A research team in the School of Social Work, led by assistant professor Kinnon Ross MacKinnon, has developed a new protocol to identify and remove fraudulent responses from web-based surveys.
The team noticed suspicious responses while recruiting for a research study on detransition experiences among LGBTQ2S+ people in Canada and the U.S. They started to apply a new method to filter out fraudulent responses from bots or scammers. Their protocol excludes questionable responses based on a list of criteria including geolocation, low reCAPTCHA scores, duplicate IP addresses and more.
The team’s method may become a new best practice for studies involving hard-to-reach populations. MacKinnon has published a template for researchers based on this new approach in the Journal of Medical Internet Research.