Abstract
The Internet has been a popular source of data amongst academic researchers for many years, and for good reason. Online data collection is fast, provides access to hard-to-reach populations, and is often less expensive than in-lab recruitment. With these benefits also come risks, such as duplicate responses or participant inattention, which can significantly reduce data quality. Very recently, researchers have become aware of another concern associated with online data collection. Bots, also known as automatic survey-takers or fraudsters, have begun infiltrating scientific surveys, largely threatening the integrity of academic research conducted online. The aim of this paper is to warn researchers of the threat posed by bots and to highlight practical strategies that can be used to detect and prevent these bots. We first discuss strategies recommended in the literature that we implemented to identify bot responses from online survey data we collected in the past six months. We then share which strategies proved to be most and least effective in detecting bots. Finally, we discuss the implications of bot-generated data for the integrity of online research and the imminent future of bots in online data collection.