The OHSU Library Data Science Institute Introduces Data Science Techniques to Librarians and Researchers

Posted by Ann Madhavan on December 15th, 2017 Posted in: Data Science, News from Network Members
Tags: Data_Science

---

Today’s Dragonfly post comes to us from Nicole Vasilevsky, Letisha Wyatt, Robin Champieux, Laura Zeigen and Jackie Wirz, with acknowledgement to funding support provided by an NNLM PNR Health Sciences Library Partnership Award, under the National Library of Medicine (NLM), National Institutes of Health (NIH) cooperative agreement number UG4LM012343 with the University of Washington. 

The Oregon Health & Science University Library in Portland, Oregon hosted the “OHSU Library Data Science Institute” (ODSI) from November 6-8, 2017 in downtown, Portland. The event was targeted towards researchers, librarians and information specialists with an interest in gaining beginner level skills in data science. The goal was to provide face-to-face, interactive instruction over a three-day workshop. The learning objectives for the training were:

• Increase awareness of key skills in data science and how these can be applied to the participants own daily practices, such as research or serving patrons
• Increase confidence with using data science techniques
• Increase the ability of participants to use or apply data science techniques in problems outlined in the course

Over 75 participants attended this event, which was held over the 3 days. Participants came from within and outside Portland, Washington, Idaho, California, British Columbia and Kansas. The topics for the workshop included topics such as an introduction to version control and GitHub, exploratory data analysis and statistics, biomedical data standards; data description, sharing and reuse; quantitative and qualitative analysis, analyzing textual data, web scraping, data visualization and mapping and geospatial visualization. All of the materials are shared and openly available via our website.

The goals of the ODSI were to:

1) to increase skills of students and information professionals (e.g., librarians and research staff) so that they may be better equipped to work with data or meet the needs of the research communities that they work with

2) provide a venue for networking and relationship-building between local research community, libraries, and active information professionals.

As an outcome of this course, the majority of our participants that identified as librarians or information professionals reported they are more aware of, can actively teach or use key skills in data science and are more aware of how these can be applied to researchers. In addition, the respondents that identified as researchers reported that they have increased awareness of and confidence using data science knowledge; that they anticipate integrating skills derived from the Institute into their workflow (experimental design, data cleaning, analysis) and that they bring this information back to their laboratory, department, and peers.

Our full webpage, which includes links to session syllabi and instructional materials.

Some lessons learned include:

• Development of curricula for a diverse audience is a daunting challenge! To address this in the future we would consider tracks or ensuring that the content is focused and targeted to a specific career field/discipline.
• A Train-the-Trainer event would help present a uniform approach towards pedagogy, hands-on-learning, and delivery. In addition, it might be helpful to host a demo day where instructors can test their content with either other instructors or a test audience.
• More coffee and tea! Data Science is fueled by warm beverages, so we need to add more.