Previous Chapter Chapter 6: Reshaping Rhetorical Space: E-learning through Online Asynchronous Discussion Next Chapter
The theme of this chapter is asynchronous online discussion (AOD) – more commonly known previously as an asynchronous bulletin board (ABB) – and its effectiveness as a medium for learning. Starting with a canonical poem as illustrating the notion of rhetorical space allows one to view this digital medium in a wider context and involves a number of important recognitions. The first of these is that a speaker is always, as Bakhtin noted, a respondent:
He [sic] is not, after all, the first speaker, the one who disturbs the eternal silence of the universe. And he presupposes not only the existence of the language system he is using, but also the existence of preceding utterances – his own and others’ – with which his given utterance enters into one kind of relation or another (builds on them, polemicizes with them, or simply presumes that they are already known to the listener). Any utterance is a link in a very complexly organized chain of other utterances. (1986, p. 69)
As Bakhtin further noted, the notion of being a respondent not only involves a relationship to ‘preceding utterances'; it also includes a relationship to those to whom one's utterance (written or spoken) is potentially addressed. We have moved here beyond the literal space of the Greek forum, to think of the rhetorical space as metaphorical – as having a temporal or historical dimension. This first recognition, then, is about audience, even though members of this audience may no longer be living. The second recognition, again using a spatial metaphor, recognizes the discursive positions available historically in relation to the topic that is subject to discussion or debate.[Page 104]
If space is rhetorical, it can also be thought of as epistemic. As Payne points out:
Space creates frameworks for conception, action, and interaction; its design – whether natural or artificial – limits and directs what we think and do, as well as with whom we do it. Space is not a neutral conduit within which social productions occur; it is itself socially produced, and as such it is shot through with the very ideologies of identity and power with which much of our disciplinary work contends. (2005, p. 485)
An asynchronous online forum, as an example of such a socially produced space, inevitably operates to foster or constrain discursive activity. In Gee's terms, it acts to elicit particular ‘ways of behaving, interacting, valuing, thinking, believing, speaking … that are accepted as instantiations of particular roles … by specific groups of people’ (1996, p. viii). Participants in an asynchronous online discussion, then, are not entering a neutral arena. Rather, they are entering a space which is value-laden and, for some, potentially intimidating.
Using a search engine like Google, it is an easy matter to enter cyberspace and download a facsimile of ‘On first looking into Chapman's Homer', the poem the English poet John Keats showed his friend, Charles Cowden Clarke, one morning in October 1816. He had walked through two miles of physical space to deliver the poem, which was published, in slightly edited form, the following December.
Digitally mediated cyberspace was not available to Keats, but rhetorical space certainly was. There are ample references to places in this poem: ‘realms', ‘states', ‘kingdoms', ‘islands'. The last eight lines read:
Yet did I never breathe its pure serene
Till I heard Chapman speak out loud and bold:
Then felt I like some watcher of the skies
When a new planet swims into his ken;
Or like stout Cortez when with eagle eyes
He star'd at the Pacific—and all his men
Look'd at each other with a wild surmise—
Silent, upon a peak in Darien.
In these lines, which pivot on the qualifier ‘Yet', there is a sense of space opening up – pure, undifferentiated extension – like a blank frame into which something wondrous (a ‘new planet', a new ocean, a new poetic) is about to enter and change forever the status/state of pre-existing knowledge. It is clear for Keats that this space is largely textual and that he is a respondent, not just a seer but also a hearer, seeing himself as engaging with time-honoured bards – Homer and [Page 105]especially Chapman himself, whose textual Homer is a very special link in a long, intertextual chain (that also includes Alexander Pope). It is a chain which includes, by implication, texts such as Robertson's History of America (published in 1777), faultily remembered by Keats who confused Cortez with Balboa. Furthermore, this chain includes the initial addressee, Clarke himself, as well as the unseen but anticipated audience with whom Keats wanted to share his ‘teeming wonderment’ at reading Chapman.
Starting with Keats is a way of suggesting that everything and nothing is new under the sun. Morten S⊘by quotes Freud – ‘Man has, as it were, become a sort of prosthetic God’ (Freud, 1962, p. 38, cited in S⊘by, 2005, ‘Formatting Cyberspace', para. 6) – in suggesting that ICT can be thought of in prosthetic terms, as an extension of the body and the senses. As a way of gaining perspective, however, it is sometimes necessary to de-heroize the current moment and think of all technologies as prosthetic (including Keats’ pen and paper). And before we pronounce on multiply-selved cyborgs at play in technologically mediated and constructed virtual worlds as something new, we might recognize that there is a sense in which the human imagination has always been a virtual capability.1 You could say that in or through his poem, Keats enters/constructs a virtual world, which effectively enables him to try on a new self.
A corollary is that technology is not just an add-on that enhances human cognition. As Walter Ong (1982) argues in respect of writing, technology has the power, directly and indirectly, to shape human thought processes (consciousness). A technology, then, is more than just an aid to learning. It shapes the cognitive processes that underpin learning. Furthermore, because the uses of technology are culturally mediated, technologically mediated learning via bulletin board systems is necessarily shaped discursively by the practices around technology privileged in a particular cultural milieu. We inhabit a text-saturated world, which we negotiate via a repertoire of textual practices that are at once cognitive, social, and themselves technologies.
There are two dimensions of Bakhtin's chain metaphor, which are relevant to any consideration of a speaker's communicative activity in a rhetorical space, whether it be a Greek forum or the digitally constructed space of an asynchronous online forum. These are reach (chains vary in length) and connection (the links in a chain can connect in differing ways).
There are two aspects to the dimension of reach: field2 and company (or membership).
