Sri Lankan women are having fun on the internet, and Sachini Perera and I were happy to talk about it at the Imagine a Feminist Internet Conference. After studying and talking about oppression, harassment, and discrimination faced by women on the internet (and elsewhere; after all, the internet is but another manifestation of our reality) it was refreshing to be able to talk about something positive. Yet, I was uncomfortable, a voice whispering in my head that this is not right. Our work had followed established ethical protocol, and it was revealing a side of this realm that allows women to act with not just agency, but pleasure. Yet, the feeling of discomfort persists even now, forcing myself to question: do the ethics of internet research based on social media platforms sufficiently address this many-faced entity?
Our work had followed established ethical protocol, and it was revealing a side of this realm that allows women to act with not just agency, but pleasure. Yet, the feeling of discomfort persists even now
The unprecedented boom in social media usage and its ability to capture many aspects and layers of the society with relative ease has made it the proverbial Holy Grail of data for researchers. Ethical considerations and frameworks for traditional (for the lack of a better term) have had decades of debates, discussions, and revisions to have Boards of Review with similar ethics regulations (although they are still being critiqued). The ethical framework of digital research, especially based on social media platforms, has seen no consensus. According to some researchers, social media as a medium for research does not demand special scrutiny, and that social media-based methodologies can be and should be evaluated using the same standards for privacy and transparency as those used in traditional research. However, as Lee points out, "current regulations have not kept pace with the diverse ways in which research can occur through social media"; one could scrape Twitter for sentiment analysis on Megan Rapinoe’s victory speech or use the webs to investigate the architecture of pornography sites. It is near impossible to predict the types of research on social media platforms, especially when expanding the definition of a text to be more than letters on a screen. There are but few discussions happening around the ethics of social media platforms like YouTube, Instagram, or TikTok where the content treads the blurred boundary of personal and public, attitude and performance; research on attitudes to phenomenon have been conducted for years, just as performance studies and research on it have been happening for years. Thus, one could always turn to decades of methodology and critiques of said methodology in order to determine the most suitable methodology for them. Yet, how are we to approach a hybrid of the two? When our data is both evidence of an attitude as well as performance, what rules then apply? Considering that the type of data involved in each is different, that each is governed by terms and conditions unique to each platform, that the purpose of each platform is different from another makes it impossible for one to homogenise these platforms. To complicate matters further, there is cross-platform posting, where the governing terms and conditions then become extremely complicated. It is but naïve to assume the ethical frameworks pertaining to traditional research sufficiently covers the intricately spun worldwide web. This, then begs the question: what is ethical research on social media platforms?
In term of Observational Research, which is what the TikTok study is, the main ethical concern is the possible involvement of human subjects. The regulations for identifying human subjects on social media research is three-fold: 1) access to social media website is not public, 2) the information gathered is private (identifiable but not private information do not constitute human subjects), and 3) information gathering requires interaction with the person who posted it online. TikTok is a publicly accessible platform; the information gathered for the study was identifiable but not private - “you control whether your User Content is made publicly available on the Services to all other users of the Service or only available to people you approve”; and no interactions were required with the content creators for the purpose of the study. Even if one were to overrule this rather limited definition of human subjects, the ethics surrounding “informed consent” on social media research relies heavily on the terms and conditions of the platform. Feminist research, however, demands that we question the ethics of our methodology, and question the integrity and responsibility of the research process. The existing frameworks, in my opinion, do not facilitate this conduct, especially with regard to informed consent.
Feminist research, however, demands that we question the ethics of our methodology, and question the integrity and responsibility of the research process.
