Introduction
In the last five years, a revolution has taken place on the Internet with the advent of what has been called “Web 2.0.”Although exemplified by the burgeoning number of wikis, RSS feeds, blogs, and user-driven Web sites, Web 2.0 has come to be understood not only as a set of technological applications but also as a philosophy (see O’Reilly, 2005). That is, Web 2.0 applications are said to embody a conceptual shift toward the Web as a platform for sharing “objects” (such as content or information), with user-generated content being just as important as content supplied by institutions. Museums interested in building community and audiences have quickly realized the potential of these new technologies and attitudes. Today most museums have an on-line presence and are planning to experiment with the forms of interactivity that we have now come to characterize as “Web 2.0” interactions. The push for experimentation comes from museum staff as well as museum visitors who “no longer accept being solely consumers of information” but want to contribute their own experiences and interpretations (Barry, 2006, n.p.).
Our study began with a simple question: amidst the hype and excitement of these new technologies, what is really happening with and through them? The considerable resources expended (including time, finances, and personnel) suggest a clear and strong commitment from museums to pursue new Web-based tools. Furthermore, there is a strong belief within the museum community that this functionality has brought about dramatic changes in the ways knowledge is shared, created, and co-created, as well as in the ways that visitors interact not only with the museum but also with each other. However, museum professionals largely agree that most of these claims are speculative or based on anecdotal data.
This paper presents early results from a large, multi-year study of the impact of Web 2.0 technologies on museum learning and practice. The larger two-part project closely examines the interactions that take place on one science museum blog, Science Buzz, associated with the Science Museum of Minnesota, while also tracing the impact of Web 2.0 technology on physical museum practice at the Museum of Life and Science in North Carolina. Here we report preliminary findings from our examination of the discursive knowledge building activities that happen on Science Buzz.
Our results to date suggest that the activity and community that forms Science Buzz are highly complex and variable. Some topics and threads attract a large number of participants with diverse levels of knowledge and perspective who participate from locations both inside and outside the physical museum. Furthermore, argumentative strategies such as making claims are the most common discursive moves on the Buzz blog. These reasoning strategies are the focus of many prior discourse-based studies of concepts like “learning” in group environments. Yet our analysis of the Buzz blog also shows a range of strategies that we associate with constructing individual identities (more common) and group identities (less common). These discourse strategies are often ignored in studies of learning or knowledge construction, but they are clearly integral to any understanding of on-line spaces made available by museums to the public.
Learning and Interaction
The relationship between interactivity and learning or knowledge-building has been interpreted and researched in several different contexts. Based on specific contextual needs, the interpretation and focus of this research is varied, but research in both science education and museum studies is useful for considering how contributors to science museum blogs like Science Buzz may learn or construct knowledge.
Research in science education has focused on the role of argumentation as a discourse type central not only to the way that scientific knowledge develops and is articulated, but also to how individuals can best learn about science interactively (e.g, Driver, Newton, & Osborne, 2000; Erduran, Simon & Osborne, 2004; Jiménez-Aleixandre, Rodríguez, & Duschl, 2000). Educators have suggested that learning science involves learning to talk and act like scientists by practising reasoning and argumentation (Duschl et al, 1999). Such active learning necessitates communicating with others in forums that allow for the negotiation of information in open “cultures of inquiry” (p. 253) where “individuals contribute ideas, thinking, and reasoning to a community-based, collaborative knowledge construction process” (Goldman, Duschl, Ellenbogen, Williams & Tzou, 2003, p. 254).
While much research on science, argumentation, and interactive learning has taken place in physical classroom spaces, research in the area of computer-supported collaborative learning environments (CSCL) has drawn attention to the advantages of on-line environments for approaches to argumentation and science learning. These include the creation of archives as a collective memory, the visualization of cognitive processes, the ability to unite groups that are geographically distributed, and the linking of different modes of communication (Andriessen et al, 2003; Zimmerman, 2005). Drawing on these theories about the nature of CSCL learning, researchers have reported on a number of structured on-line systems developed to facilitate individual and group science learning by emphasizing the formation of argumentative discourse through group conversations and writing projects (Scardamalia, 2002; DeVries et al, 2002; Bell, 2000).