In respect of field, there are two aspects (or axes):
- Depth of historical field: this refers to the temporal scope of intertextual and interdiscursive historical reference, both retrospective and anticipative. This is the historical axis of reference.[Page 106]
- Breadth of contemporary field: this refers to the range of intertextual and interdiscursive reference that can be thought of as roughly contemporaneous.
In respect of the aspect of company, we can distinguish three kinds:
- Overt company: those conversants or participants who engage in an utterance exchange and constitute a list of participants.
- Covert company: those who have opportunities to observe the exchanges of the overt company but whose presence will be unsuspected. This includes site administrators for an asynchronous learning network but also extends to ‘lurking’ members of an online discussion forum who are enrolled on a course but who choose to be non-participants.
- Implied company: those conversants or ‘addressees’ whose ‘presence’ is implied in a particular utterance.3 The writer of an article on a course reading list would be an example of this. However, should that writer be invited to participate in the discussion as a guest, he or she would become a member of the overt company.
The term connection is used here to embrace various aspects of participant activity within a rhetorical space, however ‘natural’ or technologically mediated.
The first of these relates to duration and continuity. Duration refers to the real time taken up by a discussion and is marked by a beginning and end point. In terms of an asynchronous online discussion, duration is typically marked by an imposed beginning date and an imposed completion date. Continuous discussion, in a face-to-face (FTF) classroom setting, for example, has a beginning and end point in real time and no breaks. Continual discussion, on the other hand, has a beginning and end point in real time and is intermittent. An asynchronous online discussion is an example of continual discussion.
The second and third aspects relate to an individual member's participatory behaviour. The second relates to a participant's participation rate:
- Absolute participation rate: this is the number of utterances a participant makes in a single discussion.
- Relative participation rate: this is the number of utterances a participant makes as a percentage of all utterances in a single discussion.
The third aspect relates to the concept of feedback, a verbal or non-verbal signal that acknowledges an utterance:
- Feedback spread: this is the number of participants in the overt company to whom a particular participant offers unsolicited feedback, expressed as a percentage of the number of participants in the overt company minus one.
- Feedback rate: this is the total number of feedback instances a participant produces as a percentage of his or her total number of utterances in a discussion. A percentage of more than 100 would indicate that a participant is at times acknowledging the contributions of a number of participants within a single utterance.4[Page 107]
A fourth aspect of connection has to do with ways in which the overt company collectively addresses the topic of a discussion:
- Degree of convergence: this refers to the extent to which the overt company appears to be achieving a consensus on a particular topic, that is, a kind of discursive alignment.
- Degree of divergence: this refers to the extent to which the overt company appears to be failing to realize a consensus on a particular topic, that is, a kind of discursive non-alignment. Some researchers (e.g. Hew and Cheung) use the term ‘dissonance of ideas’ (2011, p. 313).
- Degree of congeniality: this refers to the extent to which the overt company appears comfortable with divergence (Locke and Daly, 2007).
Neither convergence nor divergence are per se desirable outcomes of a discussion. However, different discourses of learning vary in respect of their valuing of either of these. Rhetorical spaces differ in terms of their tolerance and/or encouragement of dissent. As Payne puts it, ‘hegemony inherent in the production of space functions by degree', so that whether a space is totalitarian or dialogic will depend on what he calls the ‘margins of maneuver', the extent to which the discursive structures embedded allow for resistance and agency (2005, p. 489).
Finally, structuration refers to the logic or principles governing the sequence and inter-relationship of utterances within a discussion. It is clear that the factors that govern turn-taking and cohesion in an online asynchronous discussion differ from those that operate in synchronous, face-to-face discussion. These factors may be teacher-initiated design features that impact upon sequence. They may relate to the interactive behaviour and predispositions of various members of the overt company. They may relate to aspects of the platform itself. It is probable that at least in part the structuration of such discussion is characterized by rhizome-type connectivity (Deleuze and Guattari, 1987), described as follows by Semetsky:
Rhizome as embedded in the perplexity of the situation, going in diverse directions instead of a single path, multiplying its own lines and establishing the plurality of unpredictable connections in the open-ended, what Deleuze called smooth, space of its growth. (Semetsky, 2003, p. 18)
Putting it another way, considerations of the structuration of asynchronous online discussion need to be posited on the co-existence of both hierarchical and non-hierarchical principles of order/disorder.5
The preceding section enters into dialogue with Nicholas Burbules’ (2002) notion of the Web as a rhetorical place. Burbules’ prefers ‘place’ to ‘space’ because the former is socially meaningful:[Page 108]
It has an objective, locational dimension: people can look for a place, find it, move within it. But it also has a semantic dimension: it means something important to a person or a group of people, and this latter dimension may or may not be communicable to others. (2002, p. 78)
Burbules identifies two broad kinds of strategies whereby spaces become transformed into places: mapping (the development of ‘schemata that represent the space, identify important points within it, and facilitate movement within the space') and architecture (the shaping of space via ‘enduring structures') (2002, pp. 78–80). The dimensions of reach and connection, discussed above, can be seen as one way of representing the transition from space to place.
Writing about distance education in 2005, Natriello produced a range of statistics to underline its burgeoning growth, fuelled (as he argued) by a general growth in demand for education, especially among the young, and by the perceived advantages in online learning itself.
- In the 2000–01 academic year, 56% of 2- and 4-year degree-granting institutions of higher education in the USA offered collectively 127,400 different distance education courses.
- 90% of US institutions reported the use of asynchronous course delivery, 43% reported synchronous internet courses, 51% reported using two-way video and audio and 43% reported using one-way video.
- Half of China's 92,000 engineering and technology graduates obtained degrees through distance learning. (2005, p. 1886)
More recently, Allen and Seaman, reporting on changes in online education in the USA between 2002 and 2012, note that ‘the proportion of all students taking at least one online course is at an all-time high of 32.0 percent’ and totalled 6.7 million students (2013, p. 4).
