What is the value of textbooks?

If you ever find yourself with a group of other science professors, sooner or later the textbook question comes up. Usually the question is “which one do you use for X class?” The responses can range from “I’m stuck with what my department has chosen, so I have to make the best of it” to “I kind of like my book” to “textbooks are hideously expensive and totally not worth it,” but rarely does anyone give a resounding endorsement for their text. (At least not at the intro level. I think texts play a different role in more advanced classes. We’ll focus on intro texts for today.)  Personally, I am conflicted about the role of textbooks in a modern college science class.

First, the pros…

I teach using what some people would call a ‘flipped classroom’ (and others would call active learning.) A flipped classroom is one in which students take responsibility for watching lectures or reading the textbook outside of class, which then allows the professor to reduce the amount of time spent on lecture so that more active learning can happen in the classroom. Research shows that flipped classrooms lead to higher engagement and learning gains (for a review of recent research see Huelskamp, 2015).

Textbooks and other reference materials are critical to teaching in this way. I want students to have something to read before coming to class, to help prepare them for the day’s activities. I don’t expect them to understand everything they read the first time through, but I want them to have that puzzling experience before they come to class. Then, in class, they are prepared to ask questions and solidify their understanding. To ensure that they do the reading, I assign short reading quizzes before each class. These are typically not high level comprehension questions, but just make sure that they’ve looked over the chapter. The last question is always “what questions do you have on the reading?” They get a point for writing anything in this space. They get no points if they leave it blank or say “I don’t have any questions.” Which I consider to be a cop out, of course they have questions. If they think they don’t have questions, they should read it again and think about it some more. It’s like the famous quote attributed to Feynman, “If you think you understand quantum mechanics, you aren’t trying hard enough.” I think that applies to novices studying pretty much any topic in the introductory physics curriculum.

I’ve found that the reading quizzes help to prepare them to hit the ground running, so we really can spend less time on lecture and more time on active learning (problem solving, labs, etc.). If students experience that initial confusion during lecture, it’s difficult to get the questions formulated before the end of the period, and they leave confused. I often tell them that confusion is the path to understanding. A student last year replied that is only true if you come out of the confusion, which I concede is a fair point. But we have more time to get to the other side if they come to class prepared.

Additionally, I want to have a reference that students can use outside of class to check on an equation or a concept that they don’t quite understand. This is especially important as we move towards doing more active learning, and less lecturing in the classroom.

Now, on to the cons…

Unfortunately, research shows that students often don’t read the text before class. In one study, fewer than half of college biology students were reading the assigned chapters from the textbook at the beginning of the semester, and that number decreased to less than 10% as the term progressed (Burton, 2014). Additionally, they found that students had difficulty answer questions following reading assignment, indicating low levels of comprehension. The results of this study indicate that textbooks would not be effective in getting students prepared to actively engage in class material. However, studies on flipped classrooms, which require this kind of preparation and assign credit to it have greater compliance rates (see review article by Bishop & Verleger).

The study cited above demonstrated that students have issues with comprehension. Why is this? A survey of physics texts found only four of twenty-five were written at an appropriate level for students (Akinbobola, 2015). Another study found that only 38% of graduating college students could understand textbooks (Baer et al, 2006). As you can imagine, there are socio-economics factors at play here, so some groups are going to be more literate than others. Given this, what makes a textbook readable to a novice? Burton (2014) gave different groups of students readings with varying amounts of technical vocabulary. The groups with less technical language outperformed their peers on assessments.

Representations in textbooks can also be problematic, both in number and accuracy. Nyachwaya (2016) found an average of four visual representations on each page in a college chemistry textbook, and some pages requiring that up to six images be conceptually integrated. He argues that this causes a cognitive overload for students. Another study showed that only 40% of images were linked to the text (Kapici & Savasci-Acikalin, 2015). Other studies have shown that the visual representations introduce misconceptions and inaccuracies (e.g. Zeng et al, 2014).  Rybarczyk (2011) also notes that visuals in textbooks are significantly different than the representations used in scientific journal articles.

Taken as a whole, this research indicates that if we are going to use textbooks, or any other written materials such as journal articles, we need to teach students how to use them. How to read, how to look at pictures, how to work problems as they go, etc. There is just too much information, written at too high a level for our students to meaningfully process it on their own. And this disproportionally effects students of color, who we want to keep in the sciences.

Which brings us to my biggest issue: too much content. The students can’t distinguish important information and big ideas from supporting or extraneous details. For example, take the 2-D kinematics (motion) section of a typical physics textbook. In this book that I randomly pulled off my shelf, there are eight equations with boxes around them in chapter 3. In my opinion, no more than three equations (arguably two) are necessary to do projectile motion problems. Why does every textbook on earth derive the #*&% range equation?! Totally useless, except in one very particular case. It is not any harder to just start from the general equations, than to teach about a million special cases. Okay, rant over.

In any case, there are just too many topics; no introductory class covers everything in the book, but they put everything in just in case. And this creates some unrealistic expectation that faculty will actually cover all of those topics. There is loads of research supporting the “less is more” idea (see the link below from the AAAS and Department of Education). We need to move away from “a mile wide and an inch deep”; an idea that is encouraged by introductory textbooks. (To be fair, some texts now allow you to build your own book with just the chapters you cover, which is step in the right direction.)

