Content uploaded by Dimitra Tsovaltzi
Author content
All content in this area was uploaded by Dimitra Tsovaltzi on Jun 17, 2015
Content may be subject to copyright.
A Long-Term View on Learning to Argue in Facebook: The Effects
of Group Awareness Tools and Argumentation Scripts
Thomas Puhl, Dimitra Tsovaltzi and Armin Weinberger
t.puhl@edutech.uni-saarland.de, d.tsovltzi@edutech.uni-saarland.de, a.weinberger@edutech.uni-saarland.de
Saarland University
Abstract: Social networking sites pose a new arena for argumentative practices and may
provide an opportunity to facilitate, and understand argumentative processes in large scale
over longer periods of time. Through apps, group awareness tools and argumentation scripts
can be implemented in social networking sites, to provide additional, graphically visualized
information, and prompt learners to formulate or review sound arguments. This 2×2 field
study (N=105) examines how Facebook can be harnessed for argumentative learning through
group awareness tools and argumentation scripts. We measure domain-specific knowledge
outcomes, the process of argumentative knowledge construction and attitude change. The
results show a main effect for both conditions on domain-specific knowledge and
argumentative knowledge construction, but no difference to the control group on attitude
change. The epistemic quality and formal quality increased for the groups with argumentation
script. All observed changes become substantial after the fourth week of interactions.
Keywords: argumentation, group awareness, scripts, social networking sites, long-term study
Argumentative Knowledge Construction in Social Networking Sites
Social networking sites (SNS) afford collaborative processes that may be harnessed for learning (Greenhow,
2008; Greenhow & Robelia, 2009). SNS may offer a rich argumentative context that can pronounce processes of
argumentative knowledge construction (AKC). AKC is the deliberate practice of elaborating learning material
and discussion topics by constructing formally and semantically sound arguments with the goal of gaining
argumentative and domain-specific knowledge (Andriessen, 2006). With SNS being prone to trivial talk, AKC
in SNS may greatly benefit from instructional approaches of CSCL (Tsovaltzi, Puhl, Judele & Weinberger,
2014; Tsovaltzi, Judele, Puhl & Weinberger, in review). Moreover, recent research has revealed that while
argumentation skills can be fostered by controlled short-term technology-enhanced instructional approaches, no
substantial effects on domain-specific knowledge can be attained with these approaches (Wecker & Fischer,
2014). One reason could be that complex and meaningful learning such as argumentative learning takes time
(Reimann, Utz, Unterberger & Halb, 2014). Long-term interventions have been suggested before to successfully
foster argumentation skills with young adolescents (Kuhn & Crowell, 2011). However, existing approaches
have often been too coercive for learners to actively regulate their individual or group AKC processes
(Dillenbourg, 2002), and others might be too subtle to have an effect on learners' self-regulation or the co-
regulation of the group (Phielix, Prins & Kirschner, 2010). Can SNS, like Facebook, productively host
argumentative knowledge construction? How can the combination of rather coercive scripts, and less coercive
group awareness tools (GATs), foster the development of learners’ AKC?
Argumentation Scripts
Argumentation scripts are a prominent approach to foster AKC in CSCL environments (Noroozi, Weinberger,
Biemans, Mulder, & Chizari, 2012). Scripts are socio-cognitive structures that specify learners’ interactions in
collaborative learning scenarios, aiming to either activate existing internal scripts, organize, or re-organize
internally represented pieces of information, mostly declarative, that may constitute new elements of scripts
(Fischer, Kollar, Stegmann & Wecker, 2013). Internal scripts are already represented in the learners’ cognitive
system that are prone to change and adaptation. External scripts suggest collaborative processes and can change
the learners’ behavior over time through internalization (Kollar, Fischer & Slotta, 2007). In terms of
collaborative argumentation processes, the internalization of an external script can guide learners to structure,
discuss about, and reflect on their arguments, which can foster knowledge co-construction, argumentative and
domain-specific knowledge (Weinberger, Stegmann & Fischer, 2010), and structure students’ discussion of
conflicting opinions (Jermann & Dillenbourg, 1999; Felton & Kuhn, 2001). In the ArgueGraph script (Jermann
& Dillenbourg, 1999 & 2002; Kobbe, Weinberger, Dillenbourg, Harrer, Hämäläinen, Häkkinen & Fischer,
2007), differences in opinions on seminar topics where represented to the students in order to increase socio-
cognitive conflict. Students were then asked to resolve these differences in dyads and reach an agreement by
writing arguments derived for their learning materials. Discourse analysis showed that students often provided
conditional responses, in accordance with the high-order learning goal of the course.
