Understanding the difference between social media research and traditional research
If an individual posts a comment on an open-access website which is in the public domain – should market research agencies be asked to mask the identity of that poster in a MR report? Should their privacy be respected or should they be asked for permission before their comments are used?
These were just some of the questions posed on “The Great Data Privacy Debate” hosted by Greenbook on social media and data privacy last week. We at DigitalMR were delighted that our previous blog post provided the catalyst for this important debate.
The debate was chaired by Andrew Jeavons, EVP at Survey Analytics and the panellists included myself, Peter Milla: leading author on CASRO’S Social media Research guidelines, Barry Ryan: MRS Standards & Policy Manager, Adam Phillips: Managing Director of Real Research, Chairman of ESOMAR’s Professional Standards and Legal Committees, Ray Poynter: active blogger and host of the NewMR LinkedIn group and Tom Anderson: Founder and Chairman of the Next Gen Market Research (NGMR) trade association and LinkedIn group.
As researchers, we are always interested in just how representative samples are, and the make-up of the panel group was no exception. With ESOMAR, CASRO and MRS unsurprisingly pro the current guidelines for social media research vs NGMR, NewMR and DigitalMR against.
Perhaps because of this demarcation the debate did at times appear to be between trade bodies representing traditional research and those, such as DigitalMR, representing the “new wave” of social media research.
I think there was also a danger that Esomar, CASRO and MRS could be perceived as creating guidelines more suited to larger research agencies with greater resources to respond and adjust to new regulations. The flipside is that trade bodies which are in place to promote research, are perceived (perhaps unfairly) to stifle entrepreneurialism and innovation among small to medium sized agencies.
Also from a geographical perspective, there was a UK bias, which again may have favoured a more “traditional” point of view. It would also be useful for further discussions to have more input from research agencies (we were the only one!) as this debate was heavily represented by the various associations.
The debate focused on broad industry issues around privacy and the law and if codes of conduct for social media research should exist. In our opinion we should have spent more time discussing the differences between traditional research and social media research, as we believe this is where much of the confusion lies. The trade orgs still think in terms of harvesting posts from the internet as some sort of a survey involving respondents with representative samples. This is clearly not the case:
- On the issue of representation - the notion of sample may be used in the context of “sample of websites” but there is no need to limit ourselves to a sample of people who post comments on a website: WE CAN HAVE THEM ALL! This is census data as opposed to sample data.
- On the issue of respondents - there is still the archaic use of “respondent” in the MRS guide, while ESOMAR refer to the posters of comments as “users” as opposed to respondents. Posters are in no way respondents in the traditional sense of the term, and “users” is at best misleading. We can all be web users without posting or interacting with a website in any way. We feel there needs to be a lot more work in agreeing common terminology so that we can be clear that we are all discussing the same thing.
- On the issue of “respondent participation” - there was concern about alienating people who would then stop participating in research. We believe that posters (these are not respondents!) will not stop participating in internet discussions just because their comments are published in market research reports. Some of the panellists in the debate seemed to confuse the potential reduction of internet participation with respondent fatique which is a totally different issue. We are not asking people to take part in surveys here.
In the original guidelines (which have not yet become part of the code of conduct, but may well do so) the language used has been quite prescriptive – with terms such as “agencies MUST” used quite frequently. I think one successful outcome of the debate was to persuade the trade associations to reassess and refer to their guidelines as “suggestions” which are meant to protect their members. As DigitalMR is still a member of both ESOMAR and MRS this is reassuring.
In terms of traditional versus the new, Ray Poynter mentioned several times that these guidelines have been written by “dead white men” and he is probably right. We need more diversification in the mix of people who write these guidelines and codes, and we need to take into account the impact of new technology, how it is changing our world and how it affects the views of consumers. Only by having a dialogue among all stakeholders, especially consumers and end-clients can we get a fully representative view of how best to move forward.
In closing we would like to restate what DigitalMR represents in this debate:
- The codes of conduct for MR need to be revamped in some areas
- Social media research is different from traditional research and should be treated that way. Appropriate language should be used in describing social media research.
- We feel there was a lack of discussion when the ESOMAR paper was out for consultation, prior to becoming a guideline. Hopefully the MRS paper and this discussion will impact the decision of all trade orgs to see things a little differently. More research and dialogue needs to go into guidelines like this, even before they become papers for discussion.
- Finally a specific minor detail which is most important from a DigitalMR perspective is this: when using quotes in MR reports, we (MR agencies) should not be asked to mask the handle/meta data of a person who posted a comment on a public website – if that website states that posted comments can be viewed by anyone.
We applaud the industry’s willingness to get together and discuss these issues publicly, and we were pleased to be given the opportunity to contribute. I think as the debate develops it will play a key role in shaping the future of market research. There is still a long way to go in developing a deeper understanding among the key stakeholders and we look forward to playing an active role in shaping the debate over the next few months.