Anthony Pym


_Video material

Learner-Centered Distance Education: a literature review of the ‘no difference’ phenomenon 


Carmina Fallada Pouget and Anthony Pym 2000   
Paper written for a publication that has something to do with the Euroliterature project.  


Abstract: It is commonly argued that distance-education programmes are the future of higher education. This has led to considerable debate, both for and against. Some argue that it is not worth spending huge amounts of money on infrastructure, software, training and technological pedagogy. Reviewing the literature on these issues we find three basic issues: for or against qualitative research, the ‘no significant difference’ debate, and criticisms of serious flaws in research design. It is concluded that more qualitative approaches are necessary and that the object of research should be not just the various media involved in ODL, but specific social sets of goals, tasks and strategies. 
Keywords: Keywords: Open and Distance Learning, no-significant-difference, educational research methodology, electronic tools.

Open and distance learning (ODL) takes place when teachers and students are separated by physical distance and technology is used to bridge the physical gap between them in relation to the third component, namely ‘learning’ (Sánchez 1997: 11). Online courses reach wider audiences and give people the chance to study at home at any time. More important, their ‘openness’ is leading to what is called ‘internationalised education’, addressing not culturally and linguistically homogenous populations but people with different social, cultural, and economic backgrounds. These are all good reasons for pursuing ODL. 

However, it is too commonly assumed that the ODL learner is essentially the same person as the traditional face-to-face learner. This may not be the case. In a survey conducted by the Opinion Research Corporation on 667 American working adults, it was found that 50% of the respondents thought the great advantage of online courses was the chance to work from home. As has been pointed out by Willis (1995: 1), most distance-education students are older and have jobs and families. Their schedules do not allow them to attend classes on-campus. They look for other ways of taking up courses that will qualify them for better jobs. Distance education programmes thus provide a new interactive means of overcoming time and distance to reach these learners. For all these reasons, the users of ODL may quite probably be very different from the young students that populate our bricks-and-mortar university classrooms.
This difference in the learner population may be a key factor when teachers and administrators attempt to assess how effective their courses are (Knobloch 2000: 3). Do ODL courses really meet the students’ needs? Do they work better than face-to-face courses? If distance learning provides no significant advantage in learning achievement, why should capital be invested in technologies that support distance learning? 

In order to answer these questions, research is being carried out to test if students in a virtual classroom can learn as much as students in a traditional classroom. Distance educators are able to choose from a wide range of technological options but they have to know which ones are best for a particular course. In order to ensure that the technology chosen enhances the quality of education rather than degrades it, it is crucial that there be some quantitative data on how well electronic tools fulfil the students’ needs. Further, is also seems advisable for instructors to develop new skills in instruction strategies, methods of teaching, teacher-student interaction, feedback, and evaluation (Knobloch 2000: 2). 

This bundle of questions has given rise to a rich body of research and debate, mainly in the United States and almost always in online form. The following is a review of the literature under four headings: student distress, the ‘no difference’ argument, the ‘significant difference’ reply, and the ‘no more research’ issue. We will attempt to situate our own research projects with respect to each of these points. 

Studies of student distress

The issue of student distress in ODL ensues from attention to ‘student-centred learning’ in the 1980s. An example of research in this vein would be ‘Students’ Distress with a Web-based Distance Education Course’, carried out by Noriko Hara and Ron Kling in 1997 (available online in a 2000 rewrite), which consequently became the object of some debate in the American context.
Hara and Kling’s originally ethnographic study involved qualitative research on a web-based course on educational technology designed for university students and taught by a recent graduate who had previously done the same course as a student. The researchers’ original question was ‘how and how well do the students in this course manage their feelings of isolation in a virtual classroom in order to create the sense of a community of learning’. However, as the research proceeded it was found that isolation was not the major problem. Instead, ‘recurrent experiences of other types of distress such as frustration, anxiety and confusion seemed to be pervasive’. Direct observations and interviews with the students showed that their main problems were actually with computer technology, communication breakdowns, and uncertainty about what was expected of them. The teacher was found to be relatively enthusiastic but inexperienced; she tended to assume that the students could handle all the communication problems they encountered. In particular, it was found that email may not be the answer to all communication problems, since the lack of immediate feedback makes it difficult for many students to distinguish between an ironic and an angry teacher. The researchers concluded that more attention had to be paid to training teachers for ODL technology, and that serious work should be done adjusting the various media to specific educational purposes. Much of the enthusiasm for ODL was, for these researchers, clearly premature. 

