Online research offers many advantages. If well designed, a web survey/task or other data collection tool can collect robust data from large populations, incorporating validations and other tools (including anonymous cookies to minimise the risk a participant submits more than one set of responses). These techniques can ensure any data collected is more likely to be useful and reliable. Collecting data online also means that participant responses to survey questions/performance in tests can be directly added to the data set for analysis (rather than transposed later).
Another apparent ethical advantage of online research is that it can be collected anonymously so that not even the researchers know who disclosed what. This anonymity can seem especially attractive when individuals are to be asked embarrassing questions, asked about sensitive matters and/or where there are risks associated with the collected data. It might then seem easy to respond to the risk questions in an application for ethical review:
“There are no risks, no matter requiring attention because the researchers won’t have data that can be traced to an individual.” Right?
Ordinarily, if a participant used an interview to disclose significant mental anguish an experienced interviewer could suspend the data collection to counsel a participant and to urge them to seek help from someone with appropriate training. If an interviewer was sufficiently worried that there was reason to believe there was imminent and serious risk to the interviewee or others they could decide to alert the appropriate authorities.
Neither options are available to a researcher conducting anonymous online research. This enables a researcher to evade any mandatory reporting requirements to which she or he might be ordinarily subject. But, is this the right thing to do?
The same argument might be true for offline research (such as an anonymous paper & pencil survey). However the very wide scope of online data collection and the degree to which people appear very willing to disclose things online, does appear to increase the likelihood such disclosures will be made in online data collection.
I have seen this happen, where a participant used an online anonymous survey seeking views on an apparently innocuous psychological subject to disclose an intention to self-harm. Not only were the researchers unable to take any action, their powerlessness to act caused the research team (especially the junior members) a great deal of distress.
The implications of this case might be that there are some subjects and potential participant pools where data should never be collected anonymously. This might be the case even when the informed consent material urge participants to seek expert support if they are distressed. In fact in the case I am referring to above, such a statement was in fact present in the informed consent materials. Consequently, perhaps we should consider whether there are circumstances where data should initially be collected in an identified form, only to be de-identified once the researchers have determined there are no compelling reasons obligating them to act.
Of course such a mechanism raises its own ethical, risk, consent and legal considerations.
Gary Allen,
Office for Research,
Griffith University
Gary’s AHRECS profile
Gary’s Griffith University profile
This blog may be cited as:
Allen, G (2015, 7 August) The perils of anonymous online research and risk: Two hands tied behind you back. AHRECS Blog. Retrieved from https://ahrecs.com/uncategorized/the-perils-of-anonymous-online-research-and-risk-two-hands-tied-behind-you-back