It should be noted that distance learning and online learning are not the same. Many courses taught in traditional ways in traditional settings are hybridizing and incorporating online features without being fully online (Allen and Seaman, 2013). In respect of the advantages of distance learning, Natriello offered the familiar trio of: the collapse of distance as a deterrent (‘space shifting'); 24/7 program access (‘time shifting'); and a reduced need for physical structures and elaborate locations (‘resource shifting') (2005, p. 1888). In Allen and Seaman's report, the percentage of academic leaders who believed that learning outcomes for online learning matched or were better than face-to-face [Page 109]instruction stood at 77%, up from 57.2% in 2003 (2013, p. 5). In addition, there is a growing trend for Web 2.0 technologies to be utilized to enhance online learning (Diaz, 2011). However, as this chapter will discuss, there are some stubbornly persistent challenges to making such media as asynchronous online discussion work.
A ‘big picture’ view is useful for contextualizing a discussion of the affordances for learning of a particular technology, in this case an asynchronous online forum, whether this be embedded in a totally online, asynchronous learning network6 or used as an adjunct to a traditional, face-to-face course. Just how such affordances might be productively realized in practice is contingent on a range of factors. The nature of this contingency is reflected in Lankshear and Knobel's statement that:
The historical and contingent nature of discursive practice becomes readily apparent within cyberspaces. To experience oneself as engaged with others in constructing, refining and monitoring social practices which comprise amalgamations of reading-writing-imaging, values, purposes, theories, roles, identities, etc., is necessarily to envisage one's activity as simply so many representations of what social practice(s) might be, and to be aware of alternative possibilities within and outside cyberspace. It is to realize there is nothing necessary about inhabiting any particular discursive space or spaces: all such spaces are contingent and historical. This insight is central to meta-level awareness of discursive practice and the possibilities for transformative praxis predicated upon it. (1997, p. 158)
Payne's (2005) analysis of course-management software, with a particular focus on Blackboard, is a reminder of the importance of viewing any educational technology in relation to the wider, socio-contextual picture.
This picture includes recognizing that the production of Web-based software for institutions developing an online learning environment is big business. In an earlier version of this chapter (Locke, 2007), I drew attention to a press release dated 12 October 2005 announcing that Blackboard, a US company, had acquired WebCT, a software company developed in Canada but now with American owners. The announcement claimed that this combined enterprise would have a client base of more than 3,700 institutions, and termed this grouping a ‘global e-Learning “Community of Practice”'.7 In 2011, Blackboard was acquired by Providence Equity Partners, a company that claims on its website to have 37,000 clients worldwide. Meanwhile, open-source Moodle claims to have 64,385 registered sites and over 72 million users worldwide.8 Typically, the websites of these software production companies feature success stories. The modus operandi of an asynchronous [Page 110]bulletin board, featuring as part of an online learning environment, cannot be separated from the (global) marketing ambitions of companies/collectives such as Blackboard and Moodle.
Drawing on Foucault's concept of the ‘panopticon', one can view the ‘covert company’ present in a rhetorical space as those groups and interests which have managed to obtain a kind of hegemonic sway over the discursive practices permissible in an online learning environment in general and an asynchronous bulletin board in particular (1977, p. 195). In this sense, power is both diffuse and productive. In his discourse analysis of a Blackboard space, Payne drew on the geographical metaphor of the suburb and noted a tendency towards homogeneity, conformity and acquiescence to an American middle-class ideal (calling to mind the film, Edward Scissorhands), the embodiment of a ‘dominant cultural rhetoric'. ‘Blackboard', he argued, ‘serves the interests of the dominant culture by naturalizing and commodifying its space/rhetoric’ (2005, p. 495). Of particular note in his analysis was his attention to issues of surveillance and control and his insistence on the need to scrutinize the ‘degree of agency enabled’ by spaces such as Blackboard (2005, p. 499, italics in original).
In considering the overall system or framework within which an online delivery system is embedded, it is useful to distinguish between front-end users, platform administrators (for example, technical support staff in a school or faculty) and engine designers, developers and maintainers. Generally, both technical support staff and system administrators are bound by a code of conduct and are required to enter a formal verbal agreement in respect of their role vis-à-vis asynchronous discussions. Both groups can access a particular discussion if there is reason to believe that it is being improperly used or that there is a problem which may be affecting the performance/users. Both are also members of the ‘covert company', the presence of which implies a prospect for direct surveillance (in a less Foucauldian sense).
Issues of access, control, and therefore power differentiation, are necessarily implicated in asynchronous bulletin-board design. It is the course host (typically a course coordinator) who governs access to an online discussion in a class forum. It is also typically the course coordinator who can accord students the space/place to host a discussion of their own, perhaps as an assignment. So there is clearly an imbalance of power here. Course coordinators are also frequently afforded the ability to allow access to particular guests. This allows a lecturer to provide guest access to a colleague from another university, so that the colleague can participate in a particular discussion. In short, these design features privilege some over others in respect of access and have implications for ways in which the ‘overt company’ is constructed. Unsurprisingly, the academic literature related to asynchronous online discussion is more likely to focus on operational aspects than issues of power, control and surveillance.[Page 111]
How important is the duration (in real time) of a single asynchronous online discussion? Does a longer duration affect rates of participation or categories of response? On the basis of a cross-case analysis of nine naturalistic case studies of online classes, Dennen suggested that duration was a factor in increasing the number of instances of dialogue between participants. However, she wondered whether the time interval between message and response meant that the original sender would be unlikely to read the delayed response (2005, p. 136).