What is lacking from traditional studies of argumentation in formal science learning environments - especially those making use of CSCL tools - is any description of the other kinds of discursive moves that participants make. These studies tend to isolate argumentative discourse from other “social” discourse that strays “off topic” or “off-task.” Research about learning in informal environments like museums, however, shows that informal learners come with very different kinds of commitments, relationships, motivations, and learning tasks in mind. This attitude toward learning relies on the idea that science understanding and learning happens for “personal interest, need and/or curiosity” (Falk, Storksdieck, & Dierking, 2007, p. 455). As Falk, Storksdieck, & Dierking (2007) state, "Rather than framing our efforts in communicating science, we would suggest that we think about offering the public opportunities for engaging with, appreciating and better understanding the science of interest and need to them" (pp. 456-457). That is, individuals learn about science to satisfy short-term personal needs rather than long-term cultural or civic duties. Thus, it can be assumed that contributors to on-line science museum sites may not always stay “on-topic” and that the additional motivations that may accompany participation (e.g., wanting to make friends or join a community; wanting to express oneself) would likely be tied to other kinds of discourse moves or modes of interaction that vary considerably from those that accompany learners in traditional classrooms.
Existing research within museum studies pays considerable attention to the way in which individuals interact, employing visitor observation in the exhibit space and post-visit surveys to measure the impact of “sociality” on learning (e.g., vom Lehn, Heath, & Hindmarsh, 2001; Hindmarsh, Heath, vom Lehn, & Cleverly, 2005; Packer & Ballantyne, 2005). A recent study regarding the effect of interaction within the exhibit space of the Queensland Museum in Brisbane, Australia, focused on the planning steps, engagement level, learning experiences, and sustained memories of museum guests visiting in pairs or alone. Packer and Ballantyne’s (2005) conclusions about the interactions, or “social context,” of visitors in this study revealed that “solitary and shared learning experiences can be equally beneficial, but in different ways” and that “there may be a learning advantage in having access to a social context that is consistent with the learner’s preferred approach” (p. 189). While some visitors prefer experiencing the exhibit through “personal reflection without distraction” (p. 189), others valued “being able to share the experience and discuss ideas with others” (Packer & Ballantyne, 2005, p. 190).
Though this research has been productive in exploring interaction that occurs in physical exhibit space, the studies did not explore on-line interaction that revolves around museum content and related topics. Researching digital environments has always been extremely challenging due to the virtual and often anonymous nature of the interactions. Recent research on interactive technologies in museums has focused on the ability to expand the scope and reach of the museum (Keene, 1998; Anderson 1999; Parry & Sawyer, 2005), user satisfaction (Goldman & Waldman, 2002; Ockuly, 2003; Bartley & Hancock, 2006), or Web site design principles (accessibility, usability, educational value). New trends in technology experimentation and research are looking at the relationships between museums’ on-line and virtual spaces; learning theories and relationships between learning styles and Web preferences (Schaller, Allison-Bunnell, & Borun, 2005); using technologies to make the walls more permeable between permanent installations; and visitor contributions between virtual experiences and on-site experiences. However, there is a current lack of research that describes the discursive interactions that take place when museums employ Web 2.0 tools for the purposes of learning and knowledge-building.
Study Design and Methods
The study had two objects of focus, one a technological environment and the other a museum. We focused on Science Buzz as the technological environment because of its value as an exemplar of best practices on Web 2.0. For the museum environment, our focus is the North Carolina Museum of Life and Science (MLS) because of their broad experimentation with a range of technologies and practices which makes them a compelling site for an examination of changes in museum practice.
We have already shared the simple question that motivated our inquiry: our desire to understand what is really happening with and through new Web-based technologies. Our operational questions maintain this descriptive focus with a particular interest on learning and what we call the “co-construction of knowledge.”
- What is the nature of the community that interacts through Science Buzz? This is a descriptive question intended to develop a community profile based largely on how that community interacts.
- What is the nature of the on-line interaction? This is also a descriptive question intended to identify key forms of interaction. In other words, what are people doing on-line?
