 Reproducibility. We all know it's the backbone of good science, but all too often we don't have enough time or funding to carry out studies that aren't focused on discovering something new. More troubling still is that when we do try to repeat things, we often can't reproduce the results. After several high-profile failures to reproduce results in fields such as biomedical science and psychology, many scientists now think we're in the middle of a reproducibility crisis that threatens to undermine the public's faith in science. Either way, fixing the problems surrounding reproducibility are key to improving research and gaining back public trust. But where to start? It's not always obvious what reproducibility even means. There are a lot of definitions out there, and it can be confusing. To clarify matters and bring you up to speed on this critical topic is a new book from RTI International called Reproducibility, a primer on semantics and implications for research. It offers readers a comprehensive set of definitions related to reproducibility and a guide to thinking about these ideas. Take, for example, the distinction between reproducibility and replicability. Replication is an investigator performing an experiment or study again in exactly the same way, using the same reagents or subjects and getting the same result. Reproducibility, however, goes a step further and varies at least one component, such as the test subjects, measurement tools, and who is doing the work. Understanding the difference is important because a reproduced finding is generally considered more reliable than one that is just replicated. Reproduced results show that the conclusion probably cannot be attributed to chance or be the result of some one-off aspect of a particular study. The book covers the challenges of reproducibility and reviews related concepts of repeatability, accuracy, validity, and transparency, including best practices for accurate and full documentation. Armed with a shared set of definitions and a clear conceptual framework, you'll be able to communicate results more effectively within your field and across disciplines and to make sure your research is the best it can be.