For the great majority of our boys and girls, the kind and amount of education they may hope to attain depends, not on their own abilities, but on the family or community into which they happen to be born.
—Truman Commission (President’s Commission on Higher Education, 1947)
Recognizing the central role that access to higher education plays in economic mobility and a strong democracy, President Harry S. Truman commissioned the first ever federal analysis of the state of higher education in 1946 (The American Presidency Project, n.d.). The subsequent six volume Higher Education for American Democracy commission report (President’s Commission on Higher Education, 1947) highlighted major limitations of the United States’ educational system, including overcrowding and limited access to students without exceptional talent and financial resources. This report was a call to action to increase access to higher education for Americans.
To address these barriers, a series of sweeping higher education reforms were instituted. Community colleges began offering 2-year degrees and free public education for the first 2 years of college beginning in the 1940s, and continue to serve as a path or ultimate destination for postsecondary degrees for millions of Americans every year (Ginder, Kelly-Reid, & Mann, 2018). The Higher Education Act of 1965 further increased access to continued education through the advent of modern financial aid through the Pell Grant (named in honor of HEA sponsor Senator Claiborne Pell). To date, the proportion of students qualified to receive the Pell Grant remains the most commonly used proxy measure for determining the populations of low-income students at institutions of higher education (Janice & Voight, 2016). Although scholars debate whether Truman’s reform aims were ever truly achieved, particularly with respect to access (Hutcheson, 2007), the state of higher education was transformed by these federal reforms (Gilbert & Heller, 2013).
Modern higher education has become increasingly unrealistic for many students from low-income families (families whose taxable income did not exceed 150% of the federal poverty level, determined by factors such as family size and location). As of January 2019, a family of four would be designated “low-income” if the combined household income was anywhere from $38,625 in the contiguous United States to $48,285 in Alaska (US Department of Education, 2019). In 2017, Hispanic households and Black households had substantially lower median household incomes ($49,793 and $40,232, respectively) than White households ($63,704) and Asian households ($83,456; Guzman, 2018), making students from ethnic minorities disproportionally represented in this low-income family designation.
Issues of financial access are a systemic challenge to diversity in higher education. Compared to students from affluent households, students from low-income families are more likely to drop out of high school (America’s Promise, 2014), less likely to enroll in college (Carnevale, Smith, Melton, & Price, 2016), more likely to require compulsory remedial coursework during college (Radwin, Wine, Siegel, & Bryan, 2013), and are four to five times less likely to earn a 4-year degree (Lauff & Ingels, 2013; Pell Institute, 2016). Federal financial aid policies have also shifted, prioritizing scholarships and financial support for students from middle-income families, leaving students from low-income households to shoulder the burden of unmet financial need through other forms of financial aid like federal and private loans (Shapiro, Dundar, Yuan, Harrell, & Wakhungu, 2014).
To help mitigate the high cost of attendance, many low-income students choose to work through college. Research shows that working students are more likely to earn poor grades, especially as they approach full-time employment (Carnevale et al., 2016). Students from diverse backgrounds are again overrepresented in this category. Working students are more likely to be over 30 years of age, female, Black or Latino, first-generation college students, and English language learners or students who speak English as a second language (Carnevale & Smith, 2018). Students may simply choose to complete their degrees more slowly to balance work and school, but this also reveals the disparity between affluent and poor students. Part-time students are only half as likely to complete a college degree in 6 years compared to full-time students, with Black and Hispanic students disproportionately overrepresented in part-time enrollments (Shapiro et al., 2014).
The financial barriers described previously make it difficult for traditionally underrepresented populations to persist in college and earn a degree. For students entering college during 2016, 85.3% of Asian students and 78.6% of White students persist through the first year of college and reenroll during the second year, compared to only 67.0% of Black students and 70.7% of Hispanic students (NSC Research Center, 2018). This achievement gap appears to only widen over time. For students entering college during 2010, 38% of Black and 45.8% of Hispanic students had completed their 4-year degree by 2016, compared to 63.2% of Asian students and 62.0% of White students (NSC Research Center, 2017).
Taken together, these statistics on income and financial accessibility demonstrate that there are wide-spread systemic barriers for low-income and minority students pursuing higher education. One way to promote diversity within the field of behavior analysis could be to find ways to recruit and retain a more diverse group of undergraduates entering the field. Although it may not be possible for professors to directly target many of the costs associated with higher education, professors have direct control over one major hidden cost of education: required course materials and textbooks.
