As part of a broader effort to improve reporting quality, Nature and the Nature journals introduced a reporting checklist for life-sciences papers in 2013. This asked authors to reveal some key details of experimental design. Last year, this checklist evolved into a broader reporting-summary document that is published alongside manuscripts to promote greater transparency.
We have now developed two new versions of the reporting summary: one for the behavioural and social sciences, launching this week, and one for ecology, evolution and environment (EEE) research, to follow later this month. Authors will be prompted to use these documents to provide important details of study design, data collection and analysis before papers are sent out for review.
In-house editors in behavioural and social sciences across the Nature journals developed the first document to address the distinct needs of research in this field — one that is remarkably broad, and includes numerous disciplines with distinct identities. Even in the same area, research protocols can vary substantially, ranging from qualitative and interpretative methods to deductive and quantitative approaches.
This presents some challenges when deciding on priorities. Most experimental approaches value sample size, for instance. But survey-based projects must also consider whether data are collected from an appropriately representative sample. Data collection methods may be relatively easy to standardize in a laboratory, but they can vary in fieldwork when scientists use a wide array of tools, or when language or literacy barriers must be overcome.
Despite these variations, our editors think it is valuable to consider all behavioural- and social-sciences research together to try to bridge the methodological divides between and within fields. We hope that describing study elements in a standardized way across the full suite of social-sciences methodologies and data types will help our multidisciplinary readers to appreciate and understand diverse research approaches.
The reporting-summary document for the behavioural and social sciences was developed on the basis of feedback from researchers with different disciplinary backgrounds and methodological expertise, including quantitative and qualitative analysis, and lab-based and field studies. It aims to capture key elements of how studies were designed, conducted and analysed — but it does not seek to enforce a specific set of standards.
For instance, determining sample size statistically using a power analysis might be best practice in experimental psychology, but it is not a necessary or sensible step in an anthropological study of a small village, which may have low sample size but effectively encompasses the entire available population. However, in both cases, authors should be able to provide a full report of how sample size was selected. Accordingly, the reporting summary is designed to be flexible.
Similar considerations motivated the design of the reporting-summary document for EEE studies. The Nature journals are among thousands of publications and organizations that have signed up to the Transparency and Openness Promotion (TOP) guidelines (B. A. Nosek et al. Science 348, 1422–1425; 2015). The EEE reporting summary is being designed by in-house editors using the Tools for Transparency in Ecology and Evolution as a guide.
EEE studies include many distinctive features, and the reporting summary is being designed with those in mind. For example, there will be questions about fieldwork conditions and the treatment of wild animals, and authors of palaeontological work will be asked to describe specimen provenance, deposition and dating methods. We are currently integrating feedback from researchers into a final version.
The reporting-summary documents are a first step towards ensuring that the relevant communities pay systematic attention to reporting and transparency. These documents are not static, and the first iterations are intentionally broad. We look forward to receiving comments and thoughts.
Nature 555, 6 (2018)