Internet-Based Human Subjects Research

When recruiting through social media please include the following in the IRB application:

• The text of researcher-initiated posts and planned responses.
• Ensure that the language used follows the language included under the recruitment section.
• Clear description of the target population and rationale for targeting that population.
• Specification of whether data will be collected as part of the recruitment process via social media.
• Description of how you will communicate to potential subjects that information shared via social media is not secure.
• Refer to the IRB-approved template language about privacy, confidentiality and social media in the IRBNet Library. ⁸
• The method of communication (i.e. private Facebook messages), a plan for the frequency of messages, and the process for responding to incoming messages.
• Specification of which study personnel will be in charge of communicating and monitoring the recruitment account as well as any procedures to monitor the activity of those personnel.
• When using social media as recruitment strategy, use social media recruitment flier template in IRBNET Library. ⁹

Some study teams choose to recruit subjects through two-way communication via researcher-initiated social media accounts. This type of recruitment involves sustained and focused outreach to potential research subjects through direct messages on social media platforms (i.e. Facebook messages or Twitter direct messages).

Importantly, the IRB discourages study teams from establishing a social media account simply for the purposes of recruitment. The IRB will consider this approach to interactive recruitment for studies/researchers that have developed a well maintained, self-sustaining and highly-trafficked online presence

Recruitment of subjects online whether by email, chat room, discussion board, or others—must be guided by the level of privacy expected by participants. Online privacy expectations are very subjective and depend on the person/group and the nature of the information being obtained online. Some sites are "fully public" whereas others require user login or invitation by a site administrator. For instance, discussion boards (e.g., Google Groups, Facebook Groups, Linkedin, etc) may have multiple "administrators" who have more privileges than people who post. These administrators may be cautious about responding to requests for permission to analyze their discussion board posts.

Researchers must ensure they obtain appropriate permission from a website or discussion board administrator and include a copy of that approval in his/her IRB submission.

The websites of many professional societies often have guidelines for contacting (or not contacting) its members to participate in research, even if the members' names/emails appear on that organization's website.

When using a list of email addresses, please inform the IRB how you obtained the email list and if the organization owner of the email list has provided permission to use their email addresses.

If using email as a recruitment tool, the content of this email should include information about how the person was identified to be sent the email, what is involved if the person participates and an overview of any risks or potential benefits.

It should also let the person know how to inform someone if he or she wants to participate, not participate, or where to get answers to additional questions, and, of course, who is conducting the study and for what purpose. All scripts or guides that will be used for recruitment must have IRB approval prior to distribution.

Informed consent is one of the fundamental principles of ethical conduct in human subjects research. It is mandated by Federal regulations 45 CFR 46.116 and 46.117, as well as 21 CFR Subpart B.

Obtaining consent for a research project is a PROCESS and not the act of simply signing a form. Whenever required by federal regulations, informed consent must be obtained and documented. The language of the consent must be designed in a way that will be understood by the age group as well as reading and comprehension level of the target audience.

Usually, consent is documented when the research participant signs a form. Other mechanisms include audio- or videotaping the subject's agreement to participate. However, this is often difficult in internet-based/online research.

Under certain circumstances, for low risk studies, the requirement to document (obtain signatures or record the consent process) informed consent for online research activities can be altered or waived. See Florida Atlantic IRB Policy "Waiver of Informed Consent" for guidance on this topic. ¹⁰

People under the age of 18 cannot legally consent to participate in a research study. However, researchers recruiting from the internet cannot know the exact age of respondents. For this reason, it is best to limit online research studies to minimal risk research that would typically qualify for a waiver of parental consent under federal regulations (See section IV of Florida Atlantic IRB policy "Children in Research") ¹¹ unless you are able to authenticate the identity and age of your respondents through reliable means.

For internet research involving confidential and sensitive information, researchers must obtain written consent (on-line or hard copy) and within the studies procedure, include the proper methodology to verify the identity and age of the participant as well as to assure the security of the data.

When anyone who has access to the internet is a potential research subject, it may be difficult to ensure that he/she understands the consent information. One method of enhancing comprehension includes incorporating short questionnaires within the consent process to assess understanding of the information presented, and direct the subjects to additional explanatory material. Another method is to design and implement an interactive consent process tailored to the potential subjects' characteristics or primary language.

The principle of respect for persons requires that subjects be allowed to withdraw from a research study without negative consequences. Online survey instruments must explain at the outset what options are available, if any, for retrieving and discarding responses, and for some studies, it may be appropriate to provide a "no response" option for questions subjects may consider to be sensitive or intrusive.

The internet provides unique opportunities for conducting observational research in "private" settings. For example, a researcher can join a closed group (e.g. a "members only" chat room devoted to a mutual topic of interest) with relative ease to observe interactions among the members while concealing his/her identity. Such research can only be approved if the IRB determines that the deception is justified. Any research involving deception should have a plan for debriefing subjects, so they may learn about the research that occurred and have the opportunity to withdraw use of their data if they choose.