Hara and Kling’s paper has given rise to various defences of ODL methodology (summarized in recent versions of their paper). It was argued, for example, that one case study cannot be representative, that the teacher involved was exceptionally naïve, and that there was no reason to extend the findings to the whole of ODL. 

Our own research in this area does, however, give some credence to Hara and Kling’s position. In a tandem-email project involving Spanish and American learners, some students expressed uneasiness with what they saw as the external obligation to produce email messages (a reaction noted by Hara and Kling). Others were looking for extremely explicit instructions about how to use Blackboard websites and felt frustrated when they were left to discover things by themselves. Yet others were immediately concerned with the way ODL activities were to be assessed for academic credits, and were disappointed when there was no clear answer. We have thus had some students withdraw from our programme or simply ‘go very quiet’. 

It can only be concluded that there is a real need for qualitative as well as quantitative research in this area, and that Hara and Kling’s focus on students’ distress should be treated with considerable respect. 

The ‘no difference’ argument

Perhaps a more significant position in the literature is occupied by the ‘no difference’ debate, basically concerned with the idea that face-to-face learning and ODL achieve much the same results. Thomas L. Russell (1999) provides a compilation of 355 reports, summaries and papers on this general issue. Summaries of those texts are available online (see References below). For example, a 1928 doctoral dissertation on ‘Correspondence and Class Extension Work in Oklahoma’ found ‘no differences in test scores of college classroom and correspondence study students enrolled in the same subjects’ (cit. Russell 1999). The dates of the texts range from 1928 to 2000, and the very quantity of the research enables Russell to argue that, contrary to the standard research position (everyone concludes that further research is necessary), a significant body of ODL research has already been accumulated. 
It is interesting to note that, when sorted by whether the papers are for or against the ‘no significant difference’ position, only in nine years (all since 1991) has there been serious support for the argument that there is a ‘significant difference’.

The ‘significant difference’ argument
In some experiments carried out in the 1990s it was found that students learn better online and that there is indeed a significant difference between students who attend classes on campus and those who learn in virtual classrooms. These studies point out that students who attend classes online should previously be given some training on how to use technological tools. They argue that students in virtual classes tend to spend more time working with each other and that this collaboration leads to greater student-student interaction and better results (see, for example, Schutte 1996; Black 1997).  They also suggest that some students feel less intimidated in online classes than in face-to-face situations because chat sessions or email exchanges allow relative ‘anonymity’ (McCollum 1997). For such reasons, one might legitimately expect ODL learning to be quite unlike traditional face-to-face experiences. 

Both sides of this debate nevertheless seem limited in that they focus on the students’ results, without paying great attention to the kinds of issues raised by Hara Kling. The arguments are thus mainly over quantitative criteria (what should be assessed? how should it be assessed?), rather than with quantitative in-depth case studies, which in any case require methodologies unsuited to direct comparisons. 

A second obvious shortcoming in the ‘no significant difference’ debate is that there is no guarantee that we are dealing with comparable students in the first place. As has been pointed out above, the students who take ODL courses are often older, at work, or taking a degree for non-vocational reasons. This might be contrasted with the traditional university degree, which still retains many social virtues as the ‘world’s greatest youth camp’, as opposed to ODL as offering a highly developed ‘intellectual shopping mall’ (to sum up the general positions of O’Donnell 1998). For example, the Catalan Open University recently calculated that, on average, it would take each student some 15 years to complete an undergraduate degree... which can only mean that most of their ODL students are not actively studying for a degree. This in turn means that methodologies based on comparing the two teaching methods are fatally flawed from the outset, since very different student populations are involved. 
The ‘no more research’ argument