- Do these on-line interactions support knowledge building for this user community? The discourse analysis will enable a description of the interactions and knowledge building.
- Do on-line interactions support inquiry, learning, and change within the museum – ie, what is the impact on museum practice? As an institution changes its approach to interactions with the public on-line, how do practices on-site change?
Discourse Analysis
Our work with Buzz was structured by our approach to discourse analysis. The key characteristic of most Internet-based interactions - a characteristic that is so common as to be invisible - is that they are written. In fact, the explosion of Internet-based technologies should be credited with driving a renewed interest in writing (as well as reading), although it is clear from public discourse that the arguably radical changes from all this writing (e.g., anyone can blog) are not widely perceived as positive. The essential point, however, is that if one wants to describe, characterize, or understand what is happening with Web 2.0 and other on-line interactions, one must do so via writing (a statement that is particularly true if we take a broad semiotic view of writing as encompassing the use of sound, image, and video).
The fact of all this writing as the material of on-line interactions means that we need good tools for analyzing the discourse. Discourse analysis provides a structured and systematic way of reading and interpreting the interactions. Discourse analysis is commonly used to characterize and understand communication interactions (Fairclough, 1992; Dijk, 1997; Wood & Kroger, 2000; Schiffrin, Tannen, & Hamilton, 2001; Bazerman & Prior, 2004; Gee, 2005) and has been used in museums to study conversations as well (vom Lehn et al., 2001; Allen, 2002; Ash, 2003).
However, to the best of our knowledge, there are no available analytical schemes suited to the problems we are trying to solve in this study and responsive to the theory driving the inquiry. Therefore, one of the primary tasks of this study was to build an analytical tool that was descriptively accurate - that reflected and was responsive to the discourse as people used it on Science Buzz - and that was also theoretically driven by relevant work on the relationships between communication practices and knowledge building. For instance, if we look to work in formal argumentation for guidance and tools to address relationships between argument (as a style of communication) and learning or knowledge construction, most schemes are interested in formalized models of interaction and on “knowledge” as an objective truth (Toulmin, 1958). That is, the desired outcomes are very specific knowledge statements, and the discourse models for arriving at those outcomes are often rule-bound. Very few “live,” real world interactions fit within the boundaries of these interaction models (yet we all still manage to arrive at meaningful understandings of the world around us).
A second primary task is an attempt to disrupt what we see as a serious problem in how learning is thought to happen in groups. As we have mentioned, there is significant work in education that understands learning to be a function of argumentation. While we agree with this basic premise, we suggest that learning theory takes a narrow view of what counts as “argument” and therefore of what contributes to learning. Argument theory is typically concerned only with what Aristotelian rhetorical theory would understand as “logos” or reasoning; indeed, it is often concerned with formal reasoning (Toulmin, 1958; van Eemeren, 1996; Bex et al, 2003). Clearly, rhetorical theory is essential here, as it is the theoretical ground from which learning theory based on argument has been built. But rhetorical theories of how people reach understanding and agreement are layered and complex and account for issues other than logos or pure reasoning. Rhetorical theory has always been concerned with issues such as social and group norms, affect, and identity. In Aristotelian rhetorical theory - to cite just one example - these issues were attributed to either “ethos” or “pathos” and worked together with “logos” to achieve aims. Our claim, here, is that any tool for analyzing discourse at it really happens in the world (i.e., not in controlled environments) must be able to account for a more complex mode of argumentation, one that accounts for group dynamics and various identity issues.
Table 1 provides a view of the coding scheme developed for the Take Two project. The sources for the scheme are both conceptual and data driven. Conceptually, we began with two sources that were intellectually relevant to the project, to the practice of discourse analysis, and to a layered rhetorical view of argumentation. We utilized Gee (2005) for his focus on identity and activity in discourse analysis. In an effort to come to a theoretically-driven understanding of “co-construction of knowledge,” we utilized work on argumentation (e.g., Walton, 1996). But our primary resource is the work of Scardamalia (2002) and Bereiter (2002).
