Measuring transactional distance in web-based learning environments: an initial instrument development

From the ironically named Taylor & Francis journal ‘Open Learning’ (which is closed), an interesting attempt to come up with a means to measure transactional distance. Regular readers will know that I am a fan of Moore’s theory of transactional distance, a systems theory that explains some of the central the dynamics of educational systems and that can be extremely valuable in both designing and predicting the effects of distance learning, but that is susceptible to multiple interpretations and that is fuzzy around the edges. Coming up with a reliable instrument to measure it would therefore be quite useful.


“This study was an initial attempt to operationalise Moore’s transactional distance theory by developing and validating an instrument measuring the related constructs: dialogue, structure, learner autonomy and transactional distance. Data were collected from 227 online students and analysed through an exploratory factor analysis. Results suggest that the instrument, in general, shows promise as a valid and reliable measure of the constructs related to transactional distance theory. Potential refinement of the instrument and future research directions are included at the end of the article.”

There’s lots of good discussion of previous work in this paper and some fair attempts to dismantle the mechanisms and meanings of transactional distance, as well as a good research process capable of revealing some interesting insights. However, I am unconvinced by some of the very basic assumptions, so the instrument remains a bit blunt. I am a bit disappointed that one of my papers is cited for its minor criticism of the fuzziness of the theory, but the authors do not consider the major point of the paper (and a solution to much of that fuzziness) that the fundamental dynamic of transactional distance is concerned with control. I have a very strong suspicion that they might have found far more useful things in this study if they had explicitly taken that on board and tried to examine the exchange of control in the system.  Instead, they got caught in the well-known trap of seeing autonomy as a personal and unsituated characteristic, and made rough assumptions about structure/dialogue that take no account of the scale (or, as the late John Holland would have more accurately put it the boundaries) of the systems being looked at. These are not separate or separable categories – the dynamics shift according to where and when you place the boundaries. They would also have benefitted greatly from considering the various presences in the community of inquiry model, which would have made it easier to lose that very arbitrary one-to-one correspondence of teacher, student and content roles that constrains the model in quite artificial ways. Teachers are also other students, writers of content, and the creators of the surrounding physical and organizational environment. Again, the boundaries are not fixed, nor are they mutually exclusive. The most disappointing thing, though, is that that their initial hypotheses about the nature of transactional distance (which is, after all, what it was supposed to be about and that might have been a really valuable contribution, if validated) got completely lost in the process. The one thing that they really needed to show is the one thing that they did not. This is not a bad thing at all, and it is a discovery that is worthy of discussion. However, that is not quite how they see it:

“Transactional distance included learner–instructor transactional distance and learner–learner transactional distance. The original closeness, shared understanding and perceived learning did not merge; yet, the related items merged into the learner–instructor transactional distance and learner–learner transactional distance, respectively.”

This rather begs the question – if their initial model was not correct, what is that transactional distance that they are talking about and that they are attempting to measure? Their initial model, though fuzzy, was interesting and based on some thoughtful analysis but, in the final model, all they have done is to say that there are two different kinds of transactional distance depending on whether you are a learner or a teacher, without saying what they are, coming up with a sweeping sub-categorization that is just an artefact of the initial assumptions.  I think another closely related part of the problem is that they assumed at the start that transactional distance is in some way additional and separate to structure, dialogue and autonomy, rather than strictly following Moore’s meaning that it is a function of them. Their worthwhile attempt to analyze it further, by unpicking aspects of that, turned out to be fruitless because the aspects they picked were not the right ones.

This is not to suggest that the results are valueless. Far from it. This is a nicely conducted study that models a little of the complexity of learning transactions in a useful, if fuzzy, way, that explores the various meanings of transactional distance expressed in the literature pretty well, and, as well, helps to show some relatively unfruitful lines of enquiry. It’s just that it doesn’t meet the objectives set out in its own title, and it does not really do much to reduce the fuzziness of the construct that is the main problem that it set out to solve.

Address of the bookmark:

I am a professional learner, employed as a Full Professor at Athabasca University, where I research lots of things broadly in the area of learning and technology and teach mainly in the School of Computing & Information Systems, of which I am the Chair. I am married, with two grown-up children, and live in beautiful Vancouver.

Leave a Reply