ASET logo
[ OLNT'90 contents ] [ EdTech Confs ]

Open learning: Concepts and costs, successes and failures

Desmond Keegan
Department of Employment and Technical and Further Education
South Australia


Conventional education

For the purposes of this paper the definition of conventional education given by Kaye and Rumble (1979) is adequate.
The term conventional education is applied to formal classroom based instruction in a school, college or university setting, where teacher and students are physically present at the same time at the same place.
Conventional education has existed for some 2500 years. No doubt it will survive for another 2500. There are about 600 million students in conventional education today.

Distance education

Distance education or external studies has existed for about 150 years. It will probably continue for at least another 150. It may be defined as a form of education characterised by There are about 10 million students in distance education today, at least a third of them in Russia and China. Evans and Nation (1989) tell us that Mikhail Gorbachov was one of them.

The study of distance education focuses on the justification of the abandonment in this form of education of interpersonal face to face communication (previously thought to be a cultural imperative for all education both in east and west) and its replacement by an apersonal, mechanical or electronic communication created by the technology of industrialisation. The German scholar Peters (1973) characterised distance education as alienating both for staff and students, and the Ayatollah Khomeni characterised it as evil because it went counter to the Koranic tradition of the interpersonal transmission of education by the family and the ayatollahs.

Open learning

In the second chapter of Foundations of distance education (Keegan, 1990, p.20-25) various concepts of open learning are considered. In the US in the 1960s, inspired no doubt by philosophies like that of Carl Rogers, there was an outbreak of pressure for "open" education which promoted structures like University Without Walls, Antioch and Nova University, CAEL, the external degree and many experimental programmes. Many of these structures faltered or folded in the 1980s under the influence of Reaganomics but they have left a legacy today of more open approaches to tertiary education, especially in the TVI's two year and four year colleges.

In the Europe of 1968 the demand for open education took on a revolutionary aspect. The centre of the revolution was at the University Paris VIII at Vincennes (Debeauvais, 1976) but outbreaks of violence occurred in many European cities at the time.

Echoes of some of the ideas of 1968 reappear in Beyond distance teaching - towards open learning, edited by Hodgson, Mann and Snell (1987), where ideas from Illich and Freire are seen as characteristic of open learning. "Education is not neutral", they write. "Its purpose can be either to domesticate or to liberate. Education domesticates where knowledge is given to or deposited into learners where the relationship between educator and learner is that of subject to object".

In the UK since 1983 there has been a proliferation of moves towards open learning. The influences have come from administrative structures, from people and from publications. The best known structures are the Manpower Services Commission's Open Tech programme and the Open College. Among the promoters have been Roger Lewis, Mary Thorpe, David Grugeon, Malcolm Tight and Nigel Paine, coming from distance education or educational technology backgrounds in the 1970s and early 1980s. A series of publications in the last few years has included open learning how to do it kits, and books which are collections of chapters by different authors on open learning experiences. The most important are the following

In Australia the Deakin University calendar indicates that it organises three types of courses, on campus, open campus and off campus. Deakin's description of on campus and off campus programmes coincides well enough with the definitions of conventional and distance education adopted in this paper. Deakin states that open campus is a mode of study which involves both on campus and off campus study. This idea of open campus is used for the analysis of open learning in this paper. I maintain my position (Keegan, 1990, p.25) that "open learning is a term that is not to be used in an administrative context", because in MacKenzie's (1975, p.21) words
Open learning is an imprecise phrase to which a range of meanings can be, and is, attached. It eludes definition. But as an inscription to be carried in procession on a banner, gathering adherents and enthusiasms, it has great potential. For its very imprecision enables it to accommodate many different ideas and aims.


Conventional education

I can find little reference to the costing of open learning in the literature referred to above. However, the costing of conventional education is well established (Sheehan and Gulko, 1976) and can be described as
Number of faculty  =  Weekly student hours
Average class size  x  average faculty load

In the conventional university, faculty salary expense represents a major component of total costs and, to the extent that total instructional cost can be expressed as a function of faculty members, the model can be used to represent the instructional cost equation, thus

Faculty salary expense  =   WSH  x  AFS
where WSH
weekly student hours
average faculty salary
average class size
average faculty load.

Distance education

The costing of distance education is also well established (Rumble, 1982) and can be described as
T  =  F + aL + bD + gC + xS
where T
total recurrent costs
fixed recurrent costs
number of local study centres
number of courses in development
number of courses in presentation
number of students
average cost of a local centre
average cost to produce a course
average cost of presentation of a course
average cost per student.

