It would be very handy to have an agreed-upon dollar amount that we could all use when calculating savings that result from OER adoptions. Many institutions rely on an estimate of $100 per student, per course. This post explains why that is a fair estimate.

Averages don’t work for every situation, though. If you decide to calculate student savings specific to your own institution’s OER efforts, keep these two points in mind:

1. Data is squishy, so be transparent and consistent when reporting savings.
2. Saving money has a big impact on students, regardless of the method used to calculate the amount.

One-size averages

$100: it’s a nice round number that is frequently used as an average textbook cost. The Open Textbook Network (OTN) calculates student savings from open textbook adoptions by multiplying enrollment by $100, and Open Oregon followed suit to calculate the $110,600 estimated savings that resulted from OTN’s Open Textbook Workshops in Oregon’s community colleges in Fall 2016.

The Student PIRG’s 2015 publication, Open Textbooks: The Billion-Dollar Solution, discusses this $100 estimate in more detail:

“Previous estimates of potential savings per student when a course is transitioned from a traditional textbook to an open textbook have hovered between $95 – $115 in previous PIRG reports and external program analyses. In many cases, average savings is rounded down to $100 to reflect the fact that not every student pays full price for a traditional textbook (though many do) and therefore, do not ‘save’ the full price of the displaced material.”

In 2013, open education leaders Nicole Allen and David Wiley gave a keynote at the Open Ed Conference that offered additional research to support the $100 estimate. Their presentation, Save Students a Billion Dollars, cites five studies where textbook cost estimates land on either side of $100. They conclude that $100 is a reasonable median figure to use.

The National Association of College Stores, on the other hand, estimates that the average cost of a new textbook is $82 (using 2014 data). This number is accompanied by a caveat that the average includes all books assigned as textbooks, including low-cost novels and trade paperbacks. David Ernst, Executive Director of the Open Textbook Network, notes that lower-cost textbooks are less likely to be replaced by OER, which means that the NACS estimate may be less accurate for calculating savings from textbook replacements.

Custom-fit calculations

Of course, some institutions prefer to track savings with a method that relates directly to the local bookstore price of assigned materials. Local data can be powerful because it gives you a chance to showcase actual savings by students at your college.

Especially in the early stages of a textbook affordability initiative, it may be possible to establish a price of record for courses where instructors have replaced expensive textbooks with low-cost or no-cost alternatives. For brand-new courses that use OER from the get-go, a national or local average could stand in for the price of record. It is then possible to track enrollment each term and generate a dollar amount saved.

For some initiatives, though, this might become a very complex undertaking – which opens the door to alternative methods of calculating savings. A 2014 research study by John Hilton III, T. Jared Robinson, David Wiley, and J. Dale Ackerman used a local strategy to calculate textbook costs at seven colleges that participated in the Kaleidoscope Open Course Initiative. Their paper, Cost-Savings Achieved in Two Semesters Through the Adoption of Open Educational Resources, describes their method of calculating a cost for a multi-section course:

“The average textbook costs we calculated by summing together the costs of all required books per section and then divided by the total number of sections. These averages were weighted according to the number of students in each section (e.g., classes with high enrollment would affect cost data more than classes with low enrollments).”

The authors of the study calculated average costs for each college and concluded that the average student savings per course across the seven colleges was $90.61. (This study was cited by Allen and Wiley in the keynote mentioned above and contributed to their conclusion that landed on $100.)

When Open Oregon set out to answer the question, What is the cost of course materials for a transfer degree at each community college in Oregon?, a different method of determining the cost for a course was used. The study looked at bookstore prices for the highest-enrolled courses that add up to a transfer degree at each college. But rather than averaging costs across sections, the study looked at one section per course:

“If multiple sections of the course were being taught, the first one on the list was selected. This arbitrary choice accounts for the fact that students often can’t choose their courses based on textbook cost because of competing priorities such as work schedules or childcare.”

All of these methods produce valid data, but any conclusions drawn from them will need to include a clear explanation of where the numbers come from.

Disciplinary differences

Sometimes it’s informative to talk about the differences between your data points, which means taking a step back from averages and re-examining the component costs. For example, instructors in disciplines that tend to select lower-cost materials may feel that the burden of switching to OER should begin with STEM courses, where textbooks have a reputation for being the most expensive. Analyzing the data for individual courses can help determine whether this argument holds water.

Using the bookstore cost data collected for the Open Oregon study, it is possible to determine statewide average textbook costs for some of the highest-enrolled community college courses in Oregon. Data on course enrollments for each college was provided by the Office of Research and Data of the Oregon Higher Education Coordinating Commission from 2013, the most recent year available at the time of the study.

The maximum cost is the retail price at the college bookstore, and the minimum cost is the lowest price available through the store, whether used, looseleaf, rental, or embedded third-party referral. Taking an average of these two data points results in a cost estimate that takes into account both new books and the strategies that a savvy shopper can pursue through the college store.

The table below shows that there is a wide variation in average textbook costs for different courses, and also that student enrollment is just as important a factor as price when it comes to aggregate student savings. For example, the average cost of a textbook for WR 121 is $60, second-lowest in the table. Yet because WR 121 is the highest-enrolled community college course in the state, it has the second-highest potential student savings – over $2M if every WR 121 instructor switched to OER. This kind of information can guide statewide conversations about priorities.

Course	Name	Enrollment (2013)	Average max cost	Average min cost	Average cost	Student savings
MTH 111	College Algebra	19,345	$134.37	$95.50	$114.94	$2,223,417.58
WR 121	English Composition	32,956	$78.61	$43.27	$60.94	$2,008,338.64
BA 101	Introduction to Business	9,956	122.13	94.18	$108.16	$1,076,791.18
PSY 201	Intro to Psychology – Part 1	8,658	$158.55	$89.00	$123.78	$1,071,643.95
MTH 243	Statistics I	7,375	142.26	105.63	$123.95	$914,094.38
BI 101	Biology	6,471	$159.86	$97.94	$128.90	$834,111.90
COMM 111	Public Speaking	12,999	$71.07	$38.56	$54.82	$712,540.19
EC 201	Microeconomics	3,798	$152.28	$109.48	$130.88	$497,082.24
HE 250/295	Personal Health	5,398	$96.80	$53.63	$75.22	$406,010.57
Totals		106,956			$102.40	$91.10

Looking at the last row of the table, the totals are instructive as well. The average cost of the textbooks across all 9 of these high-enrollment courses is $102.40 – very close to the one-size average of $100.

The data also shows another way to understand the impact on students in 106,956 courses. The total average savings divided by the total enrollment comes out to $91.10 – very close to the weighted average of $90.61 that Hilton et al found in their study.

What’s not included

When Open Oregon Educational Resources calculates savings, the dollar amount does not account for textbook buybacks, sharing a book, pirating a book, not buying a book at all, or the other ways that students manage their limited resources outside of what’s on offer at the bookstore.

Students tend to be very concerned about book exchanges and often say that they would like more opportunities to sell their textbooks back. However, there is never a guarantee that they will be able to do so at the campus store. If a new edition is released or the instructor changes their adoption, the bookstore won’t be able to buy back the previous book. Even if there is a need to stock used copies of the book, students still need to have spent the money at the beginning of the term in order to access the course materials.

We know that students avoid purchasing textbooks that they can’t afford, and there is interesting analysis on why student expenditures are not rising along with prices. But to measure savings, the point is to record what is expected of students per the syllabus. And the bookstore is where students can spend financial aid money, veterans benefits, and so on – so it’s not useful to factor in avoidance behavior when calculating savings.