Asynchronous Discussion Tools at HGSE: Designing tools to support learning
How should discussion tools be designed to support learning? To answer this question, I created an analytical framework to evaluate discussion tools currently in use at HGSE. The framework and resulting analysis suggest tool updates that will enable HGSE discussion tools to better support learning.
The two discussion tools that this analysis works with the most closely are the Web Crossings tool, currently used by HGSE WIDE World courses, and the iSDT, the iSites Discussion Tool developed by the Harvard University iCommons group. The iSDT is a new tool and has been rolled out on a trial basis around the University. The full launch of the tool will take place for the 2006 fall semester.
These discussion tools support collaborative co-construction of knowledge, a foundational element of how people learn at HGSE. Learning theories provide insight into how these tools can best be configured to support this goal. As Dooley notes, “Learner-led threaded discussion groups can be used to help learners think about how course materials can be used in various contexts.” (Dooley 60) Garrison and Anderson add: “The collaborative yet reflective process of e-learning has great potential for facilitating critical thinking that is core to an educational experience.” (Garrison, Anderson 58)
Why build a set of recommendations around a theoretical framework?
There are myriad ways to go about critically reviewing technologies such as websites or applications. User testing (what did we see the user click on?); web statistics (what can we tell from the server logs about what the users clicked on?); and qualitative research (what did the user say about her experience?) are all valid and useful methods for gathering information about how well a given technology product serves the needs of its users. Nevertheless, these research methods do not necessarily provide a comprehensive view of how a tool is used in a particular setting. Dooley proposes that theory be leveraged as an additional tool to build understanding of how a tool is used in a specific context:
“Theory can provide a framework for understanding and explaining research findings. While research attempts to explain the what is, theory attempts to explain the why. Theory can help distance educators understand why a particular instructional method is almost always effective in a particular setting.” (Dooley 35)
In this essay, the theoretical framework is intended to help serve as a guide to looking at technology tools, in this case discussion tools, and how they support learning. Theory can provide guidance as to how effective learning principles can be used to understand and analyze educational discussion tools.
Garrison and Anderson’s notion of indicators of social, cognitive, and teaching presence served as the foundation for this analytical framework; other researchers’ works supplemented and enriched the framework, with the goal of developing an analysis of how discussion tools can be used to support learning in an asynchronous learning network (ALN).
Social Presence
As Dooley writes, “Learning is not seen as the acquisition of knowledge by individuals so much as a process of social participation.” (Dooley 48) People learn most effectively when they can connect on social levels and work together as part of a community.
In a face-to-face classroom environment, social presence is, in many ways, a given; the students and the teacher are able to communicate affect through their tone of voice, facial expressions, and body language. Every participant is a “real” person in a classroom, groups can be form to work together towards a defined goal, and a sense of community can be promoted. In an asynchronous networked learning environment, however, these modes of interaction and understanding are much more difficult, due in part to the absence of the standard social cues: tone, expression, and body language. Virtual classroom participants rely primarily on written language to convey their ideas and present themselves to the other participants, to form groups, and to establish community.
The processes of conveying ideas and self are closely linked in successful ALNs on three levels: (1) as an individual projects herself as a “real” person, (2) as participants find support in their collaborative meaning-making; and (3) as participants develop into a community of learners. A learning community displays social presence when learners can be supported on these multiple levels. This support is primarily through the choices that the participants and moderators make about how they present themselves. Specific examples of how to incorporate social presence into online knowledge sharing communities include a thoughtful and creative approach to how text is used: “Socio-emotional communication in text-based communication is possible through the use of compensating strategies, such as the adaptation of textual behaviours to reveal social and relational messages.” (Garrison, Anderson 29) Discussion tools serve to support these compensating strategies.
Participants can present themselves as “real” people using discussion tool features
Garrison, Anderson and Archer define social presence as “the ability of participants in a community of inquiry to project themselves socially and emotionally, as ‘real’ people (i.e., their full personality), through the medium of communication being used.” (Garrison, Anderson 28) Presenting the “real” self is difficult in an online context; “When the medium is the written word, establishing social presence can be problematic.” (Garrison, Anderson 28) By building on the affordances of text-based communication, though, participants can forge a sense of the “real” self. Garrison and Anderson present examples of text-based social presence, including: expressing emotions (‘I just can’t stand it when…!!!’), use of humor, self disclosure, continuing a thread, quoting from others’ messages, etc. (Garrison, Anderson 51)
An example of how participants can present themselves as “real” is to display photos of themselves. Discussion tools differ in how they support image inclusion. Currently, iSDT displays default ID photos. Web Crossings allows users to upload a photo of their choice. Web Crossings also provides the capability to allow users to post a brief biography of themselves. Clicking on their photo leads to a page with their photo, biography, and a listing of their postings. The iSDT allows users to click on a name and get a listing of posts, but does not provide support for self-selected images or biographies. The Web Crossings model of providing a bio, photo, and posts is an effective way to support social as well as cognitive presence.