Group Awareness Tools
GATs are used in CSCL scenarios to enhance collaboration processes and also to foster domain-specific
knowledge by giving students information about other learners, such as individual knowledge (Janssen &
Bodemer, 2013). GATs can reveal inconsistent knowledge or differences of views and thus enhance
controversial CSCL discussions, and impact group processes, like socio-emotional and motivational processes
that are meant to support AKC (Buder & Bodemer, 2008). GATs also make group processes or attitudes
explicit, for instance by visualizing how learners evaluate their own and the group’s cognitive and social
interactions (Phielix, Prins, Kirschner, Erkens, & Jaspers, 2011). Group awareness of knowledge distribution in
the group has been shown to counteract negative effects of overly high levels of both mutual trust and mutual
skepticism (Engelmann, Kolodziej, & Hesse, 2014). Moreover, GATs have been shown to enable group
processes that may support knowledge co-construction, like mutual modelling, (Sangin, Molinari, Nüssli &
Dillenbourg, 2011), and group conflict (Phielix, Prins & Kirschner, 2010). There are first indications of positive
effects of GATs on learning gains (Sangin et al., 2011). In particular, GATs that visualized cognitive conflict in
the form of discrepancies in the group solutions (Bodemer, 2011) or of summary of anonymous opinions (Buder
& Bodemer, 2008) have been found to enable collaborative discussion of conflicting perspectives, foster
individual learning gains, and increase critical argument elaboration during collaboration, as well as individual
and group attitude change. GATs are positively related to performance and process satisfaction (Phielix et al,
2010) and can prove particularly useful for learning in SNS such as Facebook to help group members consider
the distribution of knowledge and attitudes, spot their own positions in relation to other members. Such group
awareness may foster socio-cognitive conflict, in the sense of social tensions (e. g., why is my position different
to others?) and personal dissonance (e. g., how can I bridge two contradicting views?).
Attitude Change
Attitude change has been investigated in the context of argumentative knowledge construction in young students
(Kuhn, Shaw,&Felton, 1997), showing a tendency to more stable positions with increasing age, and in
computer-mediated communication, revealing lack of readiness to consider alternative perspectives when
attitude markers are salient (Sassenberg & Boos, 2003). Argumentative processes, like taking into account
multiple perspectives and resolving conflicting opinions appear to facilitate attitude change (Eagly & Chaiken,
1995; Erber, Hodges, & Wilson, 1993) and may ultimately lead learners to re-consider initial standpoints
(Tsovaltzi et al., 2013). Moreover, vocational training includes practicing attitude and communication at the
work place (Nasir, 2002), e.g. for medical doctors (Caralis & Hammond, 1992) and teachers (Ihmeideh & Al-
omari, 2010). Theory on attitude change supports that change is the result of dissonance that can be caused by
cognitive conflict between different attitudes of the same person (Eagly & Chaiken, 1995; Erber, Hodges, &
Wilson, 1993). Such conflicts could potentially be made salient with GATs and productively resolved with
scripts. GATs may contribute to making learners’ ideas, misconceptions, and contradictions accessible to
scrutiny (Nussbaum, 2008). Scripts have been found to effectively specify and support how learners jointly
argue to address conflicts (Weinberger et al., 2010).
We investigate the effects of a GAT and argumentation script on domain-specific learning outcomes,
learning processes, and attitude change. We conjecture that the GAT will induce socio-cognitive conflict and
will translate into argumentative processes. We also conjecture that the argumentation script will foster the
quality of arguments in the discussion. Therefore, they will both increase the domain-specific learning gains.
The combination of both factors will interact in a positive way. We also hypothesize a positive effect of the
GAT on attitude change.
Methods
We conducted a 2×2 quasi-experimental field-study with German teacher trainees (N=105) and factors GAT and
argumentation script. Participants were taking part in a seminar on “Communication and Interaction”. In the
intervention, all participants attended weekly seminars where a new theory on communication was presented
and discussed. After the seminar, the students had to fill out a questionnaire about their communication attitude.
