ArticlePDF Available

Introducing Backstage – a digital backchannel for large class lectures

  • QualityMinds

Abstract and Figures

Purpose – This paper seeks to report on the conception of a novel digital backchannel, Backstage, dedicated to large classes, aiming at empowering not only the audience but also the speaker, at promoting the awareness of both audience and speaker, and at promoting an active participation of students in the lecture. In learning settings with a large and typically passive audience, the goal of Backstage is to foster active participation and facilitate collaboration akin to small-learning groups. Design/methodology/approach – The authors present the concept of a novel kind of digital backchannel that supports different forms of inter-student communication via short microblog messages, social evaluation, and ranking of messages by the audience. The backchannel further supports immediate concise feedback to the lecturer of selected and aggregated students' opinions, making it possible to strengthen the lecturer's awareness of students' difficulties. Findings – The concept is the outcome of a joint effort between computer scientists and educational scientists, and unifies technical and usability aspects with educational claims. Originality/value – The concept combines computer-mediated communication with elements of social media that not only foster active participation and collaboration among a large and otherwise passive audience, but also provide social means to self-regulation and management of the backchannel discourse. The design of the backchannel is strongly influenced by findings of educational sciences.
Content may be subject to copyright.
Alexander Pohl*, Vera Gehlen-Baum, François Bry*
* Institute for Informatics, University of Munich
Oettingenstraße 67, 80538 München, Germany
Department of Educational Technology, Saarland University
Campus C5 4, 66123 Saarbrücken, Germany
This article reports on the conception of a novel digital backchannel, project name Backstage, dedicated to large classes
aiming at empowering not only the audience but also the speaker, at promoting the awareness of both audience and
speaker, and at promoting an active participation of students in the lecture. The backchannel supports different forms of
inter-student communication via short microblog messages, social evaluation and ranking of messages by the students
themselves. Backstage allows for the aggregation of student's opinions aiming at supporting the community building of
the audience and strengthening the students' awareness of and co-responsibility for the class work. The backchannel
further supports immediate concise feedback to the lecturer of selected and aggregated students' opinions making it
possible to strengthen the lecturer's awareness for students' difficulties. The backchannel's conception is motivated by
learning sciences' findings and theories.
This article is an extended version of (Bry et al., 2011) published at the IADIS International Conference e-Society, held
in March 2011 in Avila, Spain.
E-Learning, Social Media, Enhanced Classroom, Digital Backchannel
Fifty years ago, European higher education opened up to a much larger part of the society, turning
popular courses of study to mass education and universities to degree factories (Hartmann, 2007): Class
lectures with several hundred students has become daily life in European academia. In such classes, both a
lecturer's awareness and the students' participation in the lecture tend to dramatically decrease.
A well-recognized issue coming along with large audiences is passivity among students. This is often
seen as one of the biggest problems in learning arrangements (Renkl, 1996). With increasing group size
social barriers are raised that make members standing out of the group feel uncomfortable, e.g., when asking
a question or commenting on the discourse. Students may consider themselves as being perceived to be
incompetent in case they would frankly utter comprehension problems. As a consequence, students feeling
overchallenged typically expose a passive behavior (Schworm and Fischer, 2006). The cause of such
misjudgments is often that students are isolated from each other and have few possibilities to sense relevant,
and often similar, impressions of other students. They might, e.g., miss to recognize that they are not alone in
having comprehension problems. Computer-mediated communication by so-called dig ital backchannels can
help listeners to share their impressions and engage in collaborative activities, without interrupting the
lecture. In order to not only strengthen, but also to increase awareness of ongoing group dynamics, both
sophisticated social interactions and group communication are necessary.
Since a decade or so, digital backchannels have been spontaneously used in mostly IT and Computer
Science professional conferences with large audiences to great success. At the beginning of the decade, one
of the authors of this paper has experienced the use of an IRC1-based digital backchannel at a twice-yearly
meeting of the W3C: The backchannel enabled an audience of between six and eight hundred to quickly
agree on objections to a speaker's statement and to choose a few audience members to wave the flag. At some
Computer Science conferences, like multimedia or Semantic Web conferences, Twitter has become a
commonly used tool for commenting presentations, thus raising the audience's awareness to salient, or
questionable, paper presentations. Such a use of a general purpose digital backchannel is not unproblemat ic,
though. In her blog post ―spectacle at Web2.0 Expo fro m my perspective‖ 2 of November 24, 2009 danah
boyd reports on how, during a talk she gave, she lost the contact to her audience which increasingly moved
from the frontchannel of the talk to backchannels beyond her reach. Admittedly, part of the problem was that
the backchannels used by the audience de facto excluded the speaker.
This article reports on the conception of the digital backchannel Backstage dedicated to large classes
aiming at empowering not only the audience but also the speaker, at promoting the awareness of both
audience and speaker, and at promoting an active participation of students in the lecture.
The contributions of the article are as follows:
The description of a modus operandi for a use of a digital backchannel in large classes
The functionalities needed for a digital backchannel to support this modus operandi
The scientific underpinning of the approach with findings and theories of learning sciences
The research reported about in this article is a joint work between computer and learning scientists.
