Open Research Online Mainstreaming Open Textbooks: Educator Perspectives on the Impact of OpenStax College open textbooks

This paper presents the results of collaborative research between open textbook provider OpenStax College (OSC) and the OER Research Hub (OERRH), a Hewlett funded mixed methods open research project examining the impact of open educational resources (OER) on learning and teaching. i The paper focuses primarily on the results of two surveys that were conducted with educators using a range of OSC open textbooks during 2013 and 2014/2015. The results of this research shows that OER such as OSC enables a majority of educators to better respond to student needs whilst often making teaching easier and in some instances changing teaching practices. Although this paper does not focus on the impact of OER on students, a majority of educators surveyed perceive an increase in student satisfaction when using OER such as OSC and believe that OSC are saving students money. using the Connexions platform (now OpenStax CNX) and n=20, respectively) xxii with 38.9% of educators in 2013 and 48.9% in 2014/2015 reporting that they were more likely to contribute to Connexions as a result of using OSC (n=28 and n=22, respectively). These findings reveal that successful use of high quality resources such as OSC enhance the credibility of OER in a variety of ways: there is a perceived increased likelihood of use (or least exploration) of other OER, advocacy of open resources such as OSC with fellow educators and administrators; increased likelihood of contributing to a repository (such as OpenStax CNX), or exploration of the remixable potential of materials via OpenStax CNX or another platform. Indeed 59.2% of respondents in the 2013 survey and 89.1% of educators in the 2014/2015 survey told us they were more likely to select a particular resource when looking for OER if it had an open license allowing adaptation (n=45 and n=41, respectively). xxiii This type of impact is not confined to use of OSC materials. When the same questions were recontextualised and put to educators using Siyavula open textbooks, largely in South Africa, 90.2% of respondents reported that they were more likely to use other “free educational resources” xxiv for teaching (n=55). Almost 80% respondents reported being more likely to report errors in the textbooks (78.9%, n=45) and around 70% of respondents also said they would contribute content to the textbooks and/or volunteer for Siyavula (70.9%, n=39 and 70.5%, n=43 respectively) as a result of using the textbooks (see, e.g. Pitt & Beckett, 2014). Although the context, distribution, and use of Siyavula open textbooks is different to that of OSC (textbooks distributed in hardcopy format across South Africa by the government, for example) is important


Introduction
The high price of textbooks, which have increased in cost by more than fifteen times since the 1970s, and the impact on students in both the USA and elsewhere, is well documented (e.g. Senack, 2014Senack, & 2015. In some USA states, such as California, the cost of textbooks is reportedly higher than the cost of tuition. ii As a result of the high cost of textbooks, 7 in 10 students surveyed by the USA's Public Research Interest Group (PIRG) in early 2011 reported at least one occasion where they had not bought a textbook for a course; and those that did choose to buy proprietary 135 This paper describes research carried out by the OER Research Hub (OERRH), a research project funded by the Hewlett Foundation which examines the impact of OER through collaborative, comparative, international research (see: http://oerresearchhub.org). The OERRH was conceived as a coordinated response to the lack of evidence relating to key areas or questions about the impact of OER identified by stakeholders in the community and previous projects such as the Hewlett funded Open Learning Network (OLnet) project (see: http://www.olnet.org). The OERRH collaborated on a range of impact studies with projects, initiatives, and organisations around the world.
Using a mixed methods approach the project structured its research through the use of 11 hypotheses; a number of these were aligned to each collaboration. The OERRH conducted surveys with over 7000 respondents from over 180 countries.
As the OERRH progressed, and because of the same hypotheses being used across different collaborations, the OERRH's comparative methodology enabled findings from one survey to be compared with another. Findings against the eleven hypotheses can be found in the 2013-2014 report (de los Arcos, B., Farrow, R., Perryman, L.-A., Pitt, R. & Weller, M. 2014 This paper focuses on our work with OSC. Since 2012 OSC have provided a growing range of noto-low cost, peer-reviewed, CC-BY licensed open textbooks and report saving students over $30 million in a little over two years (see: http://openstaxcollege.org/news/our-textbooks-havesaved-students-30-million). In August 2014 OSC announced that they were diversifying into the K-12 textbook market (see: https://openstaxcollege.org/news/openstax-is-creating-digitaltextbooks-that-deliver-personalized-lessons-for-high-school-students). Since the release of their first textbook, the rate of known OSC adoptions has rapidly increased: from 40 institutions in 2012 to 350 in 2013 to 1061 by mid-October 2014 (see: http://openstaxcollege.org/news/ourtextbooks-have-saved-students-30-million). iii By mid-June 2015 OSC reported 1653 known adoptions of their textbooks at 1213 different institutions across the United States. iv

