Mitigating Fraud in Online Surveys: A Methodological Approach

Area of Science:

• Data Science
• Research Methodology
• Survey Design

Background:

• Online surveys offer efficient data collection but face challenges like fraudulent responses, particularly when using crowdsourcing platforms.
• Platforms like Research Match and Prolific present unique recruitment benefits and pitfalls.

Purpose of the Study:

• To describe strategies for enhancing an online REDCap survey to identify and mitigate fraudulent and suspicious responses.
• To present a methodology for fortifying online data collection against data integrity threats.

Main Methods:

• A two-pronged approach was employed, focusing on initial design limitations and evidence-based redesign.
• Strategies included implementing scam alert features, modifying study design, improving survey structure, and considering crowdsourcing platform specifics.

Main Results:

• An eight-step data cleaning protocol was developed and implemented in collaboration with institutional review board and REDCap experts.
• Multiple data cleaning steps identified the same fraudulent records, indicating that fraudulent data often exhibits several suspicious indicators.

Conclusions:

• Online data collection presents opportunities alongside risks of compromised data quality and wasted resources.
• Proactive, evidence-based protocols are essential for deterring and detecting fraudulent results early to ensure data integrity.