Like every Monday morning, Sue Byrd is giving the freshmen's lecture on Linear
Algebra for an audience of 280 students. She feels better because the backchannel is on
and the students begin to be used to it. A reminder would not harm, though: "If at some
point you get lost, post it on the backchannel", Sue says, "And refrain from messages on
yesterday evening's soccer." When it comes to the Laplace formula for computing the
determinant of an n-by-n matrix, there is a rush of students' messages. Sue sees that many
are posting messages and therefore awaits a few seconds for the backchannel to inform
her: "What is a matrix' signature? was asked by 37%. She recalls the concept needed
for understanding Laplace formula and goes ahead with her explanation. Meanwhile,
Bob is engaged in a private chat on the backchannel with Ann: "I thought the formula
was called Leibniz formula. Is Dr Byrd confusing names?" Ann finds the remark
interesting and suggests to Bob to make it public. He does so and within five minutes
more than 20% of the lecture’s audience support the question, leading the backchannel to
forward it to Sue who mentions that there are two distinct formulas for determinants of n-
by-n matrices. At the same time, Charly asks on the backchannel "Recall the formula for
the det of 2x2 matrix". Ann, who often tends to be rather direct, posts a public answer:
"Dr Sue explained it last week but you were once again skiing". This sharp answer gets
support, Charly's question is turned down by 80% of the students and therefore not
forwarded to the lecturer. After the lecture, Sue has a look at the backchannel's
anonymized log: She is satisfied that the communication have been limited to 18 of the 90
minutes of the lecture and to three issues - among others yesterday's victory of the FC
Bayern. She notices that the number of students more interested in FC Bayern's victory
than in determinants fits well with that of student who did not made their last week's
home work.
This admittedly simplified and idealized scenario stresses the following points:
1 Internet Relay Chat
Even though it is preferable that students immediately ask questions when they do not recall a
concept, in a large class lecture this is meaningful only if the question is relevant to sufficiently
large a number of students.
Side remarks to neighbors might be relevant to the whole audience and worth forwarding to the
lecturer. Deciding whether this is the case should be a social decision.
Knowing what students thought, or did, during a lecture is of considerable help for a lecturer.
The next section elaborates on the backchannel's functionalities outlined in the above scenario.
Yardi (2008) suggests that strengthening the collective consciousness of, and empowering, large lecture
audiences by an appropriately designed digital backchannel is possible. Appropriate forms of communication
and feedback may support students engaging in collaborative group discussionsor more specifically, fast
information exchangeand actively participating in the lecture discourse. The importance of active
participation is also highlighted by practitioners such as Arthur F. Thurnau awarded Professor of Education
at Michigan University, Deborah Loewenberg Ball, who states that ―Students' opportunities for learning are
reduced when their role is mainly that of spectators in large lecture classes‖3. Fast information exchange may
comprise asking and answering of questions or engaging in controversial discussions. Though, currently,
extensive group work is hardly possible in large lectures, well selected collaborative activities could reduce
negative effects of large class lectures. One reason is that with those collaborative activities ―inadequate
reasoning will be e xposed, […] and higher quality understanding will emerge‖ (Slavin, 1996, p. 49). In the
following, the forms of communication on the backchannel is described that may foster lecturers' and
students' awareness of group dynamics, that embraces empowerment of students and calls for active
3.1 Short Messages
In academic conferences computer-mediated communication software has been leveraged as digital
backchannels (e.g., McCarthy and boyd, 2005; Saunders et al., 2009). With digital backchannels students can
engage in group discussions during presentations without disrupting the lecturer's discourse. Due to their
success, recent efforts in e-learning have been focused on the application of digital backchannels in higher
education (e.g., Costa et al., 2009; Yardi, 2008).
Though, primarily, user-driven trends have made microblogs the privileged media for digital
backchannels (McNely, 2009), microblogs fulfill the requirements imposed by learning settings. They
provide informal group communication within the classroom appropriate to post work-related information, to
get aware of other's impressions and to build common-ground (Zhao and Rosson, 2009). The brevity of
microblog messages (typically less than 200 characters) make students write their contributions in a concise
mannerone piece of information per message. This makes contributions written and read more quickly.
The fact that students write their questions or argumentatio ns, instead of talking about them, is giving them
more time to reflect about the discourse. This could lead to a deeper understanding of arguments posted by
others (e.g., Kuhn and Goh, 2005). Concise messages could support students in dealing with the backchannel
discourse more intensely. Thus, compared to other backchannel media, like IRCs, microblogs can help to
keep students' distraction at a lower level. Being persistent, backchannel conversations can furthermore be
reviewed and rearranged specifically for reworking a lecture. We believe that the possibility to benefit from
collaboratively created lecture notes increases students’ incentives to partake. Since microblogs are easy to
use, following the principle of the least surprise, special training is often unnecessary.
Many microblogs like Twitter allow the use of a free and unbounded vocabulary for tags, so -called
hashtags that are extracted from messages and displayed in a tag cloud. Messages can be sorted out, or
retrieved, using tags. Tag clouds give an insight into the tagging vocabulary. During lectures, however, the
usefulness of free tagging vocabulary is doubtful. Since finding useful tags is a non-trivial task, e.g. avoiding
synonyms and being aware of case sensitivity, it can contribute to students' distraction and be subject to
abuse. Thus, experimental evaluations need be carried out for determining which form of tagging
vocabularies can be useful on the backchannel. In the following, we describe a limited form of tagging with
predefined and very small vocabularies.
