Background: Online survey-based research is common in nursing, but potentially fraudulent responses (e.g., bots or bad actors) threaten the validity of the data obtained from these studies. Objectives: The purpose of this methods paper was to quantify and describe the extent to which nursing investigators have protected against fraudulent submissions in online, survey-based research studies. Methods: A random sample of articles published in nursing journals and reporting the results of online, survey-based research was obtained from PubMed in June 2024. Each article was audited using a standardized audit matrix. Results: The included articles ( n = 132) published in 51 unique journals and involved 56,159 participants. Studies were primarily cross-sectional and conducted worldwide, with most originating in the United States, multiple nations, or China. Investigators mentioned screening for response validity in only 21 articles; fewer explicitly described the screening processes used. Screening strategies included reviewing open-ended responses, checking for response-set biases, reviewing completion times, using computer-assisted tools, and examining responses for implausible values. In all cases where a potentially fraudulent response was identified, investigators excluded the response from analysis. There was no significant difference in frequency of fraud screening across journal impact factor tertiles. Discussion: Fraudulent responses are an ever-present problem in survey research. The articles examined did not routinely report strategies for detecting potentially fraudulent responses or protecting data quality. Published online, survey-based studies that include methods for detecting fraudulent responses enhance reader confidence. Investigators are encouraged to develop an a priori data analysis plan that includes multiple means to identify and eliminate, or otherwise, process fraudulent responses. We suggest that investigators transparently detail the use of a standard checklist for online survey research, in addition to the Fraud detection strategies, Recruitment, Incentive, Excluded responses, Data collection (FRIED) checklist we propose in this article.
Blakeman et al. (Fri,) studied this question.