Internet Research

The Internet is widely used for research activities including data collection via electronic surveys, recruitment of research participants, and for interventional/observational activities. All uses present unique concerns for the investigator and for the IRB.

Protection of human subjects is no less important when the research involves observation of online behavior. The IRB will make every effort to ensure individuals such as members of online cancer support-groups grant consent before having their discussions used for research purposes.

With the various uses the internet provides, there are challenges that may result pertaining to data quality. Issues surrounding social networking, posting images and videos of participants online, and recruitment and compensations can also be complex and therefore require the investigator to take extra precautions to ensure that privacy and confidentiality are not breached.

Read More: Considerations and Recommendations Concerning Internet Research and Human Subjects Research Regulations (Produced by the HHS Secretary’s Advisory Committee on Human Research Protections (SACHRP)).

On this page:

Using the Internet as a Data Source

Using the Internet as a Research Tool

Guidance for Online Data Collection/Consent

Challenges of Internet Research: Data Quality

Issues in Social Networking Sites

Blog Research

Posting Images or Videos of Participants Online

Working with Children Online/ Additional Security Considerations

Recruitment and Compensation on the Internet

Literature on Internet Research

Cases of Breach of Privacy

Using the Internet as a Data Source

Research involving observation and reporting of online behavior is sometimes called data mining. The term data mining also refers to sorting through data to identify patterns and establish relationships.
Consider the following:

  • Not all content on the internet is “public information.” Access is not a justification for collecting data without consent from the subject(s).
  • Most online groups do not require individuals to participate in discussions. After obtaining IRB approval and PRIOR to collecting ANY research data, permission must be sought from the list/group/community manager, and an announcement should be made to the list/group/community of the investigator’s intention to conduct research on the group.
  • Consent must always be obtained from subjects before attempting to collect private information. However, the investigator may request a waiver of informed consent. Concern that permissions will not be granted is not a justifiable reason for the IRB to waive consent.
  • Procedures must be in place to verify that research participants are adults.
  • Depending on the sensitivity of the research/data, clarity of disclosure, and confidentiality/anonymity of subjects, the IRB may require permission and/or informed consent to be obtained in addition to a community-wide announcement.

Using the Internet as a Research Tool

The internet can be used as a method of surveying subjects via direct email, web surveys, or other electronic instruments. The internet is a non secured medium meaning data in transit is vulnerable. The potential harm resulting from breach of privacy and/or confidentiality is accentuated if the research involves data that places subjects at risk of criminal or civil liability and/or could damage their financial standing, employability, insurability, or reputation.

Guidance for Online Data Collection/Consent

  • A consent document for the internet should be written like a cover letter and should include all the elements of a regular signed consent, as appropriate. The signature line should read, “By completing the survey, you are agreeing to participate in the research.” Web based surveys should offer subjects the option to “Agree”/”Not Agree.” Online consent may not be appropriate for studies involving highly sensitive information.
  • For sensitive data transmission, the investigator should include a disclosure in the consent process, such as: “This research involves the transmission of data over the internet. Every reasonable effort has been taken to ensure the effective use of available technology however, confidentiality during online communication cannot be guaranteed.”
  • An alternative means for completing the survey should be offered where appropriate, such as printing the survey and mailing it in.
  • Survey instruments should be designed in such a way that allows participants to skip questions or provide a response such as “I choose not to answer.”
  • At the end of the survey, there should be one button to submit the data and another button to discard the data. The purpose of these buttons is to ensure that a subject may withdraw at any time and to help them understand that if they do withdraw, even after completing the survey, their data can be discarded prior to transmission to the researcher.
  • There are various protocols for transmitting data securely over the internet with SSL (Secure Sockets Layer) connections or Secure HTTP. Both SSL and S-HTTP can work interdependently or together. On an IRB application, the investigator must describe the technology chosen for implementation of the research and justify the plan based upon the sensitivity of the research.

For more information on online subject recruitment click here.