Seminar groups were accompanied by Facebook groups, where students weekly received a theory-related task
that required them to interact and discuss online within five days. The study was part of the seminar and lasted
the whole semester. Students had to fill out the communication attitude questionnaire on nine sessions and were
instructed to reflect on the seminar theories while filling in the questionnaire. On eight sessions, an
argumentation script was given as feedback to the quality of the written arguments.
Before the intervention all participants took a knowledge pre-test, and filled in a questionnaire on
socio-demographic data, subjective knowledge in the domain “Communication and Interaction”, computer and
Facebook use behavior. In addition they filled in a communication attitude questionnaire and the experimental
conditions were familiarized with the GAT. Similarly, after the intervention at the end of the semester, all
participants took a knowledge post-test and filled in another questionnaire on their subjective learning gains and
their acceptance of the learning environment.
Group Awareness Tool
Figure 1. Group Awareness Tool in Facebook.
Students receiving GAT-based support reflected on their hypothetical personal communication attitude as a
teacher by answering a case-based communication questionnaire with cases from every-day social interactions
in the school. Every scenario was followed by four Likert-scaled answers, focusing on how a teacher may reflect
on and behave in the situation. Moreover, the questions differed with respect to whether the emphasis was on
multi-perspective / flexible attitudes vs. goal-oriented / structured attitudes, following Buder & Bodemer (2008)
and Jermann & Dillenbourg (1999, 2002), both rated on a scale from 0 to 6. The result of the questionnaire was
presented to the experimental groups with a GAT implemented as a Facebook app. The app graphically
visualized the position of a student with respect to communication attitude in relation to other students. The
position in the graphic is represented as relative to the position of the whole seminar group (I vs. the others).
The GAT also gave them the opportunity to identify the person behind a position by moving the mouse over the
point on the graphic to see the name and a photo of the person (Figure 1), so the discussions were not
anonymous. The design of the GAT is oriented on the ArgueGraph script (Jermann & Dillenbourg, 1999) and
aims at increasing socio-cognitive conflict. ArgueGraph was adjusted to the Facebook learning environment.
That is, considering the Facebook dynamic, which means frequent interactions in the large group and less
discursive discussions compared to a small group CSCL scenario, we used a less coercive instruction and did
not force the students to reach an agreement.
Argumentation script
Students in the condition with script-based support received a weekly argumentation script in the form of
feedback to arguments posted in the Facebook group discussion. They had to read all arguments of the
discussion and “like” the best argument in their own opinion. Feedback was given to every group for the most
liked argument and also for the best argument in the opinion of the teacher. The feedback to these two
arguments evaluated the epistemic (theoretical concepts and relations) and the formal (reasoning and evidence)
quality of arguments for the two selected arguments. It also indicated whether an important part of the argument
was missing or illustrated how a sound argument should look like. While the structure and the ontology of the
script were standardized, it was also important to adapt the script to the different weekly tasks and the different
arguments from each group.
Instruments
Domain-specific knowledge was assessed by a course exam at the end of the semester containing definitions,
facts, and higher order discursive processes like theory-based interpretations and argumentations. The
knowledge test included 10 multiple choice items and 13 open questions with a high inter-rater reliability k=.88,
p=000. The internal consistency of the knowledge test was good (α = .69).
Process analysis was based on Weinberger & Fischer (2006) on multi-level analysis of AKC processes
and on an adapted version for argumentation in Facebook (Tsovaltzi, Judele, Puhl & Weinberger, 2014). We
measured epistemic and formal quality of arguments. The formal quality of arguments (k=.80, p=000) was
measured by the correct formal structure of the argument (e.g. if justifications were used for every argument),
the quality of the justification and the quality of the reference. We evaluated the quality of justifications by
giving a score of three points for references to scientific results, two points for examples that may be used to
explain the content of an argument, and one point for every-day knowledge that provides some potential insight
but without scientific evidence. The actual citation of references was evaluated in addition to the justification.
