A study co-authored by Dr. Shu Xu, Clinical Assistant Professor of Biostatistics at the New York University School of Global Public Health, was published by Prevention Science titled “Using Security Questions to Link Participants in Longitudinal Data Collection.”
Anonymous data collection systems are often necessary when assessing sensitive behaviors, but they can pose challenges to researchers seeking to link participants over time. To assist researchers in anonymously linking participants, this study outlined and tested a novel four-step security question linking (security question linking; SEEK) method. The method is demonstrated in SAS with two samples from a longitudinal study of adolescent dating violence.
Results indicate that the SEEK method is a feasible and reliable approach for linking multiple responses of participants over a 3-month period via anonymous online data collection. It does this without personal identifiers and passwords and tolerates errors such as recall failure and typos. SEEK may be particularly useful in prevention trials, which often seek to measure intervention effects on socially/legally sensitive behaviors. Longitudinal cohort studies with anonymous data collection are often posed to answer some of the most important questions in our field, such as the impact of preventive interventions and the identification of factors that confer prospective risk for mental and physical health problems. Given the importance of the questions these studies address and the substantial investments society makes in them, data quality is paramount. Accordingly, we believe that the strong performance of SEEK justifies the added effort it requires.Tags: Friday Letter Submission, Publish on February 14