BAAL Language and New Media SIG 2015 Workshop
Today, more than ever, data are widely accessible, visible, and searchable for research in digital media contexts. At the same time, new data types and collection methods challenge existing approaches to research ethics and raise significant and difficult questions for researchers who design, undertake and disseminate research in and about digital environments.
The aims of this workshop are to bring together researchers who use online research methods and data in different subfields of applied linguistics, to discuss ethical considerations in online data collection and analysis, to identify challenges and share solutions to ethical issues arising from applied linguistics research.
Alexandra Georgakopoulou (King’s College London)
Claire Hardaker (Lancaster University)
Annette Markham (Aarhus University, Denmark)
Stephen Pihlaja (Newman University, Birmingham)
Tereza Spilioti, email: SpiliotiT1@cardiff.ac.uk
Helen Clifford, email : firstname.lastname@example.org
CaSMa provided two contributions to this workshop:
Ansgar Koene presented: Participant Consent and Withdrawal when using publicly archived data
Abstract: In this paper we start by critically analysing the publicness of various types of online archived data and discussing under which conditions such archives relieve researchers from their ethical requirement to seek consent from, and provide opportunities for withdrawal to, participants. We will argue that, from the perspective of the participants who contribute data to online platforms, most online data cannot be classified into binary categories of public of private, but lies within a spectrum between these extremes. If we consider, for instance, the way in which many people use Twitter, we observe that twitter conservations often take the form of discourses within friend networks rather than public announcements. By analogy we might consider such conversations akin to a discussion between acquaintances in a public space, like a café. While the participants are willing to accept that their conversation can be heard by other people in same space, they would not be comfortable with the idea that someone is systematically eavesdropping and analysing their discussion. We should note in this context that the fact Twitter automatically archives such conversations is not consciously considered by most Twitter users. In the second half of our paper we take a closer look at the case of Twitter for which we will discuss a range of participant consent and withdrawal procedures. We will present the outcome of a feasibility pilot in which we surveyed the willingness of Twitter users to provide informed consent for having past tweets analysed when the specific research question is clearly explained to them. Based on this survey we will also discuss if, and how, the filtering of participants based on willingness to consent might skew the resulting data. We conclude with a set of recommendations for participant consent and withdrawal procedures to be used when accessing online archived data.
Elvira Perez Vallejos presented: Ethical considerations for online mental health communication research
Abstract: Young people experience severe and potentially long-lasting psychological difficulties, yet many perceive difficulties in communicating their concerns to professionals and only a fraction receive available support services. We propose to investigate the linguistic strategies with which adolescents present mental health concerns in online settings, the barriers they identify in communicating their experiences and how these might be alleviated. The study will utilise an interdisciplinary team, applying expertise in applied linguistics and human computer interaction to elucidate adolescents’ expressions and experiences of psychological distress. Relevant for this workshop will be the ethical considerations regarding informed consent, trust, anonymity issues, privacy and the right to withdraw. We will discuss the potential challenges regarding data access and analysis from a user centric perspective and potential solutions such as explicit opt-out/opt-in recruitment strategies. The results will inform subsequent planning and design of online research including vulnerable young people.