3.2 Modes of Communication
The conception of a digital backchannel for large lectures we developed includes public, private as well
as anonymous and pseudonymous forms of inter-personal communication which are specified by the user
within the message. Public communication is visible to the entire audience and likely to be the standard form
of communication within a lecture class. Private communication allows a student to share opinions and ideas
with others in confidence and to achieve social support: the other students can encourage her to make her
contributions public. Thus, private communication can be a stepping stone to public communication. Public
messages may, but do not need to, refer to other students using the @-prefix of IRCs and Twitter (Figure 1
and Figure 2). Private communication is only visible to students that are specified by the author of the
message (Figure 3). We therefore introduce the @>-prefix in the developed backchannel. Referring to the
scenario described in Section 2, Ann's public message to Charly and Bob's private backchannel message to
Ann is given in the Figure 2 and 3. Additionally, Backstage can be customized to allow students to post
messages anonymously, by using the symbols *** (Figure 4). Research indicates that not showing the author
of a message could lower the barrier to participate in backchannel conversations . It can hardly be stated a
priori which forms of communication are preferable for a given lecture, since experimental findings are
inconclusive (Er and Ng, 1995). Thus, the lecturer, possibly in agreement with her audience, should be able
to customize the communication to fit the requirements at hand.
Figure 1: Editor window with a public message, visible to
Figure 2: Editor window with a public message to Charly,
visible to all. Note the first symbol: @
Figure 3: Editor window with a private message to Ann,
only visible to her and the sender. Note the first symbols:
Figure 4: Editor window with an anonymous message,
visible to all. The author of the message is not shown. Note
the first symbols: ***
3.3 Collective Message Rating and Annotations
Students can approve or reject messages by means of a concise rating scheme: A positive rating exp resses
approval and a negative rating expresses rejection. By aggregating these ratings a top-k ranking of messages
can be determined that reflects the relevance of messages as considered by the audience. Such a top-k
ranking of messages provides the lecturer with a very concise content-related collective appraisal of the
relevant backchannel communication. Regarding the top-k messages lecturers can keep additional
explanations short or present them in more detail if the audience indicates interest or problems.
We believe that not only a relevant message but also messages that refer to, or are referred to by that
message, might be of interest to the lecturer. For example if an answer is considered highly relevant the
corresponding question should be likewise considered relevant. We intend to achieve this by adaptations of
eigenvector-based ranking algorithms like, e.g. HITS (Kleinberg, 1998) or the more involved PEST (Weiand
et al., 2010).
To resemble the natural decline of relevance induced by the progress of the lecture discourse, ranking of
messages should be subject to aging. Aging causes messages to climb down the ranking over time, unless
students continue to rate them (Figure 5).
Figure 5: Aging makes messages climb down the ranking over time (indicated by the two clocks at the top and the
arrows at the center), such that a more recent message may replace an older, though better rated, one (see the message at
the top in the right column; the percentages indicate the aggregated rating values for a message).
Besides, aging serves two additional purposes. First, students' interactions necessary for estimating the
relevance of a message are reduced. Second, aging makes the ranking robust against varying backchannel
activity. Otherwise, messages that are rated highly positively in peaks of backchannel activity would not be
turned down and hence congest the ranking. We intend to investigate in experimental evaluations which
aging formula should be used. Ratings of messages also allow for the determination of intra-community
status of their respective authors. The more positive ratings the messages of a student receive, the more intra-
community status the student gains (Figure 6).
Figure 6: Intra-community status of a student can be determined by the ratings of her messages. The rating of a student,
in turn, can be weighted with her intra-community status (the green check-marks illustrate approval, the red cross-marks
In turn, intra-community status can be used as a weight that is put on the student's backchannel
interactions. Students lose ground if the large part of the audience continuously rejects their contributions,
but gain influence when their contributions are mostly approved. Thus, intra-community status gives students
feedback on the value of their contributions and can be an incentive to partake. The Figure 7 shows how
students can rate a backchannel message on Backstage.
Figure 7: By right-clicking at a message in the timeline the corresponding context menu for the message is shown. Using
the context menu the user can reply to, rate (like or dislike), or annotate the message with further keywords (e.g.
important). The lecturer predefines the annotations that should be available.
Apart from rating, on Backstage messages may also be annotated, e.g., as irrelevant messages. Therefore,
a very limited keyword vocabulary for this kind of tagging is predefined by the lecturer during the
customization of Backstage. Additionally, annotations can also be automatically determined. For examp le,
the system could determine off-topic messages (e.g., specified by a database query) and annotate them
Messages can be given more than one annotation. The annotations can be aggregated, thus yielding a
degree for each keyword. Similar to rating a backchannel message, intra-community status may be involved
in the aggregation. Both annotations and rating allow for fine-grained feedback to the lecturer. For example,
the message ―Publishing board drawings online would save us time‖ could be rated highly pos itively but may
likewise be annotated as irrelevant with respect to the lecture. Especially, this examp le highlights interesting
dependencies between ratings and annotations: Messages that are annotated as irrelevant by a sufficient
number of students can be safely ignored, or only weakly considered, in the ranking of backchannel
messages. Therefore, it needs to be specified which and how annotations influence the message ranking.
One issue of rating based collaborative filtering is the so-called First-Rater-Problem (Melville et al.,
2002): A message cannot be recommended to the lecturer, or other users, unless some user has previously
rated the message. Hence, when the students are reluctant to rate, Backstage cannot provide the lecturer with
a top-k ranking. As a remedy, it has been suggested to combine collaborative filtering with content-based
filtering. We believe that for Backstage there are interesting variants of such hybrid filtering approaches
conceivable, especially when keeping the interplay between the user-centric rating and the content-centric
collective feedback to the lecturer in mind.
