Implementing Reproducible Research

In computational science, reproducibility requires that researchers make code and data available to others so that the data can be analyzed in a similar manner as in the original publication. Code must be available to be distributed, data must be accessible in a readable format, and a platform must be available for widely distributing the data and code. In addition, both data and code need to be licensed permissively enough so that others can reproduce the work without a substantial legal burden. Implementing Reproducible Research covers many of the elements necessary for conducting and distributing reproducible research. It explains how to accurately reproduce a scientific result. Divided into three parts, the book discusses the tools, practices, and dissemination platforms for ensuring reproducibility in computational science. It describes: Computational tools, such as Sweave, knitr, VisTrails, Sumatra, CDE, and the Declaratron system Open source practices, good programming practices, trends in open science, and the role of cloud computing in reproducible research Software and methodological platforms, including open source software packages, RunMyCode platform, and open access journals Each part presents contributions from leaders who have developed software and other products that have advanced the field. Supplementary material is available at www.ImplementingRR.org.

In computational science, reproducibility requires that researchers make code and data available to others so that the data can be analyzed in a similar manner as in the original publication. Code must be available to be distributed, data must be accessible in a readable format, and a platform must be available for widely distributing the data and code. In addition, both data and code need to be licensed permissively enough so that others can reproduce the work without a substantial legal burden. Implementing Reproducible Research covers many of the elements necessary for conducting and distributing reproducible research. It explains how to accurately reproduce a scientific result. Divided into three parts, the book discusses the tools, practices, and dissemination platforms for ensuring reproducibility in computational science. It describes: Computational tools, such as Sweave, knitr, VisTrails, Sumatra, CDE, and the Declaratron system Open source practices, good programming practices, trends in open science, and the role of cloud computing in reproducible research Software and methodological platforms, including open source software packages, RunMyCode platform, and open access journals Each part presents contributions from leaders who have developed software and other products that have advanced the field. Supplementary material is available at www.ImplementingRR.org.

Praise for previous editions: "Gandrud has written a great outline of how a fully reproducible research project should look from start to finish, with brief explanations of each tool that he uses along the way... Advanced undergraduate students in mathematics, statistics, and similar fields as well as students just beginning their graduate studies would benefit the most from reading this book. Many more experienced R users or second-year graduate students might find themselves thinking, ‘I wish I’d read this book at the start of my studies, when I was first learning R!’...This book could be used as the main text for a class on reproducible research ..." (The American Statistician) Reproducible Research with R and R Studio, Third Edition brings together the skills and tools needed for doing and presenting computational research. Using straightforward examples, the book takes you through an entire reproducible research workflow. This practical workflow enables you to gather and analyze data as well as dynamically present results in print and on the web. Supplementary materials and example are available on the author’s website. New to the Third Edition Updated package recommendations, examples, URLs, and removed technologies no longer in regular use. More advanced R Markdown (and less LaTeX) in discussions of markup languages and examples. Stronger focus on reproducible working directory tools. Updated discussion of cloud storage services and persistent reproducible material citation. Added discussion of Jupyter notebooks and reproducible practices in industry. Examples of data manipulation with Tidyverse tibbles (in addition to standard data frames) and pivot_longer() and pivot_wider() functions for pivoting data. Features Incorporates the most important advances that have been developed since the editions were published Describes a complete reproducible research workflow, from data gathering to the presentation of results Shows how to automatically generate tables and figures using R Includes instructions on formatting a presentation document via markup languages Discusses cloud storage and versioning services, particularly Github Explains how to use Unix-like shell programs for working with large research projects

The Practice of Reproducible Research presents concrete examples of how researchers in the data-intensive sciences are working to improve the reproducibility of their research projects. In each of the thirty-one case studies in this volume, the author or team describes the workflow that they used to complete a real-world research project. Authors highlight how they utilized particular tools, ideas, and practices to support reproducibility, emphasizing the very practical how, rather than the why or what, of conducting reproducible research. Part 1 provides an accessible introduction to reproducible research, a basic reproducible research project template, and a synthesis of lessons learned from across the thirty-one case studies. Parts 2 and 3 focus on the case studies themselves. The Practice of Reproducible Research is an invaluable resource for students and researchers who wish to better understand the practice of data-intensive sciences and learn how to make their own research more reproducible.

One of the pathways by which the scientific community confirms the validity of a new scientific discovery is by repeating the research that produced it. When a scientific effort fails to independently confirm the computations or results of a previous study, some fear that it may be a symptom of a lack of rigor in science, while others argue that such an observed inconsistency can be an important precursor to new discovery. Concerns about reproducibility and replicability have been expressed in both scientific and popular media. As these concerns came to light, Congress requested that the National Academies of Sciences, Engineering, and Medicine conduct a study to assess the extent of issues related to reproducibility and replicability and to offer recommendations for improving rigor and transparency in scientific research. Reproducibility and Replicability in Science defines reproducibility and replicability and examines the factors that may lead to non-reproducibility and non-replicability in research. Unlike the typical expectation of reproducibility between two computations, expectations about replicability are more nuanced, and in some cases a lack of replicability can aid the process of scientific discovery. This report provides recommendations to researchers, academic institutions, journals, and funders on steps they can take to improve reproducibility and replicability in science.

