By Tami Back, W&M Libraries
William & Mary Libraries has published its first open access textbook. Authored by mathematics professor Lawrence Leemis, Statistical Modeling: Regression, Survival Analysis, and Time Series Analysis supports students studying statistics, data science, operations research, economics, and industrial engineering.
Leemis has taught in the mathematics department at W&M for 30 years, and has earned numerous accolades, including the university’s Thomas A. Graves Teaching Award in 2013 and Plumeri Award for Faculty Excellence in 2010. Although not new to publishing – Leemis has published seven textbooks – this open textbook is his first experience publishing under a Creative Commons license and writing with open principles as a goal.
Leemis was able to dedicate about three years to write the book through the Library Faculty Scholar Program sponsored by William & Mary Libraries. The program, supported by private funding through the library, provides financial support and library expertise for W&M faculty working on open access projects including open educational resources (OER) and digital humanities. Creating open access content takes a significant amount of time and requires additional support in areas of preservation, copyright law, and digital literacy.
"The Library Scholar position made writing the book possible," said Leemis. "I had intended to write a book of this nature eventually, but time is always the limiting factor. Without the support from the library staff members and the Library Scholar position, the book would never have been written."
Open textbooks are published under an open access license and made available online to be used freely by students, teachers, and others. Universities are looking for ways to ease the burden of high textbook costs for students and their families; open textbooks have become a way for faculty to make their courses more affordable.
“Licensing this book as an OER will reduce some of the out-of-pocket costs for students pursuing majors and careers that require advanced statistical and mathematical courses,” said W&M’s Publishing & Open Access Librarian Rosie Liljenquist. “STEM textbooks tend to have higher purchase costs than their humanities counterparts.”
Leemis wrote the textbook to enable a single-semester class to survey three important mathematical topics: regression, survival analysis, and time series analysis. Most undergraduates who major in quantitative disciplines, such as statistics, data science, or industrial engineering, receive minimal exposure to these topics. However, regression plays a crucial role in machine learning; survival analysis is critical to actuarial science, biostatistics, and reliability; and time series analysis is used routinely in economic and science applications.
"The three topics covered in the book are each traditionally covered in a full-semester course at most universities," said Leemis. "My goal was to include a light introduction to all three topics in one semester in order to expose students to these commonly used statistical modeling techniques. In addition, instructors can pick and choose topics that might be of interest to them because it is an open educational resource."
Leemis worked closely with Liljenquist and other members of the library staff to create the open access textbook.
"Rosie was instrumental in helping make this open textbook a reality," said Leemis. "She has a very thorough knowledge of copyright law and Creative Commons licensing. She helped me navigate those aspects of the project. She also provided much needed encouragement as the deadline for completing the project approached."
Positive feedback for the book and the support provided by the Library Scholar position have come from Georgia Tech, Roosevelt University, Baylor University, Penn State University, and William & Mary. Overseas interest has come from WHU--Otto Beisheim School of Management in Germany and University of Pretoria in South Africa. The book has been submitted to several journals for review.
“In the changing research landscape, academic libraries are stepping into publishing roles for various reasons, including increased accessibility. The Library Scholar Program has been instrumental in building a future for open access publishing at William & Mary,” said Liljenquist.
Leemis’ newest textbook can be accessed freely on the library’s institutional repository, W&M ScholarWorks, at https://scholarworks.wm.edu/oer/3/. Since its publication on the site in late July, the book has been downloaded 718 times.
The work is also available through Pressbooks -- a versatile, user-friendly publishing platform that supports open education initiatives, institutional publishing programs, curriculum development projects, and more. Access to the e-pub can be found here: https://viva.pressbooks.pub/statisticalmodelingleemis/.