3.4 Collective Feedback to the Lecturer
Backchannel co mmunication can be distinguished into process-oriented and content-oriented
communication, among others. Process-oriented communication deals with the analysis and steering of the
frontchannel discourse, e.g., pace of lecturing. Content-oriented communication deals with the public
frontchannel and backchannel discourse, e.g., critics or appraisal (Cogdill and Kilborn, 2001). Both kinds of
communication may contain relevant feedback for lecturers during their discourses. However, this would
involve comprehending the backchannel discourse while lecturing, a task that is often infeasible. To give
lecturers quickly comprehensible and concise feedback during lecturing makes aggregation. Suppose that
collective feedback on the lecturing pace is sought for. The lecturer customizes the backchannel with a field
―pace‖ and two notifications, ―too slow‖ and ―too fas t‖. During the lecture students can notify the lecturer
regarding her lecturing pace. These notifications are aggregated and graphically displayed at the lecturer's
dashboard very concisely to give the lecturer the possibility to quickly grasp the audience's feedback and
react on it.
In order to avoid giving unconsidered feedback to the lecturer, students explain their feedback in a
backchannel message. As mere messages, other students can rate and annotate them. These ratings can
further be considered in the aggregation of the feedback (Figure 8).
Figure 8: This figure shows two messages that were annotated with "Question". A student approves the two messages.
These messages, together with their ratings, slip in the lecturer’s aggre gated overview of the backchannel, that may
underly aging.
The Figure 9 to Figure 12 show the way collective feedback to the lecturer is currently given on
Backstage. It has been shown that assistance in how to give feedback and comments, among other things, is
necessary, (e.g., Stegmann et al., 2007). Thus, it could be helpful to add further instructions to Backstage as
described in Section 4. Similar to the relevance ranking of messages, aggregated feedback to the lecturer
underlies aging, in order to resemble the progress of the lecture discourse. Depending on the customization of
the backchannel, the impact of students' feedback to the lecturer can be weighted with their intra-co mmunity
status. That is, feedback of students with high intra-community status is considered more important than that
of students with low intra-community status. Arguably, whether to use egalitarian or status-weighted
aggregation depends on various factors, for example, the type of feedback and the constitution of the
audience. Further research regarding this topic is necessary.
Figure 9: As a feedback that is aggregated and displayed
on the lecturer’s dashboard the user selects the category
Figure 10: After selecting the category of the feedback
(note the flag-icon near the top right corner) the user
writes a message providing additional information for the
Figure 11: The public message and its annotation are
visible to all. Thus, users may rate the message. The
ratings may be considered in the aggregation of the
collective feedback to the lecturer.
Figure 12: At the lecturer’s dashboard the aggregated
number of messages annotated with Question, Remark and
Off-Topic, among others, are displayed.
3.5 Performance Feedback to the Students
Beside the lecturer, also students may obtain feedback by data that is acquired both during and after a
lecture. In addition to the data of their backchannel comments, students are assessed in short surveys to gain
information regarding their current knowledge, motivation and progress. Giving a student an overview on the
data can contribute to her reflections and learning (e.g., Hattie and Timperley, 2007). Topic-related
knowledge can be tested by short quizzes consisting of questions and sets of possible answers. When such a
quiz is opened by the lecturer, students choose from the possible answers for a specified amount of time
(Figure 13). This can be comb ined with discussions on the backchannel. When the specified time runs out the
lecturer closes the quiz. The aggregated result is then displayed at the student's and the lecturer's dashboards
(Figure 13). On the basis of these results, further backchannel or frontchannel discourse can emerge. Not
only do quizzes allow students and lecturers to get an overview of the individual and collective learning
performance, but also does it break up the lecture discourse. In his review on the effects of large classes,
Cuseo (2007) remarks that students attention and concentration drastically declines after ten to twenty
minutes of continuous lecturing. Additional to multiple choice quizzes which usually comprise answers that
are either correct or incorrect, further quizzes with different kinds of students’ responses are conceivable. For
example, quizzes may assess the students’ opinions or ask for keywords best des cribing the subject at hand.
After lectures, students may fill out session-related surveys that poll interest, individual imp ressions
regarding the lecture, and the perceived usefulness of the backchannel discourse. Also, the data retrieved
from these surveys are processed and displayed on the dashboards. Moreover, the data are also relevant
datasets for experimental evaluations.
Figure 13: (Back) The student's view of the quiz is shown after the lecturer has opened the quiz. (Front) The lecturers
view of the quiz' aggregated results is shown after the lecturer has stopped the voting. The correct answer and the bar
indicating the number of students who have chosen the corresponding answer is highlighted in green. The incorrect
answers (and the respective bars) are highlighted in red.
One issue that might arise with such a highly customizable digital backchannel is to find the appropriate
fine-tuning to fit the lecture setting at hand. Though the realization of Backstage is focused on easy
customization, the approach needs to be thoroughly tested. Also, Backstage is to be deployed into real lecture
settings in order to evaluate its effectiveness for s tudents’ learning. These evaluations are done on the basis
of data that are collected on Backstage as well as on the students' knowledge gain. It is difficult to determine
whether the benefits of using Backstage for learning as described above results in an imp rovement of
students’ grades. Yet, we a rgue that an improvement of students grades might be too restricted an indicator
for learning success. The availability of the digital backchannel to students can contribute to a more
enjoyable and active learning environment and thus supports learning on a more mediate level. We intend to
provide a thorough discussion of this topic in the near future.
