12:20:30 We're here today to talk with you about a project that was a collaboration between the University of Michigan library, and the National Center for Institutional Diversity on the University of Michigan campus.
12:20:39 My name is Rachel with Brooke, I'm the data curation librarian at the University of Michigan library, and I was the PI on this project.
12:20:48 My name is Amanda Vera and I was a graduate student when I was working on this project as a researcher and I now work for the Friends of the public like.
12:20:58 Hi, My name is Megan's ago she I'm a former staff member of the National Center for Institutional Diversity at the University of Michigan and current manager for faculty diversity initiatives at Boston University.
12:21:10 If you'd like to follow along with us our slides are also available online at this tiny URL backslash Open Data da, and we'll try to add this URL to the chat, also.
12:21:25 Today we're going to talk a little bit about our research, very briefly about some of our results from that. The toolkit that we're working on building, and also a little bit about the partnership between the library and see ID, and potential applications
12:21:36 of the toolkit.
12:21:39 So in terms of the origin of this project, overall, while funders and journals and some cases disciplines are increasing their expectations for effective data management, and ultimately data sharing the prerequisites for doing this appropriately I really
12:21:53 complex, and many researchers still aren't getting formal or effective training or as much training as they need. This has, especially problematic implications for researchers doing diversity scholarship, which we'll talk about also, which often focuses
12:22:07 on and affects underrepresented or vulnerable populations.
12:22:12 This project was supportive generously by a 2019 catalyst fund from lyricists, and the assessment and revision of the toolkit is also supported by 2020 a diversity research grant.
12:22:24 And as we mentioned this is a collaboration and it came out of an existing partnership between the University of Michigan library, and the National Center for Institutional Diversity, which Megan will introduce a little bit more in a moment.
12:22:35 And the goal is to better understand the types of support scholars still need to enable data practices that help them stay in alignment with diversity, equity inclusion and accessibility considerations.
12:22:47 And it's also shaped our choice of sample population for the network to Megan, talk about as well.
12:22:52 We scope this for a particular sample but we do think that the resources we identified, especially for the toolkit would be useful for other scholars as well.
12:23:02 And finally for our research design, we use the mixed methods design scoping review interviews, I'm going to survey.
12:23:09 The purpose of the study again was to identify perceived gaps and effective practices that diversity researchers were already. We're already using that we might be able to share to propose resources that could help improve satisfaction and confidence
12:23:24 and making data decisions throughout a project, and then to prioritize the types of toolkit resources that will be most useful for diversity scholars.
12:23:34 There's more detail available about our process, and our findings in the report will link to at the end here and some manuscripts we're working on.
12:23:53 But basically our overarching research question was, what implications do diversity, equity inclusion and accessibility considerations have for best practices in each stage of the data life cycle. So da, and then the data life cycle where the main framing ideas for the project.
12:23:59 Although how we talked about these over the course of the project shifted as well.
12:24:03 And this is just a diagram of the data life cycle image that we use during our project.
12:24:11 And we use this to represent how we understand and thinking about the research data life cycle and then we'll break it down a little bit more later. But basically, the first two circles here starting at the top.
12:24:23 Roughly mapped to before a project starts the next three are during a project and the last three are taking place after a project is complete, and they just represent the different stages and different types of actions people might take with their data
12:24:37 throughout the project
12:24:42 as Rachel mentioned this project really considered what diversity scholarship is and how we can form resources to support diversity scholarship throughout the data life cycle.
12:24:52 The framework we used to define diversity scholarship comes from the National Center for Institutional Diversity or NC ID, which is also at University of Michigan NCI D defines diversity scholarship as research or scholarship that broadly seeks to inform
12:25:08 our understanding of historical and contemporary issues of social inequality illuminate the challenges and opportunities that arise when individuals from different backgrounds and frames of reference come together, inform our understanding of systems
12:25:22 of power and privilege and their interactions with groups.
12:25:26 Highlight the experiences of disenfranchised populations and foreground knowledge systems assets and resources and cultural strength of members of historically marginalized communities.
12:25:37 This is kind of an abridged version of the framework which is actually quite expensive. So if you're interested in learning more about it you can go to LSA you mesh.edu slash MC ID or a more full definition.
