigate these, the objective was to suggest ideas for best practice for those currently in the field and new scholars entering. The online symposium included ten papers, each offering personal insights from speakers at various stages of their academic careers. With the event’s focus on the researcher rather than the research, the openness and honesty of those presenting generated a tangible sense of community support, with the experiences of the presenters echoed throughout the day by attendees.
In a plenary which set the spirit of the day, Nuria Lorenzo-Dus (Swansea University) shared her personal experiences working on two on-going multidisciplinary, multiagency research projects investigating online child sexual grooming. She outlined the complexities of defining what constitutes ‘distressing data’, adding the proviso that sensitivities to different types of data can arise at any time, often unexpectedly. She challenged the generally-held assumption that textual data is less triggering than visual input, with examples given of how the visualisations of the participants were found to cause distress, and how subtle, less obviously problematic language was often particularly upsetting. In multidisciplinary teams, the assumptions of non-linguists as to the extent language needs to be analysed in context were also highlighted. Nuria outlined the formal on-boarding processes and informal measures in place for researchers in her teams as well as the emphasis she places on limiting their immersion in the data and their reflexive practices.
The challenges of distressing data
The presentations described a range of negative impacts of analysing distressing data, including the effect they can have on the speakers’ world view. Dasha Dayter (University of Basel) and Sofia Rüdiger (University of Bayreuth) shared their experiences of looking at the misogynistic, discursive behaviour of pick-up artists (PUAs), a group associated with the ‘manosphere’ which often promotes violence against women. In looking at the data, they developed feelings of distrust of others and concerns over personal safety. Encroaching cynicism and an altered world view were similarly discussed by the MANTRaP team. Alex Krendel (Lancaster University), Jessica Aiston (Lancaster University), Mark McGlashan (Birmingham City University) and Veronika Koller (Lancaster University) shared their individual experiences of analysing extreme misogyny through their work on data from involuntary celibate (incel) forums and other anti-feminist communities aligned with the manosphere. Feelings of paranoia, physical discomfort, and ‘psychological whiplash’ were noted, along with an awareness they were judging aspects of their own behaviour and interpersonal relationships through the lens of the participants. Intrusive thoughts were also discussed by Kristina Pahor de Maiti (University of Ljubljana), who explained how the interpretation of quantitative, rather than qualitative, data impacted her team. Her work with linguists and annotators of socially unacceptable online discourses showed how the personal circumstances of the researchers can trigger different negative responses.
The triggering of strong negative emotions was not limited to data from socially unacceptable participants but also in relation to data which generated empathy. In his research on the intrusive thoughts of people who suffer from sexuality and gender-related obsessive-compulsive disorder (OCD), Elvis Coimbra-Gomes (Queen Mary University of London) described how he managed the distress of his participants while safeguarding himself. Regulating personal emotions in order to protect participants and elicit upsetting information were similarly discussed in Sarah Turner’s (Coventry University) paper. Her research, which analysed vivid accounts of grief from pregnancy loss and the death of a child, posed particular challenges in terms of emotion management. Jennifer O’Donovan’s (University of Edinburgh) experiences analysing online comments from the 2018 Irish referendum on legislating for lawful termination of pregnancy raised the issue of having to repeatedly interact with the data years after collection. She detailed the challenges of managing both empathic and negative reactions to the data, while highlighting the difficulties posed by having to revisit these for current and future publications.
Dealing with distressing data
Measures to mitigate the effects of working with distressing data were suggested by presenters, with some commonality in the strategies employed. One of the most prevalent related to the benefits of working with others. Dasha and Sofia described ‘being in it together’, allowing them to employ sarcasm and jokes to help mentally process their research. They also set up an ‘atrocious quotes’ file to store particularly upsetting examples of PUA data. The MANTRaP team ensured time was given to team members to articulate their responses to the data through an ‘honesty round’. For researchers who are working alone, Frazer Heritage (Birmingham City University) introduced CDSupport: an initiative he has set up in the form of an informal buddy-system. The CDSupport network matches researchers working on similar data and encourages them to check in with each other to offer peer support.
A reliance on physical exercise was shared by speakers as a way to personally manage emotions. Andrea Vaughan (UCL Institute of Education) focused on the strategies she employed during her research on the language of suicide and suicidal ideation online. These included using the Wellness Wheel, a tool which offers a holistic method of monitoring reactions and making healthy choices. She found adopting a sports training approach to time management useful to break up continual exposure to distressing data and to differentiate between recovery and rest. Mental strategies for processing data were also discussed, for example, (de)humanising participants and mentally fictionalising their accounts. The question of neutrality arose, with many speakers challenging the expectation of maintaining an objective stance to their data. As another outlet, Elvis shared his method of watching horror films to ‘let the fear out’, which proved to be a surprisingly common one among the attendees.