Building an Argument |
Exploring New Ideas |
Building a Writer’s Identity |
Building a Community Identity |
|---|---|---|---|
|
|
|
|
Table 1: Argument Coding Scheme for Take Two Project
The large body of work on argument and learning - and in rhetorical theory - reinforces the importance of discourse acts like making claims and the other issues in our “building an argument” category. Scardamalia and Bereiter’s work on knowledge building provides the foundation for how we operationalized both “learning” and “co-creation of knowledge.” Specifically, we referenced their twelve principles of knowledge building for the practices that we call “exploring new ideas,” but also to identify practices we have listed as important for “community identity” (e.g., invitations and constructing connections). Rhetorical theory provides the conceptual grounds for our focus on individual and community identity as essential, but as will become visible in our discussion of results, the Buzz discourse itself makes obvious the importance of identity performances to any meaningful description of on-line activity there. The content and shape of this coding scheme has been significantly shaped by the data itself. As an understanding of argument that leads to outcomes like “understanding,” “learning,” or “knowledge,” this scheme is innovative and productive.
Sampling
Because of its size, Science Buzz provided us with a significant sampling problem. The Buzz blog has been running for 3 years and has over 1,500 threads. Coding all of the discourse was not a pragmatic possibility. We made a number of decisions to arrive at a sample. As a first step, we chose to look at threads in which the initial post was followed by 15 or more comments that came from 3 or more unique responders. Isolating threads that met this criteria still left us with thousands of individual posts, and because we coded at the T-Unit level (roughly a sentence), we were left with many thousands of units to characterize. This was still too many for human coders to handle given the time frame of the study, and so we were required to make additional sampling decisions. There is actually very little published consensus for arriving at a coding corpus in situations where a power analysis is not relevant, and so we detail our decision-making here in the interests of scholarly disclosure.
Collins et al (2007) write, for instance, that “the criteria for sample size in qualitative research are not based on probability computations but represent expert opinion” (p. 6) and go on to write the following:
It should be noted that the issue of sample size in qualitative research is a controversial one. However, as noted by Sandelowski (1995), a general rule is that sample sizes in qualitative research should not be too small that it is difficult to obtain data saturation, theoretical saturation, or informational redundancy. At the same time, the sample should not be so large that it is difficult to undertake a deep, case-oriented analysis. Teddlie and Yu (2007) referred to this balancing act in qualitative sampling as the representativeness/ saturation trade-off.
In the most methodologically detailed studies that we could find, O’Connor (2007) analyzed 10% of a larger sample (roughly 300 of 3,000 texts), but no explanation is provided for that 10% decision. In Juzwik et al’s (2006) analysis of some 1,500 articles, her team coded all of them but were only coding abstracts using a scheme that characterized discourse at units larger than the T-Unit, resulting in many fewer units of analysis than in our study. Therefore, in order to achieve “data saturation” and to achieve a situation in which we began seeing “informational redundancy,” or the phenomena of interest represented in its variation, we chose to subject 20% of posts with more than 15 comments and 3 unique responders to coding. We additionally compared this sample to the average number of posts per year on Buzz to ensure that our sample accurately reflected Buzz blog threads from the early, middle, and later stages of Buzz.
Results of Discourse Analysis of Science Buzz
In this section, we provide a general overview of our results, organized in relation to the major headings in the coding scheme. Focusing on the nature of discursive activity within the sample of Buzz blog threads, these results begin to offer answers to the first and second of our analytical questions related to community and activity. The percentages reported here are based on the number of T-Units per blog thread that could be coded. That is, the percentages reflect the number of T-Units with a particular code in the larger sample of T-units.
Building an argument
The Buzz blog is a site where informal argumentation happens. Lots of it. Perhaps not surprisingly, argumentation through making claims, contributing evidence, calling on authority, and using other argumentative strategies happens more frequently than any other single discursive category for which we coded. Slightly over sixty percent of our total sample was coded as “Building an Argument.” At the same time, however, across threads there is a high degree of variation separating the most argumentative threads from the least argumentative. Although the thread containing the least coded argument units still contains a good deal of argumentative discourse, some threads are obviously more oriented toward traditional argumentation than others are. The least argumentative thread, which was 20.72% argumentative, for example, is a thread about gephyrophobia, or the fear of crossing bridges, which was written not long after the collapse of the I-35W Mississippi River Bridge in Minneapolis/St. Paul. The most argumentative thread, in which 96.15% of T-Units were associated with argumentation, discussed whether “Eskimos have a hundred words for Snow.”
