Title: Empowering lecturers - how to produce Low-Cost-MOOCs
Author: Loviscach, Jörn; Mulligan, Brian; Uhl, Matthias
Affiliation: University of Applied Sciences Bielefeld, Germany; Institute of Technology Sligo, Ireland; Institute for Economic Education Oldenburg, Germany; Erasmus+ LoCoMoTion Project
Strand: Open education and MOOCs
Empowering Lecturers -- How to Produce Low-Cost-MOOCs
Loviscach, Jörn; Mulligan, Brian, and Uhl, Matthias, University of Applied Sciences Bielefeld, Institute of Technology Sligo, Institute for Economic Education Oldenburg, Erasmus+ LoCoMoTion Project, joern.loviscach@fh-bielefeld.de, mulligan.brian@itsligo.ie, uhl@ioeb.de.
Abstract
Whereas MOOCs have been praised by many as a democratization from the perspective of students, the production of such courses can be regarded as highly elitist. The majority of such courses is created by a tiny number of high-ranked academic institutions worldwide. The reason for this distribution is that the making of a MOOC is a resource-intensive endeavor involving time, money, staff, locations, technical equipment, and expertise. The Erasmus+-funded project LoCoMoTion (Lowering the Cost of MOOC Production) aims to change this by providing a concise guideline on how to realize such a project as economically as possible. Hence, the international project team produced the MOOC “moocs4all”, which teaches just this. This paper introduces techniques that we recommend, shares results from the test run of the moocs4all course that took place from June to August 2015, present lessons learned, and outlines our plans on how to improve the course.
Keywords: MOOCs, low-cost production, video production, patterns, templates, open pedagogy, scalability.
Introduction
MOOCs are expensive because of the labor and partially because of the technology involved. The main reason behind this is the use of video as primary medium. One estimate amounts to costs of USD 4,300 (Hansch et al., 2015) per hour playing time of high-quality educational video. Lecturers seldom possess the didactical and technical expertise for realising such a project on their own. Thus, producing and running a MOOC typically requires a team of a range a specialists, which can dramatically increase the costs, in particular if the workflow is not (yet) mature.
Due to this expensive nature MOOCs started out as a playground for the big names in higher education. In reverse, producing a MOOC could even be seen as a sign for belonging to a small circle of world-class institutions which can afford this (Loviscach and Uhl, 2014).
The idea that this type of courses hints at the future of learning seems weird when one considers that digital technology generally has had a democratizing effect (think desktop publishing or music production in the home studio). In many domains, consumers have become producers (or “prosumers”). Nearly everyone, nearly everywhere connected to the internet can contribute. On the side of content production, however, only a small number of educational institutions achieves worldwide visibility. When it comes to offering courses on a global level, the playing field for institutions and lecturers is not at all leveled. So in terms of sustainability, producing MOOCs in the style seen on platforms such as Coursera and edX does not promise to become a broad phenomenon and practice for the future. MOOC production of this kind simply is too expensive to be suitable for the majority of educational institutions.
Strategies for cost reduction can change this and turn massive and open online teaching into a option that can be considered far wider. One option for a better use of time and money is to introduce lecturers to lean ways of video production. Nobody needs a production team any more; this requires, however, the lecturer to acquaint him- or herself with technology (which gets easier to use day by day) and pedagogy. Another option is to introduce a lean and streamlined workflow for a small team. In the following, we provide details on both these options.
MOOCs should not be traditional lessons on video, but constitute a novel format in its own right. This implies specific requirements, which have to be addressed for the course to be educationally useful. The technical part is the centerpiece for making MOOC production affordable and lowering the entrance fee for lecturers. For an engaging learning experience, however, technology is only one vital ingredient. Other vital ingredients consist in a well-adapted presentation of the content and the use of didactics tailored to the needs and possibilities of the medium. Good online teaching is very different to even a high-quality traditional lecture on tape. And finally, assessment and accreditation have to be taken care of if the MOOC is to be accepted for academic credits.
Publicising low-cost approaches
The LoCoMoTion project aims at providing lecturers and members of education institutions with the knowledge on how to realize online teaching in a lean, cost-efficient way. For this reason, the project collects and documents low-cost approaches to producing and delivering MOOCs. The LoCoMoTion team members have been involved in online education and video production for years and share knowledge, tips and tricks, and lessons learnt. In addition, the team conducts hands-on research, for instance concerning the production of videos, which are a staple of this type of course. By testing different ways to produce video, the project team tries to establish practically tested how-to guidelines.