12:25:54 So the National Center for Institutional Diversity supports programming and resources for scholars committed to diversity, and to diversify the Academy, and that can come in many forms so and sometimes it's through their research sometimes it's too they're
12:26:06 their teaching sometimes it's through their service, and CIZ had an existing working relationship with the library but this partnership was really born out of the recognition of the need to create resources and support for specifically diversity scholars.
12:26:23 We hope that one of the takeaways you will take from this presentation is that this partnership can serve as a model for future collaborations to address the real needs of certain populations.
12:26:37 Because it's not through this partnership MC ID and a library entity was able to provide the library with access to their sample population, and they were also able to draw on the expertise of NC IT staff who are all very well versed in higher education
12:26:55 research and diverse and di, so the sample population for this study were members of the diversity scholars network or DSM, the DSM is a scholarly community of over 900 diversity scholars across the country and some international diversity scholar is
12:27:15 somebody who engages with our framework for diversity scholarship, the DSM consists primarily of assistant professors that also includes other faculty and researchers affiliated with an institution.
12:27:30 And they represent a wide variety of disciplines and methodologies, they do not come from any one single field of study.
12:27:40 Finally, data members are supported by EMC ID.
12:27:45 Through its, providing of opportunities to build community and network to have their scholarship highlighted and promoted publicly by end to end, which also provide some programming and workshops and publication grant opportunities specifically targeting
12:28:10 scholars and members of the DSM. Now I'm just going to talk really briefly about some of our findings from the survey portion of the project. We send a survey to the diversity scholars network and its entirety, almost 900 people, and we got about 140
12:28:25 completed surveys that we were able to use back. And our first research question we were looking at for this portion look at what factors might correlate with diversity scholars perceived likelihood of using VDI specific data tool kit, and whether they
12:28:36 would be likely to use a toolkit at all. And what we found was that there were not any player demographic correlations, or correlations between demographics and how likely respond and thought they would be to use the tool kit but almost everyone said
12:28:49 that they would be likely to try and use a toolkit was confirmed the need for further research or support in this area. And the main barrier that they anticipated, unsurprisingly, was limitations on time and resources.
12:29:07 And that's not just time and resources to use the resources that are in the toolkit, but also to navigate the toolkit itself and identify the most useful resources.
12:29:15 The second research question we were investigating was where in the data lifecycle stages diversity scholars felt the least and the most comfortable with data life cycle actions.
12:29:27 So, what data stages were most challenging, or most uncomfortable or most comfortable, and we asked respondents to choose from the stages of that data life cycle that I showed before to choose one that was important and comfortable for them where they
12:29:41 were comfortable with the actions they would need to take in that portion of the project. And then one that was important, but that was uncomfortable.
12:29:57 The full table of data is at the end of these slides but one result that was especially interesting to note is the two stages that were most often marked as uncomfortable were data sharing and data archiving and preservation.
12:30:01 And this makes sense these come at the end of a project and they're greatly affected by how well it was planned for and managed throughout the project.
12:30:08 For example, if consent was collected in a way that allows the researcher to share the data as they want or need to, at the end of the project. And these are also areas that are often already targeted by libraries it and other centralized campus resources
12:30:22 as know needs are gaps.
12:30:25 Sometimes with departments.
12:30:26 So there's an ongoing effort already to support these stages on many campuses, but more support is still needed.
12:30:35 Finally, we also looked at and asked what types of resources diversity scholars would most like included in a toolkit. And we provided specific examples for each type listed here.
12:30:48 This table shows the number of respondents who stated that each kind would be most used would be useful to them, they could choose as many as they wanted, and we ranked it in descending order here.
12:30:59 And some of the specific examples we provided as well as some other resources are already available on our website, which will share at the end of the presentation.
12:31:08 At this point and is going to talk a little bit more about the toolkit itself.
12:31:15 So far with our toolkit we are currently looking at platforms for publishing our resources. We have curated about a dozen or so resources that apply to each stage of the research data life cycle.
12:31:27 We are particularly interested in platforms that are accessible and sustainable after we're done with the research.
12:31:36 And since accessibility and sustainability our diversity principles we aren't interested in losing those core principles in the presentation and regions of the resources.