Challenges of Internet Research: Data Quality

Sample Bias – Exclusion of Subjects

  • Internet samples make it problematic for researchers who attempt to generalize to broader groups.
  • Internet users vs. non-users vary on demographic, social, and psychological dimensions.
  • Dropout is an issue.
  • Longitudinal data collection is problematic online because email addresses and membership in online forums can quickly change.

Control over Data-Collection Setting

  • Researchers can lose control over the environment in which the research is conducted when they conduct surveys or experiments online.
  • Behavioral monitoring and verification of subjects’ identities, age or gender present increased challenges.
  • Online subjects may simply invest less time and energy in the research task than those involved in a telephone survey or laboratory experiment.

Issues in Social Networking Sites

  • Voluntary participation and identity disclosure is violated when researchers participate in discussion groups and bulletin boards and/or record and analyze text, but do not identify themselves as researchers (Engel & Schutt, 2010).
  • Posting patient protected health information is a violation of the Heath Insurance Portability and Accountability Act (HIPAA).
  • Posting pictures of patients (including those taken in international sites) is a potential violation of patient privacy.
  • The participants should know what information the researcher has/will collect; boundaries should be clear regarding what information will be gathered.
  • Knowing the researcher – if participants were to look at the researcher’s Facebook profile or conduct a “Google” search on her prior to the first study visit might this impact how they answer questions (Brooks & Churchill)?

Blog Research

To ensure that privacy is protected when using blogs for research:

  • Assess the blogger’s intent to be more or less private.
  • Ensure that the information used does not contain identifiable information. If a blogger uses a pseudonym used elsewhere on the internet, it may lead someone to information that can help identify the blogger.
  • To respond to higher levels of privacy, you may want to exclude certain content. Obtain consent from the blogger or mask certain details to minimize risk.
  • Work with the IRB to come up with a plan to help clarify how to calibrate privacy protections for blog material (i.e. exclude password-protected blogs).

Posting Images or Videos of Participants Online

Posting images or videos online can have ethical ramifications. It is good ethical practice to let the people identified in the picture know that you are photographing them and inform they of any plans to post online.

Participatory Photography

  • Online networks, such as PhotoVoice, have increased opportunities for artistic expression through participatory photography. Photo Voice employs participants to take images of their surroundings/neighborhoods as a way to advocate for social change.
  • Using images created of or by research participants can become complex as issues of consent, anonymity, and copyright come into play.

Tips for Photographing Subjects

  • Research participants may be actively involved in gathering data as they take photos to address different topics regarding social change.
  • Researchers should tell participants if they are required to report any photos revealing child or elder abuse or the likely prospect of harm that a subject may inflict on themselves or others.
  • Investigators must learn what methods of consent are acceptable to the IRB prior to photographing or otherwise recording subjects.
  • Researchers legally own the photos.

Working with Children Online/ Additional Security Considerations

  • Investigators communicating with children over the internet are subject to the Children’s Online Privacy Protection Act (COPPA) in addition to human subject regulations (45 Part 46 Subpart D). Investigators must not collect personal information from a minor without verifiable parental consent.
  • As appropriate, technology may be used to help screen out minors, such as programs that check for Internet Monitoring software or Adult Check systems.
  • The data file used for data analysis should be free of IP addresses or other electronic identifiers. If IP addresses are collected by the survey tool, the addresses should be deleted from the downloaded data file.

Recruitment and Compensation on the Internet

The text and context of the recruitment message for any Human Subjects Research must be reviewed and approved by the IRB before the research begins. This includes posting to a blog or message board, mass e-mailings, and web pages created for recruitment of participants.

Tips for Compensating Anonymous Subjects

  • Consider using gift certificates from online retailers (and displaying the unique certificate redemption number to subjects after they complete the survey), in lieu of requiring identifiable information in order to mail out compensation.
  • Do not link compensation to contact information.
  • Use an intermediary service to manage your compensation.

Literature on Internet Research

Cases of Breach of Privacy