As a new technology, lecturers and students need to be guided in how to use the backchannel. Especially,
more fine-grained guidance is necessary on how students communicate on the backchannel. Studies suggest
that many students have difficulties in correctly formulating questions or argumentations. Moreover, findings
show that students create arguments and counterarguments on a rather superficial way (e.g., King, 2002) or
do not take the arguments of others into account (Stegmann et al., 2007). One way to overcome these
problems is to give students concise instructions on tasks like asking questions or fo rmulating argu ments
(e.g., King, 1997; Stegmann et al., 2007; Weinberger et al., 2010). This can be accomplished by using so -
called scripts which have been shown to be effective (e.g., Steg mann et al., 2007; O'Donnell and Dansereau,
1992; Weinberger et al., 2010). Especially when dealing with novices, learning in digital environments
should be guided (Kollar et al., 2007). When becoming more advanced backchannel users, these scripts may
gradually be faded out (Wecker and Fischer, 2007). The fading of scripts can be supported by data that is
collected on Backstage (cf. Section 3.5) and thus allows for the adaption of scripts to the student’s individual
needs. It is reasonable to include various kinds of scripts into Backstage to support students when this seems
necessary. Also, this kind of support could be helpful, based on the findings of Aleven and colleagues (2003),
since they found out that students are often not aware of needing help.
Harry and colleagues (2009) present a special purpose backchannel that supports Question -and-Answer
sessions after conference presentations. During a presentation listeners can post their questions on the
backchannel that can be rated by other listeners to either approve their relevance or to reject them. According
to ratings a ranking of the eight most positively rated messages are determined and projected on the wall. The
panelist may select the best rated questions to answer and tally them on the backchannel. Questions
previously posted and tallied cannot be posted again. The rating scheme, though comprehensible, considers
aging of messages and the activity of the audience. That is, in the ranking highly ranked old messages may be
replaced by less highly ranked but recent messages. The article presents empirical findings on the use of the
backchannel in various, mainly industrial conferences. The article also describes experiences made with the
backchannel during those conferences .
The backchannel only allows directed communication from the audience to the panelist. Thus,
collaborative activities are only marginally supported. This is also highlighted in the paper: Listeners
complained about the missing possibility to reply to other listeners. The backchannel does not provide intra-
community status and, thus, the backchannel interactions are egalitarian. While this may be reasonable in
presentations with short-existing audiences, in lectures with long-existing audiences intra-community status
may be worthwhile.
Ebner (2009) has investigated how presentations can be improved by microblogging channels. The
experiment was conducted at the E-MEDIA 2008 conference using Twitter as the microblogging platform.
During presentations the Twitter timeline was projected on the wall via a separate projector. This made it
possible for the audience to follow both presentation and the backchannel discourse on Twitter at the same
time. Mainly, the backchannel was used to share links as a supplement of the presentation, to comment on the
presentation, i.e. reacting on the statements of the speaker, as well as on the quality of the slides. The article
highlights Twitter as a way to persist both the presentations as well as the discussions on the microblogging
platform. On the basis of their findings, the authors conclude that the use of microblogs improves learner-
lecture interactions in large classrooms, since it may provide solutions to major problems in learning settings:
Feedback lag (students hesitate to give feedback), Students' apprehension (students fear to ask) and the
Single-Speaker Paradigm (the fact that only one person speaks leads to less activity).
The approach described in Ebner (2009) focuses on the mere group communication by using the general
purpose microblogging platform Twitter. While general purpose microblogging platforms have proven to be
useful in daily life they lack several aspects that we believe to be important in learning settings. Although
they may enhance collaboration by group communication, they usually do not take the outstanding role of the
lecturer into account: They do not provide means of immediate feedback to the lecturer, and neither do they
provide the possibility to socially agree on relevant content all messages are considered equally important.
This makes it often difficult for lecturers to get aware of the relevant thoughts and impressions of the
The concept of the aforementioned quizzes are reminiscent to Audience Response Systems (ARSs) that
have already shown to be effective for learning purposes in various ways (Kay and LeSage, 2009). ARSs
basically allow usually co-located audience members to interactively engage in polls during presentations. In
several studies, lecturers reported that students appreciate ARSs and that participation of students is
increased. Students reported that they are more interested and engaged in the learning discourse when
Audience Response Systems are used. Hence, similar to ARSs, quizzes may help to increase students'
attention, participation and engagement.
This paper presents the modus operandi and functionalities of a digital backchannel that increases
awareness of both lecturer and students in large class lectures. It embraces the empowerment of students
while rewarding the lecturer with instant feedback giving an insight into the dynamics of the audience.
Important aspects of the modus operandi are motivated by findings from learning sciences.
The project reported about in this paper is currently redesigned to account for the integration of
presentation slides which, as an additional operation mode of Backstage, can be published on the dashboards.
Since the integration of slides allows for a better structuring of the backchannel communication and further
collaborative interactions of students, the redesign affects the large part of the user interface and workflows.
The extended version of Backstage is presented in a companion paper in the near future. In Fall 2011 we
intend to thoroughly test Backstage in a field study jointly with educational scientists.
During a first phase from the Fall of 2009 up till the Spring of 2010, the design of Backstage as reported
about in this paper has been worked out in a joint work of computer and learning scientists. The second
phase, from Spring to Fall 2010 was devoted to implementing Backstage relying as much as possible on
standard web technologies.
Aleven, V., Stahl, E., Schworm, S., Fischer, F., and Wallace, R., 2003. Help seeking in interactive learning environments.
Review of Educational Research, Vol. 73, No. 3, pp. 277-320.
Bry, F., Gehlen-Baum, V., and Pohl, A., 2011. Promoting Awareness and Participation in Large Class LecturesThe
Digital Backchannel Backstage. In Proceedings of the IADIS Int. Conf. e-society, Avila, Spain, pp. 27-34.
Cogdill, S., and Kilborn, J., 2001. Backchannel: Whispering in digital conversation. In Proc. of the 34th Hawaii
International Conference on System Sciences (HICSS), IEEE Press.
Costa, C., Beham, G., Reinhardt, W., and Sillaots, M., 2009. Microblogging In Technology Enhanced Learning: A Use-
Case Inspection of PPE Summer School. Workshop at the European Conference on Technology Enhanced Learning
(ECTEL), Maastricht, The Netherlands.
