Recruiting etiquette in cyberspace is different from conventional methods
It might be best to respect the web’s culture
|
(Editor’s note: This month, IRB Advisor presents the second part of a series on Internet research and the unique challenges it poses to IRBs while striving for optimal privacy and ethical protection of human subjects. In this issue, we look at how to handle privacy and other issues that arise during the recruitment of subjects for a study involving Internet communication and research. Also included is a summary of a task force report on the ethical and legal aspects of cyberspace human subjects research. In the January issue, we covered ways that Internet research can harm subject privacy and questions to ask Internet researchers.)
There may be no regulatory restrictions preventing an investigator from contacting a group of people talking in an Internet chat room to ask them to participate in a research study, but if it’s just plain bad Internet manners, then why not suggest a different strategy?
"We all get junk mail at home, and we as users seem to feel differently about that because it’s easy to throw away," says Mick Couper, PhD, senior associate research scientist for the Institute for Social Research at the University of Michigan in Ann Arbor. "Because the culture of the web has evolved differently, people have a strong feeling against spam and mass mailings to e-mail addresses," he says. "To send out questionnaires in the U.S. mail requires more effort — people, time, and money — and the financial constraints are therefore stronger," Couper says. "To send out messages in the U.S. mail, I have to set up an operation, spend money, and the financial constraints are stronger," he explains.
Besides e-mailing potential research subjects, there are other Internet methods that can be used for recruitment, including inviting people into a study on a chat room or listserv; setting up a web site that will be used to recruit subjects; and providing a link on an organization’s web site to a survey that can be printed and returned by mail.
IRBs will need to consider how much oversight is necessary when researchers use the Internet for recruitment. Here are some suggestions for what to consider:
• Consider how subjects will be invited to join Internet surveys.
"If I was an IRB member, one question I would ask a researcher is: How are people being invited to the survey, and how are they being told about how they are being invited?’" Couper says. Some IRBs do not object to a researcher recruiting members off a listserv or chat room as long as the survey request does not conflict with the listserv’s or chat room’s stated rules, he says.
Participants have choices
The important thing is to make certain that participants are volunteers who understand that they have a choice of not answering questions; and that if they do choose to answer, their answers will remain confidential, says Andrew Cockburn, PhD, director of institutional research compliance/biological safety at West Virginia University in Morgantown.
When researchers establish a web site that will be used to recruit subjects, an IRB might consider this part of the advertising and review it, says David Forster, JD, MA, CIP, director of regulatory and legal affairs for the Western Institutional Review Board in Olympia, WA. The Western IRB has a policy that the IRB will review all web sites used for recruitment purposes, including parts of the site that are unrelated to the proposed research, Forster says. "That’s different from what we do for other forms of media," he says. "We may ask to see advertising, but we don’t ask to see nonresearch-related ads; but because of the connectivity of the Internet, we ask to see the whole web site." This may be reviewed by one IRB member or by the entire board, and it may extend to reviewing the web site’s links, although that falls into a gray area and typically would be handled on a case-by-case basis, he adds.
Another strategy is to require Internet research participants to enter an identification number and password when they log onto a web site. This way they are not asked to enter their e-mail address, which would identify them, Couper says. Then researchers would use only the identification number to send them follow-up e-mail or reminders or to offer incentives for continued participation, he adds. "But as far as the investigator is concerned, the link to e-mail addresses is removed as soon as the data collection is over." In this type of scenario, the Internet offers even greater privacy protection than telephone or in-person surveys, Couper explains. "If we knock on their door, we know where they live; if we call them, we know their telephone number, but it can be set up on the web so that the researcher does not know where or who the person is."
• Address the use of pseudonyms and web quotes.
When an investigator is conducting chat-room research, there may be opportunities to include comments made on-line by participants. Several issues pose a problem for an IRB when research is expected to include anonymous comments.
First, IRBs will need to decide whether they will permit investigators to use the pseudonyms that chat-room participants might use while on-line, says Amy Bruckman, PhD, assistant professor in the college of computing at the Georgia Institute of Technology in Atlanta. "Pseudonyms frequently are traced back to real people because people will use the same pseudonym over the same period of time," Bruckman says. "People may choose to use a real name as a log-in or use the same name so persistently that if they at any point reveal their real name in connection with the pseudonym, it is publicly archived somewhere."
Bruckman has conducted research that involves observing chat rooms on-line and has found that chat-room participants sometimes are angry by what they see as an intrusion. By using their personally chosen pseudonym, an investigator may be further intruding on their privacy, she suggests. "Not everything that upsets people is unreasonable, but it adds to the argument that you can’t just walk in and record the chat room," Bruckman says.
Privacy different on the web
Another issue that can arise involves the use of quotes from web-based sources, including chat rooms. If an investigator interviews a subject either by telephone or in person and then changes the subject’s name and identifiable features when using the subject’s quotes, then this is considered ample confidentiality protection. However, if the same quote is drawn from a web-based source, such as a chat room or bulletin board, it may not offer the same privacy, Bruckman explains.
"Even if you change the names and pseudonyms of people involved, someone who does a full-text search may be able to go back to the original posting," Bruckman says. "Anytime you quote someone, it may be findable by a search, and so you are identifying them even if you don’t use their name." An IRB may decide that if a study involves low risk to subjects, then this potential identification is a negligible concern. But in other cases, the IRB might decide it poses too great of a risk and that it is not justified, Bruckman says.
• Decide on a policy regarding identifying and contacting children subjects.
IRBs will find that there are no clear-cut answers regarding protecting children during Internet research, whether or not the children are the intended subjects. "If you have a site where access to children is not appropriate, then you have to find some reasonable way to exclude children from participation," Bruckman says. "And that’s tough."
Previously, an investigator might have asked for a credit card number as a means of determining whether the participant is an adult. Now, many minors have access to credit cards, so it’s unclear whether that method will work, Bruckman says. When Bruckman and colleagues asked for informal federal guidance on this issue, the answer was that it is not satisfactory to simply ask Internet participants to state whether they are age 18. This means each IRB should develop a policy or guidance on an acceptable method of identifying a subject’s age over the Internet.
The other issue involves obtaining parental consent when children are sought as subjects of an Internet study. Bruckman and fellow investigators settled that question by having children print out a permission form, ask their parents to sign it, and mail it back. "That’s a real hassle to us, but we decided we had to do it," Bruckman says. "We lose 90% of the children who are interested, but I do think it’s the right approach."
• Advise researchers on when and how to obtain informed consent from on-line subjects.
When signed informed consent is deemed necessary, this should include information about how the use of the Internet has a risk of a breach of confidentiality, says Michael Gallo, PhD, associate professor at the Florida Institute of Technology in Melbourne.
Couper says he would argue that if an investigator establishes a web site in which participants voluntarily enter, review, and decide to answer questions, then that constitutes informed consent. "Every step in the process of doing a web survey is voluntary, and people can opt out at any point," Couper adds. "You can get a waiver of informed consent, and many of the web surveys I do fall into that waiver category." Also, most Internet surveys probably fall into the category of involving minimal risk to respondents, Couper says. "The questions are often mundane, and the threat of disclosure is very small, and thus informed consent may not be necessary," Couper explains.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content