Informed consent, especially from a feminist research point of view, is important because it shows respect for persons and allows them to control what happens/does not happen to them (and by extension) and their data. While more important when working with communities at risk, it is no less important in the current context. The general understanding among most researchers is that when the user agrees to the terms and conditions of a platform which explicitly states that the information is available to third parties and is publicly available, it serves as consent. TikTok explicitly states that:
By posting User Content to or through the Services, you waive […] any and all rights of privacy, publicity, or any other rights of a similar nature in connection with your User Content, or any portion thereof. (2019)
Furthermore, it informs the user that “you control whether your User Content is made publicly available on the Services to all other users of the Service or only available to people you approve”. Ethically, then, the use of data on these platforms pose no concern. However, Fonow & Cook suggest that feminist research should also recognise the potential for the respondents to be exploited; seeking cover under the terms and conditions of a platform could easily come under this. Studies have repeatedly shown that users rarely read through the terms and conditions of any application before agreeing to them; in fact -
If I have implicit knowledge that my “participants” may not be fully aware of what they agreeing to, am I upholding my integrity as a feminist researcher? The terms and conditions are rarely available in the vernacular; the content reads on the border of legal jargon. It’s almost as if these Terms and Conditions were written to breed fatigue, exhaust the reader to consenting without too much critical engagement. As a feminist researcher who is tasked with questioning inherent power structures and their exploitative nature, can I in good conscience say my practice was ethical? How can I reconcile the adherence to the regulations when common sense demands that these frameworks do not address social inequalities? Am I not being instrumental in perpetuating an agenda that I am trained to question?
As a member of the “born digital” generation, my personal use of social media is simultaneously public but personal. I often forget that it is but a tool, as the boundaries between what is “online appropriate” and the life I live are not distinctly marked. When I share content on social media, I am aware of the community I share it with, yet forget that by the virtue of a user agreement that I barely read, my content can be accessed by anyone; worse, I have given explicit permission for my content to be used by anyone for any purpose. I doubt I am alone in this behaviour. If we are multitudes, then the legal framework that supposedly allows any individual to access and analyse the information would not complement the way users use the platforms. Thus, assumptions made in creating this ethical framework for online research become null. Of course, there are other ethical issues embedded within such research: if the researcher is a part of the discourse community, where does the participant end and the researcher begin?
A social media platform has its own discourse community. Within this, there would be smaller communities brought together through hashtags or location tags. If the researcher is a part of this discursive community, it could be that there are bigger moral questions to grapple with. If the researcher has interacted with the content creators, for example, then the dynamic between the researcher and the subject would be different. Furthermore, a discourse community is based on the members having a shared interest (among other characteristic features). Thus, there is an implicit trust among the members which may be violated via research being conducted without explicit consent. While some platforms like Tinder explicitly involve all individuals who access its content to the community (You cannot proceed without selecting if you are interested in men, women, or both – thus implicitly getting you to be a part of a community), some other platforms allow people without an account to access public content. TikTok is one such platform. Being able to access the content without having created an account yourself, you are not bound the moral obligations of a discourse community. Yet, if the content creators are aware of this fact is a question.
If the researcher is a part of this discursive community, it could be that there are bigger moral questions to grapple with.
Since presenting the paper in February, I have been grappling with the ethics of my own practice. On one hand, the content creators chose to make it public: it no different from analysing a music video, I tell myself. But would I want my public videos (if I ever make any) to be the material for research, I ask myself, and the answer is a resounding no. A friend remarked that they can paint a better picture of an individual by observing their online behaviour than through a conversation. After all, do we not “stalk” our potential dates on the internet before agreeing to meet them for a drink? If our personal is thus embedded in our social media existence, it is imperative that our understanding of an ethical code of conduct and research framework reflect it. I do not have an alternative framework to offer. All I have are questions and a sense of deep discomfort with existing practices that demand probing.
 Gelinas, L., Pierce, R., Winkler, S., Cohen, I. G., Lynch, H. F., & Bierer, B. E. (2017). Using social media as a research recruitment tool: Ethical issues and recommendations. The American Journal of Bioethics, 17(3), 3-14.
 Lee, S. S. J. (2017). Studying “friends”: The ethics of using social media as research platforms.
 Moreno, M. A., Goniu, N., Moreno, P. S., & Diekema, D. (2013). Ethics of social media research: common concerns and practical considerations. Cyberpsychology, behavior, and social networking, 16(9), 708-713.
 McCormick, M. (2012). Feminist research ethics, informed consent, and potential harms. The Hilltop Review, 6(1), 5.
 Townsend, L., & Wallace, C. (2016). Social media research: A guide to ethics. Aberdeen: University of Aberdeen.
 Fonow, M., & Cook, J. A. (2005). Feminist methodology: New applications in the academy and public policy. Signs: Journal of Women in Culture and Society, 30(4), 2211-2236.
 Kim, D., & Vorobel, O. (2016). Discourse Communities: From Origins to Social Media. Discourse and Education, 1-15.
 Swales, J. (1987). Approaching the Concept of Discourse Community.