Textbook Affordability and Open Educational Resources
The cost of textbooks has increased over 1,000% since 1977, far outpacing the cost of any other consumer goods or services (Bureau of Labor Statistics, n.d.). A variety of studies have shown that students report sometimes (up to 60%) or regularly (around 25%) going without their required materials due to cost (Chae & Delaney, 2018; Florida Virtual Campus, 2012, 2016). Most students report paying for some or all their textbooks out of pocket because the rising cost of tuition and fees is not being offset by increases in financial aid (Florida Virtual Campus, 2016), meaning that students must either go without required course materials or find some other way of accessing the materials (e.g., working extra hours, sharing books, or finding alternative editions of the material). These studies also indicate that textbook costs are associated with other challenges. Students report taking fewer credits, choosing classes strategically to lower textbook costs, dropping classes due to the cost of the books, or withdrawing from the course due to low grades associated with not having the required materials, meaning that financial resources are a direct barrier to student success.
To address the rising cost of textbooks, open educational resources (OERs) have swiftly gained popularity in higher education. As of 2017, OERs have been adopted in 16.5% of large introductory undergraduate courses, consistent with the use of comparable commercial textbooks (Seaman & Seaman, 2017). OERs tend to be digital and free of cost to users, meaning that students can quickly and easily download their course materials, without encountering paywalls, through repositories such as the University of Minnesota’s Open Textbook Network (2018) and Rice University’s Openstax initiative (2018).
There is no consistently used definition of OER (Wikicommons, 2016), yet most definitions focus on material (a) being free of cost to the user and (b) including permission for the user to retain, adapt, redistribute, and create derivative products (e.g., Hewlett Foundation, 2016; SPARC, 2018a). These later rights are referred to as the 5 Rs: the rights to retain access to the work indefinitely, to reuse the work in a variety of ways, to revise the content, to remix the content by combining with other material to make something new, and to redistribute the work, including any derivative versions created, to others. To better understand the permissions associated with OERs, as well as common misconceptions about the nature of open materials, it is prudent to briefly review relevant copyright licenses.
Understanding Copyright Licenses
Traditionally speaking, copyright refers to an author’s or owner’s right to control the way the content is used, as well as any revenue generated by that content (17 U.S.C. § 102, 1990). Since the founding of the United States, works either fell under copyright law, with all rights reserved to the author, or existed in the public domain and could be used, adapted, or redistributed by anyone without permission. To use, adapt, or create a derivative work from copyrighted material, a creator would need to secure the permission of the original author.
Critics argue that copyright law has overwhelmingly favored the economic rights of the creator with little thought given to other values, such as promoting fairness and innovation, increasing the pace of scientific discovery, and forming a just and attractive culture (Fisher, 2001; Gordon, 1989; SPARC, 2018b). With the goal of improving education, the fair use doctrine (17 U.S.C. § 107, 1992) permits instructors to use copyrighted material in the classroom without first seeking the permission of the copyright holder. It is important to note that there is a variety of factors that influence whether it is acceptable to use copyrighted content for teaching (e.g., the nature of the work, the profit for the user, the financial impact to the author, the substantiality of the portion used relative to the whole of materials), and there are no specific rules describing what type or amount of use is covered by fair use doctrine. For additional advice, readers are encouraged to review the guidance provided by the US Copyright Office (2014).
Nevertheless, there was no way for creators to share some but not all permissions protected by copyright with their audience until Creative Commons licenses were created and released in 2002 (Creative Commons, n.d.). Creative Commons (hereafter referred to as CC) licenses permit all users to retain and redistribute the work. There are four additional conditions specified by these licenses: attribution (i.e., use permitted only if the author is credited for the work), share-alike (i.e., use permitted only if the licensee distributes the work under a license identical or less restrictive than the source’s license), noncommercial (i.e., use permitted only if the licensee uses the work or derivatives and does not profit from them), and no derivative works (i.e., the licensee may use the work but not make derivatives or remixes of the work). From these rights, six regularly used CC licenses were developed: Attribution/CC BY, Attribution-ShareAlike/CC BY-SA, Attribution-NonCommercial/CC BY-NC, Attribution-NonCommercial-ShareAlike/CC BY-NC-SA, Attribution-NoDerivatives/CC BY-ND, and Attribution-NonCommercial-NoDerivatives/CC BY-NC-ND (see Table 1).
Even though work shared under a CC license is free to access, it may not be free of copyright restrictions. The variations of CC licenses give content creators a way to communicate to users exactly what permissions they wish to share and those they do not. For example, a CC BY-NC-SA license would permit users to access, retain, and adapt the copyrighted material, but the licensee would be required to share the material under a similar or less restrictive open license and could not use the copyrighted material for commercial gain.