Open communication is promoted in successful ALNs
Open communication in this context means that participants are able to fluidly exchange ideas and build on each other’s knowledge construction. The two tools analyzed in this essay support open communication in that they allow students to respond to each other’s statements, but do not have specific “quoting” features to easily integrate others’ thoughts into a response. For example, some discussion tools incorporate a “quoting” feature that allows the user to format a quoted chunk of text by highlighting the text and clicking the “quote” button. This would perhaps put the text inside of <blockquote> and <em> tags to visually separate it from the commentary content.
Using real names supports social presence
Lastly, when participants use their real names in a discussion social presence is more supported. As Etzioni notes,
… f2f [face-to-face] communication systems are significantly superior to most CMC [computer-mediated community] systems as far as identification, accountability, and authorization are concerned. However, it should be noted that there are no design difficulties in providing a much stronger basis for interpersonal knowledge in CMC systems. In fact, several forums already demand that participants use their true names, and others verify such claims. (Etzioni)
Both of the discussion tools in this analysis support the use of real names.
Task-oriented social interactions should be supported
Working together towards a common goal becomes more successful when social presence is in place. Shields, Gil-Egui & Stewart report:
Case studies conducted by Jarvenpaa and Leidner (1998) on virtual teams of college students whose members never met before and lived in different countries suggest that those teams that manage to balance task-oriented communication and social-oriented communication are likely to perform better than teams with little or no social-oriented communication. (Shields, S., Gil-Egui, G., & Stewart 128)
This means that when people can integrate socially oriented postings into an online discussion, they are more likely to be able to successfully work towards their learning goals. Discussion tools can support social interactions by not limiting posting topics. For example, all users can create new discussion branches; this is not functionality that is limited only to course moderators.
Another way that discussion tools can support social interactions is by providing a “break-out” or “sandbox” area for participants to use. Professor Stone Wiske provides an example of how this type of informal discussion space provided particular benefit to our course:
Our T526 class lobbied hard for a “watercooler” to complement the spaces for more academic exchanges and several members said it added to their sense of involvement and commitment to learning. From my perspective, it was great to have a place where students could raise and address topics of their own choosing away from the places where I was trying to promote a more focused exchange. (Personal exchange, May 31, 2006)
This example shows how providing mechanisms to support informal exchanges can support the learning process. An ALN should not, though, promote the use of alternate identities in these different discussion spaces. As Greenspun et al note,
[P]roviding a breakout and reassemble chat room is useful, but that it needs to be tightly integrated with the rest of the online community and that, in particular, user identity must be preserved across all services within a community. (Greenspun et al)
Informal social exchange provides an opportunity for members of a community to present their “real” selves, particularly when user identity is preserved.
Learning happens in communities
Another theory suggesting causation between social interaction and learning is that of situated cognition. Dooley et al write:
… [L]earning involves participation in a community of others who are practicing the behaviors to be learned. Learning is situated in communities of practice and cannot be analyzed in isolation from either the practice or the community. There is an interconnectedness of persons, learning, practice, participation, and the social world. (Dooley 48)
For effective learning, people must be able to be present in a group of learners. This is another example of how individual behavior and tool affordances work in concert to support learning. Tool affordances to support community development include controllable access levels to discussion boards.
Cognitive Presence
Cognitive presence is the second of Garrison and Anderson’s three categories of learning indicators. They clarify cognitive presence as follows:
“… specifically in the context of this discussion, cognitive presence means facilitating the analysis, construction, and confirmation of meaning and understanding within a community of learners through sustained discourse and reflection largely supported by text-based communication.” (Garrison, Anderson 55)
Garrison and Anderson use Dewey’s model of practical inquiry as the framework for building a discussion of the “analysis, construction, and confirmation of meaning and understanding.” These actions, though intrinsically behavioral, can be bolstered by effective discussion tool affordances. Again, effective and well-designed discussion tools advance learning experiences.
