Two years ago I taught undergraduate Econometrics and Data Analysis (Econ 131) for the first time, and just used the same textbook as the instructor before me did without much consideration for alternatives. In the short term, this saved me time, but the book turned out to be a poor fit for my class and my students mostly hated it. This summer I’m teaching the class for the fourth time, and I finally made a serious search for a good replacement. Here’s what I learned about textbook selection during the process:

1. Ask around for ideas and don’t constrain yourself to colleagues at your institution. Friends from grad school, colleagues at other universities, Twitter friends, and even Facebook friends might have some good suggestions. Take a look at the syllabi of instructors you might not know but respect. Ask publishers for review copies and access to potential books online (e.g., through Cengage). You’re going to be stuck with this book for the whole semester and you will spend time making it work. It’s worth casting a wide net to find the best match.

2. Actually read big chunks of your chosen book to make sure it motivates the material well and explains it clearly.

3. Look for a lot of overlap between what you’re teaching in class and the content of the book. For some classes this is surprisingly hard. Most schools split the material we cover in Econ 131 into two classes: one on probability and statistics, and one on linear and more advanced regression methods. There is a wide selection of decent books for both subjects but none that covers both at level I’d like.

4. Be ready to supplement your book(s) with other material including:
   • accessible contemperorary research–This is useful for showing applicability of material to highly motivated students.
   • newspaper articles–This is a cheap way to show the class that they should care about what you’re teaching them, and it isn’t just a tortuous Ivory Tower exercise.
   • chapters/sections from other books–These are critical in my case since the “perfect” textbook doesn’t exist, but it can also be useful when you want to give students more or less technical treatments of the material. Make sure to pay attention to your institution’s policy on fair use before you go crazy posting scanned pdf’s.

   The first three times I taught Econ 131, I used a business statistics book (Aczel’s Complete Business Statistics) and supplemented it with more advanced material on econometrics. My undergraduates felt the book was long-winded and spreadsheet-heavy. They wanted stronger mathematical foundations and a book that could serve better as a reference in the future. This year I’m using a traditional undergraduate econometrics book (Stock and Watson’s Introduction to Econometrics) that has 100 pages of “review” material on probability and statistics. Considering we spend half the semester here, it’s not enough, and I’ll be relying on excerpts from Hogg and Tanis’ Probability and Statistical Inference for more depth at the beginning of the term.

5. Look for an alternative perspective to your own. When I was an undergraduate, I hated classes where the professor wrote the book. If students don’t understand something in lecture, they turn to the book for a different take on the material. Your textbook should also go into considerably more depth than you cover in class.

6. Pay attention to price. If older editions will suffice, they are often available for a lot cheaper. Students will be way more likely to buy (and thus read) books that they can afford.

7. Think carefully about what’s required reading and what’s optional reading. Students will often allocate a fixed amount of time to read before class. Do you want them to read one chapter carefully, or skim three chapters? Reading less material carefully is almost always better.