The role institutions play in mitigating the impacts of analysing distressing data was discussed throughout the day. Mirroring points put forward in Nuria’s plenary, Tim Grant and Sarah Atkins offered their perspectives from the Aston Institute for Forensic Linguistics (AIFL). They discussed a range of distressing data in the context of the support they could secure from their institution and external support networks. Recognising the potential for distress was a key driver in the processes in place at AIFL. These included providing researchers with access to a specialist psychologist who could provide individual support and offer advice on how to implement ‘work hygiene’ measures, which ensure there is a clear delineation between work and non-work spaces at home. Ye Bin Won (Georgetown University) focused on the lack of institutional support for undergraduate students working with distressing data. Through the context of her analysis of incel data, she discussed the culture of stoicism in the sphere of extremism research, which can prevent researchers seeking help. Mitigation strategies advocated include integrating mental health literature and well-being training into undergraduate courses, and investing in mentorship programmes. The roundtable discussion saw David Wright (Nottingham Trent University) and Emily Powell (Cardiff University) join speakers from the day to discuss the main themes emerging from the presentations, with a particular focus on those concerning institutional responsibilities. Talking points included the need to move well-being higher up the agenda and the challenges of investing time and money to doing so. Suggestions were also put forward as to how researcher well-being could be addressed more explicitly in ethics sections of publications.
Final reflections
It was apparent from the discussions that distressing data can affect individuals differently as their life experiences change. There was a common acknowledgement that one approach to mitigating this distress is unlikely to suit everyone and approaches need to be adaptable. The necessity of institutionally-based measures to safeguard researchers, without needing to depend on the proactivity and tenacity of senior academics or project leaders, was also recognised. Finally, along with these discussions, the presenters reiterated how rewarding and motivating it can be working on projects which include difficult linguistic analysis. It is hoped that, in normalising these conversations on the effects of distressing data, we can make it easier for researchers to, in Nuria’s words, ‘put their own oxygen masks on first’ when they do so.
Speaker abstracts can be accessed via the link below and more reactions to issues raised can be found on Twitter at #BAALDDD.
Kate Barber
For the final Programme of the Seminar, please click here
The book of ABSTRACTS is available from this link
REGISTRATION IS NOW OPEN
Focus on the Researcher: Dealing with Distressing Data
14th May 2021
Please REGISTER here
Linguistic studies of online extremism, including racist, misogynistic, and homophobic discourses; forensic linguistic data and hate speech; and the analysis of discourses on particularly upsetting or traumatic personal experiences all make an increasingly valuable contribution to the field of applied linguistics and beyond.
However, undertaking these studies poses a risk to the mental, emotional, and potentially, physical wellbeing of the researchers involved.
This online symposium attempts to consolidate the experience of those working in the field and to identify practical strategies to ensure the welfare of present and future researchers. The effect and necessity of processes implemented by institutions in the form of, for example, risk assessments, training, and providing mental health support will also be discussed. It is hoped that, by sharing best practice and discussing resilience strategies, we can begin to establish specific recommendations on safe and effective working methods for linguists dealing with distressing data, and, while doing so, build a community of support.
Along with papers from researchers, the day will include:
Confirmed Keynote Speaker: Professor Nuria Lorenzo-Dus (Swansea University)
Roundtable Discussion
Launch: Dr Frazer Heritage introducing the CDSupport Network
THE CALL FOR ABSTRACTS IS NOW CLOSED
Submission Guidelines
We invite contributions from scholars working with data which is inherently distressing, shocking, upsetting, or which, more generally, could challenge researchers' mental wellbeing. This may include (but is not limited to) research into extreme online materials such as those connected to the far-right, misogynistic groups or terrorist groups; forensic linguistics and data on crime, policing, and online grooming; and studies on distressing and/or traumatic personal experiences.
We particularly welcome submissions from PhD and ECR researchers who are dealing with distressing data.
Abstracts should be a maximum of 250 words (exc. references). Please include details of the type of data you are analysing; the challenges you faced when dealing with the data; and either what you did to address those challenges or what you would have done in retrospect. Any details or examples of particularly effective institutional processes or practice would also be welcome. The focus of the paper should be the challenges of the data and the necessary strategies to overcome them not the results of your research.
Send abstracts to: [email protected]
Presentation Format
Each presentation in a panel will be allocated 15 minutes with 20 minutes at the end of the panel for questions.
Key dates:
Deadline for papers: January 25th 2021
Notification of acceptance: By the end of February 2021