Triggering event
The “triggering event” is the first phase of practical inquiry. This stage is intended to provide the foundation and stimulus for the ensuing learning experience. Garrison and Anderson write, “While the responsibility of the teacher is to initiate this phase of inquiry, this can be structured in a more open manner by framing the issue and eliciting questions or problems that students see or have experienced.” (Garrison/Anderson, 59)
In many asynchronous learning environments, including that which is integrated into T526, this triggering event takes the form of a discussion prompt. In this case the prompt is the result of collaboration between the professor, teaching fellow, and student moderator. The prompt helps inspire and guide the week’s discussion.
Instructors can leverage discussion tool affordances to post triggering prompts. One way is to put the prompt text into the first posting and ask students to respond to this message. This method is effective in that it does offer a place to put the text. What it doesn’t allow for, however, is ongoing and immediate access to the prompt. If the prompt were displayed persistently, perhaps in one frame of the browser window, the participants would see it as they explore the subsequent posts and compose their own responses.
Exploration
Exploration is just what it sounds like: “This means to first understand the nature of the problem and then to search for relevant information and possible explanations.” (Garrison, Anderson 59) Asynchronous discussion formats are particularly valuable for this endeavor. First, to understand the nature of the problem, discussion participants can work together to develop a collaborative understanding. When the problem is defined, and searching for and sharing information becomes paramount, discussion tools can again provide support. Participants can share the results of their independent analysis through text-based discussion postings. They can also share documents, either through entering hyperlinks to websites or through uploading documents to the discussion tool and attaching them to specific posts. Images can also be shared by directly displaying them in the post; other participants can then view the image without having to open another file. (Although not taken from either of the tools under review, see http://www.photo.net/bboard/q-and-a-fetch-msg?msg_id=000gTu for an example of images displayed directly in posts.) A variety of supporting documentation can then be integrated into the course discussion, strengthening and enriching the exchange.
These multiple modes of information also allow for different learning styles. For example, visual learners can strengthen their explanations with easily viewed images and auditory learners can share links to multimedia resources. This is not to say, however, that asynchronous discussions can be successful without written dialogue; expanding the framework to include alternate means of representation strengthens, rather than dilutes, the discussion.
Integration
After ideas are shared and explained in the exploration phase, the next phase is to pull concepts together into a cohesive framework. This integration phase is essential for learning and is well served by asynchronous discussions. Garrison and Anderson note:
[I]t may well be during this phase of the inquiry that the characteristics of e-learning come to the fore. The asynchronous and explicit nature of text-based communication may well facilitate deep and meaningful learning outcomes. (Garrison, Anderson 60)
Discussion tools can support integration in many ways. The most important as well as the most straightforward way is to allow multiple posts to be viewable at one time. Participants have a difficult time integrating strands of thought if they cannot read others’ postings; their learning experience is diminished when they are not able to view and connect ideas. Also, when participants compose messages to post to the discussion board, their idea formation is strengthened when they are able to view others’ postings as they are in the process of writing.
Another powerful way of enhancing participants’ learning is to allow them to move threads and posts (in their own view) to allow visual connection of ideas. For the tool designer, this mode of enhancing integration is more programmatically challenging; while linking and displaying images are affordances of HTML, dragging and rearranging text areas is not. Nevertheless, the ability for learners to be able to connect ideas in this way would be highly valuable.
Resolution
The resolution phase allows for learners to assess the knowledge that they have collaboratively developed. They can ask themselves, does this make sense? Is this correct? Is this the best solution? Developing a resolution, though, is challenging:
In good educational environments, as in real life, resolution is seldom fully achieved. Inevitably, results of the resolution phase raise further questions and issues, triggering new cycles of inquiry and, thereby, encouraging continuous learning. (Garrison, Anderson 60)
This difficult task can be supported by technology. Technical mechanisms in current online tools for resolution include (1) rating of postings by other members (see www.slashdot.org; a thread might have 200 responses; the top 10 rated ones represent some sort of community consensus), and (2) a wiki so that multiple participants can edit a single document.
Even without these additional capabilities, the essential affordances of a discussion tool support resolution. As Bransford writes,
An added advantage of networked technologies for communication is that they help make thinking visible… By prompting learners to articulate the steps taken during their thinking processes, the software creates a record of thought that learners can use to reflect on their work and teachers can use to assess student progress. (Bransford)
Being able to review how thoughts have changed over time allows the community to better assemble ideas of how they have been able to communicate, both for reflection as well as assessment.