Open learning

The basic costing of open learning might be expressed thus
  +  T   >   WSH  x  AFS
where T = technology (T might be similar to T in the distance education equation or totally different depending on the system). If T is very small, say the cost of a video cassette recorder or an overhead projector plus the tape or transparency used with T, then the cost will always be higher than the lecturer without the technology, but the cost can be absorbed. If T is large then open learning will always be costlier than conventional education unless ACS or AFL can be increased because WSH and AFS are regarded as constants .

ACS or AFL can be increased by refusing open learning students the right to attend some or any lectures, thus freeing staff to lecture to other groups of students to justify the cost of T.

As there seems to be so little analysis of costing of open learning in the literature, it may perhaps be of value to examine some examples of distance education costings to see it they can shed any light mutatis mutandis on future costs of open learning.

The fixed costs of the first Arts Foundation course offered at the OU was £162.558. A comparable course in a conventional university (CU) had fixed costs of £401. The variable costs of the OU course were £56 as compared with £117 at the CU so that the OU course had to have 2658 students on it in order to have an average cost equivalent to that of a CU course.

Another OU course, D281 on geography, had a fixed cost of £73.702 compared with a CU fixed cost of £671. Its variable cost was £109, while the comparable variable costs in CUs was only £87. Here, the OU course could never be more cost efficient than the CU offering.

Yet another OU course in science had a fixed cost of £95.007 against a CU cost of £666. Its variable cost of £100 was lower than the CU equivalent of £218, hence it needed a student population of 800 to have average student costs equivalent to that of a CU course (Rumble, 1987).

These examples of distance courses being dearer or cheaper than their conventional counterparts exemplify the following principles of distance education costing which can be applied to open learning.

Hopes, successes and failures

ETTI, Vol. 26, No. 2. May 1989

For over 20 years Programmed Learning and Educational Technology has been a leading influence in the field of educational technology. Recently, under the driving leadership of Chris Bell it has changed its name to ETTI (Educational and Training Technology International) and seems set to be a leading authority on educational technology and open learning in the 1990s.

The May 1989 issue (Vol 26, No 2) is devoted to educational failure and one of the major articles is Student failure in open learning (Whitlock, 1989). The whole issue is worth studying as a background to the hopes for educational technology and open learning in Australia in the 1990s.

The theme is that we can learn much more from a careful study of those projects which fail, than we can from congratulating ourselves on those that succeed. The emphasis is on a constructive analysis of the failure, examining the reasons why and drawing lessons for the future. Worth quoting is the following by Professor John Tiffin (1989) who, with Combes, was responsible for the launching of many educational TV systems in Africa and Latin America in the 1970s.

Why did educational television fail?

It must be over ten years now since I got a sad letter to say that the ETV experiment in Ghana was all over. Today Brazilians talk of the cemetries of educational television and I have visited the ruins of ETV projects from the jungles of Amazonas to the hills of Porto Alegre. What happened to educational television in Sierra Leone, Pernambuco, Argentina, Chile, Fernando Po, Uganda and Ethiopia? The USA poured millions into giant instructional broadcasting projects in El Salvador and Nicaragua, but the revolution they got was not in education (p.136).
Kone and Jenkins (1990) comment
Samoa, El Salvador and now the Ivory Coast: all the large schools television projects of the 1960s and 1970s have gone. The Ivory Coast scheme, the largest and most ambitious of all, closed down quietly in the early 1980s, and very little information is available on the decline and fall (p.86).

The collapse of UMA

One of the flagships of open learning in the 1960s and 1970s was the University of Mid America. Based at Lincoln, Nebraska, it was a high profile institution, frequently written up in the literature, often referred to by administrators, which was in the vanguard of the use of new communications technology in education.

I would describe it as an open learning materials development agency masquerading as an open university. As an open university the model was fatally flawed and those who knew their distance education theory well should have been able to forecast that trouble was on the horizon (McNeil and Wall, 1983).

UMA collapsed in 1982. It was a costly open education failure. It was still being written up as an ideal model for open education in the mid 1980s long after it had been disbanded and its staff dispersed.

