PUBLICATION BIAS - THE PROBLEM THAT WONT GO AWAY
PUBLICATION BIAS - THE PROBLEM THAT WONT GO AWAY
复制标题
DOI：
10.1111/j.1749-6632.1993.tb26343.x
复制
发表时间：
1993-01-01
期刊：
影响因子：
--
通讯作者：
MIN, YI
中科院分区：
文献类型：
--
作者：
DICKERSIN, K;MIN, YI
发表偏倚是指研究者、期刊审稿人和编辑倾向于发表拒绝零假设的研究结果，这一倾向越来越受到科学界的关注。如果存在发表偏倚，那么现有信息的质量就会受到影响，从而影响未来的研究和现有知识的应用（例如，病人护理）。发表偏倚研究关注的是是否存在发表偏倚，如果存在问题，问题的大小以及与之相关的风险因素。这些问题并不容易解决，因为传统的流行病学设计（病例对照，队列，横断面，实验）很难应用。在这种情况下，研究受试者是启动的研究，而不是人，关注的结果是出版物，而不是发病率或死亡率。
Publication bias, the tendency on the part of investigators, journal referees, and editors to favor publication of results of studies rejecting the null hypothesis, has been of increasing concern to the scientific community. If publication bias exists, the quality of information available is compromised, influencing both future research and application of available knowledge (eg, patient care).Research on publication bias is concerned with whether it exists, the size of the problem if there is one, and risk factors with which it is associated. The problems are not simple to address as the traditional epidemiologic designs (case-control, cohort, cross-sectional, experimental) are difficult to apply. In this case, the study subjects are initiated studies, not persons, and the outcome of interest is publication, not morbidity or mortality.