It’s a strange time to be a data scientist.
News about Cambridge Analytica’s ‘problematic’ (we prefer the term morally bankrupt) approach to data analysis has left the 23 million Canadians on Facebook feeling suspicious and defenseless. And the community of social, behavioural and data scientists who work in marketing and communications have never looked more like the enemy.
It’s important to note that the firm’s alleged impact on electoral upsets such as the 2016 US Presidential election and the Brexit vote may be exaggerated. When witnessing a complex and unexplained phenomenon, never trust the guy with a simple answer – especially when he’s got something to sell you.
Regardless of impact, the ends of unethical data collection do not justify the means. It’s more important than ever to talk about ethics in research, especially as it pertains to social media. We are all, together, collectively, creating a massive data set – one like or photo at a time. This data set has tremendous power for those who can find insights and opportunities in it.
But with great power, comes great responsibility.
Whose obligation is it to develop and enforce ethics in this quickly changing space? Can we regulate Facebook and the social media giants into better protecting data and privacy, as proposed recent Canadian legislation attempts to? Is it up to the individual user of these platforms to secure and safeguard their own content? How can ethicists and academics hold the business research community accountable?
These are big questions that need ongoing discussion.
In the meantime, however, there is no excuse for unethical research methods, and no place for unethical researchers.
As a group of data scientists, analysts and strategists who build marketing strategies, we work in some similar ways to Cambridge Analytica. We too analyze social behaviour, profiles and conversations, at scale, using data scientists, analysts and anthropologists, in order to unearth insights that can make communications and marketing activities more effective. We look at the conversations happening on Facebook, Twitter and other platforms.
But we put ethics and user privacy at the core of our approach, which means our activities are not invasive or predatory for the communities we study.
In fact, data-driven consumer research can produce tangible benefit for these communities. For a healthcare client, these techniques allowed us to find and understand the experience of subjects who would be unlikely to participate in more traditional research methods such as focus groups or surveys. We discovered that these subjects were not just hidden from traditional research, but that they were also hidden from support organizations and critical patient care. This information led directly to the development of better support networks and services.
So how can data scientists and the organizations working with them navigate the complexities ethically?
To start, ethics need to be proactive not reactive. Our team has developed a set of ethical guidelines that we and any contractors must review and institute from the start of a research project. We outline how to get the data and treat the data once it’s sourced using the Association of Internet Researchers’ (AoIR) guiding principles of human dignity, autonomy, protection, safety, maximum benefit and minimum harm. We also review stringent academic ethics protocols and adapt them to this business environment. For example, we conceal our research subject’s identity by using pseudonyms and paraphrasing. We aggregate our data to conceal profile details and make identifying individuals nearly impossible. We follow users’ privacy settings first and foremost, and do not look for ways to break them.
The internet is a changing landscape, with new social platforms and new rules created daily. To be successful, you must adapt to these changes, a process which may involve revisiting your ethical guidelines from time to time.
Any research project that doesn’t put ethics first is built to fail.
- The Association of Internet Researchers (AoIR)
- Social Media Research: A Guide to Ethics by Dr. Leanne Townsend and Prof. Claire Wallace
1 Annette Markham and Elizabeth Buchanan, Ethical Decision-Making and Internet Research: Recommendations, from the AoIR Ethics Working Committee Version 2.0 (2012): 4.
——— Written by Kayla Musyj, former Qualitative Research Analyst, NATIONAL Public Relations, and Becca Young, Senior Vice-President, Strategy and Integration, NATIONAL Public Relations.