There have been moments in my career as a data analytics instructor that I have considered writing my own textbook, just so I could have one that works. When I started in 2017, Samford University was one of the first schools to seriously reshape the undergraduate business school curriculum in response to the increase in demand for analytics skills. The pickings for appropriate textbooks were slim. Students in my class have already taken “business statistics”, which is a class I had to take as an undergraduate as well. I was trying to smash together business case studies, analytics that was more advanced than basic stats but also not beyond the undergrads, all while using a software program for applications.
I am pleased with what I see in my review copy of the new book by Saltz & Stanton Data Science for Business with R
Publisher : SAGE Publications, Inc; 1st edition (March 3, 2021)
The book begins with instructions for using R that do not assume any prior knowledge of programming. If I use this book, then I can just have one book to follow instead of tacking on software instructions to an analytics book. The authors use business examples throughout to motivate the coding.
Much of the first 9 chapters cover issues you might discuss in ye olde Business Stats combined with getting up to speed with R coding. As an instructor, I would move more quickly through these chapters.
In Chapter 10, the authors finally introduce modeling. Much of chapter 10 is about linear regression, along with some other machine learning concepts. Incidentally, Scott Cunningham has put his book Causal Inference online for free. Instructors could pull examples and R code for regressions from Scott’s book to supplement chapter 10 of the Saltz & Stanton book.
By my definition, only 4 of the 15 chapters are about predictive analytics. Those chapter are pretty dense. I already know what decision trees are, so Chapter 11 makes perfect sense to me. I can see where my job as the teacher would come in to explain this clearly to undergraduates. That’s fine. That’s why I’m there. I like the idea that an analytics teacher could bring in supplemental materials and examples for the analytics concepts instead of the software part of the class.
I predict that this book will be very successful.
Personally, I would skip Chapter 14 on Shiny web apps. Depending on how the timing works out, in one semester, I might also skip one more of the final chapters, to ensure that the modeling concepts we do cover are taught thoroughly.