One clear descriptive statement that can be made about the discourse on Science Buzz is that it is highly argumentative, and within the category of argument, claim-driven. Almost half of the time, in fact, writers use the Buzz Blog as a space for making individual claims which we understand as statements of policy, fact, or value. Claims come in a variety of forms on the Buzz Blog, sometimes stating results of recent scientific research, sometimes taking a side in a scientific fact or discovery, sometimes articulating much simpler beliefs that undergird an argument, sometimes speculating about possible future effects of current science problems, sometimes linking events in the past to events in the present. For example, in “The Chicken and the Egg” thread, a thread that at the time of writing contained 568 comments and delved into pretty much all there is too know about chickens and eggs, claims from individuals might be observations about why chickens might not be laying: “Chickens will also stop laying if they’re ‘clucky’” or “Your chicken may be sick!!!” However, participants also provide information as well, such as when one contributor referenced the Migratory Bird Treaty Act in order to respond to another contributor’s question, stating that “The only birds that don’t fall under this act are pigeons, sparrows, and starlings, since these species are not native to the U.S.”
While claims are a dominant argumentative move, they are not the only move visible. Furthermore, they are not the strongest moves possible: claims absent citations of evidence or nods to authority often make for weak arguments. Buzz writers vary in the extent to which they back up their claims with evidence or citations of authority. One particular sample thread about potential changes to U.S. coins did not contain any direct citations of evidence, while another thread in which commenters detail the lifestyle changes they’ve made to conserve energy contained a full 33% of T-Units citing evidence. Similarly, calls to authority ranged from a low of less than one percent of T-Units coded to a high of 31%. While there is much variation in these traditional strategies of supporting claims, we also used a coding category called “argument-other” to capture argumentative strategies outside more canonical categories (e.g., the use of “rhetorical questions” in a chain of reasoning). The use of our argument-other coding category suggests that writers on the Buzz Blog are using other strategies to develop and support the claims they make. We believe that additional analysis will reveal useful categories of argumentative work on Buzz.
Building the writer’s identity
The Buzz blog is also a site where writers build and articulate individual identities, with this category being used to describe over one-fourth (25.22%) of the total sample. As previously described, Buzz writers most often built arguments in very direct ways - by positing direct claims about the topic at hand rather than, for example, indirectly citing evidence. However, identity-building moves were more frequently indirect. That is, writers established who they were more often by articulating affect or by using technology than they did by overtly articulating a role that they play, citing their education or values, or by calling on their status or lack of status. Of the 25% of T-Units that were associated with building individual identities, 12.13% of T-Units were labeled as use of affect, 5.67% as using technology, 3.93% as articulating values, and the remaining 3 percent were distributed among articulating a role, status, education, or place.
Generally, building identity appears to be more an implicit or indirect result of the site and the interaction that happens within it than something that writers directly attempt to take part in. For example, contributors rarely introduce themselves to conversations within the thread by telling others about their life outside of the Buzz Blog. Instead, identities emerge particularly through infusing their writing with emotion, humor, and other markers of affect, as is particularly notable in threads like, “When you die you become a colossal squid.” Even invocations of status, roles, and education are often indirect and emerge in conventions of this kind of conversation. For example, many identity building moves happen through the signature lines writers use at the ends of posts, as in the case of one particular contributor who declares her education and her expert status by signing off as “University of Minnesota – Poultry Specialist.”
While the number of direct citations of role, status, or place is relatively infrequent, individuals are more likely to articulate individual roles or experiences than to identify with the roles or experiences that others have articulated. That is, they are more likely to deploy discursive moves that are individual identity performances rather than moves that we understand as “community building” (see below). Notably, it also seems likely that a closer coding and examination of the “other” category may provide us with new categories of individual identity-building that lie outside our current scheme.