In computational science, reproducibility requires that researchers make code and data available to others so that the data can be analyzed in a similar manner as in the original publication. Code must be available to be distributed, data must be accessible in a readable format, and a platform must be available for widely distributing the data and code. In addition, both data and code need to be licensed permissively enough so that others can reproduce the work without a substantial legal burden. Implementing Reproducible Research covers many of the elements necessary for conducting and distributing reproducible research. It explains how to accurately reproduce a scientific result. Divided into three parts, the book discusses the tools, practices, and dissemination platforms for ensuring reproducibility in computational science. It describes: Computational tools, such as Sweave, knitr, VisTrails, Sumatra, CDE, and the Declaratron system Open source practices, good programming practices, trends in open science, and the role of cloud computing in reproducible research Software and methodological platforms, including open source software packages, RunMyCode platform, and open access journals Each part presents contributions from leaders who have developed software and other products that have advanced the field. Supplementary material is available at www.ImplementingRR.org.

Development Research in Practice leads the reader through a complete empirical research project, providing links to continuously updated resources on the DIME Wiki as well as illustrative examples from the Demand for Safe Spaces study. The handbook is intended to train users of development data how to handle data effectively, efficiently, and ethically. “In the DIME Analytics Data Handbook, the DIME team has produced an extraordinary public good: a detailed, comprehensive, yet easy-to-read manual for how to manage a data-oriented research project from beginning to end. It offers everything from big-picture guidance on the determinants of high-quality empirical research, to specific practical guidance on how to implement specific workflows—and includes computer code! I think it will prove durably useful to a broad range of researchers in international development and beyond, and I learned new practices that I plan on adopting in my own research group.†? —Marshall Burke, Associate Professor, Department of Earth System Science, and Deputy Director, Center on Food Security and the Environment, Stanford University “Data are the essential ingredient in any research or evaluation project, yet there has been too little attention to standardized practices to ensure high-quality data collection, handling, documentation, and exchange. Development Research in Practice: The DIME Analytics Data Handbook seeks to fill that gap with practical guidance and tools, grounded in ethics and efficiency, for data management at every stage in a research project. This excellent resource sets a new standard for the field and is an essential reference for all empirical researchers.†? —Ruth E. Levine, PhD, CEO, IDinsight “Development Research in Practice: The DIME Analytics Data Handbook is an important resource and a must-read for all development economists, empirical social scientists, and public policy analysts. Based on decades of pioneering work at the World Bank on data collection, measurement, and analysis, the handbook provides valuable tools to allow research teams to more efficiently and transparently manage their work flows—yielding more credible analytical conclusions as a result.†? —Edward Miguel, Oxfam Professor in Environmental and Resource Economics and Faculty Director of the Center for Effective Global Action, University of California, Berkeley “The DIME Analytics Data Handbook is a must-read for any data-driven researcher looking to create credible research outcomes and policy advice. By meticulously describing detailed steps, from project planning via ethical and responsible code and data practices to the publication of research papers and associated replication packages, the DIME handbook makes the complexities of transparent and credible research easier.†? —Lars Vilhuber, Data Editor, American Economic Association, and Executive Director, Labor Dynamics Institute, Cornell University

2017 PROSE Award Honorable Mention The PROSE Awards draw attention to pioneering works of research and for contributions to the conception, production, and design of landmark works in their fields. Featuring peer-reviewed contributions from noted experts in their fields of research, Reproducibility: Principles, Problems, Practices, and Prospects presents state-of-the-art approaches to reproducibility, the gold standard of sound science, from multi- and interdisciplinary perspectives. Including comprehensive coverage for implementing and reflecting the norm of reproducibility in various pertinent fields of research, the book focuses on how the reproducibility of results is applied, how it may be limited, and how such limitations can be understood or even controlled in the natural sciences, computational sciences, life sciences, social sciences, and studies of science and technology. The book presents many chapters devoted to a variety of methods and techniques, as well as their epistemic and ontological underpinnings, which have been developed to safeguard reproducible research and curtail deficits and failures. The book also investigates the political, historical, and social practices that underlie reproducible research in contemporary science studies, including the difficulties of good scientific practice and the ethos of reproducibility in modern innovation societies. Reproducibility: Principles, Problems, Practices, and Prospects is a guide for researchers who are interested in the general and overarching questions behind the concept of reproducibility; for active scientists who are confronted with practical reproducibility problems in their everyday work; and for economic stakeholders and political decision makers who need to better understand the challenges of reproducibility. In addition, the book is a useful in-depth primer for undergraduate and graduate-level courses in scientific methodology and basic issues in the philosophy and sociology of science from a modern perspective. “A comprehensive, insightful treatment of the reproducibility challenges facing science today and of ways in which the scientific community can address them.” Kathleen Hall Jamieson, Elizabeth Ware Packard Professor of Communication, University of Pennsylvania “How can we make sure that reproducible research remains a key imperative of scientific communication under increasing commercialization, media attention, and publication pressure? This handbook offers the first interdisciplinary and fundamental treatment of this important question.”Torsten Hothorn, Professor of Biostatistics, University of Zurich Harald Atmanspacher, PhD, is Associate Fellow and staff member at Collegium Helveticum, ETH and University Zurich and is also President of the Society for Mind-Matter Research. He has pioneered advances in complex dynamical systems research and in a number of topics concerned with the relation between the mental and physical. Sabine Maasen, PhD, is Professor for Sociology of Science and Director of the Munich Center for Technology in Society (TU Munich) and Associate Fellow at Collegium Helveticum (ETH and University Zurich). Her research focuses on the interface of science, technology, and society, notably with respect to neuroscience and its applications.