Current Research
The Open Education Group's review project, which compiles peer reviewed research on OER,  • Use of OER leads to improvement in student performance and satisfaction.
• The open aspect of OER creates different usage and adoption patterns than other online resources.
• OER adoption at an institutional level leads to financial benefits for students and/or institutions.
The breakdown of educator respondents for both sets of survey is as follows: ix

Sample Overview
The

139
The following table illustrates users (either current or previous) by textbook title (note that some textbooks were not available at the time of the 2013 survey and are indicated by N/A): x or Introduction to Sociology. xii "Current use" as one semester or less is further supported by a small number of educators commenting that the timing of the survey was too early to make real assessment of the impact of the materials. However, although these respondents could not perhaps yet comment on any impact on their students, they were familiar with the resource and had reviewed it.

First Encounters with OpenStax College
All respondents were asked, in both sets of surveys prior to filtering, an open textbox question regarding how they first became aware of OSC textbooks. Personal recommendation was an important route for the first exposure to OSC: in 2013 over a quarter of respondents who had used or are using OSC told us that they first heard of OSC materials via a colleague, peer, etc.
(27.6%, n=21). Similarly just under 20% of respondents in our 2014/2015 survey told us that they had first found out about OSC via this route (18.4%, n=9). As our survey findings were filtered so that only those who had, or currently used, OSC materials were included, it can be surmised that positive, personal recommendation (e.g. anything from explicit recommendation or at the very least having the resource highlighted to them as something worth examining) is an important first exposure to this brand of open textbook, encouraging people at the very least to review (and in this instance adopt) the textbook.
These findings support that of Clements & Pawlowski (2012)

Increasing Access to Resources
"Access. Access. Access. We are at a college that predominantly serves lowincome students. OpenStax provides students a great text at the best price (free) or for a nominal fee if one wants a printed edition. It has solved the problem of students not having or delaying getting the text due to financial concerns." (Educator, Fall/Winter 2013) Wiley (2014aWiley ( , 2014bWiley ( and 2013 has written extensively about the problem of "disappearing ink" and the impact of restricted usage or high cost online or hardcopy resources on students being 142 able to "own" resources and other associated materials (such as their own online notes) beyond their course duration. Ironically the problem appears to be compounded by educational institutions themselves: "Specifically, campuses have initiated a number of programs like textbook buyback, textbook rental, digital subscription programs, and DRM-laden ebook programs, each of which results in students completely losing access to their required textbooks at the end of term." (Wiley, 2014b) In asking educators about the impact of using OSC on their students, the consequence of a free or low-cost resource being used in class and available both online and/or in low-cost print format had a range of reported impacts. xvi In addition to cost savings, some educators highlighted how immediate access to the resource enabled pre-and post-course use of the textbooks, which as one educator described it, was "a great excuse-defying tool." Access to materials in a range of formats enables every student to participate in their preferred way, without the need to request or be dependent on financial aid to help pay for proprietary textbooks. It is less about digitalization making a difference here but about cost and openness. The fact that whatever format a student prefers to use, access to OSC is either no or low cost (as in the instance of a printed version of a textbook). Compared with proprietary e-materials (which require a student to pay for them and are often accessible for a restricted time period) the OSC textbook has no barriers to use: anyone with an interest in a subject can utilize and share open materials such as OSC.
"They are able to access the textbook and start doing homework immediately rather than being delayed until weeks after the start of the course due to lack of finances." reported that "over 20 percent of teachers described the advantages of OER in terms of student access to materials at the very beginning of the course" (p9).

