The first six years of meta-research at PLOS Biology

Meta-research involves the interrogation of every stage of the research lifecycle, from conception to publication and dissemination. Looking back over the first six years of PLOS Biology Meta-Research Articles highlights the important insights that can be obtained from such “research on research”.

and career stage can influence collaboration patterns [2], and how long it will be until men and women are equally represented in various fields [3].
2. The choice of research topic. Although all of biology is open for investigation, the topics selected for research are clearly non-random. What influences which literature, organisms, tissues, cells or genes are targeted for study? One study asked what determines the research effort invested in each of the~19,000 human genes [4]; another quantified the importance of including non-English language literature in biodiversity surveys [5].
3. The design and methodology of research studies. Once any given topic has been selected, the research and subsequent analysis can be performed in myriad different ways, many of which can seriously impact the strength of support for the conclusions. Studies of these choices have included general issues such as the benefits of sample heterogeneity in preclinical work [6] and the effects of statistical model choice on association studies [7], but also the conduct of specific techniques such as functional MRI and RNA sequencing.

The description and interpretation of results.
The distillation of the complex collection of actions that constitute a research project must somehow be compiled into the stereotyped format of a research manuscript. Meta-researchers have systematically examined the role of "spin" in the presentation of scientific results [8] and the way in which microscopy images are prepared for publication [9].

The publication and dissemination of results.
The impact of the study on other workers in the field, on the general public, and on the body of scientific knowledge depends on a set of events that start with a completed manuscript file and pass through posting a preprint, submission to a journal, peer review, revision, publication, press coverage, social media and indexing. Many aspects of this process have been scrutinised, including the misuse of scientific preprints by right-wing social media groups [10] and the effects of article titles on subsequent press coverage [11].
6. The assessment of researchers and funding decisions. These published outputs of scientific research, for good or ill, are used by numerous agents to judge the performance of specific researchers and their institutes and to apportion funds accordingly. In turn, these decisions shape the research community of the future. We've published papers that address the way that the influence of individual papers can be measured [12] and that propose novel ways of distributing research funding [13].
Over the years, as editors, we have had to decide what does or does not count as "metaresearch". Our understanding of this concept has been shaped by the sheer diversity of manuscripts that have come to us and we anticipate that the next six years will take meta-research in increasingly varied directions. Only by auditing and experimenting with the way we approach discovery can we hope to address some of the issues that pervade research culture today and that threaten to undermine public trust in scientific findings. The launches of initiatives such as the RoRI (Research on Research Institute), the UKRN (UK Reproducibility Network), METRICS (Meta-Research Innovation Center at Stanford) and QUEST (Quality, Ethics, Open Science, Translation), among others, are testament to the growing recognition that science must actively strive to keep its own house in order. The past six years have emphasized the growth and importance of meta-research; "research on research" is clearly here to stay and we look forward to seeing it happen.