- Understand the motivations for open sharing
- Understand important considerations before making data public
- Understand how to make research findable
- Understand how to get credit for your work
Now that we’ve created our project, uploaded all our materials, documented our entire process, we’re at the point where we need to decide if we want share our work, and if so what parts of it do we want to share? There are many different reasons why you might want to share your work. Many funder and some journals now require that data and code be shared upon publication or the conclusion of a grant. Additionally there is evidence to suggest that papers with open data get cited more than papers that do not. Finally, sharing materials and data makes it easier for others to try and reproduce your work, and to build off your work. So, there are both internal and external reasons that may motivate sharing some or all of your research products.
As we talked about before, everything on the OSF is private by default. You can make either the entire project, or specific subproject or components of the project public at any point you want by clicking on the public button of each project/component you wish to make public. Now that we’re finished with our study, we need to decide how much we want to make public. One important thing to consider before making anything public is whether it contains anything that you can’t make public for legal or ethical reasons. That might include materials that you used in your project that are perhaps copyrighted, or data-files that are not anonymous and so your IRB has told you that you can only make public the anonymous dataset. So, we need to make sure we have organized our project in such a way that it will allow us to make public only the parts we want, while keeping some parts private if that is necessary.
If you remember, I mentioned that our datafile had geotags, so we can’t actually make the raw_data file public. We’ve already create a de-anonymized, ‘cleaned’ data file, which we can make public, but right now I have both those files in the same ‘data’ component, so I can’t make one public and keep the other private.
Activity (Question): What could I do so that I could share my clean data file but not reveal my raw data file with the geolocation variable in it?
One thing I could do would be to create another component under Data and move my raw data file to that component. Any file can be moved by simply clicking and dragging it to whichever part of the project you want to move it to.
Activity: Now that I’ve shown you how to move files, discuss how much of your project you can/want to make public. Make those sections of your project public, and if you need to reorganize any parts of your project to accomplish that, please do that now.
![shared project]
I already mentioned that all projects, subprojects/components, and files on the OSF have GUIDs. So, if I want to send people to my project I can send them that GUID, or put it in my paper to help people navigate to my materials. However, we also want to make sure that our materials can be found by people who don’t already know it exists. We want to make sure it is discoverable.
Work posted on the OSF can be discovered in a couple of different ways. The first is through the search function within the OSF itself. We can search for anything we want, a project title, a general topic, a person’s name, and the system will bring up anything related to our search.
You’ll see it will also make suggestions about ways to potentially refine your search.
Public project on the OSF are also indexed by google, so they can be discovered outside the system. So, for example, if we go to google and search for ‘Daniel Lakens effect size’ his OSF project related to effect sizes comes up.
Finally, information that has been registered on the OSF is discoverable through the SHARE notification service. If we go to ‘osf.io/share’, this will allow us to search through the contents of over 50 different providers. For SHARE specifically, only public registrations show up in the search results, which is another reason to register projects when you want to release a stable version of them.
![share image]
Part of making our work discoverable is making sure the right people find it through searches, so we want to make sure we understand how are work is being searched. Currently, the search on the OSF will search through contents in the wiki, project/component names, and tags. But it won’t search through individual files. So, if you think there are a few important key words that people might use to search for your project that don’t appear in the wiki or project name, where do you put this information? We can add tags to our project to make it more discoverable. You’ll notice this ‘tags’ area to the right of the screen. I can put whatever descriptors or keywords in here that I want to make it easier to discover my project for searching for related words/topics.
Activity: And add some tags that you think might be relevant to your project.
So a big part of academia and the job market is being able to show that your work has impact. The classic way many fields often talk about this is citation counts. Now, we know this isn’t a great metric, particularly for newer papers, but since there is still a heavy emphasis places on it, we want to make sure that whatever we make open can be cited by other people who used it. We talked before about how you can get DOIs for your work through public registrations. Every project and component on the OSF also has an automatically generated citation which uses the GUID of the page, so as long as you have the correct contributor list on your project, everyone can get credit for it.
[citation widget]
But, since citation counts have limitations, what else could we use to measure the impact of our work? Well, two things we might be interested in are views and downloads. For all public projects we track how many people are visiting a page, and we track how often all files are downloaded. Now, your projects aren’t particularly old, so they probably won’t have very many views. But, if we search for ‘Daniel Lakens’ and click on this project, we can see how views and downloads can give vastly different information than citation counts. So, this is an OSF project associated with an article published in Frontiers, which is an open access journal. The paper and supplemental materials were also made publicly available on the OSF. Now, last time I check, this paper has had a bit under 100 citations, which isn’t bad, but it’s not a huge amount of impact. If we click on the files tab, that will allow us to see the download counts for all the files. You can see that the paper has been downloaded around 1500 times, which is a ton.
[download counts]
Additionally, if we go to the statistics tab, we can see how many people are viewing the cite and where in the world they are viewing it from. So, these view and download numbers are telling a much more impressive story than the citations; a lot of people are interested this project because they find the materials useful.
[statistics page view]
- Recap of general things they’ve learned