The focus of discussion this week was on elements of the open science process to consider at the beginning of the research process. We started with a brief discussion of how the benefits of open and high powered research outweigh costs (Lebel, Campbell, & Loving, 2017). I then provided a tutorial on how to use the Open Science Framework (OSF) to manage the research workflow. Here is a link to a useful webinar. When I talk to people that are not familiar with using the OSF they are confused regarding the differences between “projects” and “registrations”, and think the OSF is primarily used to “pre-register” study hypotheses. I share my belief that whereas the OSF is useful for pre-registrations, it is even more useful as a free web platform for managing the research process with collaborators. In fact, a researcher could choose to keep all project pages private (i.e., not share with the public) and never pre-register anything at all, but use the OSF to efficiently manage his or her research workflow. That seems unlikely to happen, but the point is that projects on the OSF are dynamic with a great deal of functionality over the course of the research process. In our lab we use the OSF to store research materials (scales, procedures, data files, code, ethics applications, and so on), but also more and more to document communications between collaborators during the research process. And that is where open notebook comes in.
Open notebook is “…the practice of making the entire primary record of a research project publically available online as it is recorded.” At first blush this does not seem all that different from basic “open science” principles whereby the researcher publicly shares research materials, specifies hypotheses in advance, outlines a data analytic plan, and then later on shares data, meta-data and code (all topics to be covered in future classes). But the concept of open notebook also includes sharing the laboratory notebook, something that often contains communications between collaborators, as well as “dear diary” types of entries, that shed light on the decision making process throughout the research process. An open notebook can take on many different forms, and there are some excellent examples from the medical bio-sciences here, here, and here. These three examples take the form of dedicated web pages with regular updates, and I admire their commitment to “extreme open science”.
Rather than create a dedicated website for our lab notebooks, I wanted to develop an approach that uses what our lab already uses on a daily basis—the OSF. That is where all of our research materials are stored for our projects already, but what has been missing to date is the nature of the communications between colleagues during the research process that results in particular decisions being made. And decisions need to be made regularly as new issues arise that were not considered earlier.
Along with current graduate student Nicolyn Charlot, we are trying out the following open notebook approach. First, rather than using email for basic communications for a current project (this one), we decided to make use of the “comments” function that is available for every project page (the “word bubble” icon located at the top right of the screen). That way our messages are documented over time as they occur and are embedded with all of our research materials. With notifications for comments set to “on” (under the settings tab), we receive emails when a new comment has been added. Second, because many of our decisions are made during lab meetings and not via email, we decided to briefly document the decision making process following each lab meeting in a shared google doc file that is linked to a component nested within the main project page titled “Open Notebook”. Here is the open notebook for our project. It is within this component where we communicate using the comments function (click on it and see what we have so far), and where we keep the shared file. Our goal will be to have an “Open Notebook” component for all new projects going forward as a way to document the decision making process, and to add a more personal element to the project pages beyond simply being a repository of files.
I am curious to see how it works out.