No need for fury: Ethics and internet research

After all the fuss about Recursive Fury, I did some reading on ethics as it relates to internet research. There is quite a bit of material available. Below is some of what I found. These are from international literature and I'm not claiming they would apply in all jurisdictions.

The short answer is, however, that if what I've read is any guide, there was no breach of ethics by the authors of Recursive Fury in preparing or writing the paper, and none by its publication. (I can't say the same for its withdrawal by Frontiers. I didn't research that aspect however the behaviour of the "Chief Editor" did not come across as very professional to say the least.)

No diagnosis

Complaints about Recursive Fury were largely centred on two themes.  Firstly, that the study "pathologised" people who made comments.  This took the form complaints such as:

• diagnosing people in absentia as having mental disorders, - claimed by Anthony Watts at WUWT
• naming people in a science paper as having a psychological affliction without their consent - a similar claim in the same WUWT article

Yet the paper itself contained no diagnoses of individuals, let alone one that attributed a "mental disorder" or "psychological affliction". (Not that I've noticed Anthony Watts being in absentia, nor would I think too many people would consent to having a psychological affliction.)

No human subjects

Another objection raised related to whether informed consent was necessary.  Whether or not informed consent is required from an ethical standpoint, the first condition is that the research relates to "human subjects".  The National Science Foundation has a policy: 45 CFR Part 690: Federal Policy for the Protection of Human Subjects, in which it defines "Human Subject" as follows:

 (f) Human subject means a living individual about whom an investigator (whether professional or student) conducting research obtains 

(1) data through intervention or interaction with the individual, or

(2) identifiable private information.

Intervention includes both physical procedures by which data are gathered (for example, venipuncture) and manipulations of the subject or the subject's environment that are performed for research purposes.

Interaction includes communication or interpersonal contact between investigator and subject. 

"Private information" includes information about behavior that occurs in a context in which an individual can reasonably expect that no observation or recording is taking place, and information which has been provided for specific purposes by an individual and which the individual can reasonably expect will not be made public (for example, a medical record). Private information must be individually identifiable (i.e., the identity of the subject is or may readily be ascertained by the investigator or associated with the information) in order for obtaining the information to constitute research involving human subjects.

In the case of Recursive Fury, although some of the textual material came from the blogs of the researchers the comments arguably were not as a result of any such communication or interpersonal contact but rather were because of the publication of the "moon landing" paper, LOG12.  The research certainly did not meet the definition of "intervention". There was no venipuncture or any manipulation of the subject or the subjects' environment. Nor did it meet the definition of "private information".  The blogs from which the information was gathered were not merely public blogs, rather than private, they were much publicised public blogs and, according to Alexa rankings, were the more popular blogs in their sector of interest and, indeed, on the internet as a whole.  Blog owners did not require readers to log in.  There was no expectation of privacy. In addition, it is obvious from reading the blogs that many of the blog owners (and commenters) crave publicity, promoting new blog articles on Twitter for example.  One blog owner boasts of wide readership.

Therefore the complaint that prior consent was required fails before passing the first hurdle, that of "human subjects".

Waiving requirements

Given that the textual analysis did not involve human subjects as per the definition above, the following waiver does not apply.  Still it's worth reading.  This is from the same National Science Foundation policy:

(d) An IRB may approve a consent procedure which does not include, or which alters, some or all of the elements of informed consent set forth in this section, or waive the requirements to obtain informed consent provided the IRB finds and documents that: 

(1) The research involves no more than minimal risk to the subjects;

(2) The waiver or alteration will not adversely affect the rights and welfare of the subjects;  

(3) The research could not practicably be carried out without the waiver or alteration; and  

(4) Whenever appropriate, the subjects will be provided with additional pertinent information after participation.  

(e) The informed consent requirements in this policy are not intended to preempt any applicable federal, state, or local laws which require additional information to be disclosed in order for informed consent to be legally effective. 

(f) Nothing in this policy is intended to limit the authority of a physician to provide emergency medical care, to the extent the physician is permitted to do so under applicable federal, state, or local law. (Approved by the Office of Management and Budget under Control Number 9999-0020.)

A strong argument could be mounted that requirements (1) and (2) would have been met in this study and (3) would arguably apply as well. That's if the policy had applied to this research, which it didn't because the entire policy only relates to "human subjects".

Public vs Private

Here's a snippet from a paper by a PhD student in Sweden, which discusses the distinction between public and private in the context of internet research ethics (my paras and bold italics).

Without reducing the complexity of problems and previous arguments, it should be noted that the private/public distinction is a recurring theme in IRE [Internet Research Ethics] debates. More precisely, in online contexts, the boundaries between private and public appear blurred (Bromseth, 2002; Löfberg, 2003; Mann, 2003; Thorseth, 2003; Sveningsson Elm, 2009). This makes it difficult for researchers to assess the sensitivity of information and situations, It also makes it difficult to determine when research requires informed consent.

Attempts to solidify the private/public status of online phenomena have been made from various ontological and epistemological viewpoints, but arguments tend to follow one of two lines of reasoning: online phenomena can be considered public either

(1) if publicly accessible or

(2) if perceived as public by participants (Bromseth, 2002; Chen, Hall, & Johns, 2004; McKee & Porter, 2009; Sveningsson Elm, 2009).

According to the first argument, online phenomena are essentially public if they can be accessed by anyone with an open Internet connection. Moreover, public discourse must always be open for scholarly analysis and critique, and, in lack of restricted entrance, there is no need for consent or even anonymizing. The second and often counter-posed view holds that, though something may be accessible, the general public (including researchers) may not be the intended audience.

Researchers must therefore base their ethical decisions on a community’s purpose and participants’ expectations of privacy. Without taking consideration to personal privacy, researchers might instigate feelings of intrusion and exposure, or attract unwanted attention to online communities. My description of the above positions is, of course, a simplification.

In the context of the above, the material for Recursive Fury was public in that there was no restricted entrance nor was there any expectation of privacy on the part of people commenting.  Quite the contrary. An examination of comments shows that some of the same people posted their ideas on multiple public high profile blogs, indicating the the authors wanted as many people as possible to read them.  Indeed since Recursive Fury was published, some commenters have been posting their objections / conspiracy ideations on every site they can find which mentions the words "Lewandowsky" or "Recursive Fury".  (In some cases, so quick is their response that one might surmise they have a Google alert notification set up for these words.)

---

Rosenberg, A. "Virtual world research ethics and the private/public distinction." International journal of Internet research ethics 3, no. 12 (2010): 23-36.

Lewandowsky, Stephan, Klaus Oberauer, and Gilles E. Gignac. "NASA Faked the Moon Landing—Therefore,(Climate) Science Is a Hoax An Anatomy of the Motivated Rejection of Science." Psychological science 24, no. 5 (2013): 622-633.

Lewandowsky, Stephan, John Cook, Klaus Oberauer, and Michael Marriott. "Recursive fury: conspiracist ideation in the blogosphere in response to research on conspiracist ideation." Frontiers in psychology 4 (2012): 73-73.