The central tool for spreading the knowledge about inexpensive MOOC production the team uses is an open online course called moocs4all on the edX Edge platform. Being a European endeavour by scientists and lecturers from five different countries (Ireland, Germany, Netherlands, Spain, United Kingdom), the collaboration is largely coordinated by web communication. The first (beta) version of www.moocs4all.eu went public in June 2015, with the quizzes and the optional peer assessement to be completed in August 2015. Being aware that production processes are constantly evolving -- as is online teaching in general --, the team is going to update the course again. Feedback and insights from the first run will be used for improving the course, that is: its content, its production techniques, its didactics, and its institutional setting.
Low-cost Techniques
A major decision to make is which technical platform to run the MOOC on. Well-known platforms in the cloud such as Coursera, edX, Udacity, MiríadaX, FutureLearn, and FUN can only be used by affiliated universities (which in the case of Coursera and edX requires steep fees) to publish courses. Other platforms in the cloud such as Udemy, Eliademy, Canvas Network, iversity, mooin, iMooX, and mooc.house are open to many more potential producers of MOOCs, but may want to greenlight or even inspect the course before publication.
In order to rid oneself of such interventions and in order to keep full possession and control of the material (videos, quizzes, etc.) and the data (user tracking, comments, etc.), one can install a MOOC platform on one’s own or rented servers or in rented webspace. There is a range of open source software available for this, such as open edX and Google Course Builder. One may even tweak a standard learning management system (LMS) such as Moodle into a MOOC platform, as has been accomplished in the case of mooin. Some institutions have decided to develop their own platform, such as openHPI. Finally, one can combine different web services for constructing an online course, which may be the leanest approach of all. The best-known example for this is CCR08 (Connectivism and Connected Knowledge 2008), which is regarded as the first connectivist MOOC. Such a mashup of web services leads to pieces of content and discussion scattered across the Web. This complexity of the user experience may not be adequate for all learners.
The platform is not just the place where all content is stored and can be accessed by anyone. Rather, the platform is the central communication tool, starting with registration, which lets you build a closed community focused on the course’s topic. The history of interactions and the results of tests taken by every member of this community gets stored for enabling an efficient handling and evaluation of the course. This storage of data is also a prerequisite for being able to hand out certificates.
It must not be overlooked, however, that a platform -- in particular a well-established one -- also is a communication tool in another way, too: It makes a course visible and findable, it attracts learners through advertisements such as regular mailings, and it may add to a course’s reputation and credibility by displaying it in a list of courses from renowned institutions. For courses that are produced in inexpensive manner and hence probably cannot be found on such platforms, the usual social networks such as Facebook, Twitter, LinkedIn, Google+ as well as blogs and mailing lists plus search engine optimization of the course’s landing page are indispensable for (hopefully viral) marketing.
Online courses raise the question of privacy. In traditional education, term papers and test results are well hidden from prying eyes and cannot easily be subjected to automated data analysis. This is drastically different from MOOCs where each and every move of a learner can lead to a trace in the data. Hence, a platform operated by a brick & mortar university may look more trustworthy in terms of privacy that a faceless commercial platform “somewhere in the cloud”. A related aspect to consider is which rights an author has to sign away when using a platform hosted by a third party. For instance, it may not be possible to reuse videos of one’s own MOOC.
Deciding which platform to use (or whether to create a mashup) is only one vital decision to make. Another decision to make that may have as many consequences is from where to source the material of the course. The most efficient way to obtain videos, problems, etc. for use in the course may be not to produce them but to find existing material that can be reused for free, for instance, thanks to an appropriate version of the Creative Commons (CC) license. This idea formed the basis of many courses run on the Peer-to-Peer University (P2PU).
In case one decides to produce one’s own videos, a complete studio staffed by several persons will typically turn out to be too expensive. The most inexpensive approach consists in screen recordings that only show slides or (as in Salman Khan’s videos) text and diagrams written and drawn by the lecturer who talks at the same time, which can be heard in the audio track of the video (Noper, 2012). This can even be accomplished with inexpensive tablet apps such as Explain Everything. In case the talking head of the lecturer is to be shown in the video, a web camera plus the free PowerPoint add-on Office Mix turns out to be a straightforward solution.
On the hardware side, the most important piece of equipment is a decent microphone. (The built-in microphones of notebook computers and tablets are abysmal.) The sound must be pristine, whereas.the audience may tolerate poor picture quality. A gadget worthwhile to build from a display or a tablet and material from the home improvement store is a teleprompter. It enables a lecturer to read text from a (mirrored) computer screen while looking straight into a camera.
Here are four core principles we found helpful to keep video production lean:
Finally, low-cost production also requires examination of the entire lifecycle of the courseware. With a MOOC, one can contribute to the OER (open educational resources) movement, in particular use appropriate CC licences that enable other producers to reuse the courseware and hence save them cost and effort. Looking at the lifecycle is also beneficial for the original producer him- or herself and brings up challenges like these: How to correct errors or to update the courseware, in particular for a second run of the MOOC? How to reuse the courseware for other MOOCs or for on-campus or paid-for online courses from one’s own institution? One answer to such questions may be create videos that are highly focussed in terms of content and are neutral in terms of visual style.
