I'm about to pick my Calculus book for next year. I want to recheck a few things, but I think I've made my selection.

The winner is Gilbert Strang's Calculus. Allow me to summarize its good points:

• It is available online for anyone. This is great because I can see everything and my students can access it from anywhere without lugging a great beastly book. I wish more textbooks would do this!
• The hard copy is relatively inexpensive.
• It's an incredibly readable book. I got caught up in a few of the discussions as I perused the book. This is a book that the untrained reader can understand.
• It recognizes the value of technology. Unfortunately, the technology is out of date, but I have a baseline for my own creations. Were I to rewrite these pieces, I'd be less specific to one technology and be more general.
• It covers all of the material it should cover. (and then some)
• It covers the graphical and mathematical link.
• It is well reviewed.

Now, no textbook is perfect. This one suffers several faults. The first is the technology section really is out of date. The second is that it doesn't seem to do enough with numerical methods in Calculus. (This is linked to the older technology.) Finally, I fear that this may be its final version.

Since I've given links to Amazon for the other books, here is Strang's Amazon link.

Other Textbooks

I did look at a few other books. I will disregard the really bad books and focus on the others that were in the running. I actually have hopes that I'm wrong in some way. Here they are, in no particular order.

Finney, Demana, et al with Calculus: Graphical, Numerical, Algebraic

This was a good book I learned about through recommendation. It does as the title promises. Problems in the book cover all three methods of problem solving. The book lends itself well to technology as well. In fact, the authors claim it was one of the first to use graphing calculators. (I don't know enough to evaluate that claim.)

The reading is concise and clear. It is much more "mathematical" than Strang's book, and this was a strike against it. Were I teaching a college course, this would be less of a strike. This is a nice smaller book. A college could make a one-year Calculus sequence from it.

I also saw a disconnect between the reading and the problems. While I do believe students should apply what they have learned, I think that the book expects them to make too many leaps.

Larson, Hostetler, Edwards Calculus of a Single Variable

This book was quite mathematical. Since I am teaching high school students, I see that as a drawback. To prepare students for college, they need to understand the material. I don't think a high school student could easily understand this book.

Other than that, what a great book! It's well written, it's thorough, the problems fit well with the readings, and technology may easily be integrated. Were I to criticize, I might say that it does not have enough of numerical problems.

Each section is followed by many, many problems. The teacher has great flexibility in choosing what to assign. If a class needs lots of practice, it is there. If not, there are some great applications in the chapter. The teacher can differentiate with this book.

Stewart with Calculus

I used this book as a student (an older version). It has changed a bit. Stewart seems to enjoy issuing frequent revisions of his text. He also issues several versions of it. It's a popular book, and I found the most reviews with this text, especially when I checked older versions.

Honestly, I am biased. I remember my frustration with this text. It is at a very high level. In my first college Calculus course, the professor would start class with a prayer. This was followed with, "Any questions on the homework?" That was it. He did not teach at all. This forced me into an intimate acquaintance with the textbook. It is hard to read. I remember my mind swimming around delta and epsilon early on. While reviewing the book as a teacher, my mind did the same thing!

I think Stewart would be good for someone with a strong mathematical background. Of all the books I looked at, Stewart was by far the worst offender in this regard. High school students need context and they need concrete explanation.

Experts solve problems differently than students. To the expert in Calculus, delta and epsilon, limits, and derivatives all go together. To the student, these are three unique topics. Stewart is written well for those who automatically think mathematically. This is what took it out of the running for me.

I did like its selection of problems and its layout.

Saxon Calculus

I saved this book for last. It failed to be a finalist for reasons unrelated to the book. I'll just be mean and identify this as belonging to the company that wanted to "investigate me." It should not be this difficult to get an examination copy. By the way, it's a week later, and they're still not done investigating. Sorry folks, you're out of the running.

Now I do know Saxon. It is worth analyzing their methods because they really are different from anything else out there. I know without looking that this book will cover everything I could possibly need. The company is meticulous about that.

Saxon is based on review and repetition. I may learn to do derivatives of linear equations today. I will practice one or two, but the rest of my homework will be from earlier sections. I also know that I will probably find the derivative of a linear function in one problem on every homework problem from now to the end of the year.

Saxon books are not arranged in units. There is no order. Each section jumps from topic to topic with no relation to the first. Math is learned by practice.

I really liked this when I taught a group of low-skill math students in Algebra I. Repetition was great for them. It's boring or even insulting to more advanced students, at least to the degree that Saxon takes it.

Students are taught to recognize cue words in word problems and to run through mechanical solutions. I don't think that Saxon works well for the better student who needs more.