While controversy on the efficiency of ODL still continues, some have found significant cause to argue that the research carried out in this field is largely flawed and that the methodologies and conclusions are inadequate. In a 2000 paper on ‘Measuring Learning Effectiveness: A New Look at No-Significant-Difference Findings’, Joy and Garcia suggest that instructors should carefully interpret the results of the studies that compare different media, since fundamental design flaws abound. In particular, Joy and Garcia detect a common failure to control ‘time on task’ (how long students actually take to complete a task), and to constitute adequate experiment and control groups. Treatment periods (the time taken for the observations) were mostly too short, and it was frequently assumed that all instructors were equally able to teach in all media. In sum, many research projects failed to control major variables and were thus able to find whatever they set out to look for. Joy and Garcia further suggest that, given the range of complex variables involved, it may be wrong to ask which medium of instruction is the best. The more legitimate question would seem to be ‘what combination of instructional strategies and delivery media will best produce the desired learning outcome for the intended audience’. 
Such a conclusion would seem to contradict the position of Russell, who basically argues that there is already a significant body of research in this field. Clearly, there can be no question of going back and starting from scratch. Yet it is equally clear that many wrong questions have been asked, and many proposed answers are thus misleading. This, in turn, should provide a strong argument for pursuing research on a revised footing. 


From our brief survey of the copious literature available on ODL research, we retain the message that qualitative research is uniquely suited to the complexity of the many variables involved, whereas a purely quantitative focus on assessing student performance is likely to lead into cul-de-sac dialectics.

Our second conclusion is that much must be done to assess the impact of having different students and different instructors involved in ODL activities. Since these human variables are notoriously difficult to control, their diversity must be accepted before it can be countered. In many cases, a period of controlled prior training, for both instructors and learners, may be necessary before research as such can begin.

Third, and in the same vein, researchers must be prepared to look at packages of media and strategies, rather than at isolated ‘pure’ forms of ODL and its negation. This especially concerns focusing not just on the media but on the specific goals to be achieved for each task. To take a simple example, a brainstorming session might seem suited to chat exchanges, whereas an activity aimed at improving formal written language would be better suited to email or even pen-and-paper. In most cases, however, there is much collateral learning going on. The chat session can teach students a great deal about language use (usually more than about whatever content is involved), and the greater reflection time involved in email should improve the quality of the ideas exchanged and thus the conceptual focus of any brainstorming. Either way, it is the combination of medium and goal that must be assessed.

In sum, given the complexity of this domain and the many doubts raised by much previous investigation, the best research projects may be the actual teaching activities themselves, to be assessed interactively by the teachers and students involved. In questions of ODL, there can perhaps be little fundamental difference between doing it and studying it. 


Hara, Noriko & Kling, Rob. (2000). ‘Students’ Distress with a Web-based Distance Education Course: An Ethnographic Study of Participants’ Experiences’.

Joy, Ernest H. II, & Garcia, Federico. (2000). Measuring Learning Effectiveness: A new Look at No-Significant-Difference Findings. Journal of Asynchronous Learning Networks. 

Knobloch, Neil. A. (2000). Distance Learning: Is It Working? 

McCollum, Kelly (1997). ‘Students Taught Online Outdo Those Taught in Class’.

O’Donnell, James (1998). Avatars of the Word. From Papyrus to Cyberspace. Cambridge MA, London: Harvard University Press.

Russell, Thomas L. (1999). The ‘No Significant Difference Phenomenon’. Fifth edition. North Carolina State University Office of Instructional Telecommunications. 

Sánchez-Mesa, Domingo, ed. (1997). Crosscultural and linguistic perpectives on European open and distance learning. Granada: Universidad de Granada.

Schutte, Jerald. (1996). The Intellectual Superhighway or Just Another Traffic Jam?

Willis, Barry. (1995). Strategies for Learning at a Distance. In Distance Education at a Glance: A Practical Guide.

Last update 26 December 2000  

© Anthony Pym 2014
URV. Av. Catalunya, 35
45002 Tarragona, Spain
Fax: + 34 977 299 488