PDF(Pubmed)
Abstract：
Changes in statistical practices and reporting have been documented by Giofrè et al. PLOS ONE 12(4), e0175583 (2017), who investigated ten statistical and open practices in two high-ranking journals (Psychological Science [PS] and Journal of Experimental Psychology-General [JEPG]): null hypothesis significance testing; confidence or credible intervals; meta-analysis of the results of multiple experiments; confidence interval interpretation; effect size interpretation; sample size determination; data exclusion; data availability; materials availability; and preregistered design and analysis plan. The investigation was based on an analysis of all papers published in these journals between 2013 and 2015. The aim of the present study was to follow up changes in both PS and JEPG in subsequent years, from 2016 to 2020, adding code availability as a further open practice. We found improvement in most practices, with some exceptions (i.e., confidence interval interpretation and meta-analysis). Despite these positive changes, our results indicate a need for further improvements in statistical practices and adoption of open practices.
摘要：
Giofrè等人记录了统计做法和报告的变化。PLOSONE12(4),e0175583(2017),他在两个高级期刊（心理科学[PS]和实验心理学杂志[JEPG]）中调查了十种统计和开放实践：零假设显著性检验；置信区间或可信区间；多个实验结果的荟萃分析；置信区间解释；效应大小解释；样本大小确定；数据排除；数据可用性；材料可用性；以及预先注册的设计和分析计划。这项调查是基于对2013年至2015年间在这些期刊上发表的所有论文的分析。本研究的目的是跟踪PS和JEPG在随后几年的变化，从2016年到2020年，增加代码可用性作为进一步的开放实践。我们发现大多数实践都有所改善，除了一些例外(即，置信区间解释和荟萃分析)。尽管有这些积极的变化,我们的结果表明需要进一步改进统计实践和采用开放实践.
