100 Questions (and Answers) About Research Ethics. Emily E. Anderson

design that has only a single interaction with participants, such as a one-time interview or survey.

      Second, consider how you are going to collect data. Collecting anonymous data is not possible when you (or another member of the research team) meet in person with a participant to conduct an interview or survey, for example. By conducting a face-to-face interview, you know what the person looks like and therefore can identify him or her, even if you do not know the participant’s name or have any other identifiers. Depending on the topic of the study, being identified as a participant (even without any disclosure of information discussed) could be potentially stigmatizing. For similar reasons, if a researcher wants to collect data anonymously, participants can neither be video recorded nor have their pictures taken because they can be visually identified. Data from audio-recorded interviews are also not considered anonymous because participants’ voices are unique, like fingerprints, and therefore considered identifiable.

      Third, consider the kinds of data to be collected. For data to be anonymous, you cannot collect any information that can identify a participant. This information includes, for example, participant names, email addresses, and date of birth. Eighteen specific identifiers are listed in the Health Insurance Portability and Accountability Act (HIPAA) Privacy Rule (USHHS, 2015b). The Family Educational Rights and Privacy Act Regulations (FERPA) also provide a list of direct and indirect identifiers (USDE, 2017). Some identifiers may be unique to a particular study participant, such as a description of a tattoo, and when known, could identify the participant. A combination of identifiers when viewed together, such as ethnicity, sexual orientation, and age, could also reveal the identity of a study participant in some situations, especially when research is conducted in small towns or communities.

      Data that were originally collected with personal identifiers can become anonymous data, in theory, if all personal identifiers are removed from the data and any documents linking identities and data are destroyed.

      Ultimately, it may be difficult to collect data that are truly anonymous. Often researchers want to know identifying information to provide context to the data or to maintain long-term contact with participants. If you need to collect participant identifiers but are concerned about the negative implications of others potentially discovering the identities of participants in your research study, use strict procedures to protect the confidentiality of study data and consider obtaining verbal consent, so that participants’ names are not linked to the study though their signature on a consent form.

       More questions? See #19, #23, and #25.