Teaching Presence
The third foundation of successful learning environments is that of teaching presence. Garrison and Anderson define teaching presence as “the design, facilitation and direction of cognitive and social processes for the purpose of realizing personally meaningful and educationally worthwhile learning outcomes (Anderson et al 2001).” (Garrison, Anderson 29) Thus, learning environments are successful when the learners and moderators can adapt the components to support their learning needs.
Flexible display of information
This notion of adaptability is important for adult learners in an online learning environment. As Dooley notes, “Distance learners should support themselves and their instructors and take responsibility for their differences, educations, and means of motivation.” (Dooley 72) Garrison and Anderson also support this reasoning:
If e-learning is to be a collaborative, constructivist process, then students must have some influence on what is studied and how it is approached…This is best accomplished when a teacher can both design and organize the educational experience in a way which will introduce effective responsiveness to developing needs and events.” (Garrison, Anderson 68)
Offering control over online learning environments facilitates individual meaning-making and self expression to the fullest extent, to the support of greater learning outcomes. Many of the tool affordances noted as supporting cognitive presence provide this control over online learning environments. For example, multiple posts should be viewable at one time.
Instructional design indicators: Contextual information
Garrison and Anderson list “Instructional design and organization indicators” as examples of how teaching presence can be established (Garrison/Anderson 68); these are primarily actions that participants perform to craft the conversation. Again, though, there is an opportunity for the discussion tool to support this human behavior. One way to provide context and “establish time parameters” is to clearly label posts with the date and time at which they were added. Another ways to establish context is to add a “new” indicator to unread posts. For example, there might be a small image next to the titles of new postings. Also, discussion tools can be built so that the visited links are a different color from visitor links; this is universal across the web and would be well understood by discussion tools users. The most essential way to indicate how posts are related is to display how they branch; each posting is displayed as a subpost to the one to which it responds. This way, users have a better understanding of how different postings relate to each other.
Direct instruction indicators: Knowledge sources and technical concerns
Garrison and Anderson also provide indicators of direct instruction (Garrison, Anderson 71). Direct instruction is perhaps the closest example listed to “normal” teaching; essentially, this is the core of what the facilitator is communicating to the participants. Two indicators that rely on tool affordances are “Knowledge can be incorporated from multiple sources” and “Responding to technical concerns.” For knowledge to be incorporated from multiple sources, the discussion tool should include affordances for uploading and attaching documents to posts, as well as providing capabilities for easily displaying images and creating hyperlinks. “Technical concerns” can be addressed by contextual “help” information as well as by providing information about how to access technical assistance, whether live or virtual. The moderator should be able to focus on higher value teaching activities.
Discussion tools at HGSE
This chart provides a comparison view of how Web Crossings (as implemented by WIDE) and iSDT support social, cognitive, and teaching presence.
| Web Crossings | iSDT | |||||
| Social Presence | ||||||
| Participants are "real" people | ||||||
| Participant photos are shown with their posts |  |
ID photos are shown, but cannot be replaced | |
Participants can upload the photo of their choice | ||
| Photos link to participant bios, previous posts | |
Photos are not linked, but participant’s name leads to a list of their previous posts |  |
Participants’ photos lead to bio, list of posts (note: example of dual social and cognitive affordances) | ||
| Built-in functionality to support quoting other posts in reply (supporting open communication) | |
Built-in quoting feature | |
No built-in quoting feature | ||
| Use of real names is supported | |
Real names are used and supported | |
Real names are used and supported | ||
| Task-oriented social interactions are supported | ||||||
| "Sandbox" feature for social interactions | |
No sandbox feature | |
No sandbox feature | ||
| Learning communities supported | ||||||
| Can control access levels to discussion forum | |
Controllable access levels | |
Controllable access levels | ||
| Cognitive Presence | ||||||
| Phases of practical inquiry are supported: Triggering event | ||||||
| Persistent prompt display | |
Prompt is not displayed persistently | |
Prompt is not displayed persistently | ||
| Phases of practical inquiry are supported: Exploration | ||||||
| Text-based, branching postings | |
Posts are text-based, branching | |
Posts are text-based, branching | ||
| Hyperlink functionality (without having to know HTML) | |