In a study of duration effects, Im and Lee (2004) mapped changing patterns of interaction over time among their student participants, using, for analytical purposes, five categories of content (topic-related, learning-related, related to discussion management, related to social interaction, and technical management-related) and developed a model for staged development in an online learning community:
- S1 Social bond formation (‘the first stage of learning community development, where participants introduce themselves and get to know each other');
- S2 Information sharing (‘where participants feel comfortable in exchanging and sharing knowledge and information');
- S3 Advanced stage (‘where participants apply advanced metacognitive skills such as awareness, reflection, and evaluation'). (Im and Lee, 2004, p. 158)
They found that while synchronous discussion was more useful in promoting social interaction, asynchronous discussion was more useful for task-oriented communication. Indeed, the synchronous discussion content failed to lead to ‘meaningful learning … but remained at the level of the social bond formation stage as the course progressed’ (2004, p. 166). While duration appeared to have little impact for synchronous discussion, it was clearly a factor for asynchronous discussion, with stage 2 postings (information exchange) decreasing over time and stage 3 postings (advanced stage) becoming more prevalent by the conclusion of the discussion (2004, p. 163). A more recent study by Hew and Cheung, however, found no correlation between duration and ‘higher level knowledge construction occurrences’ (2011, p. 303). However, they did speculate that the disparity between this finding and earlier research might be due to ‘the complex interplay of other factors such as facilitation techniques’ (2011, p. 315).
A number of researchers have explored the advantages and disadvantages for learning of synchronous (continuous) and asynchronous (continual) discussion. For instance, Barnett-Queen, Blair and Merrick (2005) found that most of their participants rated both the quantity and quality of online discussion more highly than traditional face-to-face discussion, and regarded the former as enabling a greater understanding of course content. A number of studies comparing asynchronous online discussion with face-to-face discussion in classrooms suggest collectively [Page 112]that ‘the critical success factors for asynchronous collaborative learning may be different than in FTF environments’ (Heckman and Annabi, 2005, ‘Background', para. 2). Markel, drawing on a case study involving students in a graduate course on ‘Professional Problems of Teachers', argued that in slowing down students’ time for reflection, asynchronous discussion fostered a process of deep learning through acts of writing free from ‘the tyranny of the ever present “now” of the face-to-face classroom’ (2001, ‘Summary', para. 1). Similarly, Vess, comparing the impact of asynchronous discussion in a fully online course with its impact in a face-to-face course enhanced with asynchronous discussions, found students stating a preference for online debating because they found it easier to participate and it allowed research and preparation time prior to posting a message (2005, p. 360).
Other studies concern themselves with comparing synchronous and asynchronous discussion online. Lim and Tan, based on findings from a case study using a bulletin board as the basis for a focus group discussion, argued that asynchronous discussion allowed more reflection time, more opportunities for discussion management and more time for conflict resolution (2001, p. 58). Poole (2000), drawing on findings from a case study involving 14 students enrolled in a two-unit graduate course on ‘Social Perspectives of Technology in Education', reported that students ‘preferred the more time-independent communication facilitated through the bulletin board’ (2000, p. 175).
A number of research findings suggest that this time-independent aspect of asynchronous online discussion has the potential to mitigate the effects of certain inequalities. Researchers such as Sorensen and Baylen (2004) and Hlas, Schuh and Alessi (2008) have shown how an online discussion environment can facilitate the participation of non-native speakers because there is more time to both articulate and respond to the ideas of others. In their case study of an online course involving both Pākehā (European) New Zealanders and Chinese nationals, Locke and Daly found that a large majority of students said they found it easier to express their views in an online discussion than in a face-to-face classroom situation. All Chinese students on the course (taught in English) took this position, with one commenting: ‘In the Class Forum, non-native English speakers can express themselves better than face-to-face learning because online learning give [sic] them time to properly put their ideas into English’ (2006, p. 6).
In setting up their study of 40 pre-service teacher education students, Im and Lee were mindful of studies suggesting that in traditional classroom settings, male students appear to dominate discussion (2004, p. 157). Over a 13-week period, they gathered and analysed 336 asynchronous discussion postings from 39 students, and 2820 synchronous postings from 21 students (across five separate discussions). In respect of gender, they found female students to be more active than males in online discussions, suggesting that a ‘more egalitarian atmosphere’ and ‘social distance’ had lessened male dominance (2004, p. 166). However, they also found differences between genders in respect of initiating postings – males preferred to initiate postings than respond to the postings of others (2004, p. 162).[Page 113]
What motivates students to participate in an asynchronous online discussion? Hew, Cheung and Ng (2010) reviewed 50 empirical studies in order to identity factors inhibiting student participation in asynchronous online discussion, and identified seven main factors:
- not seeing the need for online discussion
- the behaviour of other participants
- personality traits
- keeping up with the discussion
- not knowing what to contribute
- lack of critical thinking skills or being content to merely answer queries
- technical aspects. (2010: 573)
Factor 2 included a lack of response from other students, and supported Pena-Shaff, Altman and Stephenson's (2005) finding (from a course-based case study involving 34 students) that feedback from other students was a major factor influencing participation and interaction. So too was feeling intimidated by certain kinds of student behaviour, and the perceived lack of interest and involvement of the instructor. The latter is consistent with findings by Swan (2001) and Dennen (2005) that dialogue is likely to be fostered when instructors are more involved (see next section).
Based on their review, Hew, Cheung and Ng developed a set of guidelines for enhancing participation corresponding to each of the above inhibiting factors. These were:
- Making online content curriculum-related; making the online discussion ‘mandatory’ or incentivizing it; spelling out purpose and expectations; and setting ‘deadlines’ or limiting time frames for participation’ (2010, p. 578);
- Instructor involvement, the use of ground rules and the use of ‘controversial topics’ (2010, p. 578);
- Using ‘note starters’ (e.g. ‘tell me why …') and ‘combining high- and low-profile students in the same group’ (2010, p. 578);
- Finding ways of representing discussion threads visually, and restricting the number of postings (e.g. one idea per message);
- Using ‘open-ended questions’ and scaffolding types of response (2010, p. 578);
- Modelling ‘Socratic questioning to enhance students’ critical thinking skills'; instructor modelling of a range of facilitation techniques; asking students to perform roles such as ‘summarizer’ (2010, p. 578);
- Ensuring students are prepared, and making use of ‘easy navigation functions’ (2010, p. 578).