Approaches to Pedagogy
The lean production of MOOCs means using as little time, money and personal energy as possible while still achieving a pedagogically valuable result. xMOOCs and cMOOCs constitute two broad classes of approaches to teaching and learning. The former resembles a traditional lecture, with the content being prepared in advance, whereas the latter can be more challenging in terms of interaction and communication during the course, requiring significant contributions by the learners. Hence, it can be expected (see Margaryan et al., 2015) that participants of cMOOCs are in advantage when it comes to applying new knowledge and skills. By providing social networking tools and encouraging their use, one may be able to achieve the same for a cMOOC.
Concerning lean production, cMOOCs benefit from the high volume of contributions typically required from the learners. This implies that less material has to be produced upfront. A similar reduction of costs may be possible in an xMOOC for instance by asking learners to demonstrate learning by creating teaching materials to be peer reviewed by their classmates. Typically, however, the audience of a course in a STEM field will expect and demand a substantial amount of expository teaching being delivered to them.
In both the xMOOC and cMOOC styles of teaching, the learning path should be designed in terms of objectives that are formulated as competencies. Again templates -- in this case, patterns -- are valuable resources for the design of the learning path, see for instance the MOOC Design Pattern Project, (2015).
The lean way for organizing the steps is using word processing or spreadsheet software, or even better a collaborative web-based application such as Google Docs, which facilitates updates and online discussions. A specialized project management tool is not necessary. Define clearly what, who, how, when and where things shall be done for using the given resources most efficiently. The most important questions are “What?” and “Who?”. After the answers to these are fixed, the person responsible for the respective item can look into the when, how, where and why.
The issue most discussed concerning MOOC pedagogy is the retention of participants. One may hope to boost this by more personal (or pseudo-personal) communication. Lecturers need to be present on the forum, which is time-consuming unless one can share this job among a group of lecturers. In addition, lecturers can provide more seemingly personal communication such as telling stories about the MOOC or oneself, for instance through making-of videos and filmed fireside chats. Such videos may even look more authentic when recorded in an inexpensive handmade style.
Institutional Services
A MOOC may or may not be a one-off product. A lecturer, a department or a university may want to produce a number of such courses, for instance as part of a marketing strategy or of an outreach project. It may even be the case that a department wishes to develop a series of MOOCs using staff who are neither particularly skilled in IT nor interested in developing such skills. For a one-off MOOC, the best approach may be simplification, in particular to reduce the technical and pedagogical expertise required. If, however, several courses are produced in sequence, simplification may give way to specialisation: Certain tasks such as video editing or the creation of quizzes occur in such amounts that it may become efficient to assign a specialist (possibly a student assistant?) who deals exclusively, skillfully, and efficiently with a single one of these.
Ultimately, this thinking leads to an assembly-line model: a MOOC factory as implemented by Udacity and (under this very name) by the Ecole polytechnique fédérale de Lausanne (EPFL). The assembly-line model ensures a substantial level of quality and reduces the amount of training and labor required from highly paid lecturers. However, it may not be appropriate for a lean production. For instance, the lecturer knows what went wrong in a video recording. A video editor, however, has to watch the entire recording -- and may still overlook mistakes concerning technical terms or errors hidden in mathematical equations. This demonstrates a predicament: Having a team of specialists enables everybody to do what he or she is best at: content, technology, instructional design, or visual design. However, the larger the team, the larger the amount of communication overhead. It should hence be borne in mind that once an institution decides to provide institutional services and support to MOOC developers, costs can quickly increase, so it is important to create simple lean processes which may include relatively simple pedagogical models and the use of templates.
Institutional support may start with providing technology -- including cameras, graphics tablets, visualizers, and editing software -- and may reach as far as keeping do-it-yourself recording booths or even fully automated studios. These tools should come with IT support to enable lecturers to stay focused on teaching. Depending on the volume of courses produced, the support may consist in maintenance but may extend to personnel taking over recording and video editing.
The institutional support needs to be transparent so that lecturers know where and when to get which kind of help. One can think about producing an FAQ (frequently asked questions) or even a MOOC that explains the production workflow, technology, and didactics.
Assessment and credentialing
How can we measure what someone learned and how can a learner demonstrate the growth in skills and knowledge? And how can we do so at little cost? The massive number of participants of a MOOC requires assessment methods that can be scaled at minimal marginal costs. Closed questions can be graded by machine. Open response assignments can only be used with self or peer grading, e.g., letting an essay be rated by fellow students according to a rubric provided by the lecturer.
