Retracted papers clinging on to life: An observational study of post-retraction citations in psychology

Self-correction is assumed to be a defining feature of science. However, science’s ability to correct itself is far from optimal as shown, for instance, by the persistent influence of papers that have been retracted. In this study, we investigated citation patterns for 140 papers retracted in psychology due to data fabrication, scientific misconduct, or error, according to the Retraction Watch Database. After retraction, 88 (63%) of these papers received at least one positive citation (median = 2, interquartile range = 4, min = 0, max = 89 positive citations). These results demonstrate the enduring influence of erroneous or flawed data, even when they have been formally declared as invalid. To ameliorate this problem, we propose several procedures and tools that may enhance the discoverability of retracted papers, including standardization of publisher retraction notices and meta-data, issuance of retraction notices within researchers’ reference management software, and automated pre-publication screening for citations to retracted papers.