Cuseo, J., 2007. The empirical case against large class size: adverse effects on the teaching, learning, and retention of
first-year students. Journal of Faculty Development, Vol. 21, pp. 5-21.
Ebner, M., 2009. Introducing Live M icroblogging: How single presentations can be enhanced by the mass. Journal of
Research in Innovative Teaching, Vol. 2, No. 1, pp. 108-111.
Er, M. C., and Ng, A. C., 1995. The anonymity and proximity factors in group decision support systems. Decision
Support Systems, Vol. 14, Issue 1, pp. 75-83.
Harry, D., Gutierrez, D., Green, J., and Donath, J., 2008. integrating backchannels with physical space. In
Extended Abstracts on Human Factors in Computing Systems (CHI); ACM, Florence, Italy, pp. 2751-2756.
Hartmann, M., 2007. Eliten und Macht in Europa ein internationaler Vergleich [Elites and Power in Europe an
International Comparison]. Campus Verlag, Frankfurt am Main.
Hattie, J., and Timperley, H., 2007. The Power of Feedback. Review of Educational Research, Vol. 77, No. 1, pp. 81-112.
Kay, R. H., and LeSage, A., 2009. A. Examining the benefits and challenges of using audience response systems: A
review of the literature, Computers & Education, Vol. 53, pp. 819-82.
King, A., 1997. ASK to THINK-TEL WHY: A model of transactive peer tutoring for scaffolding higher level complex
learning. Educational Psychologist, Vol. 32, No. 4, pp. 221-235.
King, A., 2002. Structuring peer interaction to promote high-level cognitive processing. Theory into Practice, Vol. 41,
No. 1, pp. 33-39.
Kleinberg, J., 1998. Authoritative sources in a hyperlinked environment. Journal of the ACM, Vol. 46, No. 5, pp. 604-
Kollar, I., Fischer, F., and Slotta, J. D., 2007. Internal and external scripts in computer-supported collaborative inquiry
learning. Learning and Instruction, Vol. 17, pp. 708-721.
Kuhn, D., and Goh, W. W. L., 2005. Arguing on the computer. In T. Koschmann, D. Suthers & T. W. Chan (Eds.).
Computer Supported Collaborative Learning 2005: The Next 10 Years! In Proceedings of the International
Conference on Computer Supported Collaborative Learning. Lawrence Erlbaum Associates, Mahwah, NJ, pp. 125-
134 .
McCarthy J., and boyd d., 2005. Digital backchannels in shared physical spaces: experiences at an academic conference.
In Extended Abstracts, Human Factors in Computing Systems (CHI); ACM, Portland, USA, pp. 550-553.
McNely, B., 2009, Backchannel Persistence and Collaborative Meaning-Making. In Proceedings of the 27th ACM
International Conference on Design of communication (SIGDOC), New York, USA, pp. 297-304.
Melville, P., M ooney, R., and Nagarajan, R., 2002. Content-Boosted Collaborative Filtering for Improved
Recommendations. In Proceedings of the 18th National Conference on Artificial Intelligence (AAAI), Edmonton,
Canada, pp. 187-192.
O'Donnell, A. M., and Dansereau, D. F., 1992. Scripted cooperation in student dyads: A method for analyzing and
enhancing academic learning and performance. In R. Herz-Lazarowitz, and N. Miller (Eds.). Interaction in
cooperative groups: The theoretical anatomy of group learning. Cambridge University Press, pp. 120-141.
Renkl, A., 1996. Träges Wissen: Wenn Erlerntes nicht genutzt wird [Inert knowledge: when what is learned is not used].
Psychologische Rundschau [Review of Psychology], Vol. 47, pp. 62-78.
Saunders, N., Beitrao, P., Jensen, L., Jurczak, D., Krause, R., Kuhn, M., and Wu, S., 2009. Microblogging the ISMB: A
new approach to conference reporting, PloS Computational Biology, Vol. 5, No. 1.
Schworm, S., and Fischer, F., 2006. Academic help seeking. In H. M andl, and H. F. Friedrich (Eds.). Handbuch
Lernstrategien [Handbook of Learning Strategies]. Hogrefe Verlag, Göttingen.
Slavin, R. E., 1996. Research for the future Research on cooperative learning and achievement: What we know, what
we need to know. Contemporary Educational Psychology, Vol. 21, pp. 43-69.
Stegmann, K., Weinberger, A., and Fischer, F., 2007. Facilitating argumentative knowledge construction with computer-
supported collaboration scripts. International Journal of Computer-Supported Collaborative Learning, Vol. 2, No. 4,
pp. 421-447.
Wecker, C., and Fischer, F., 2007. Fading scripts in computer-supported collaborative learning: The role of distributed
monitoring. In Proceedings of the 8th Conference on Computer-Supported Collaborative Learning (CSCL), New
Jersey, USA.
Weiand, K., Kneißl, F., Furche, T., and Bry, F., 2010. Term-Propagation over Wiki Structures as Eigenvector
Computation. In Proceedings of 5th Semantic Wiki Workshop (SemWiki), Heraklion, Greece.
Weinberger, A., Stegmann, K., Fischer, F., 2010. Learning to argue online: Scripted groups surpass individuals
(unscripted groups do not). Computer in Human Behavior, Vol. 24, No. 4, pp. 506-515.
Yardi, S., 2008. Whispers in the classroom. Digital Youth, Innovation, and the Unexpected. MIT Press, Cambridge, pp.
Zhao, D., and Rosson, M ., 2008. How might microblogs support collaborative work?. Workshop on Social Networking in
Organizations. San Diego, USA.