No built-in hyperlink functionality | |
No built-in hyperlink functionality | ||
| Supporting documents can be attached to posts | |
Documents can be attached to posts | |
Documents cannot be attached to posts | ||
| Functionality to display images in posts (without having to know HTML) | |
No built-in image display functionality | |
No built-in image display functionality | ||
| Phases of practical inquiry are supported: Integration | ||||||
| Multiple posts viewable at one time (also transparency of actions) | |
Multiple posts viewable at one time | |
Only one post viewable at a time | ||
| When writing response, can view message to which you are responding | |
Can view messages while responding | |
Cannot view messages while responding | ||
| Threads/posts can be rearranged (also transparency of actions) | |
Threads/posts cannot be rearranged | |
Threads/posts cannot be rearranged | ||
| Phases of practical inquiry are supported: Resolution | ||||||
| Posts can be given a rating | |
Posts cannot be given a rating | |
Posts cannot be given a rating | ||
| Wiki/collaborative editing area available | |
Wiki/collaborative editing area not available | |
Wiki/collaborative editing area not available | ||
| Thinking is made visible | ||||||
| Previous threads/posts remain available throughout span of course | |
Previous threads/posts remain available throughout span of course | |
Previous threads/posts remain available throughout span of course | ||
| Teaching Presence | ||||||
| Flexible display of information | ||||||
| Multiple posts viewable at one time (also Integration phase of inquiry) | |
Multiple posts viewable at one time | |
Only one post viewable at a time | ||
| Threads/posts can be rearranged (also Integration phase of inquiry) | |
Threads/posts cannot be rearranged | |
Threads/posts cannot be rearranged | ||
| Contextual information about postings | ||||||
| Established time parameters | |
Posts are time-stamped | |
Posts are time-stamped | ||
| Posts have “new” indicator | |
Posts have "new" indicator | |
No "new" indicator | ||
| Posts displayed in threaded format indicating how messages are related | |
Posts in threaded format | |
Posts in threaded format | ||
| Links to post change color or otherwise indicate when they have been viewed | |
Links to posts do not change color to indicate visited status | |
Links to posts do not change color to indicate visited status | ||
| Knowledge can be incorporated from multiple sources | ||||||
| Hyperlink functionality (without having to know HTML) | |
No built-in hyperlink functionality | |
No built-in hyperlink functionality | ||
| Supporting documents can be attached to posts | |
Documents can be attached to posts | |
Documents cannot be attached to posts | ||
| Functionality to display images in posts (without having to know HTML) | |
No built-in image display functionality | |
No built-in image display functionality | ||
| Responding to technical concerns | ||||||
| Contextual "help" information | |
No contextual "help" information | |
No contextual "help" information | ||
Some of these features would be very straightforward to implement, such as changing the CSS (Cascading Style Sheets) so that visited posting links will appear in a different color. Other features, such as rearranging threads and posts, would require more advanced programming. The decisions about how to make changes going forward depend on priorities and resources; neither of these constraints is addressed in this document.
Conclusion
Discussion tools are powerful components to learning experiences; considering how these tools support learning, and adapting the tools to meet this need, allows them to be even more valuable for HGSE and for Harvard University. Educational theory and research can help designers understand how to build and adapt tools; this, in addition to working directly with users (participants, administrators, and faculty) provides a thorough grounding in how to think about building software to support learning.
Works Cited
Andersson, E., Greenspun, P., Grumet, A. Software Engineering for Internet Applications http://philip.greenspun.com/seia/scaling
Bransford, J.D., Brown, A.L. & Cocking, R.R. (Eds.) National Research Council, How People Learn. Chapter 9 “Technology to Support Learning”, pp. 194-220. Washington, DC: National Academy Press, 1999. http://books.nap.edu/html/howpeople1/ch9.html
Dooley, K. E. with Murphy, T. Chapter III Learning Theories, pp. 31-55 and Dooley, K.E. with Wilson, S. Chapter IV Adult Learning Principles and Learner Differences, pp. 56-75. In Dooley, K.E. Advanced Methods in Distance Education: Applications and Practices for Educators, Administrators and Learners. Hershey, PA, USA: Information Science Publishing, 2005
Etzioni, O., "Face-to-Face and Computer-Mediated Communities, A Comparative Analysis" in The Information Society Vol. 15, No. 4, (October-December 1999), p. 241-248.
http://www.gwu.edu/~ccps/etzioni/E31.html
Garrison, D.R. & Anderson, T. (2003). E-Learning in the 21 st Century: A Framework for Research and Practice. London: RoutledgeFalmer.
Shields, S., Gil-Egui, G., & Stewart, C.M. Chapter V “Certain about Uncertainty: Strategies and Practices for Virtual Teamwork in Online Classrooms,” pp.116-141. In Distance Education Evolution: Case Studies. Monolescu, D. (Ed.) Idea Group Inc. 2003.