CC licenses are also nonexclusive and nonrevocable, meaning that any works obtained under a CC license may continue to be used under that license indefinitely (creativecommons.org, n.d.). A nonrevocable license offers great benefits to instructors and course designers because it means that OER users can—at a minimum—continue to use, retain, and redistribute the material as found perpetually and without requiring the permission of the author. By comparison, instructors seeking permission to use copyrighted material for educational purposes under fair use doctrine would have to clearly specify what material would be used, how much of the material would be used, who the material would be shared with, how long the instructor would be able to share that material with users, and so on, and permission to use the work could be revoked at any time.
Educational material shared with a CC license also helps mitigate costs associated with new editions and challenges associated with textbooks going out of print or changing publishers. For example, Research Methods in Psychology, an OER research methods textbook originally published with a now-defunct commercial publisher, was released under a CC license. Because of that license, a new author was able to adapt and revise that work and rerelease it under a CC license with attribution to the first author. Later, additional authors adapted the same textbook to better suit students with more diverse or specific needs. In short, CC licenses helped protect the rights of the author while permitting the robust reuse and adaptation of this textbook, resulting in four different, high-quality versions of this educational material.
OERs and Student Success
The impact of OERs on the educational experience of students and instructors has received recent attention in scholarly research. Studies have explored topics such as student and instructor perception of the quality of open and commercial textbook materials (Bliss, Robinson, Hilton, & Wiley, 2013; Brandle et al., 2019), barriers to the adoption of OERs (Seaman & Seaman, 2017, 2018), and student performance when using OERs compared to using commercial materials (Clinton, 2018; Gurung, 2017; Jhangiani, Dastur, Le Grand, & Penner, 2018). A 2016 meta-analysis of studies on the efficacy and perception of OERs reports that most studies show favorable perceptions of the quality and effectiveness of OER products from students and faculty (Hilton, 2016). This meta-analysis also reports that students using OERs tend to do as well as or better than students in the same courses using commercial textbooks. It is unclear whether this difference is due to factors like alignment between course materials and course content or early access to educational materials (cf. Grimaldi, Basu Mallick, Waters, & Baraniuk, 2019). However, Hilton’s (2016) meta-analysis also highlighted weaknesses in the available research, including a lack of control for teacher and student effects, a lack of large-scale pre- and postcourse performance metrics, and a lack of random assignment to OER vs. commercial use conditions.
Fischer, Hilton, Robinson, and Wiley (2015) compared the academic performance of 16,727 students across 10 educational institutions to determine what effect the transition from commercial to OER textbooks had on student performance. The authors evaluated student performance in 15 courses where instructors moved from commercial textbooks to OERs, choosing student performance metrics available through institutional records, including course persistence (i.e., completing the course), earning a passing grade (C+ or higher), overall course grade, and the number of credits taken both during a semester when using an OER textbook and in the following semester. Researchers reported that students in courses using OERs were more likely to persist through the course, with 6% student withdrawals using OERs compared to 21% student withdrawals using commercial textbooks.
Fisher et al. (2015) also report that for most courses, there were no statistically significant differences in passing grades or overall course grades (9 courses and 10 courses, respectively), but performance did improve in a small set of courses (more passing grades in 5 courses and higher overall grades in 4 courses). In only one course did students perform better using commercial textbooks. There were also indirect benefits, with students using OERs enrolling in more courses during the same semester than peers using commercial products, t(8101) = 27.81, p < .01, as well as in the subsequent semester when credit loads in the prior semester were held constant, F(1, 6440) = 154.08, p < .01. Lack of disaggregated student data about factors such as information about student enrollment status (part time vs. full time) or financial need was a significant limitation of this study.
Colvard, Watson, and Park (2018) conducted a similar study, which included variables such as enrollment density and financial need that were not clearly evaluated in previous studies. Researchers compared performance, measured by final grades and grades of D, F, or W, for students taking eight courses before and after the adoption of OERs. Greater access to student performance data and institutional metrics permitted the researchers to also evaluate the impact of OER adoption on student performance specifically for (a) low-income students, (b) non-White students, and (c) part-time students. Results indicated that there was a statically significant overall improvement in final course grades when using OERs relative to using commercial textbooks, t(21, 820) = −15.95, p < .001, as well as a reduction in DFW grades by 2.68%. The gains achieved by adopting OERs disproportionately improved performance of Pell recipients, F(1, 21,818) = 9.348, p = .002, non-White students, F(1, 19,012) = 10.374, p = .001, and part-time students, F(1, 21,818) = 59.68, p < .001.
These previous studies indicate that adopting OERs improves performance, especially for students traditionally underrepresented in higher education. Increasing access to low-cost and easily accessible educational materials in behavior analysis has been called for informally (OpenBehavioralScience.org; Pavone, 2018), yet it is unclear to what extent these materials exist to be adopted. This review aims to assess what types of behavior-analytic OERs exist with special attention paid to whether the materials could be used to support a course in behavior analysis.