The quality of references was rated with two for a scientific reference (e.g. journal paper) and one for a link to a
relevant internet site (e.g. wikipedia). The epistemic quality (k=.76, p=000) was measured by the quality of the
used concepts or theories and the relations between them. We evaluated the quality of theories advocated in the
arguments by giving a score of three points for a related theory to the discussed topic, which is the
theory/theories of the week, two for using the theory of the current week, and one for subjective theories that
have no scientific background. A relation drawn between current and other theories was also rated with three
and added to the overall score for epistemic quality. The posts from the second seminar sessions in the Facebook
group were not analyzed because of a different weekly task. The students were given a task to reflect on non-
verbal communication and post a personal experience they had in this week. Therefore there was no discussion
in this week and no arguments to analyze.
Communication attitude change was measured using the results of the communication questionnaire.
The questionnaire was based on the situation assessment test for consulting (Keller, Bruder & Schmitz, 2010)
and used scenarios to assess the communication attitudes of the participants. Contrary to the elaborated
scenarios in the consulting questionnaire, the scenarios in the communication attitude questionnaire are concise
for students to fill in on a weekly basis. Students were asked to reflect on the communication situations in the
questionnaire and indicate how they would behave as teachers in the sketched situations. For example, a
scenario presented the situation in which most of the students failed in a test. The teacher could either handle
this situation by simply ascertaining that they did not study enough, or by discussing with them what they
thought was the problem. Furthermore the teacher could offer to repeat the test, or discuss different possibilities
with the students such as repeating the test or talking about the situation with the school principle. There are no
correct answers to the questionnaire, but different theories from the ones taught in the seminar could support the
different options (Puhl, Tsovaltzi & Weinberger, in press). Students expressed their attitude by indicating their
agreement on the Likert scale described in the Section Group Awareness Tool.
Findings
There were no significant differences in the socio-demographic data, the Facebook use and the knowledge test
prior to the intervention. Log data confirmed that all participants spent time looking at the graph as an indication
of reflecting upon it.
Process Analysis
Using an ANOVA with repeated measures (sessions 1 and 3-8), the process analysis showed a significant effect
for the interaction between time and epistemic quality, F(6;606)=3.81; p<.001; ηp2=.10, and between time and
formal quality, F(6;606)=1.88; p=.015; ηp2=.053. Descriptive statistics also show that the epistemic quality did
not change in the control group throughout the semester but increased for both groups with argumentation script.
The formal quality also only increased for argumentation script and decreased for the control group (Figure 2).
Figure 2. Epistemic and formal quality of arguments.
To identify when the argumentative processes took shifted, we did simple contrasts comparing each
session from the 3rd to the 8th time with the first session. On the epistemic dimension we found a decrease for
three groups from the first time of measurement to the third. Only the experimental group with argumentation
script increased the epistemic quality of their arguments. Until the fifth time of measurement all groups are close
together and simple contrasts were not significant. Substantial and significant changes occurred from the first to
the sixth time of measurement, F(3;101)=5.70; p=.001; ηp2=.15, and all contrasts are significant until the end of
the interventions. A similar process was observed on the formal quality of arguments. Descriptive statistics
show that all groups are close until the fourth time of measurement, after that they diverged and simple contrasts
are only significant between the first and the last time of measurement, F(3;101)=5.98; p=.001; ηp2=.15.
Figure 3. Multi-perspective and goal-oriented communication attitudes.
Using an ANOVA with repeated measures, we found significant main effects on communication
attitude change over the seminar period for both multi-perspective, F(8;800)=2.98; p=.003; ηp2=.03 and goal-
oriented dimensions, F(8;800)=14.2; p<.001; ηp2=.12, but no significant effect for the interaction between time
and group. Descriptive statistics also showed that attitudes increase on the multi-perspective dimension for the
experimental groups and independent from the groups. Simple contrasts showed a significant main effect for the
factor time from the first date of measurement to the second, F(3;100)=4.33; p=.04; ηp2=.04, and also the
changes from date one to eight and one to nine are significant, F(3;100)=5.38; p=.004; ηp2=.08 (Figure 3). The
attitudes decrease on the goal-oriented dimension for all groups and simple contrasts are significant for the
factor time as of the fifth time of measurement until the last date, F(3;100)=16.58; p=.000; ηp2=.25.