Exploring New Ideas
Based on our coding, we cannot claim that the Buzz blog is a place where writers generally explore new ideas. Only 1.81% of T-Units were assigned to this code. Although the discursive moves that we associate with exploring new ideas are found infrequently in terms of percentage, they are present in all but four sample threads. This code seems to work less to identify a common discursive move across the sample than it does to point us to generative “sweet spots” within each individual thread. It is likewise notable that writers were most likely to explore a new idea by introducing their own individual ideas than by improving on another’s idea or by engaging in the kinds of conversations about definition or issue that constitute rhetorical stasis. For example, phrases coded with the exploring new ideas category often were prompted by genuine questions that the individual contributor had related to the topic, like “Is “Eskimo,” as it is applied to Eskimos (by themselves), a post-contact term?” or they brought a previously unconsidered point of view into the conversation, as was the case with the question, “What’s with health insurance?” that emerged in the thread about whether or not to get a flu shot. From these kinds of examples, writers often explore new ideas as the result of individual questions or ideas that seem to come “out of the blue,” but at the same time invite participation from others. These “exploring” discursive moves seem to promote and energize the conversation that comes after them.
Building Community Identity
Finally, the Buzz blog is a site where writers engage in building community. All threads contained community identity building moves, and over ten percent of sampled threads (11.36%) were coded as contributing to building community identity. Interestingly, much like in the case of building individual writer’s identities, Buzz writers employed some types of community building strategies much more than others. Specifically, writers constructed community identity much more frequently by inviting participation or by constructing connections to individuals or to prior statements, than they did by articulating shared experiences or roles. When Buzz writers were building community, they most often appeared to be doing it as a direct result of the conversation at hand, not as an overt attempt to establish a relationship with another individual or group within the thread. For example, writers often asked questions that were more than just “rhetorical” argument builders - these questions genuinely invited response from others. In just one thread related to global warming (“Siberia is Warming Up”), community-building invitations ranged from questions like, “What does the melt mean?” to “Will we survive?” to “when will globle [sic] warming start to impact the world ?”
In sum, community-building moves most prevalent in our sample were generally not moves of articulating group identities or boundaries, claiming relationships, or claiming that the Buzz community constitutes a “we.” Instead, community-identity building happened more frequently in terms of momentary acknowledgments placed in the context of the conversation about the topic or controversy being discussed, as individuals articulated personal questions and esponded to other individuals directly, as well as to the questions and comments that other contributors had previously posted.
Conclusion
Our analytical work must be seen as preliminary. Our analysis to date suggests that argumentative strategies such as making claims are most common on the Buzz blog. These forms of reasoning strategies are the focus of much of the discourse-based studies of concepts like “learning” in group environments. Yet what our analysis of the Buzz blog also shows is a range of strategies that we associate with constructing individual identities (more common) and group identities (less common). Although it is preliminary, our work has identified patterns in on-line activity yet to be identified in museum studies. Moreover, we have paid attention to discourse moves that are often ignored in studies of learning in both formal and informal settings. We suggest that the range of identity performances is fundamental to on-line community dynamics and integral to any understanding of activity and any judgments made about the nature and status of the learning or knowledge building in those communities.
Based on the descriptive data shared above, we continue to contemplate whether and how the co-construction of knowledge takes place on Science Buzz. Based on the aforementioned theories of knowledge construction from Bereiter and Scardamalia, discursive acts related to exploring new ideas and creating a group of people working together are crucial. While these moves were not dominant on Science Buzz in terms of the frequency, they were present across threads. Community building moves, in particular, appear related to the activity of discussing current and past science topics in a forum like Science Buzz. That is, building community was not an end in itself, but rather an activity in conjunction with discussing or debating ideas. Critically, building community in the service of discussing ideas is foundational to any co-constructing activity, and we see the presence of community building moves as an especially positive feature of what is happening as a result of the implementation of Web 2.0 technology in this particular on-line museum site. Because we believe these to be important issues for museums today, our ongoing analytical work will focus on the dynamics described here.
Acknowledgements
This project is made possible by a grant from the U.S. Institute of Museum and Library Services. The Institute of Museum and Library Services is the primary source of federal support for the nation’s 122,000 libraries and 17,500 museums. The Institute's mission is to create strong libraries and museums that connect people to information and ideas.