The failure of the Open Tech

The Open Tech programme was the flagship of open learning in the UK in the 1980s. It was discontinued in 1987, much to the surprise of many enthusiasts, some of whom still seem to think it survives today. Fay (1988) gives the following analysis of the "fatal contradictions" of the Open Tech programme.
The aims of open learning (OL) are in themselves quite innocent and laudable in so far as they attempt to operationalise ideas of access; ideas of choice; ideas of accreditation of both formal and informal learning; ideas of progression between the range of options. In practice, however, and with the benefit of hindsight, there have been deep contradictions which now add up to a bleak prognosis for OL. Take Open Tech (OT) as an example since it has been the market-leader in the field. Is it too much to say that OT has proved an expensive white elephant or, rather an expensive herd of white elephants? As Twining (1987) has pointed out, OT has suffered from a number of diseases which, caught singly, are nasty and 'together - as Open Tech proves - they are fatal' (p.10)
Those who wish will read Fay's article and Twining's study in detail to analyse their reasons for the failure of the Open Tech. The following are some of the reasons given. In all, the Open Tech reached 25,018 of its target audience of 3,500,000 at a cost of $100,000,000. Fay (1988) makes it clear that he feels that it was important that the Open Tech programme should fail.
The fact that it was the Manpower Services Commission, traditionally so hostile to the education system, which was behind so much of open learning and wholly responsible for Open Tech, has now proved serious. The Manpower Services Commission is notoriously associated with a pump-priming, project-based methodology which makes it difficulty for either it or anyone else to set up lasting systems.
Today the Open College in the UK has learned some of the messages of the failure of Open Tech, but even so, in 1990 it remains a fragile structure.

Two way audio, two way video

The SA Department of Employment and TAFE interactive video conferencing project at present provides two way video and two way audio between Adelaide, Clare and Barossa at distances of 75 km, 110 km and 135 km. This technology stands on the borderline between conventional and distance education because a lecturer and three groups of students are linked electronically in such a way that a question, comment or joke by any participant is shared by all other participants. The elements of the SA system are The two codecs at Adelaide (A), plus one each at Barossa (B) and Clare (C), take the PAL signal from one of the classrooms, digitise it, transmit it by telephone, and decode the incoming digital data into a PAL signal for display on the large monitors at the other two sites.

Telecom had in place in SA 2 Mb/s lines which are dedicated in their entirety 24 hours per day to the project. A 2 Mb/s line has capacity equivalent to 32 STD calls, that is 32 x 64 kb/sec data streams. The installation will be upgraded from the older 2 Mb/s video conferencing technology to the newer 384 kb/s technology available as part of the ISDN service. This will offer a reasonable compromise between video image quality and cost effectiveness. 384 kb/s is the equivalent of six telephone channels. Six is probably the optimum, although the newer codecs may operate with as few as two channels (one for video and one for audio), if desired in order to economise at the expense of picture quality.

The codec compresses the signal from its PAL equivalent of about 100 Mb/s. It digitises the image, analyses it and transmits data describing changes from the previous image rather than the whole of a new image. The spatial resolution is reduced by 50% horizontally and 50% vertically. This means that the 400 pixels of the horizontal dimension are reduced to 200 and only 200 are transmitted, and the 625 lines of vertical resolution are reduced to 312 for transmission. At the other end each line is doubled so that the 400 pixels are restored at the receiving end but are the double of the 200 pixels transmitted and any information contained in the 200 pixels not transmitted is lost. Similarly the 625 PAL horizontal lines are restored by doubling the 312 lines transmitted. This effect reduces the PAL signal to the quality of resolution of a VHS recording which has about 200 pixels per vertical line and 312 horizontal lines.

The codec reduces the PAL transmission time of 25 pictures per second by half. Thus the total compression is to one eighth of the original signal as 50% x 50% x 50% is an eighth. Each picture is analysed in the codec's memory. It looks at each location on screen to see if there is change in the adjacent fields and selects for transmission only the changing picture and leaves in storage the static half.

Thus if the lecturer moves around a lot, all his or her movement has to be transmitted in the allowed 12 pictures per second and the background is not retransmitted. If the camera is panned or zoomed, the codec is being asked to transmit all the picture at 12 pictures per second. It will do it but at low resolution so that the picture does not stabilise until panning stops. Similarly, videotapes are usually satisfactory for transmission, but if they include relatively complex movement, for example a lecturer walking though a supermarket, then the result can be harsh on the eye.

A vision switcher is needed in the system to provide smooth transition from one camera to another. In addition all cameras at all sites in the system should be genlocked, otherwise the image is disrupted when cutting between cameras.

An acoustic echo canceller is needed to eliminate feedback in the audio and to prevent the lecturer getting the uncomfortable effect of hearing his or her own voice. The echo canceller suppresses any signal from the microphone down the line and when it has transmitted it mutes the incoming line slightly so that the muting cancels the echo effect. However, the echo cancellers may lose control if an excessively loud sound is received. Hence the system needs to be fitted with a compressor/limiter for control of the audio signal. The signal is fed through the limiting amplifier before it is fed to the echo canceller.