Knowledge outcomes
We found a significant effect for the experimental groups in domain-specific learning gains, F(3;102)=11.11;
p<.001; ηp2=.25. Three participants did not take part in the final exam. The descriptive statistics show an
increase in learning gains from the control group to the experimental groups (Table 1).
Table 1: Domain-specific knowledge, mean and standard deviation
Group
Control
(n=19)
GAT
(n=29)
ARG-Script
(n=29)
GAT+ARG-Script
(n=25)
M (SD)
34.08 (6.53)
39.81 (5.42)
41.34 (4.38)
42.52 (4.33)
Besides significant main effects for the two factors (GAT: F(1;98)=11.24, p=.001, ηp2=.10;
Argumentation script: F(1;98)=23.44, p=.000, ηp2=.19), we also found a significant effect for the interaction
between GAT and argumentation script, F(3;102)=4.89; p=.029; ηp2=.05. The factor argumentation script shows
higher learning gains (Figure 4).
Figure 4. Domain-specific knowledge.
The epistemic quality of arguments from the last time interval correlates with domain-specific
knowledge r(102)=.24, p<.05, and formal quality of arguments also correlates with domain-specific knowledge
r(102)=.29, p<.01. Correlations between communication attitudes and learning outcomes are not significant.
Conclusions and implications
As hypothesized, the GAT and the script guided learners to acquire more domain-specific knowledge and
improved learners' quality of arguments. The argumentation script seems to outperform the GAT with respect to
both of these variables. The combination of both instructional approaches shows the best results on all variables.
Notably, the effect of GAT and script on the quality of arguments seems to set in only after 4-5 weeks of the
seminar, indicating that it takes time to develop strategies for building high quality arguments and respective
knowledge. By the same token, it may be very difficult in short-term interventions to raise quality of arguments
and even more difficult to increase domain-specific knowledge (Wecker & Fischer, 2014). It has been argued
that learners are overwhelmed in short-term learning environments to consider both, the quality of arguments
and acquire domain-specific knowledge. Long-term interventions, in contrast, can realize AKC that facilitates
both, argumentative and domain knowledge. Consequently, the relation between AKC processes and outcomes
appears to develop over time.
Learners' attitudes shifted on the multi-perspective dimension after the first seminar week from the
baseline to the second measurement time. Since the first intervention took place after the second measurement
time, the change from first to second time cannot depend on the two factors, but rather on the content of the
seminar. The communication theories from the first session specifically address the multi-perceptivity of
understanding messages and the students were always instructed to reflect on the seminar theories while filling
out the communication attitude questionnaire. Over the entire seminar period, all experimental groups incl.
30
34
38
42
46
no GAT GAT
no Argumentation
Script
Argumentation
Script
control group developed on the multi-perspective dimension. All groups declined on the goal-oriented
dimension over time. While these attitude changes may derive from the seminar theories independent of GAT or
script, it goes to show that learning and attitude change are not linear processes. Over the seminar period there
were no significant differences between the groups, which is contrary to prior findings showing that groups with
higher coercion awareness tools surpassed groups with lower coercion awareness tools (Puhl et al., in review).
This potentially means that interventions of group awareness and higher coercion (e.g. scripts that prompt
learner’s to reflect on the visualized group awareness revealing conflictual ideas and attitudes) can rather
influence attitudes.
This study extended CSCL research on scripts and group awareness tools to SNS. It thus provided
insights into long-term effects of these instructional approaches of CSCL. While these approaches do not
produce the expected short-term effects on domain-specific knowledge (Wecker & Fischer, 2014), this study
shows that such effects can still emerge in the long run.
In contrast to popular views, SNS can serve as a platform for argumentation and learning. The results
in this study reveal that learning to argue may take a social environment, in which groups develop shared
argumentative practices over time. SNS may offer themselves as such an environment. Two main directions
may be taken from here. First, SNS can host long-term studies that can contribute to understanding and
facilitating argumentative learning in social communities. Second, massively used online fora like SNS offer the
possibility for technological interventions such as the ones investigated here. SNS may be leveraged to develop
practices of argumentation on a large scale.
References
Andriessen, J. E. B. (2006). Arguing to learn. In K. Sawyer (Ed.), (Tran.), Handbook of the Learning Sciences
(443-459). Cambridge: Cambridge University Press.