References:
Allen, S. (2002). Looking for Learning in Visitor Talk: A Methodological Exploration. In G. Leinhardt, K. Crowley, & Karen Knutson (Eds.) Learning Conversations in Museums. Mahwah, NJ: Lawrence Erlbaum Associates, 259-304.
Andreissen, J, M. Bake, & D. Suthers (2003). Arguing to Learn: Confronting Cognitions in Computer-Supported Collaborative Learning Environments. Dordrecht/Boston/London: Kluwer Academic Publishers.
Ash, D. (2003). Dialogic inquiry in Life Science Conversations of family groups in a museum. Journal of Research in Science Teaching 40(2): 138-162.
Barry A. (2006). Creating A Virtuous Circle Between a Museum’s On-line And Physical Spaces. In J. Trant and D. Bearman (eds.). Museums and the Web 2006: Proceedings. Toronto: Archives & Museum Informatics. Last updated February 28, 2006, consulted 2008. http://www.archimuse.com/mw2006/papers/barry/barry.html
Bartley E. and J. Hancock (2006). A Question of Interactivity. In J. Trant and D. Bearman (eds.) Museums and the Web 2006: Proceedings. Toronto: Archives & Museum Informatics. Last updated March 7, 2006. Consulted 2008. http://www.archimuse.com/mw2006/papers/bartley/bartley.html
Bazerman, C, & P. Prior (2004). What writing does and how it does it: An introduction to analyzing texts and textual practices. Mahwah, NJ: Lawrence Erlbaum.
Bell, P. (2000). “Scientific arguments as learning artifacts: Designing for learning from the web with KIE”. International Journal of Science Education 22(8): 797-817.
Bereiter, C. (2002). Education and mind in the knowledge age. Mahwah, NJ: Lawrence Erlbaum Associates.
Bex, F, H. Prakke, C. Reed, and D. Walton (2003). Towards a formal account of reasoning about evidence: argumentation schemes and generalisations. Artificial Intelligence and Law 12: 125–165.
Collins, K.M.T., A.J. Onwuegbuzie, Q.G. Jiao. (2007). A Mixed Methods Investigation of Mixed Methods Sampling Designs in Social and Health Science Research. Journal of Mixed Methods Research 1: 267-294.
De Vries, E, K. Lund, and M. Baker (2002). “Computer-mediated epistemic dialogue: explanation and argumentation as vehicles for understanding scientific notions”. The Journal of the Learning Sciences 11(1): 63-103.
Dijk, T. A. V. (1997). Discourse as interaction in society. In T. A. V. Dijk (Ed.), Discourse as social interaction. London: SAGE, 1-37.
Driver, R., P. Newton, & J. Osborne, (2000). Establishing the norms of argumentation in classrooms. Science Education 84(3), 287–312.
Duschl, R. A., K. Ellenbogen, & S. Erduran (1999). Understanding dialogic argumentation among middle school science students. Invited paper at the annual meeting of the American Educational Research Association (AERA), Montreal.
Duschl, R., K.M. Ellenbogen, & S. Erduran (1999). Middle school students’ dialogic argumentation. In Komorek, M., Behrendt, H., Dahncke, H., Duit, R. Gräber, W. & Kross, A. (Eds.), Research in science education: Past, present and future. Proceedings of the Second International Conference of the European Science Education Research Association (ESERA), Volume 1/2, pp. 420-422.
Duschl, R., & J. Osborne (2002). Supporting and promoting argumentation discourse. Studies in Science Education 38, 39–72.
van Eemeren, F.H., R. Grootendorst, and F. Snoeck Henkemans (1996). Fundamentals of Argumentation Theory. Mahwah, NJ: Erlbaum Associates.
Erduran, S., S. Simon, & J. Osborne (2004). TAPping into argumentation: Developments in the use of Toulmin’s Argument Pattern in studying science discourse. Science Education 88(6), 915-933.
Fairclough, N. (1992). Discourse and social change. Cambridge, UK: Polity Press.
