Data management and accountability in behavioral and biomedical research

Abstract

Regulations recently enacted by the Public Health Service and the National Science Foundation to address misconduct in scientific research were designed primarily to curtail deliberate forms of misconduct, such as fabrication or falsification of findings; however, researchers may also be held accountable for inadvertent deficiencies in data management. This article examines some of the problems in data quality control, documentation, and data retention that can occur when computers are used in scientific research. It focuses on deficiencies that could make it difficult to verify the integrity of research data or to reproduce statistical analyses. Strategies for prevention of data management problems are recommended.