Cost savings
Although it is not clear how many educators we surveyed were motivated by financial concerns to adopt OSC, it is of note that impressions of cost saving are positive and support reported actual cost savings made through the use of OER. xvii Over 85% of educators in both sets of surveys told us they believed students had saved money by using OER (88.3%, n=68 in 2013 and 86%, n=43 in 2014/2015).
A smaller number of educators in both sets of surveys believed their institution benefited financially from using OER: 59.2% in 2013 and 62% in 2014/2015 believed that their institution saved money in some way by using OER (n=45 and n=31, respectively). In comparison to the overall OERRH educator sample that showed that 73.1% (n=264) and 47.4% (n=172) of educators thought that students and institutions respectively had saved money by using OER, xviii OSC using educators seem more confident in their belief that this type of OER saves money. However, as

Wiley and Hilton III (2012) state calculating institutional cost savings is very dependent on how
OER are remixed and a range of factors need to be taken into account. This reason might also explain why a small but significant number of respondents told us they weren't sure whether their institution had saved money by using OER (18.4% in 2013 and 26% in 2014/2015 n=14 and n=13, respectively). This lack of certainty is reflected by the comments that 2013 educators provided for both the "yes" and "don't know" response options. As one "Don't Know" respondent told us: "I sure think that if the institution more fully made use of open educational resources that we could benefit financially: by retaining more students who otherwise have to drop out because of the high cost of textbooks; by providing 144 higher quality and more diverse and accessible learning and teaching resources which would be a great financial benefit."

(Educator, Fall/Winter 2013)
Whilst similarly reflecting on the complexity of calculating institutional financial savings, one "yes" respondent more generally remarked: "The school district can allocate the resources to other areas rather than textbooks." The number of educators who were unsure about institutional cost savings arguably highlights an apparent lack of certainty amongst some respondents as to if and how institutional cost savings could be calculated and/or what constitutes cost savings. This could arguably indicate a possible lack of transparency and/or lack of available data on the financial impact of textbooks beyond student cost saving, the latter which remains largely the focus for impact data to date. xix There are a small and growing range of tools available to assist with cost saving calculations available, however these appear to not yet be in widespread use. xx

The Impact of Exposure to Open Textbooks
Positive experiences with an open textbook (e.g. OSC) or other OER has potentially far reaching consequences both in terms of facilitating the "mainstreaming of OER" but also in terms of changing educator practice (see below). In both sets of OSC questionnaires xxi , educators were asked about the likelihood of carrying out a range of actions as a consequence of using OSC materials.
Over 95% of educators in both surveys reported that they were more likely to recommend OSC materials to fellow educators as a result of using them.

Impact of OER on Educators
"I simply swapped one textbook for OpenStax. Aside from assigning homework in an online system (not through OpenStax), I have not majorly changed my teaching practices." (Educator, Fall/Winter 2013) "Not at all, except that I had to adapt to slightly different notation." (Educator, Fall/Winter 2013) The self-reported impact of using an OER like OSC on educators varied from no reported or uncertain impact to triggering widespread transformation of practice. In some instances, as above, reported impact on educator practice was purely practical at the time of the survey: a matter of exchanging one textbook for another with changes to notation and/or online homework systems. In other instances, educators, as below, reported more radical impacts on their own 146 practice. Each response was thematically categorized and revealed a range of impacts on educator practice: There are two main impacts reported by educators on their own teaching practice: around 30% of educators in both surveys reported that using OSC materials makes their teaching easier xxv while around 25% of respondents in both surveys reported that using OSC had enabled innovation or changed their pedagogical approach. In the remaining reported cases impact was currently unknown (as they were just starting to use the text), OSC materials were directly benefitting the educator themselves in terms of confidence or up-to-date resources, or there were no reported impacts. A small number of educators made other comments in response to this question; e.g.
gave feedback on the materials more generally.
Although these sets of surveys did not explicitly ask respondents how they reused or remixed Students can be exposed to a wider range of teaching styles than those contained in one set of resources through the use of OER, particularly if educators choose to remix content and draw together a range of sources to teach with. This is possible because any obligation to use a costly resource extensively is removed. Similarly, the following educator notes the "internal guilt" they felt from being contractually bound to use particular resources and the way in which they could show their empathy with students' situation by taking action to alleviate the cost of textbooks: "It allow[s] me to choose the best affordable materials in a class. I am not bounded to use big commercial publishers' products. I do not take bribe from them and it gives me freedom from internal guilt. It allows me to show how I care about the students in a practical and beneficial way." (Educator, Winter 2014 148 Similarly, another educator described how taking back "control" through using OSC involved breaking the "hold" that previous textbooks had and in some sense "liberated" them: "I am teaching the way I want to teach, in the order and flow that I want. I am free of any text book. The book is a resource...The book no longer drives the course. I produce the curriculum. The book is my servant. I am not its servant." (Educator, Winter 2014 Although this paper does not focus on educator perceptions of the impact of OSC on students, some respondents noted the impact of changing practice on students in response to this question.
Echoing the "excuse defying" comments of the fellow respondent noted earlier, the following educator notes that by using OSC materials, the relationship to, and between, students has changed.
"The availability of OpenStax textbooks online has allowed me to give more responsibility to my students. They cannot use the excuse that they left their book at home. Their textbook is always available and with our class web page, their assignments and handouts are also always available. Easy home access has given us more time to spend working on problems collaboratively in class. In the past when assigning problems as homework, if the student were to get stuck, he/she was stuck. Now, working collaboratively, students answer each other's questions and I am there to help as well." (Educator, Fall/Winter 2013) This could be seen as a representative example of an impact of OSC on students that over 55% of educators in both surveys strongly agreed or agreed with: the idea that either OER/OSC in the classroom "develops learners' increased independence and self-reliance" (58.1%, n=43 in 2013 and 56.8%, n=25 in 2014/2015). xxvii Providing resources that students can access 24/7, in perpetuity, and in multiple formats means students have no "excuse" not to participate whenever they choose to. Another response saw one educator reflect on their own practice, copyright issues, and a growing "interest" in creating OER: Arguably a consequence of this and, for example, student costs savings, it is of note that around 65% of 2013 and 69.8% of 2014/2015 respondents strongly agreed or agreed that use of OER such as OSC in the classroom increased learners' satisfaction with the learning experience (66.2%, n=49 and 69.8%, n=30 respectively). The relationship between increased "control" over resources and teaching practices noted by a range of educators was also expressed by one educator in terms of "…greatly increased … enjoyment…" xxix and it would be interesting to see if increased teacher satisfaction impacts positively on students and if so, how?