Bodemer, D. (2011). Tacit guidance for collaborative multimedia learning. Computers in Human Behavior,
27(3), 1079–1086. doi:10.1016/j.chb.2010.05.016
Buder, J., & Bodemer, D. (2008). Supporting controversial CSCL discussions with augmented group awareness
tools. International Journal of Computer-Supported Collaborative Learning, 3(2), 123–139.
doi:10.1007/s11412-008-9037-5
Caralis, P.V., Hammond, J.S. (1992). Attitudes of medical students, housestaff, and faculty physicians toward
euthanasia and termination of life-sustaining treatment. Critical Medical Care, 20(5):683-90.
Dillenbourg, P. (2002). Over-scripting CSCL. In P. A. Kirschner (Ed.), Three worlds of CSCL: Can we support
CSCL (pp. 61-91). Heerlen: Open University of the Netherlands.
Eagly, A., & Chaiken, S. (1995). Attitude strength, attitude structure and resistance to change. In R. Petty and J.
Kosnik (Eds.), Attitude Strength. (pp. 413–432). Mahwah, NJ: Erlbaum.
Engelmann, T., Kolodziej, R., & Hesse, F. W. (2014). Preventing undesirable effects of mutual trust and the
development of skepticism in virtual groups by applying the knowledge and information awareness
approach. International Journal of Computer-Supported Collaborative Learning, 9(2), 211–235.
doi:10.1007/s11412-013-9187-y
Erber, M. W., Hodges, S. D., & Wilson, T. D. (1995). Attitude strength, attitude stability, and the effects of
analyzing reasons. In R. Petty and J. Kosnik (Eds.), Attitude Strength. (pp. 433–454). Mahwah, NJ:
Erlbaum.
Felton, M., & Kuhn, D. (2001). The development of argumentive discourse skill. Discourse Processes, 32(2),
135–153. doi:10.1207/S15326950DP3202&3_03
Fischer, F., Kollar, I., Stegmann, K., & Wecker, C. (2013). Toward a script theory of guidance in computer-
supported collaborative learning. Educational psychologist, 48(1), 56-66.
Greenhow, C. (2008). Connecting informal and formal learning experiences in the age of participatory media:
Commentary on Bull et al. Contemporary Issues in Technology and Teacher Education, 8, 187–194.
Greenhow, C., & Robelia, B. (2009). Informal learning and identity formation in online social networks.
Learning, Media and Technology, 34(2), 119–140. doi:10.1080/17439880902923580
Ihmeideh, F. M., & Al-omari, A. A. (2010). Attitudes toward Communication Skills among Students’ -
Teachers’ in Jordanian Public Universities, 35(4).
Janssen, J., & Bodemer, D. (2013). Coordinated computer-supported collaborative learning: Awareness and
awareness tools. Educational Psychologist, 48:1, 40-55
Jermann, P. & Dillenbourg, P. (2002). Elaborating new arguments through a CSCL script. In Andriessen, J.,
Baker, M., Suthers, D. (eds.), Arguing to Learn: Confronting Cognitions in Computer-Supported
Collaborative Learning environments, 1—6. Kluwer Academic Publishers. Printed in the Netherlands.
Jermann, P., & Dillenbourg, P. (1999). An analysis of learner arguments in a collective learning environment.
Proceedings of the 1999 conference on Computer support for collaborative learning - CSCL ’99, 265-
273.
Karpinski, A. C., Kirschner, P. a., Ozer, I., Mellott, J. a., & Ochwo, P. (2013). An exploration of social
networking site use, multitasking, and academic performance among United States and European
university students. Computers in Human Behavior, 29(3), 1182–1192. doi:10.1016/j.chb.2012.10.011
Keller, S., Bruder, S., & Schmitz, B. (2010). Lernberatungskompetenz situativ messen – Validierung eines
Situationsbeurteilungstests. Vortrag auf der 74. Tagung der AEPF (Arbeitsgruppe für Empirische
Pädagogische Forschung), Friedrich-Schiller-Universität Jena, Jena.
