The ethical use of social media research

x Sponsored content on Research Live and in Impact magazine is editorially independent.
Find out more about advertising and sponsorship.

If you are collecting information on a person’s attitudes, beliefs, habits, views or demographic characteristics, should it matter that you are collecting it on a virtual platform rather than on a physical street?  

Social media is a primary avenue for interaction and, as a result, is valued by clients and researchers for the wealth of data and the insights it can deliver. It covers platforms ranging from blogs or microblogs – such as Twitter – to video-, photo- or image-sharing sites (YouTube, Pinterest and Instagram, for example), and from online forums such as Mumsnet, to social networks including Facebook, LinkedIn or Google+.

It also encompasses avatar-based social spaces, as well as virtual worlds and online gaming spaces, such as Second Life and The Sims. These provide a vast data store, with opportunities for researchers. However, use of all these platforms still needs to meet fundamental legal rules in robust data-protection legislation, and ethical principles and safeguards in the Market Research Society’s Code of Conduct.  

Ensuring that researchers’ ethics apply equally to the virtual, digital world as to the physical world continues to be difficult. The lure of large-scale data sets that are often easily accessible, with valuable nuggets of information about what people are currently thinking, has brought new and evolving challenges. MRS guidance on online and digital research continues to reassert the fundamentals requiring that researchers, among other things: 

• Obtain informed consent when using identifiable participant data, and ask permission to use techniques that uniquely identify participants, such as cookies, unique IP addresses, digital fingerprinting or browser profiling. 
• Minimise the collection of unnecessary data that can often be captured in social media monitoring, text analytics or sentiment analysis.
• Ensure that any ‘sentiment’ data collected – for example, use of phrases or expressions – does not include identifiable data and that identifiers, including in the ‘metadata’ (or information about the data), are not used. 
• Avoid surreptitious, misleading or unsolicited data-collection techniques. Researchers should declare their presence, their role, the identity of the organisation they work for, what information they intend to collect, what it will be used for, and who will have access to it.  

But do these platforms raise new ethical concerns? If so, what can – and should – we do about it?  

Ethical concerns arise at all stages of the research process. They start at the recruitment stage, when finding participants for research projects and collecting primary data, both actively and passively. This extends into analysing secondary data generated online, such as by tracking data streams through automated monitoring of sentiments or specific desk research, and reporting in a meaningful – but privacy-enhancing – way. 

But obtaining consent for social media research is an area that continues to raise tricky ethical questions. Researchers need to follow the terms of use of each platform and meet user expectations.  

Some platforms, such as Twitter, are publicly accessible, while others, such as Facebook, require a password. Although boundaries may be clearer in private spaces, does the fact that information is in public spaces mean users intended – or were even aware – that it could be used in this way?  

Publication of one comment, tweet or post is magnified and amplified when used by researchers with other data sets. We need to determine where we draw the line because, although publication in this way may not have been intended by ordinary citizens, thought leaders who brand and market themselves as such may have lower expectations of privacy. Is carrying out or using research that identifies thought leaders or key influencers on Twitter allowable in these circumstances? 

A way forward?  

The Ipsos MORI social media report Wisdom of the Crowd identified some of the inherent concerns. It also threw out a series of challenges to researchers, regulators and owners of social media platforms to address some of the thorny ethical and methodological issues. Some other initiatives – such as the draft Data Science Framework – are also starting to produce results.  

One thing is certain, there are no easy or straightforward answers. Understanding and addressing the ethical dilemmas and challenges in social media research requires careful consideration of the interplay between several factors: the type of platform; the terms and conditions of use, and the licence granted to the researcher; the type of data being collected; the likely use of the data; and the approach to reporting and publication – all need to be reviewed in assessing ethical use. 

Consideration of the level of harm and risk at all of these stages will determine the likely level of risk posed by the research and the safeguards that need to be put in place.  

MRS is developing a framework to offer commercial researchers a workable approach to ethical decision-making. This will help them to ask the right questions enabling them to make ethical choices in social media research and reflect public expectations. 

Social media research guidance needs to recognise evolving technologies while encouraging and developing a research culture that continues to respect participant privacy, regardless of the platform they choose to use. 

Dr Michelle Goddard is director of policy and standards at MRS