12:31:58 So imagine that this tool tipping can be utilized at any phase of the research lifecycle. And we hope that I can call into question how justice issues are handled or not.
12:32:09 Yes, address during research practice.
12:32:13 We also hope these resources can help frame and model practices for future research inside and outside of academic setting. One important thing to note about our resources and many of them come from outside of academic research practice.
12:32:29 And these resources can help to inform those performing research work with vulnerable populations and preparing data for open access deposits. Such considerations might not be addressed and basic courses or preparation offered by repositories, or academic
12:32:42 institution. And as we know it's especially important when working with diverse groups because they are more likely to be exploited or raised or are raised, or even face backlash, in their own communities, or from outsiders.
12:32:55 If their data is reverse engineered.
12:32:59 Something else to notice that our context changed in the middle of our project timeline, one part due to coven 19, but also was nice by increasing momentum in the Black Lives Matter movement, as well as more publications that underlying the need for data
12:33:16 justice work including notable works like data feminism and the feminist a manifesto.
12:33:26 So Rachel mentioned the research data life cycle.
12:33:37 Earlier on but I'll touch on it a little bit more now.
12:33:33 The first stage consists of finding data or secondary research and data planning this leads to data collection data processing and data analysis or active data management.
12:33:41 And the final stages of the data research data life cycle researchers engage in data curation data sharing and data archiving creation.
12:33:50 Researchers might not engage in all of these stages but these are stages that they couldn't get out their research and even after.
12:33:58 And some of these stages might return back to the beginning and lead to more data finding or data planning, we use these different stages to create a framework for how our tool kit is structured and we also used it to structure some of our additional
12:34:11 research we did before, planning our toolkit, different resources are specifically tagged to the stages to facilitate guidance in specific areas. Some of our resources more broadly address the life cycle, but the structure corresponds with our scoping
12:34:22 review where we were able to name specific concern related to dia throughout the lifecycle.
12:34:34 And so some of our proposed resources are listed here. These resources are good exemplars of practices that we think need to be implemented throughout the research data life cycle or in specific stages.
12:34:47 One research resource we like in particular with the Chicago neon handbook for community research titled Why am I always been researched this guidebook lays the groundwork for specific questions researchers need to ask all data planning it touches on
12:35:00 creating relationships with research participants and agreements that needed to be discussed before research against to ensure equity throughout the process.
12:35:09 Some of these considerations include who will control the data during and after the project informed consent and how to build mutually beneficial research projects.
12:35:16 They also address approaching these conversations from the perspective of researchers funding bodies and those who are being researched.
12:35:23 So it's really good at addressing these concerns from multiple angles and approaching these questions and building these relationships before data collection begins helps to mitigate issues later on in the research data life cycle, but could be returned
12:35:36 to later if need be.
12:35:39 Another resource we liked is the data ethics Canvas by Open Data Institute, the ethics Canvas walks us through a series of questions to interrogate and tension potential users and the content of research participants to consider when preparing data for
12:35:53 sharing. So, this corresponds with the data sharing stage of the life cycle.
12:35:57 And it could be used by any researcher who in tons or may want to share their data outside of their project.
12:36:08 So we have about two dozen resources that we're currently working on curating and assigning to different data lifecycle stages in order to put into the toolkit.
12:36:17 Once we have a working version of that which will hopefully be soon will make that available and will solicit input for it.
12:36:23 We're also working on writing several manuscripts to help share the results of our research, and we do plan to share other product research products like our bibliography and our data as well.
12:36:33 Once we can get those prepared.
12:36:36 And then, in terms of the toolkit, and dissemination and closing the loop we're currently in conversations with NCI de to discuss the best way to share the toolkit with their researchers, the diversity scholars network, and to close the loop, then by
12:36:49 hearing from them what is useful about it and what may need to be improved. We were also welcome feedback from those outside of the data, scholars, the diversity sellers network we're planning to make this toolkit openly available, obviously.
12:37:02 And then, this also gives us additional context to work with and see it and continue building that relationship to potentially support their scholars and other ways over growing our con contextual knowledge of each other's work.
12:37:15 And what other opportunities there might be for support.