Another issue here is the relative participation in discussion of teacher/tutor and student. In a content analysis comparison of learning processes between online and face-to-face discussions, Heckman and Annabi found that student-to-student [Page 114]interactions tended to predominate in asynchronous online discussions whereas instructor–student exchanges tended to dominate in classroom face-to-face discussions (2005, ‘Discourse process', paras 2 and 3). Vess's (2005) research with tertiary history students found, similarly, that ‘Students in the fully asynchronous class … were inclined to continue a discussion thread, while those in face-to-face discussions generally tended to respond only to the instructor's questions and not to each other.’ Her research also confirmed Heckman and Annabi's claim that ‘student-to-student interactions engendered by asynchronous discussion often demonstrate a great many cognitive indicators of involvement and elaboration’ (2005, p. 362). How and how often instructors should participate in an asynchronous online discussion is a vexed question and will be addressed in the next section.
The effectiveness of grades in encouraging participation is also a vexed question (Dennen, 2005), as is the how of assessment (Andresen, 2009). A finding from Goodwin, Graham and Scarborough's study is pertinent here:
The tutors and many of the active students argued for more marks to be assigned to the group work to force greater participation by all students. The question remains as to what percentage of marks should be allocated but there is no doubt that unless some marks are awarded, participation will be very low. (2001, p. 46)
But is the simple award of marks going to solve the problem of ensuring an adequate rate of student participation? Hammond's selective literature review found conflicting views, sometimes favouring summative assessment, sometimes suggesting that summative assessment increased the number of postings without improving their quality (2005, p. 16). Hew, Cheung and Ng also reported on studies suggesting that awarding grades for online discussion led to perfunctory participation, and concluded that ‘the mere giving of marks to increase contribution may not be a good strategy'. They wondered about using ‘an evaluation rubric that spells out different marks for different specific categories of contributions’ (2010, p. 582), a possibility I will return to in the next section.
On the basis of this study, then, one can conclude that asynchronous discussion in an online setting is better able to facilitate and support learning for a greater range of students than synchronous discussion. However, there are other aspects of rhetorical connection that have a bearing on learning: structuration, feedback and convergence.
As indicated previously, structuration refers to the logic or principles governing the sequence and inter-relationship of utterances within an asynchronous online discussion, and in this sense embraces aspects of feedback and convergence. [Page 115]One model for theorizing these principles is the community of inquiry (COI) framework, first developed by Garrison, Anderson and Archer (2000) and taken up by later researchers (e.g. deNoyelles, Zydney and Chen, 2014; Wang, 2015). In a stocktake of the literature couched in relation to their framework, Garrison and Arbaugh described the model (see Table 6.1) as having three dimensions or ‘presences’ – social, teaching and cognitive – each with its corresponding categories of activity and ‘indicators’ (2007, p. 158).
In terms of this model, social presence is ‘the ability of learners to project themselves socially and emotionally, thereby being perceived as “real people” in mediated communication’ (2007, p. 159). The focus here is on social values such as cohesion, collaboration, communication, trust, emotional sharing, and has a particular relationship to what I have called congeniality. Cognitive presence is defined as ‘the extent to which learners are able to construct and confirm meaning through sustained reflection and discourse’ and is operationalized as a process with four phases: (1) ‘a triggering event; (2) exploration; (3) integration; and (4) resolution’ (2007, p. 161), aiming at developing higher-level thinking. Cognitive presence is viewed as the most challenging for course designers and as calling for a particular kind of teaching presence, which enables ‘the design, facilitation, and direction of cognitive and social processes for the purpose of [Page 116]realizing … worthwhile learning outcomes’ – a presence viewed as having three prongs: ‘(1) instructional design and organisation; (2) facilitating discourse …; and (3) direct instruction’ (2007, p. 163). (For a comparable attempt at developing a research framework for analysing the structuring processes underpinning asynchronous online discussions, see Heckman and Annabi .)
Wang (2015) builds on studies influenced by the COI model, suggesting a new assessment presence he defines as ‘an ongoing process of evaluating and improving student learning using explicit and measurable performance expectations, criteria, and quality standards’ (2015, p. 600). Characterized by such activities as evaluation, focused feedback and feed forward, and the utilization of detailed and comprehensive rubrics, Wang's work – underpinned by a belief that effective assessment is facilitative of learning – can be viewed as an attempt to deal systematically with messy debates around the pros and cons of assessment in relation to asynchronous online discussions. ‘Assessment presence is characterized by the activities of evaluation and grading with constructive feedback and recommendations. The important indicators of assessment are rubrics with expectations, rules, protocols, and scoring and ranking criteria’ (p. 600).
Feedback, convergence and structuration, as characterizing patterns of interaction, operate among the overt company in a rhetorical asynchronous online discussion space in different ways from traditional classrooms (Andresen, 2009). Immediate and non-verbal feedback is not readily available to asynchronous online discussion participants. (Emoticons can be thought of as quasi-non-verbal feedback.) Yet it is clear that feedback patterns in asynchronous online discussions have a bearing on such things as convergence and structuration (Dennen, 2005). Locke and Daly identified feedback as a factor contributing to the success of an online course whose transcripts they analysed. For the students who participated in the course, the feedback spread averaged 80 and the feedback rate averaged 187, indicating that all students frequently incorporated multiple feedback instances in single utterances across a range of discussions. These measures, they concluded, equated with a high degree of responsiveness in all participants (2006, p. 7).
