作者: Jennifer J. Ware , Marcus R. Munafò
DOI: 10.1111/ADD.12673
关键词: Research design 、 False positive paradox 、 Public economics 、 Publication 、 Publish or perish 、 Novelty 、 Scientific literature 、 Psychology 、 Social psychology 、 Research question 、 Incentive 、 Medicine (miscellaneous) 、 Psychiatry and Mental health
摘要: Background and Aims The low reproducibility of findings within the scientific literature is a growing concern. This may be due to many being false positives which, in turn, can misdirect research effort waste money. Methods We review factors that contribute poor study an excess ‘significant’ published literature. Specifically, we consider influence current incentive structures impact these on practices. Results The prevalence attributable number questionable practices, ranging from relatively innocent minor (e.g. unplanned post-hoc tests) calculated serious fabrication data). These practices driven by pressure publish), alongside preferential emphasis placed journals novelty over veracity. There are potential solutions reproducibility, such as new publishing formats emphasize question design, rather than results obtained. has minimize significance chasing non-publication null findings. Conclusions Significance chasing, unfortunate consequence ‘publish or perish’ culture preference among for novel findings. It likely top–down change implemented those with ability modify structure funders journals) will required address problems reproducibility.