Limitations to this Study
There are limitations to the approach that was adopted for our survey work. In the instance of the 2013 surveys incentivisation (two $250 Amazon giftcards could be won in a prize draw), whilst increasing participation in the survey may have attracted respondents only interested in the prize draw or more inclined to give positive responses. Although there is no evidence for this bias, it is the case that educators who participated in all sets of surveys were self-selecting. To mitigate bias it was made clear that any information relating to the prize draw (e.g. name and email) was separated from survey responses before being passed to OSC for the prize draw.

150
An arguable limitation to this study is that it is both self-reported impacts of OER on educators and, particularly in the instance of future intended actions (e.g. it is more likely that someone will do X due to Y), there is no follow-up to see what occurred next or if instructors did, for example, use more OER in the future. Moreover, the research findings reported above only represent the experiences and thoughts of a small number of OSC users: as noted earlier there are 1,653 known adoptions as at mid-June 2015 (and likely hundreds more adoptions or use that is not reported to OSC) whereas our research makes use of data drawn from a total sample of 127 respondents.
The surveys also highlight further avenues for research, which were either omitted (e.g. the hypotheses on educator reflection) or not included (e.g. the length of time using OSC materials).
Whilst some of these were explored in interview there is scope for improvement in the questionnaire structure. In addition, our research highlights the need for more in-depth, focused and systematic study of how educators use OER, their motivation for doing so and whether any pedagogical change occurs.

Conclusion
Open textbooks such as OSC have a positive impact on both educators and students. The findings above show that educators report increased student participation as a result of using OSC, in addition to cost savings. Whilst some educators are deliberately searching for open material themselves, for others personal recommendation and a "trusted source" of open material appears to act as a shortcut to finding resources that are of high quality. In the instance of the OSC users we surveyed, there appears to be a potential "domino" effect in terms of successful use; almost all of our survey respondents were more likely to recommend using OSC materials to others as a result of using the books themselves. This finding and the wider range of "advocate" style behaviour that our survey results revealed show the potential of even a brief introduction to a resource such as OSC. However, further focused longitudinal research with a broader range of participants would be needed to develop these initial findings.
Calculating the cost of OER adoption at an institutional level remains complex and there was uncertainty amongst educators we surveyed regarding the calculation of institutional cost savings.
Providing tools such as those highlighted earlier, and promoting the use of these, would help gauge potential costs savings and provide useful, localised evidence to help initiate further discussion with regard to policy change.
In summary, the above results show that OER such as OSC enables a majority of educators to better respond to student needs, which can make teaching easier, and in some cases, engender a change of pedagogical approach. The educators surveyed believe that OER such as OSC are saving students money and perceive an increase in student satisfaction. The role of personal introduction to OER also appears an important one, particularly when recontextualised within the findings related to "advocate" style behaviour reported above.