Sorensen and Baylen's (2004) study comparing patterns of communication of the same students from a ‘hybrid’ class across face-to-face (FTT) and asynchronous online discussion settings constructed a feedback typology with five categories to code types of communication:
- Initiating: for example, ‘stating an opinion or insight to get the conversation started';
- Supporting: for example, ‘sharing evidence to support a position';
- Challenging: for example, ‘offering different opinions';
- Summarizing: for example, ‘when a participant states in a concise way the essence of someone else's remarks';
- Monitoring: for example, ‘statements that keep the group on task and focus the discussion on the topic'. (2004, p. 119)[Page 117]
A further coding system (Initiate-Response-Reply Framework [IRR]) was employed for response levels, which were coded as an initial posting (IP), response to a post (RP), reply to a response (RR), or reply to a reply (RR#). Level 4 responses were deemed to be high, and Level 1 low, as indicators of interactions among participants (2004, p. 120).
For asynchronous online discussions, Sorensen and Baylen found that patterns 1 and 2 (initiating and supporting) dominated. They also found more evidence for initial levels of response patterns: ‘Students were most likely to respond to an initial posting (level 2), and much less likely to reply to a response (level 3) and not likely at all to reply to a reply (level 4). Thus, the interactions appeared much less like a discussion, in which conversation builds upon previous responses, and more like a question-and-answer scenario’ (2004, p. 124).
The authors drew a number of implications from their research:
- If metacognition is going to occur, students may need to learn and apply new communication patterns (such as challenging, summarizing and monitoring).
- There is a case to be made for designing online tasks that demand that students engage in both synthesizing and challenging.
- Use can be made of models or exemplars of expected online role behaviour, and these expected roles clearly defined and illustrated.
- It is best to choose topics that lend themselves to high-level patterns of communication and interaction.
- Instructors ‘need to pay attention to instructional design principles that enhance the learning environment. The use of online discussion should allow learners to focus on key components of what they are learning. They must be able to connect what they know prior to this experience and to make those connections to what they are currently learning’ (2004, pp. 124–5).
In respect of convergence, Sorensen and Baylen drew attention to difficulties in consensus-building in asynchronous online discussions. Of course, the nature of consensus-building and its desirability take us into the realm of educational philosophy. However, calling to mind Hammond's (1997) view that much online debate is serendipitous and hard to structure, Sorensen and Baylen noted a leadership void (related to teaching presence) and lack of structure (related to cognitive presence) as potential disadvantages of asynchronous online discussions and related to difficulties in ‘consensus building’ (2004, p. 118).
Another issue related to convergence and congeniality is communicative competence and style (an aspect of social presence). In this regard, discourse analysis with an emphasis on pragmatics can identity factors which foster or hinder the kind of collaboration that facilitates online learning. In their analysis of asynchronous online discussion transcripts from a course involving both Chinese and European students, Locke and Daly (2006) identified the important role played by such positive politeness strategies as: the use of in-group address forms; [Page 118]the intensification of interest in another; and the assertion of common ground. The latter played a key role in the achievement of convergence. They also found some striking differences between Chinese and European students in respect of modes of group address and in the management of such speech acts as expressing disagreement and responding to a challenge. They concluded that politeness practices specific to Chinese culture had contributed to congeniality in the group as a whole (2006, pp. 7–9). Without mentioning politeness specifically, Hew and Cheung found in their Singaporean study that showing appreciation and offering encouragement enabled the relaxed expression of opinions, even dissenting ones (2011, p. 316).
The role of the instructor in facilitating a particular kind of structure in asynchronous online discussion is a major focus of research, involving issues of professional identity, productive pedagogical practice and – last but not least – workload. It is worth noting that in their recent survey, Allen and Seaman found that the percentage of academic leaders who believed it took more faculty time and effort to teach online increased from 41.4% in 2006 to 44.6% in 2012 (2013, p. 5). In a review article addressing workload issues, Dunlap asserted that ‘facilitating discussions is the single most time-consuming and effort-intensive component of an online course’ (2005, p. 21). She addressed this issue by drawing on the literature to bring together a range of instructional strategies designed to help teachers achieve ‘presence’ without their being constantly online. She divided these strategies into three categories: ‘course orientation and management; assessment of learners during online activities; and discussion facilitation and management’ (2005, p. 18). The first of these was emphasized by Goldman (2011), who viewed the establishment of ground rules as mitigating workload intensification.
A radical approach to workload and to issues of structuration and convergence (or lack of it) was proposed by Eleuterio and Bortolozzi, who worked from the premise that systems such as asynchronous bulletin boards, as ‘passive tools designed to merely organize discussions in threaded structures, with no mediating capability to promote interaction, or handle disagreements and conflicts among the group', are demotivating for students and beyond the mediational capacity of tutors (2004, p. 13).
The Amanda method allows for ‘very little or no interference of a human mediator’ (p. 13):
The AMANDA method consists of launching a set of issues for group debate and then redistributing the corresponding answers and argumentations among the participants to be collectively validated along successive discussion cycles. At each discussion cycle, the method detects disagreements and proposes new interactions among the group, so that the focus of the discussion is intentionally controlled and the debate advances according to specific interaction objectives. New discussion cycles are successively opened until the discussion cannot be advanced any further or until the discussion time expires. (2004, p. 14)[Page 119]
Their report of field experiments noted improved student motivation, brought about, the authors explained, because the method shaped the discussion into a ‘regular and disciplined activity', with participants ‘receiving “personal” discussion forms to work on and being challenged to argue over conflicting positions from their peers’ (2004, p. 20). The writers themselves put the word ‘personal’ in inverted commas. It is certainly a challenging thought that an algorithmically based system may be superior to a human tutor and that student motivation may be enhanced by replacing or supplementing the teacher in the overt company by a machine.