For awarding credible certification, ID verification used to be needed. Options are having the ID checked through a webcam and/or to (as Coursera claims to be doing) recognize a learner’s typing pattern. Another possibility for reliable certificates are proctored exams taking place at universities or in testing centers. A side effect of the verification is a high finishing rate (for instance, 50 % rather than 5 %), which may simply be due to the learners not wanting to lose their financial investment in the testing fee. However, due to the manpower involved, standard proctored testing involve costs which increase linearly with scale and hence require some payment by learners.
In the past years, however, new options appeared that do without the manual verification of an ID document and hence can save costs. LinkedIn enables users to “endorse” each other concerning specific skills. In addition, LinkedIn has teamed up with Coursera to display MOOC certificates in a user’s profile. Internet-based badges (such as those powered by openbadges.org) can be displayed on one’s web page, in social networks, and can be accessed through special websites. In addition, many students and employees alike display a portfolio of past work on the net. Given the labour that a user needs to put into creating his or her network presence over years, these novel, entirely digital means to display one’s learning may be quite reliable, even though they are wide open to fraud for those who do not refrain from putting in time and money for doing so.It could be argued that many of the low-cost methods for validation of learning, being proposed or developed, fall short of the level of reliability of proctored written examinations, but it can also be argued that they are a significant improvement over the certificates of “attendance” or “completion” awarded in many work-based learning courses. It is likely that in the short term such low-cost alternatives to credentials will need to be corroborated by traditional assessment methods such as proctored examinations or interviews, but it is reasonable to hope that techniques to reduce the cost of assessment will emerge as demand for accreditation of independent learning increases in the future.
Preparing and delivering the Moocs4all beta course
An important objective of the LoCoMoTion project was to get the target audience involved in the development as early as possible to gain feedback on the low-cost approaches being recommended as well as the mode of delivery. This objective led to a rapidly developed “beta” MOOC consisting of five sets of short videos and associated quizzes covering the main issues. This was hosted on the edX Edge platform which is openly accessible but lacks a public, invitingly designed portal with course listings; hence, the course had to be marketed through social media, mailings, conferences, and personal contacts by the developers themselves.
Approximately 100 persons enrolled for the beta run of the course so that this pilot run of the course should not be described as massive but simply as an Open Online Course (OOC). As is known from most MOOCs, only a relatively small number of participants were active. The discussion fora were quiet and the developers were able to respond to all comments themselves (which may be counterproductive with a larger course, where you want to limit the work done by the lecturer and rather have the learners discuss among themselves).
The drop-out pattern followed the typical MOOC pattern with a low completion rate. Although responses to the course survey were largely positive, the response rate was too low to gain significant insight into improving the course.
Conclusions and Next Steps
Although the number of enrolments was modest, the general feedback was that the overall objective -- the reduction of costs in MOOC production -- was worthwhile and needed to be pursued further. Initial feedback from users and discussion among the developers indicated that the next steps in the project should include:
This is indeed along the lines originally envisaged in the project plan. The team will proceed to carry out the above tasks and to deliver the first version of this course as a MOOC during 2016. Details will be published on moocs4all.eu.
References
Hansch, A., Hillers, L., McConachie, K., Newman, C., Schildhauer, T., & Schmidt, P. (2015). Video and Online Learning: Critical Reflections and Findings from the Field. HIIG Discussion Paper Series No. 2015-02. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2577882 (last checked on 2015-09-17)
Margaryan, A., Bianco, M., & Littlejohn, A. (2015). Instructional quality of Massive Open Online Courses (MOOCs). Computers & Education 80, 77--83.
MOOC Design Pattern Project (2015): Design Patterns. www.moocdesign.cde.london.ac.uk/outputs/patterns (last checked on 2015-09-17)
Noper, M. (2012). One man, one computer, 10 million students: how Khan Academy is reinventing education. Forbes, Dec 12, 2013, http://goo.gl/yD7zx8 (last checked on 2015-09-17)
Uhl, M., & Loviscach, J. (2014). Abstrakte Räume und unterschwellige Signale. Neue Sichten auf das Phänomen „MOOC“. In K. Rummler (Ed.). Lernräume gestalten -- Bildungskontexte vielfältig denken (pp. 310--316). Münster, Germany: Waxmann. http://2014.gmw-online.de/310/ (last checked on 2015-09-17)
The LoCoMoTion project is funded by the the Erasmus+ programme of the European Union.
((EU logo))
The European Commission support for the production of this publication does not constitute an endorsement of the contents which reflects the views only of the authors, and the Commission cannot be held responsible for any use which may be made of the information contained therein.