At its meeting on October 30, the Secretary’s Advisory Committee on Human Research Protections (SACHRP) heard a presentation from B.R. Simon Rosser, University of Minnesota School of Public Health, on “The Evolution of Consent in Low-Risk Studies: Lessons from Online Survey Research.” He suggested that SACHRP rethink how researchers handle informed consent for low-risk studies conducted over the internet.
Rosser observed that several norms of online behavior make the informed consent process for research tricky. It is not uncommon for people to give consent or accept terms of service agreements without reading them, or to misrepresent their identities (for example, giving a fake name or email to avoid getting spammed). He suggested that for research conducted online, the consent process can be better tailored to the environment. It already differs in some ways from pen-and-paper consent (prospective participants are screened for eligibility prior to consent, for example), but studies of the drop outs during the consent process suggest it could still be improved. Twenty to thirty percent of prospective participants dropped out during the consent process, citing reasons including length, hassle, inadequate payment, inconvenience, and privacy concerns.
The most common risk of this type of research is that participants will be asked a question that makes them uncomfortable. But unlike when a survey is conducted in person, Rosser observed, online, a participant can just “click off” and end their participation.
Rosser posed the question of whether traditional “consent” is necessary at all for minimal-risk online research (with notable exceptions, including research involving deception or vulnerable populations). He suggested a model in which prospective participants are shown a screen with information on the length of a survey, the topic of the questions, any payment for participation, and a stipulation that participants must be at least 18, and are given the option of viewing more information or going straight to the survey (implying consent).