The Amanda project notwithstanding, most studies posit a central role for the teacher as both facilitator and designer and thus having the key role in structuration (Andresen, 2009; Nandi, Hamilton and Harland, 2012). In respect of facilitation, Im and Lee argued on the basis of their research that ‘Tutors should take on a variety of roles to successfully promote online discussion, such as guiding students in the online discussion, providing prompt input and feedback, and offering summaries of the discussions’ (2004, p. 167). In his systematic review of 62 case studies, Hammond suggested as best practice that instructors should ‘show teaching presence but encourage critique and divergence; fade as appropriate; have an administrative role …; have a pastoral role …; suggest activities and roles to generate debate; and take responsibility for monitoring the nature and scope of discussion and group processes’ (2005, p. 18). In their study of instructor questioning, Noce, Scheffel and Lowry found that students were more likely to respond to questions that developed coherence through ‘instructor uptake and authentic questions’ (2014, p. 81), where uptake refers to a technique where an instructor takes up a student comment for reflective attention.
The role of the tutor (pre-discussion, during-discussion and post-discussion) was also the major focus of research carried out by Lim and Cheah (2003), who used questionnaires, focus group interviews and analyses of discussion records with a sample of 700 pre-service teacher education students and four online tutors. They found that tutor involvement varied across the discussions and that students identified a number of areas where tutors could improve. On the basis of this, they identified six important tutor roles:
- Setting meaningful tasks;
- Providing guidelines in respect of online discussion ‘technicalities’ (for example, length and structure of messages);
- Actively participating (answering queries, offering feedback, playing devil's advocate);9
- Maintaining the focus of discussion;
- Drawing conclusions and offering expert knowledge, especially in the post-discussion stage;
- Suggesting further resources for learning (especially Web-based ones). (2003, pp. 43–4)
In another study focusing on online instructor roles, Havard, Du and Olinzock claimed that ‘The roles of the instructor are critical in the implementation of [Page 120]discussion strategies and design of student tasks for significant learning in online collaborative environments.’ Their research examined three productive roles for the instructor – class member, initiator and discussant – and linked these with ‘three strategies for dynamic online discussion – flexible peer, structured topic, and collaborative task’ (2005, p. 134).
Other studies (e.g. Poole, 2000; Hew, Cheung and Ng, 2010; Nandi, Hamilton and Harland, 2012) have highlighted the advantages of having students take on a leadership role such as discussion moderator. Hew, Cheung and Ng found that students in their study responded more to other students’ postings when they were in the role of facilitator, and that most students preferred to contribute when the discussion was facilitated by a peer. They identified two factors that motivated students to contribute to student-facilitated discussion – ‘familiarity with the student facilitator and mutual obligation’ (2010, p. 593). While endorsing the key facilitative role of the instructor, Dennen made the point that instructors can be too dominant (thereby creating a workload rod for their own backs) (2005, p. 145).
Does the size of the group matter? A study by McCarthy, Smith and Deluca (2010) had mixed results in comparing the effectiveness of different-sized groups. Redundancy was less reported by small-group participants, and the latter were more likely to call for more instructor participation. Full-group discussions challenged the generation of new messages and increased the number of threads, i.e. the discussion was potentially richer.
Relevant to structuration in relation to course design is a distinction Hammond makes in suggesting that: ‘Cooperative learning involves the completion of a task by breaking it down into subtasks that team members solve independently, whereas collaborative learning involves team members working together to develop a joint solution to a problem’ (2005, pp. 13–4; see also Klemm, 2005, p. 1). An approach used in Lundin's (2003) research was to design asynchronous learning activities to support synchronization in collaborative learning. Based around the use of a Net-scenario (a multimedia-enhanced story to support collaborative role-play for learning), the study tested three types of ‘synchronization point': locked scenes, written instructions and collaborative production.
The locked scenes were effective in synchronizing all participants but slowed down the group's progress and did therefore make the most active participants lose motivation. The written instructions were successful in gathering the active group but failed to engage the less active participants. The collaborative production was successful in gathering the active group in collaborative activities and to engage the less active participants individually. (2003, p. 441)
Though the research was based in a workplace, it has relevance for tertiary online learning environments.[Page 121]
In his review article, Hammond stated that ‘Researchers express broad agreement that the argument for using asynchronous online discussion rests in a commitment to interaction between learners and adherence to a social constructivist approach to teaching and learning', focusing on the value of social interaction and learner involvement in active meaning-making (2005, p. 18). A view of how rhetorical space might be shaped in an asynchronous online discussion will depend on the philosophy that underpins both platform and programme design. Hammond notes that most of the 62 papers he drew on ‘had an action research element to them’ and were focused on ‘improving curriculum design or instructor practice rather than establishing the value of asynchronous online discussion per se’ (2005, p. 15). So much is reassuring, but needs to be viewed from the perspective that these studies tend to be single cases and are a drop in the ocean when compared with Natriello's statistics referred to previously. They don't provide an overview of what is generally occurring in online settings using asynchronous bulletin boards.
In the same review, Hammond further noted a general absence of focus on the characteristics of particular software programmes (2005, p. 17). Payne's (2005) critique of Blackboard is a rare exception. Such critiques are useful in rendering visible the discursive underpinnings of particular asynchronous learning networks, and online learning environments in general. As suggested earlier, they provide the necessary act of social contextualization in terms of which issues related to asynchronous online discussions can be framed. They also suggest that a move towards increased commodification may be at odds with a commitment to social constructivist pedagogy.