Finally, we need to talk about how textbooks are stupidly expensive. I recently listened to a podcast on Planet Money (link below) about how this is at least partially due to the fact that we who pick the books are not the ones who pay for them. This is something economists call “the principal agent problem.” Textbook publishers have realized that students like to buy used books, so frequently release new editions to prevent students from using books for more than a couple of years. If you are skeptical on any of these points, note that some of the exact same textbook companies that are gauging our students make high school textbooks that are affordable and designed to be used for more than a couple of years. (Because high schools choose and buy the books themselves, they don’t have the principal agent problem.)

To address the cost concern, a recent large statistical study compared open source textbooks to traditional textbooks. They found marginal gains for chemistry, and no differences in learning for physics or earth science (Robinson et al, 2014). The researchers discuss that this reinforces the idea that online textbooks will not be replacing teachers any time soon; we still need teachers in the classroom to achieve learning gains. Other studies have shown that multimedia textbooks do improve learning gains (e.g. Sadaghiani, 2012).

My experience with textbooks

As my husband (who is also a physics professor) points out, a textbook is organized in a way that makes sense to a physicist, but not in a way that makes sense to students. They don’t understand why the topics are presented in the way they are (which is sort of a historical progression, a fact that is not obvious from reading most texts). For the algebra-based intro physics classes, we have scrapped the textbook entirely. We have moved to a more project-based/issues-based approach in which each unit is motivated by a particular interesting problem. For example, there is a unit on photovoltaics that covers series and parallel circuits, blackbody radiation, photoelectric effect, and semi-conductors. Needless to say, these topics are not typically in the same chapter in a textbook. (For the non-physicists reading this, that would be approximately chapters 15, 31, 32 and not in the text, respectively.) When we first started this, we had the students jumping around and reading sections from different chapters every night. But the text becomes less useful if it doesn’t follow the same structure of the course, and when you aren’t using half the book it becomes hard to justify the cost.

With this course, I want them to have a reference and something to read to prepare for class, but I don’t like the traditional text. What we use now is a combination of popular science articles, chapters from popular science book, sections from textbooks, and things we’ve written. Ideally, we would develop all of our own reference materials that are specific to our learning objectives. I think a good reading assignment would be short and to the point, emphasizing big ideas and important equations, leaving the examples and details for class. Two or three pages max. Longer than that, and they will miss something important. But this is a slow process. Each semester we add a couple more readings, but I think I’ll need a sabbatical to finish the project.

On the other hand, I do use a textbook in my calc-based physics class, which is mostly taken by physics and chemistry majors. These students do okay with a more traditional approach to teaching physics (by which I mean sequence of topics, the class is still primarily based on active learning). I use the Knight textbook, which I think does a good job of limiting the number of topics and emphasizing really important concepts. As discussed above, I have the students complete reading quizzes to hold them accountable for completing the assignment. It’s also easier to justify the cost when you know they will save the book to use as a reference throughout their undergraduate careers.

Once again, I have no easy answers. I think each professor has to weigh the pros and cons and the wishes of the department. I’m lucky enough to have the autonomy to decide if to use a book, and which one to use. But not everyone has that choice, so we do the best with what we’ve got and try to meet the needs of the students as best we can.


AAAS. Trimming the Overstuffed Curriculum.

Akinbobola, A. O. (2015). Guidelines on How to Read a Physics Textbook and the Assessment of the Readability of Recommended Physics Textbooks in Secondary Schools in Osun State of Nigeria. Journal Of Education And Practice, 6(6), 32-39

BAER, J.D., COOK, A.L. AND BALDI, S. 2006. The Literacy of America’s College Students. Washington, DC: American Institutes for Research.

Binns, I. C., & Bell, R. L. (2015). Representation of Scientific Methodology in Secondary Science Textbooks. Science & Education, 24(7-8), 913-936.

Bishop & Verleger. (2013). The Flipped Classroom: A Survey of the Research.

Burton, R. (2014) Readability, Logodiversity, and the Effectiveness of College Science Textbooks. Bioscience: Journal of College Biology Teaching, 40(1): 3-10.

Department of Education. Research & Reports.

Huelskamp, D. (2015). Flipping the Collegiate Science Classroom: A Review of the Research. Global Education Journal, 2015(1), 61-72.

Radio Lab. (2014). Why Textbook Prices Keep Climbing.

Robinson, T. J., Fischer, L., Wiley, D., & Hilton, J. I. (2014). The Impact of Open Textbooks on Secondary Science Learning Outcomes. Educational Researcher, 43(7), 341-351.

Sadaghiani, H. R. (2012). Controlled Study on the Effectiveness of Multimedia Learning Modules for Teaching Mechanics. Physical Review Special Topics – Physics Education Research, 8(1), 010103-1-010103-7.

Zeng, L., Smith, C., Poelzer, G. H., Rodriguez, J., Corpuz, E., & Yanev, G. (2014). Illustrations and Supporting Texts for Sound Standing Waves of Air Columns in Pipes in Introductory Physics Textbooks. Physical Review Special Topics – Physics Education Research, 10(2), 020110-1-020110-24.