... smartphones among students made them an obvious tool for communications in an earlier pilot study 1 . They were used to run a backchannel system that provided a supplementary virtual space in which students could interact with the lecturer and other students by sharing their thoughts and engage in collaborative activities without interrupting the current discourse (Pohl et al., 2011). ...
... The demonstrable usefulness of digital backchannel systems to supplement live demonstrations and online learning environments for real-time interaction between students and lecturer has been reported in the literature (Yardi, 2006;Cogdill et al., 2001;Pohl et al., 2011). However, less has been reported about the impact of students' emotions and opinions in their feedback during a lecture in the teaching and learning processes (Russell, 2003). ...
... Mobile backchannel systems such as Hotseat (Aagard et al., 2010), Backstage (Pohl et al., 2011), ActiveClass (Ratto et al., 2003), and ClasCommons (Du et al., 2012) were investigated. They are applications that allow students to use their smartphones to ask questions, vote on questions from their peers, respond to polls set by the lecturer, and give feedback to the lecturer during a lecture session. ...
When presenting to a large group of students, either in an amphitheatre or through an online platform, effectively connecting to the audience – understanding how well the audience is following the presentation and taking appropriate actions promptly if they experience difficulties – is a serious challenge. Backchannel systems are sometimes deployed to address this issue by allowing audience to give feedback to the presenter without interrupting the current discourse. However, these systems are not designed to immediately aggregate and present the audience's feedback to the presenter in a meaningful way that is easy to quickly digest. To fill this gap, we have explored a proof-of-concept method for analysing the emotions and sentiments from the audience's feedback in real time and displaying to the presenter a morale graph showing a trend of the audience's overall reaction over time. This allows a presenter to effectively connected to their audience in real time, knowing whether their presentation is going well and what issues their audience may have in common at any specific moment. We have further implemented this method in an educational context, using a prototype backchannel system, known as ClasSense, for a lecturer to effectively connect to their students. This paper presents the evaluation of this system, which shows that lecturers accept and prefer the morale graph based user interface developed over other backchannel user interfaces that display all posts in chronological order. Students also positively expressed their agreement that the system not only makes their feedback an important part of the class but also increases their interactions with the lecturers. This is further confirmed with a Markov chain predicting the probability that students’ and lecturers’ survey results lead to their overall positive sentiment towards the tool. The flexibility of the ClasSense system suggests it may also be suitable in contexts other than education.
... One such approach to encourage interrogative questioning is the use of "back channels" which allow interactions between staff and students alongside the traditional lecture (Baron, et al. 2016). Facilitating engagement in this fashion is seen to both increase the volume of questions from the class to staff but can also facilitate intra-student communication to build a community (Bry, Gehlen-Baum, & Pohl, 2011;Fortes & Tchantchane, 2010;Yong-Su Pohl & Gehlen-Baum, 2011). ...
Conference Paper
Full-text available
... One such approach to encourage interrogative questioning is the use of "back channels" which allow interactions between staff and students alongside the traditional lecture (Baron, et al. 2016). Facilitating engagement in this fashion is seen to both increase the volume of questions from the class to staff but can also facilitate intra-student communication to build a community (Bry, Gehlen-Baum, & Pohl, 2011;Fortes & Tchantchane, 2010;Yong-Su Pohl & Gehlen-Baum, 2011). ...
... Accordingly, this promotes interaction between learners and lecturer, but importantly between learners who may not normally interact with each other outside the backchannel. This leads to an increased number of questions and comments from the class and promotes the development of a learning community (Bry, Gehlen-Baum, & Pohl, 2011;Fortes & Tchantchane, 2010;Yong-Su Pohl & Gehlen-Baum, 2011). ...
... Efforts to develop a system that can harvest student live feedback on the lecture content and use it for pedagogical refinement is in its infancy. There are some systems such as Google-glass based classroom feedback system [28], Live Interest Meter (LIM) [29], Classroom Feedback System (CFS) [30] and Backstage [31] that have been designed to collect live feedback. Specifically, Rivera-Pelayo et al., [29] designed LIM app which enables the audience to poll on a single variable (namely, confusion) using a mobile device. ...
Conference Paper
We have developed a web-based feedback system that students can use to report difficult, easy, engaging, and boring sections of a lecture in real-time. Such feedback can identify potentially problematic lecture content, track cognitive-affective dynamics in a classroom, and assist instructors in retrospective self-evaluation. We use a mixed-method approach within a design-based research (DBR) framework. In this paper, we discuss initial development and implementation of the feedback system, which constitutes the first cycle of DBR in the project. We discuss potential use-cases of such data and follow the DBR framework to identify strengths and weaknesses in the current prototype and its implementation. Based on such an analysis we come up with a list of modifications for the next cycle of DBR.
... The functionality of digital backchanneling in the secondary English classroom during a performance-based lesson regarding Shakespeare's Macbeth was described as beneficial to both the educator and student. This tool fosters increased student engagement, low-risk student participation, learning community development, practicing digital citizenship skills necessary for respectful and responsible online communication, formative assessment of understanding, and a timely response to students' questions regarding the text (Mueller, 2009;Pohl, Gehlen-Baum, & Bry, 2011;Ugoretz, 2005;Yardi, 2008). Digital backchanneling, through the use of BackChannel Chat, is one strategy for maximizing student engagement during a performance-based lesson in the secondary English classroom. ...