Future research might well focus on those two aspects of rhetorical reach – ‘field’ and ‘company'. As the rhetorical space of an asynchronous online discussion is discursively constructed, then (in Foucauldian terms) certain privileged discourses will serve to regulate what can and cannot be said in respect of the field. Analyses such as Payne's are important ways of identifying the nature and degree of the hegemonies operating in a particular online learning environment. It is interesting to note that Payne's antidote to the social conformity he sees operating in the suburban Levittown of Blackboard is another kind of space, an ‘urban ideal’ he finds aptly expressed in a passage from Christy Friend:
In cities, complex networks of production, distribution, transportation, exchange, entertainment, and communication are coordinated across a wide, diverse geographic area. At the same time, individuals retain connections to multiple groups, some of which are small and homogeneous, and others comprised mainly of strangers. The play of identification and difference is thus continually present, yet continually shifting. (Friend, 1999, p. 660, cited in Payne, 2005, p. 502)[Page 122]
An asynchronous online discussion meeting such a description would be characterized by the encouragement of self-reflexivity, diversity, resistance to and critique of naturalized ‘common sense', and the valuing of difference.
Questions of what constitutes an ideal rhetorical space can be couched in terms of how one envisages the composition and qualities of the overt, covert and implied company. Questions such as, ‘Whose space is it? Who is privileged in this space and who is not? Who is excluded? Who is represented, and how, and for whose ends?’ are useful ones for discourse analyses of asynchronous online discussions and their contextual online learning environments (Payne, 2005, p. 503). So are questions such as: To what extent are participants enabled to enlarge the overt company? How broad is the implied company in a discussion? How is the covert company to an asynchronous online discussion and its online milieu constituted?
Notions of ‘instructor presence’ in an asynchronous online discussion can usefully be linked to Bakhtin's (1986) idea of addressivity discussed earlier. In Bakhtin's terms, a student message-writer in an asynchronous bulletin board is both respondent to previous utterances and addressor to those in a position to respond to his or her message. A key focus of future research should be to find ways of helping student participants to move beyond viewing the online teacher as the dominant member of the overt company to seeing themselves as entering into dialogue with a larger company: the overt company in total, the implied company, and even the covert company as reflected in the discursive underpinnings of the socio-cultural environment the asynchronous online discussion is contextualized by.
Hammond's systematic review calls for research that focuses on the qualities of participating learners, a critical feature of any overt company. In his consensus of best practice, he includes the following qualities: an appreciation of the benefits of group work; ICT competence and access; lack of face-to-face (FTF) access; a willingness to critique authority; text-based communication as a preferred learning style; a willingness to interact publicly both constructively and critically; and some fluency and proficiency in the language of the forum (2005, pp. 18–19). Research needs to focus on ways of developing proactively the desirable qualities of asynchronous online participants, perhaps using the Community of Inquiry (COI) model discussed earlier.
There is plenty of scope for research to be done on issues related to structuration. There are no simple answers or generalized prescriptions in respect of such questions as: How involved should the online teacher be in discussion? How directive should the teacher be in respect of their stance? How much intervention should they provide? All teachers face questions in respect of instructional design. It is clear that the discourse patterns of asynchronous online discussions differ markedly from the teacher-centred discourse of traditional FTF discussion in classroom settings. The Community of Instruction model of Garrison and Arbaugh (2007) is a useful framework for designing solutions [Page 123]to structuration issues and addressing the widespread concern with workload intensification.
Finally, there are a number of ethical considerations that need to be borne in mind in the development of future research agendas, on the understanding that all teaching is an ethical undertaking and that there is always an ethical dimension to the educational philosophy one subscribes to. In respect of asynchronous online discussions in particular, there are issues at the macro level related to surveillance, for example, the nature and limits of the covert company and the discursive pressures operating in a rhetorical space, that need to be constantly kept in mind. There are also ethical issues at the micro level, related to how members of the overt company manage their interactive behaviour and how course designers and teachers assign, construct and reward specific roles and behaviours. Finally, when courses attract culturally diverse participants, modes of cultural inclusivity, reflected in participant behaviour and environmental design, need to be explored, so that difference is viewed as a resource and not a deficit.
1. In Imagined Communities, Benedict Anderson argues that ‘All communities larger than primordial villages of face-to-face contact (and perhaps even this) are imagined’ (1983: 6).
2. My use of the word ‘field’ here is roughly synonymous with Hallidayan grammar's use of the term ‘field of discourse', the general sense of what a text is on about, referring to ‘what is happening, to the nature of the social action that is taking place’ (Halliday and Hasan, 1985: 12).
3. The term utterance is being used in the Bakhtinian sense of a unit of speech communication (spoken or written) determined by a change of speaking subject (Bakhtin, 1986).
4. I concede that these are crude measures and overlook qualitative differences in types of feedback, depth of feedback and considerations such as delay (how much ‘time’ has elapsed between the feedback message and the original message in a thread).
5. In an interesting discussion, Heckman and Annabi suggest that ‘the traditional Socratic questioning and feedback functions are to some extent dependent on the linear nature of the FTF dialogue’ and that instructors in ABB contexts may need to ‘choreograph’ discussions differently from those in FTF mode (2005: ‘Discussion', paras 7 & 8). Considerations may also need to be posited on the assumption that besides ‘individual’ intelligence some kind of ‘collective’ intelligence is at work in an asynchronous online discussion as it begins to reflect an emerging ‘community of practice’ (see Lévy, 1997; Wenger, McDermott and Snyder, 2002).
6. An asynchronous learning network (ALN) is a learning environment which involves a student in both self-study and in significant, asynchronous interaction with teachers and other students via networked computers and other technologies (mobile phones, ipods), without the need for synchronous interaction (see Goodwin, Graham and Scarborough, 2001).
7. The press release was hotlinked from the Blackboard index page and retrieved on 9 March 2006 from www.blackboard.com/company/press/release.aspx?id=767025
8. See moodle.org/stats/. These figures relate to an access date of 20 August 2014.
9. Students in this study expected their tutors to visit discussions daily. The authors quote a number of studies highlighting the importance of tutor participation. See also Goodwin, Graham and Scarborough, 2001; Markel, 2001.[Page 124]