Kirschner, P. a., & Karpinski, A. C. (2010). Facebook® and academic performance. Computers in Human
Behavior, 26(6), 1237–1245. doi:10.1016/j.chb.2010.03.024
Kobbe, L., Weinberger, A., Dillenbourg, P., Harrer, A., Hämäläinen, R., Häkkinen, P., & Fischer, F. (2007).
Specifying computer-supported collaboration scripts. International Journal of Computer-Supported
Collaborative Learning, 2(2-3), 211–224. doi:10.1007/s11412-007-9014-4
Kollar, I., Fischer, F., & Slotta, J. D. (2007). Internal and external scripts in computer-supported collaborative
inquiry learning. Learning & Instruction, 17(6), 708-721.
Kuhn, D., & Crowell, A. (2011). Dialogic argumentation as a vehicle for developing young adolescents’
thinking. Psychological Science, 22, 545-552.
Nasir, N.S. (2002). Identity, goals, and learning: Mathematics in cultural practice. Mathematical Thinking and
Learning, 4, 213–248.
Noroozi, O., Weinberger, A., Biemans, H. J. A., Mulder, M., & Chizari, M. (2012). Argumentation-Based
Computer Supported Collaborative Learning (ABCSCL): A synthesis of 15 years of research.
Educational Research Review, 7(2), 79-106. doi: 10.1016/j.edurev.2011.11.006
Nussbaum, M. E. (2008). Collaborative discourse, argumentation, and learning: Preface and literature review.
Contemporary Educational Psychology, 33(3), 345–359. doi:10.1016/j.cedpsych.2008.06.001
Phielix, C., Prins, F. J., & Kirschner, P. A. (2010). Awareness of group performance in a CSCL-environment:
Effects of peer feedback and reflection. Computers in Human Behavior, 26, 151–161.
Phielix, C., Prins, F. J., Kirschner, P. a., Erkens, G., & Jaspers, J. (2011). Group awareness of social and
cognitive performance in a CSCL environment: Effects of a peer feedback and reflection tool.
Computers in Human Behavior, 27(3), 1087–1102. doi:10.1016/j.chb.2010.06.024
Puhl, T., Tsovaltzi, D. & Weinberger, A. (in review). Blending Facebook into seminars for practicing
argumentation: a long-term perspective on the effects of group awareness tools and scripts. Computers
in Human Behavior.
Reimann, P., Utz, W., Unterberger, R., Halb, W. (2014). Capturing Learning over Time for Supporting
Pedagogical Decision Making: A Process Modeling Approach. In Elena Barbera, Peter Reimann (Eds.),
Assessment and Evaluation of Time Factors in Online Teaching and Learning, (pp. 264-293). Hershey,
PA: Information Science Reference.
Sangin, M., Molinari, G., Nüssli, M.-A., & Dillenbourg, P. (2011). Facilitating peer knowledge modeling:
Effects of a knowledge awareness tool on collaborative learning outcomes and processes. Computers in
Human Behavior, 27(3), 1059–1067. doi:10.1016/j.chb.2010.05.032
Sassenberg, K., & Boos, M. (2003). Attitude change in computer-mediated communication: Effects of
anonymity and category norms. Group Processes & Intergroup Relations, 6(4), 405–422.
doi:10.1177/13684302030064006
Tsovaltzi, D., Judele, R., Puhl, T., & Weinberger, A. (in review). Scripts and Group Awareness Support in the
Service of Learning in Facebook: How does CSCL compare to Social Networking Sites? Computers in
Human Behavior.
Tsovaltzi, D., Puhl, T., Judele, R. & Weinberger, A. (2014). Group Awareness Support and Argumentation
Scripts for Individual Preparation of Arguments in Facebook. Computers & Education, 76, 108-
118,doi.org/10.1016/j.compedu.2014.03.012.
Wecker, C. & Fischer , F. (2014). Where is the evidence? A meta-analysis on the role of argumentation for the
acquisition of domain-specific knowledge in computer-supported collaborative learning. Computers &
Education 75, 218-228.
Weinberger, A. & Fischer, F. (2006). A framework to analyse argumentative knowledge construction in
computer-supported collaborative learning. Computers & Education, 46, 71-95.
Weinberger, A., Stegmann, K., Fischer, F. (2010). Learning to argue online: Scripted groups surpass individuals
(unscripted groups do not). Computers in Human Behavior, 26(4), 506–515.
