There are a growing number of surveys recruiting and collecting data from participants solely through online methods, without any contact via telephone, mail, or in person. A concern for any such survey is the assurance of effective antifraud measures such that no one person is able to enter the study multiple times. The purpose of this paper is to share an antifraud method from an innovative study, the National Cancer Institute's QuitTXT Smoking Cessation Evaluation (QuitTXT). This study included online recruitment with a non-probability sampling method, data collection via Web and cell phone, and monetary incentives. In QuitTXT, antifraud efforts centered on detecting duplicate phone numbers, email addresses, and IP addresses to prevent a single individual from enrolling into the study multiple times. We developed procedures during the recruitment phase of the study, along with a post-recruitment method for detecting and eliminating multiplicity when lapses in the screener antifraud methods occurred. This retroactive fraud detection is useful for any study recruiting and implementing a study entirely online, in particular for studies without a rigorous antifraud screener process.