Three video distribution amplifiers are necessary for the routing of video. Unlike audio, which can be fed to many receivers at a time, video can be fed to only one at a time. If you need copies of a video signal you have to use a video distribution amplifier which generates the copies of the signal you need to feed to the number of sites in the system.

One ELMO graphics camera on a stand is highly desirable at each site. The technology has been with us for 30 years but is now a vital adjunct to two way video as it has a zoom lens and a toggle. It has two main functions.

Two small colour TV monitors are set into the lectern in front of the lecturer so that he or she can see what is being sent to the two distant classes - the two large silent TV screens in front of the lecturer at the rear of the electronic classroom show the two distant classes of students at B and C.

In a further refinement these two colour monitors in the lectern can be fitted with touch sensitive screens which will give computer control of the cameras in the room. The cameras behind and above the lecturer's head are adequate for showing the A group of students to the students at B and/or C while the lecturer is lecturing. But when the lecturer starts discussing with the class, questioning the students and receiving questions from them or a discussion takes place between a student at A with a student at either B or C, what is needed is a camera with a longer focal length lens and a computer controlled remote pan and tilt head so that when the student starts speaking the lecturer can touch the face of the student in question on the monitor in the lectern and activate the computer controlled mechanism to tilt the camera and pan onto the face of the student until the student's face fills the screen and the other students are no longer on camera.

The present disposition of the four genlocked cameras in the electronic classroom is as follows. A video camera is placed in the back wall of the classroom directly opposite the lecturer. It is remote controlled for pan and tilt and, as this is a no technician system, it is set before the class starts in accordance with the height, weight and agility of the lecturer to provided a stable head and shoulders shot of the lecturer throughout the class. On the wall of the electronic classroom behind and above the lecturer's head are two genlocked cameras, one providing a medium angle and the other a wide angle shot of the ten to twenty seats, desks and students that form the class at site A. Each desk has an inbuilt voice activated microphone with no controls . On the lecturer's desk to the left of the lectern sits the fourth camera, the ELMO. To the right sits the VCR.

The front and back walls of the electronic classroom each have two silent large domestic TV screens. Although the variations are numerous, the two screens on the wall behind the lecturer show usually the graphics or the classes at sites B and C for the students at A to see and the two TVs behind the students at site A show the lecturer the classes at B and C.

As it is too awkward for the lecturer to be constantly pressing buttons when two way interchanges with students occur, a quad split screen is automatically activated when discussions occur so that the screen is split into four quarters. The third party gets to see the quad split as it is important that all should see both questioner and the one who answers. In the control room beside the electronic classroom is the audio desk so that C gets audio from A and B, B from A and C and A from C and B. (Control is not really an apt word as the real control is in the control panel on the lecturer's lectern.) As this is a no technician system, these controls are preset.

There is one speaker per site and two telephones for fault identification. The electronic classroom is furnished with special fluorescent lighting, with special phosphorous coating on the tubes. Without these, fleshtones suffer from a greenish tint on colour TV.

Two way audio, one way video systems

The IBM ISEN system attempts to unite a lecturer to groups of 12 students in each of 11 sites by satellite around Australia (refer also Cheng in this volume). It does not attempt to unite students to lecturer visually nor student to student visually but it can do so orally. The hub of the Australian system is at the IBM centre in Sydney. The IBM ISEN system in the US links up to 300 sites by satellite, but is not linked to the IBM Australian system.

In Adelaide there is a pleasant classroom headed by two TV screens, one for the lecturer in Sydney and the other for graphics. Each desk has a small unit comprising a microphone, a coloured button for the student to press if he or she wishes to ask a question and another button which is activated if the lecturer takes the question, two red lights for multiple choice answers and 5 buttons for multiple choice questions. There is an automatic control unit in the room next door comprising decoder, multiplexers and other video and audio technology. There is a satellite dish and a satellite uplink.

The students at any site can press the question button and if and when the lecturer chooses, the question will be taken. Alternatively the lecturer can ask a question of student 5 at Adelaide or student 11 at Brisbane or get student 5 at Adelaide to talk to student 11 at Brisbane (none of whom can see each other).

At Sydney the equipment for the lecture room is much more complex than at Adelaide College of TAFE. There are two cameras in the ceiling for graphics, a camera opposite the lecturer, a wall of seven TV monitors opposite the lecturer to show the various options available to him or her for sending by satellite. There are also an electronic pen, slide projector, PC, VCR, and a computerised results screen, giving all the results of the answers to the multiple choice questions from up to 132 students . The Sydney control and lecturing room is supported by a full broadcast studio and computer graphics laboratory. There are no students at the Sydney studio.