Falk, J.H., M. Storksdieck & L.D. Dierking (2007). “Investigating public science interest and understanding: Evidence for the importance of free-choice learning”. Public Understanding of Science, 16(4).
Gee, J.P. (2005). An Introduction to Discourse Analysis: Theory and Method. Second Edition. London: Routledge.
Goldman, S. R., R.A. Duschl, K. Ellenbogen, S. Williams, & C. T. Tzou (2003). “Science inquiry in a digital age: Possibilities for making thinking visible”. In H. van Oostendorp (Ed.) Cognition in a Digital World (pp. 253-284). Mahwah, NJ: Erlbaum.
Goldman, H. & M. Waldman (2002). “There's something happening here, what it is ain't exactly clear”. In D. Bearman and J. Trant, (eds.).Museums & the Web 2002: Proceedings. Pittsburgh, PA: Archives & Museum Informatics. Last update not provided, Consulted 2008. http://www.archimuse.com/mw2002/papers/haleyGoldman/haleygoldman.html
Hindmarsh, J., C. Heath, D. Vom Lehn, and J. Cleverly (2005). “Creating Assemblies in Public Environments: Social Interaction, Interactive Exhibits and CSCW”. Comput. Supported Coop. Work 14, 1 (Feb. 2005), 1-41.
Jimenez-Aleixandre, M., A. Bugallo, & R. Duschl. (2000). “Doing the lesson” or “doing science”: Argument in high school genetics. Science Education 84(6), 757–792.
Juzwik, M., S. Curcic, K. Wolbers, K.D. Moxley, L.M. Dimling, & R.K. Shankland (2006). “Writing Into the 21st Century: An Overview of Research on Writing, 1999 to 2004”. Written Communication 23(4): 451-476.
Keene, S. (1998). Digital Collections: Museums and the Information Age. Oxford: Butterworth-Heinemann.
Ockuly, J. (2003). What Clicks? An Interim Report on Audience Research. In D. Bearman and J. Trant, (eds.). Museums & the Web 2003: Proceedings, Toronto: Archives & Museum Informatics. last updated March 13, 2003 Consulted 2008. http://www.archimuse.com/mw2003/papers/ockuly/ockuly.html
Packer, J., & R. Ballantyne (2005). “Solitary vs. shared learning: Exploring the social dimension of museum learning”. Curator, 48, 177-192.
O’Connor, P. (2007). “Doing boy/girl and global/local elements in 10-12 year olds’ drawing and written texts”. Qualitative Research 7: 229-247.
Scardamalia, M. (2002). “Collective Cognitive Responsibility for the Advancement of Knowledge”. In B. Smith and C Bereiter (Eds.) Liberal Education in a Knowledge Society. Chicago: Open Court Publishing.
Schaller, D. T., S. Allison-Bunnell & M. Borun (2005). Learning Styles and Online Interactives. In J. Trant and D. Bearman (eds.). Museums and the Web 2005: Proceedings, Toronto: Archives & Museum Informatics. last updated May 16, 2005. Consulted 2008. http://www.archimuse.com/mw2005/papers/schaller/schaller.html
Schiffrin, D., D. Tannen & H. E. Hamilton (eds.) (2001). Handbook of Discourse Analysis. Oxford: Blackwell.
Toulmin, S. (1958). The uses of argument. Cambridge: Cambridge University Press.
vom Lehn, D, C. Heath and J. Hindmarsh (2001). “Exhibiting Interaction: Conduct and Collaboration in Museums and Galleries”. Symbolic Interaction, vol. 24, pp. 189-216.
Walton, D.N. (1996). Argumentation schemes for presumptive reasoning. Mahwah, NJ: Lawrence Erlbaum Associates.
Wood, L. A., & R. O. Kroger (2000). Doing discourse analysis. Thousand Oaks, CA: Sage.
Zimmerman, H.T. (2005, September). Adolescent argumentation in online spaces. Presentation at First Congress of the International Society for Cultural and Activity Research (ISCAR), Sevilla, Spain. Last update not provided. Consulted January 2009. http://depts.washington.edu/cogstudy/everydaycognition/documents/iscar2005/Zimmerman-online-arguments.pdf