Full-text available
Engaging students during the reading of a complex text is a difficult task for any educator. Digital backchanneling, a Web 2.0 technology, is one tool that can promote: student engagement, low risk participation, learning community development, digital citizenship, an opportunity for formative assessment, and timely response to students' questions. The purpose of observing the effect of digital backchanneling in an English 11 classroom was to measure the level of students' engagement with the reading of Shakespeare's Macbeth. Student engagement increased to 100% when digital backchanneling was introduced during a performance-based lesson.
Full-text available
As with most sectors, the Covid-19 pandemic has been a hugely disruptive force in higher education, with teaching and learning severely impacted. Out of necessity, there is an ongoing need for educators to employ emergency ‘hybrid’ teaching strategies in the hope that geographically dispersed cohorts may continue to be provided with equitable and meaningful learning experiences. A key finding to emerge from our exploratory research is that even when synchronous hybrid teaching is planned out with respect to nurturing an inclusive community of inquiry (CoI), inequities remain when it comes to learning. Whilst instigating informal ‘backchannels’ of communication can help reduce inequities, in-class students remain at a distinct advantage over online students because of their physical and social proximity to teaching staff. Therefore, synchronous hybrid delivery mode, whilst beneficial as an interim/emergency measure, may not be a suitable permanent delivery option for those wishing to maintain a strong and equitable CoI. Where adopting a hybrid approach becomes potentially unavoidable in the future, two recommendations designed to help strengthen a CoI are provided.
It is very important to know users’ behavior to design and build effective interactive systems, tools, or applications. The behavioral study not only helps to assure the success of any design or product but also helps other researchers from various related areas. In this study, we have systematically collected and analyzed the behavioral data for smartphone usage by 1711 students of 188 academic institutions throughout India. We have observed students’ behavior on smartphone usages both inside and outside the classroom. We conducted the study focusing on two aspects: to find the behavioral differences on the smartphone usage based on the gender, and academic level; and to identify the most frequently performed smartphone activities by the students inside and outside the classroom. Although there are few similarities with the existing related studies, we have found many dissimilarities as well. It is expected that the findings of the study will help many researchers from various fields including HCI, Mobile HCI, Behavioral Science, Psychology, and Education.
Full-text available
Microblogs, a relative new phenomenon, are a new communication channel for people to broadcast information that they likely would not do otherwise with existing channels (e.g., email, phone, IM, or weblogs). Microblogging has become quite popular quite quickly, raising the possibility that it may become a new informal communication medium at work. If so, the exchanges it supports may have a variety of impacts on collaborative work, including improved information sharing and expertise seeking, building and maintaining common ground, and enhancing the feeling of connectedness among co-workers. This project is aimed at gaining an in-depth understanding of why people microblog as a social activity, exploring the potential impact of these communication activities on collaborative work, and designing microblog tools that may help to at enhance collaborative work. Author Keywords Microblog, collaborative work, informal communication.
Feedback is one of the most powerful influences on learning and achievement, but this impact can be either positive or negative. Its power is frequently mentioned in articles about learning and teaching, but surprisingly few recent studies have systematically investigated its meaning. This article provides a conceptual analysis of feedback and reviews the evidence related to its impact on learning and achievement. This evidence shows that although feedback is among the major influences, the type of feedback and the way it is given can be differentially effective. A model of feedback is then proposed that identifies the particular properties and circumstances that make it effective, and some typically thorny issues are discussed, including the timing of feedback and the effects of positive and negative feedback. Finally, this analysis is used to suggest ways in which feedback can be used to enhance its effectiveness in classrooms.
Knowledge, although seemingly available, is often not used for solving problems, that means, it remains "inert". Three types of explanations for this phenomenon can be found in the literature. Meta-process explanations assume that the relevant knowledge is available, but it is not used because of inadequate meta-processes (e. g., lacking metacognitive control). Structure deficit explanations suppose that the deficit is rooted in the structure of the knowledge itself, that is, the knowledge is not available in a form that allows for its application. In situatedness explanations, the traditional concepts of knowledge and transfer are questioned. One basic assumption is that knowledge is fundamentally situated, that is, context-bound. In recent years, instructional models have been developed which try to remedy the inert knowledge problem and which take into account many important aspects that have been focussed in the different types of explanations referred to above. However, in order to reach a sufficient understanding of the inert knowledge phenomenon, the cooperation of different psychological subdisciplines is called for.
This article presents an inquiry-based model of mutual peer tutoring called ASK to THINK-TEL WHY®©. The model is described along with its purpose, its theoretical and research bases, and how it is used by tutoring partners to mediate each others' learning. Unlike tutoring systems used to promote learning at the comprehension level, this model is designed to promote higher level complex learning, namely the construction of new knowledge. The model emphasizes reciprocal tutor-tutee roles, supportive communication, and (elaborated explanation and questioning skills. When in the tutoring role, students learn to use different kinds of questions to prompt their partners to make corresponding responses. Tutors also learn to sequence their questioning in a particular way. Thus, during this transactive process, partners scaffold each other's thinking and learning to progressively higher levels. Mechanisms involved in this form of peer-mediated learning are discussed along with results of research on the model's effectiveness.
The network structure of a hyperlinked environment can be a rich source of information about the content of the environment, provided we have effective means for understanding it. We develop a set of algorithmic tools for extracting information from the link structures of such environments, and report on experiments that demonstrate their effectiveness in a variety of contexts on the World Wide Web. The central issue we address within our framework is the distillation of broad search topics, through the discovery of "authoritative" information sources on such topics. We propose and test an algorithmic formulation of the notion of authority, based on the relationship between a set of relevant authoritative pages and the set of "hub pages" that join them together in the link structure. Our formulation has connections to the eigenvectors of certain matrices associated with the link graph; these connections in turn motivate additional heuristics for link-based analysis.