For many decades the attempt to develop successful multimedia educational systems has been intense. Any realistic assessment of the developments in the last three decades would have to be that the results have been disappointing. In the last two years there has been focus on a new development which many hope will succeed where so many others have failed. This development is referred to as hypermedia, hypertext, hypercard and CD ROM.

Hypercard is a package developed by Bill Atkinson, one of the original Apple team in California, and introduced by Apple. Similar systems run on other machines. These systems are called WIMP environments (windows, icons, mouse and pull down menus). The goal is to produce a mechanical device that is an aid to human associative memory leading to what is being called the creation of "artificial reality systems."

Hypermedia is computer delivered text or illustration that allows one to move quickly and in an ordered manner to what is needed by what is being called "non-linear text".

Hypercard is a useful referencing system for conventional education. It costs only $2 to reproduce a CD ROM Hypercard disc containing 255,000 A4 pages of data and only $5000 to produce. Hypercard is also a valuable adjunct to computer based training. Its characteristics include a talking dictionary with sophisticated moving graphics, laser vision for digitised TV and advanced animation technology.

In distance education when the student or the system provides a machine the equivalent of a MacPlus, the student can be logged on to a bulletin board system. With software like Timbuktu, the tutor can give immediate feedback on student homework, by correcting errors on the student's screen. By linking Hypercard to a duct telephone line one can duplicate simply many of the possibilities of tele conferencing and computer conferencing.

In the development of open learning materials, Hypercard is a simple and powerful authoring package. It has advanced graphics, sophisticated voice production and sound production capabilities. It has extensive stacks of documentation, both data and graphics. Hypercard can scan into text any illustration in a book, a photograph, a slide or an OHT and can combine these illustrations electronically with text.

Students can learn WIMP environments in a very short time in comparison with other kinds of computer operating environments. Event driven WIMP environments mean that the student or user can alter the course of the program at any time, whereas normal writing of code offers each element of choice only when the person who wrote the programme chooses.



Debeauvais, M. (1976). L'universite ouverte: Les dossiers de Vincennes. Grenoble: Presses Universitaires.

Dubin, R. & Taveggia, T. (1968). The teaching-learning paradox. Eugene: University of Oregon.

Evans, T. & Nation, D. (1989). Critical thinking and distance education. London: Farmer.

Fay, P. (1988). Open and student centered learning: Evangelism and heresy. Journal of Further and Higher Education, 12(1),3-15.

Hodgson, V., Mann, S. & Snell, R. (Eds.) (1987). Beyond distance teaching - towards open learning. SRHE and OU Press.

Kaye, A. & Rumble, G. (1979). An analysis of distance teaching systems. Milton Keynes UK: Open University.

Keegan, D. (1990). Foundations of distance education. London and New York: Routledge.

Kone, H. & Jenkins, J. (1990). The programme of educational television in the Ivory Coast. Educational Media International, 27(2), 86-93.

Lewis, R. (Ed.) (1984). Open learning in action. CET.

McKenzie, O. et al. (1975). Open learning. Paris: UNESCO.

McNeil, D. & Wall, M. (1985). The University of Mid America: An analytic perspective. ICDE Bulletin, 2, 34-51.

Paine, N. (Ed.) (1988). Open learning in transition: An agenda for action. London: Kogan Page.

Peters, O. (1967). Distance teaching and industrial production: A comparative interpretation in outline. In D. Sewart, D. Keegan and B. Holmberg (Eds.) (1985), Distance education: International perspectives, 95-113. London: Routledge.

Rumble, G. (1982). The cost analysis of learning at a distance: Venezuela's UNA. Distance Education, 4 (2), 101-131.

Rumble, G. (1987). Why distance education can be cheaper than conventional education. Distance Education, 8(1), 72-94.

Thorpe, M. & Grugeon, D. (1987). Open learning for adults. Harlow, Essex: Longman UK.

Tiffin, J. (1989). The failure of success and the success of failure. Educational and Training Technology International, 26(2), 136-140.

Whitlock, Q. (1989). Student failure in open learning. Educational and Training Technology International, 26(2), 141-144.

Please cite as: Keegan, D. (1990). Open learning: Concepts and costs, successes and failures. In R. Atkinson and C. McBeath (Eds.), Open Learning and New Technology: Conference proceedings, 230-243. Perth: Australian Society for Educational Technology WA Chapter.

[ OLNT'90 contents ] [ EdTech Confs ] [ ASET Home ]
This URL:
© 1990 The author and ASET WA Chapter.
Last revised: 30 Apr 2003. HTML editor: Roger Atkinson
Previous URL 12 Jul 1998 to 30 Sep 2003: