Content uploaded by Paivi Häkkinen
Author content
All content in this area was uploaded by Paivi Häkkinen on Oct 06, 2014
Content may be subject to copyright.
312
K. MÄKITALO, A. WEINBERGER, P. HÄKKINEN,
& F. FISCHER
UNCERTAINTY-REDUCING COOPERATION SCRIPTS
IN ONLINE LEARNING ENVIRONMENTS
Abstract. Online learning courses can create new interaction situations for participants who have not
previously worked with each other. Initially, there is some degree of uncertainty between participants in
these interaction situations. According to the uncertainty reduction theory, low uncertainty increases the
amount of discourse and decreases information seeking. Thus, uncertainty may influence online discourse
and learning. However, the relation of uncertainty reduction to learning outcomes has not yet been
investigated systematically. Cooperation scripts may reduce uncertainty, and therefore enhance learning.
A cooperation script, which aims to reduce uncertainty at a cognitive level, was chosen for this study. The
participants were 48 students in their first semester of Educational Sciences and they were randomly
grouped into triads. The amount of discourse, information seeking and individual learning outcomes in
two conditions (with uncertainty-reducing script and without uncertainty-reducing script) were
investigated. The results indicate that the uncertainty-reducing script in fact increased the amount of
discourse and decreased information seeking activities. The results revealed, however, that the unscripted
and more uncertain condition led to better learning outcomes. These results are discussed against the
background of the uncertainty reduction theory.
1. INTRODUCTION
People participating in online learning courses do not necessarily know each other
and are not sure how to act in these new learning environments. In any initial
interaction situation there is some degree of uncertainty (Berger & Bradac, 1985;
Berger & Calabrese, 1975). A previous study (Mäkitalo, Pöysä, & Häkkinen, 2003)
shows that uncertainty occurs when participants are not sure how others are feeling,
reacting and thinking in online interaction situations. Immediate feedback and non-
verbal cues are missing in online discourse (e.g., Järvelä & Häkkinen, 2002;
Rochelle & Pea, 1999) which can increase uncertainty between participants.
Uncertainty can appear at least at two different levels: at the socio-emotional and at
the cognitive level (Mäkitalo et al., 2003). At the socio-emotional level uncertainty
can occur, for example, when participants do not get immediate feedback on how
others are reacting to their messages, whether they agree or disagree with one’s
suggestions. At the cognitive level, then again, participants might be uncertain about
the content of their contributions: Are their contributions relevant in terms of the
issue and with respect to other messages? Do the learning partners understand the
content of the message?
In this paper, our aim is to investigate the phenomenon of uncertainty in
discourses of online learning environments with the help of the uncertainty
reduction theory (Berger & Bradac, 1985). The important point of this theory is that
a high degree of uncertainty might hinder the participants to communicate
effectively with each other with respect to attaining shared goals, e.g., solving a
problem together. In the field of communication science, Berger and his colleagues
(Berger & Bradac, 1985; Berger & Calabrese, 1975) have developed the uncertainty
K. MÄKITALO, A. WEINBERGER, P. HÄKKINEN, & F. FISCHER
313
reduction theory, seeking to explain how uncertainty affects communication in
relationships. The important components of this theory, which we are applying to
online learning, are (1) uncertainty, (2) the amount of discourse and, (3) information
seeking. Berger and his colleagues indicate that as the amount of verbal
communication in initial interaction situations increases, the level of uncertainty
decreases. Further, as uncertainty is reduced, the amount of verbal communication
increases. They also point out that information-seeking behaviour is increased in
highly uncertain situations. As uncertainty declines, information-seeking behaviour
decreases (Berger & Bradac, 1985; Berger & Calabrese, 1975).
In online learning environments, specific forms of discourse are considered
important for learning. Not just any kind of discourse, but specifically the kind of
discourse where participants ask certain types of questions, evaluate suggestions,
elaborate explanations, hypothesise and summarise the on-going discourse, is seen
to be effective for individual learning (King, 1999). On the one hand, uncertainty
promotes uncertainty reduction strategies, e.g., information seeking, which might
enhance learning. The uncertainty reduction theory states that uncertainty-reducing
interactions enable participants to communicate more effectively with others and
help to achieve shared interaction goals (Berger & Bradac, 1985). On the other hand,
low uncertainty levels increase the amount of discourse, which might also promote
learning (e.g., Cohen, 1994; Jeong & Chi, 1997; Rochelle, 1992). With respect to
learning, we are applying the uncertainty reduction theory to explore interaction
with uncertainty-reducing cooperation scripts.
Cooperation scripts have been considered as facilitators of collaborative learning
activities (O’Donnell, 1999). Scripts can specify, sequence and assign collaborative
learning activities in online learning environments (Kollar, Fischer, & Hesse, 2003;
Weinberger, 2003). A script may, for instance, provide collaborative learners with a
strategy to solve the task and may thus reduce uncertainty. On the basis laid by
Berger and colleagues (Berger & Bradac, 1985; Berger & Calabrese, 1975), we
assume that uncertainty declines at the cognitive level when the degrees of freedom
decrease with respect to the necessary steps or concepts in problem solving
processes. By scripting cooperation, we could reduce the number of alternative sub-
task solutions in online learning environments, which should decrease uncertainty
and therefore increase the amount of discourse and promote learning. In this case,
the focus is on the cognitive level of uncertainty. Reducing the number of
alternatives on the cognitive level by scripting we expect that participants would
focus on the topic and therefore be able to reach a high quantity and quality of
discourse which also should promote individual learning.
2. AIM OF THIS STUDY
In this study we test the following hypotheses:
The amount of discourse hypothesis
H1: The uncertainty-reducing script increases the amount of discourse.
According to previous studies, increased discourse improves learning. Berger and
colleagues point out that when the number of alternatives increases, also uncertainty
UNCERTAINTY-REDUCING COOPERATION SCRIPTS
314
increases. Inversely, reducing the number of alternatives will decrease uncertainty.
Based on these claims we are assuming that it is possible to reduce uncertainty by a
specific task strategy, and therefore we expect higher amounts of discourse.
Information seeking hypothesis
H2: The uncertainty-reducing script decreases information seeking.
According to the uncertainty reduction theory, low uncertainty decreases
information seeking. Berger and Bradac (1985) assume that in highly uncertain
situations participants are seeking more information about the others. We, therefore,
expect the uncertainty-reducing script to lessen information seeking.
The next two hypotheses focus on the effects of uncertainty with respect to the
individual learning outcomes.
Effects of an uncertainty-reducing script on individual learning outcomes
Information seeking and individual learning outcomes hypothesis
H3a: Uncertainty should lead to uncertainty reduction activities, e.g., information
seeking, and this is related to better learning outcomes.
For example, asking specific questions is believed to foster collaborative activities
and therefore enhance learning (King, 1999).
Amount of discourse and individual learning outcomes hypothesis
H3b: The uncertainty-reducing script should lead to a higher amount of discourse
and this is related to better learning outcomes.
This hypothesis is based on the assumption that the amount of discourse correlates
positively with learning (e.g., Cohen, 1994). According to Berger and colleagues,
when uncertainty decreases, interaction increases. They also suggest that when
uncertainty level is low, interaction becomes more effective between participants.
3. METHOD
3.1. Participants
The participants of this study were 48 students in their first semester of Educational
Sciences in an introductory course. The students were randomly grouped into triads
(n=16) and each group was randomly assigned to one of the two experimental
conditions. The first experimental condition was the uncertainty-reducing script
condition. The second experimental condition was the unscripted uncertainty
condition. Time-on-task was held constant in both conditions.
3.2. Learning environments in two experimental conditions
Students worked together by applying theoretical concepts of Weiner’s attribution
theory (1985) to problems, which were presented in the online environment. These
three authentic cases were the central elements in the online learning environment.
At first, students analysed the problems by writing their own initial analysis. Then
they discussed about the analysis via online discussion boards within the group of
three students, and at the end they wrote their own final analysis. In the online
environment there were three online discussion boards, one for each problem. The
K. MÄKITALO, A. WEINBERGER, P. HÄKKINEN, & F. FISCHER
315
collaborative learning session lasted for 80 minutes. In the unscripted uncertainty
condition students were not provided any support to solve the three cases. In the
uncertainty-reducing script condition, participants were guided to apply theoretical
concepts to problems with the help of prompts (see Figure 1). The prompts were
aimed to support the participants’ identification of the relevant problem information
and adoption of the concepts of the attribution theory to problem information. These
prompts included questions and proposals for pedagogical interference regarding the
problem.
Figure 1. Prompts of the uncertainty-reducing script to apply the concepts of Weiner’s (1985)
attribution theory to the problem cases.
3.3. Procedure
Students were placed in three separate rooms and they were communicating via the
online learning environment. First of all, the students’ prior knowledge was tested
individually by means of a problem case. This test was used to control
randomisation. Second, the students were given 20 minutes to read the text about
Weiner’s (1985) attribution theory. Third, the online learning environment was
shortly introduced to the students. Fourth, an 80-minute collaboration phase started.
At the end of the session the students took a post-test based on another problem case
measuring their knowledge.
3.4. Data sources and instruments
The written discussion data consisted of the 16 groups’ discussions (including three
problem solving cases) from eight groups of two different conditions. Two coders
reached an acceptable agreement in identifying the units of analysis (87%). The data
was aggregated and analysed at the group level. We used word counts in order to
measure the amount of discourse. Information seeking was analysed with the help of
the ‘Coding system of a multi-level analysis of knowledge co-construction’
Case information, which can be explained with the
attribution theory
Relevant terms of the attribution theory for this case:
- Does a success or a failure precede this attribution?
- Is the attribution located internally or externally?
- Is the cause for the attribution stable or variable?
- Does the concerned person attribute himself/herself or
does another person attribute him/her?
Prognosis and consequences from the perspective of the
attribution theory:
Case information which cannot be explained with the
attribution theor
y
:
UNCERTAINTY-REDUCING COOPERATION SCRIPTS
316
(Weinberger, Fischer, & Mandl, 2002). The post-test based on a problem case
measured individual learning outcomes. Applicable knowledge was measured with
the amount of adequate relations between theoretical concepts and case information.
The effects were tested with a t-test for unpaired samples for statistical significance.
4. RESULTS
Next, we present the main results based on the proposed hypotheses. Table 1 shows
the absolute numbers with respect to the amount of discourse, information seeking,
and individual learning outcomes.
The amount of discourse hypothesis
H1: The uncertainty-reducing script increases the amount of discourse.
The amount of discourse was higher in the uncertainty-reducing script condition
than in the unscripted condition (see Table 1). A t-test revealed a significant main
effect for the uncertainty-reducing script (t(14)=-2.67; p<.05). The results indicate
that the uncertainty-reducing script increased the amount of discourse, and therefore
the results support this hypothesis.
Table 1. Means and standard deviations for the amount of discourse, information seeking,
and individual learning outcomes for the unscripted and scripted groups
Groups
Variables
Unscripted
M (SD)
Scripted
M (SD)
Amount of discourse 17.44 (4.95) 24.85 (6.10)
Information seeking 6.75 (5.85) 3.00 (3.38)
Individual learning
outcomes
10.75 (6.20)
5.00 (2.00)
Information seeking hypothesis
H2: The uncertainty-reducing script condition decreases information seeking.
The results show that learners sought information in the uncertainty-reducing script
condition less often than in the unscripted uncertainty condition. These results
support the hypothesis; the information seeking scores should be lower in the
uncertainty-reducing script condition (see Table 1). A t-test revealed a marginally
significant main effect of the uncertainty-reducing script (t(14)=1.57; p<.10).
Effects of an uncertain-reducing script on individual learning outcomes
There were significant differences between the learning outcomes in the unscripted
uncertainty condition compared to the uncertainty-reducing script condition
(t(14)=2.50; p<.05). The results show that individual learning outcomes reached
higher scores in the unscripted uncertainty condition than in the uncertainty-
reducing script condition (see Table 1). These results seem to support the
hypothesis, which indicates that uncertainty leads participants to use uncertainty
reduction activities, e.g., information seeking, and therefore lead to better learning
outcomes.
Information seeking and individual learning outcomes hypothesis
K. MÄKITALO, A. WEINBERGER, P. HÄKKINEN, & F. FISCHER
317
H3a: Uncertainty should lead to uncertainty reduction activities, e.g., information
seeking and this is related to better learning outcomes.
The results support this hypothesis. High uncertainty increases information seeking
and the results also reveal that the unscripted uncertainty condition leads to better
learning outcomes than the uncertainty-reducing script condition.
Amount of discourse and individual learning outcomes hypothesis
H3b: The uncertainty-reducing script should lead to higher amount of discourse and
this is related to better learning outcomes.
The results show that the uncertainty-reducing script increases the amount of
discourse, but the results also show that the scores for individual learning outcomes
were lower in the uncertainty-reducing script condition than in the unscripted
uncertainty condition. Therefore, the results do not support the hypothesis that high
amount of discourse would be connected to better learning outcomes.
5. CONCLUSION
The first hypothesis concerning the amount of discourse is supported by the results.
The amount of discourse increased in the uncertainty-reducing script condition as
was suggested by the uncertainty reduction theory. Also the second hypothesis was
supported by the results. Information seeking decreased in the uncertainty-reducing
script condition. With respect to individual learning outcomes, the results support
the hypothesis which claims that uncertainty should lead to uncertainty reduction
activities, e.g., information seeking, and therefore to better learning outcomes.
The results of the first and the second hypotheses are supporting the ideas of the
uncertainty reduction theory. According to the uncertainty reduction theory, low
degrees of freedom with respect to the necessary steps or concepts in problem
solving processes will decrease uncertainty, which fosters high amount of discourse.
The uncertainty reduction theory also states that a high uncertainty level leads the
participants to use uncertainty reduction activities such as information seeking in
order to reduce uncertainty. Based on the results, we can assume that by using the
uncertainty-reducing scripts, designers can influence discussions that will more
easily take place in online learning environments (e.g. Suthers, Hundhausen, &
Girardeau, 2003).
The results concerning individual learning outcomes indicate that the students
who worked in the unscripted uncertainty condition had better learning outcomes.
This does not support the idea that the amount of discourse would be closely
connected to better learning outcomes. Apparently, the quality of discourse is more
important than it’s quantity with respect to learning outcomes. This seems to include
that discourse should provide for some degree of uncertainty, in order to facilitate
specific critical interactions, such as information seeking. In the uncertainty-
reducing script condition participants may have been too certain about the quality of
their contributions. According to Cobb (1995), interaction should be multivocal,
containing more than one perspective in order to reach an effective level of
interaction (e.g., Clark, 2000). It might also be that the prompts used in the scripted
conditions constrain too much the students’ degree of freedom to participate into
UNCERTAINTY-REDUCING COOPERATION SCRIPTS
318
discourse (e.g. Sawyer, 2004). In collaborative learning, information seeking can
enhance discourse and take it to higher levels (King, 1999). These results raise the
question in what ways uncertainty can facilitate and hinder learning. Further studies
are needed to explore uncertainty in different kinds of script conditions, for example
scripts that facilitate information seeking or other uncertainty reduction activities.
A major goal of our study was to shed light on the relationship between
discourse and learning in online learning environments. The uncertainty reduction
theory assumes that there is always some degree of uncertainty and uncertainty is
typically perceived as a barrier to more successful interactions. One of the critical
points of this theory is that it assumes that everyone feels uncertain. Furthermore, it
does not provide any tools to measure the level of uncertainty directly. Further steps
will include the development of a tool for measuring uncertainty. Expanding the
uncertainty reduction theory to online learning, our study shows that some degree of
uncertainty may actually enhance learning. For designers of scripts, it might be
important to know that a certain level of cognitive uncertainty in some conditions is
beneficial to collaborative activities and learning. Thus, further studies are needed to
explore, for example, socio-emotional uncertainty-reducing scripted conditions (e.g.,
Weinberger, 2003; Wosnitza & Volet, 2003). More vital problems concerning
interaction in online learning environments may occur at the social level rather than
at the technical level, for example, or even at the cognitive level as our study
suggests (see also Gunawardena, 1995). In practice, teachers and mentors can help
students recognise the barriers of uncertainty they may face when participating in
online learning environments.
AFFILIATIONS
PhD student Kati Mäkitalo, University of Jyväskylä, Institute for Educational
Research, Jyväskylä, Finland & Knowledge Media Research Center, Tübingen,
Germany (kati.makitalo@ktl.jyu.fi)
Dr. Armin Weinberger, Knowledge Media Research Center, Tübingen, Germany
(a.weinberger@iwm-kmrc.de)
Prof. Päivi Häkkinen, University of Jyväskylä, Institute for Educational Research,
Jyväskylä, Finland (paivi.hakkinen@ktl.jyu.fi)
Prof. Frank Fischer, Knowledge Media Research Center, Tübingen, Germany
(f.fischer@iwm-kmrc.de)
This study is supported by the grants from the Academy of Finland and from the
Finnish Cultural Foundation and Deutsche Forschungsgemeinschaft.
REFERENCES
Berger, C. R., & Bradac, J. J. (1985). Language and social knowledge: Uncertainty in interpersonal
relations. (2nd ed.). London: Edward Arnold.
K. MÄKITALO, A. WEINBERGER, P. HÄKKINEN, & F. FISCHER
319
Berger, C. R., & Calabrese, R. (1975). Some explorations in initial interaction and beyond: Toward a
developmental theory of interpersonal communication. Human Communication Research, 1, 99-112.
Clark, J. (2000). Collaboration tools in online learning environments. ALN Magazine, 4(1),
http://www.aln.org/index.htm.
Cobb, P. (1995). Mathematical learning and small-group interaction: Four case studies. In P. Cobb, & H.
Bauersfeld (Eds.), The emergence of mathematical meaning: Interaction in classroom cultures (pp.
25-129). Hillsdale, NJ: Lawrence Erlbaum.
Cohen, E. G. (1994). Restructuring the classroom: Conditions for productive small groups. Review of
Educational Research, 64(1), 3-35.
Gunawardena, C. N. (1995). Social presence theory and implications for interaction and collaborative
learning in computer conferences. International Journal of Educational Telecommunications, 1(2/3),
147-166.
Jeong, H., & Chi, M. T. H. (1997). Construction of shared knowledge during collaborative learning. In R.
Hall, N. Miyake, & N. Enyede (Eds.), Proceedings of Computer-Supported Collaborative
Learning’97 (pp. 124-128). Toronto, Ontario, Canada.
Järvelä, S., & Häkkinen, P. (2002). Web-bases cases in teaching and learning the quality of discussion
and a stage of perspective taking in asynchronous communication. Interactive Learning
Environments, 10(1), 1-22.
King, A. (1999). Discourse patterns for mediating peer learning. In A. O’Donnell, & A. King (Eds.),
Cognitive perspectives on peer learning (pp. 87-115). Mahwah, NJ: Lawrence Erlbaum.
Kollar, I., Fischer, F., & Hesse, F. W. (2003). Cooperation scripts for computer-supported collaborative
learning. In B. Wasson, R. Baggetun, U. Hoppe, & S. Ludvigsen (Eds.), Proceedings of the
International Conference on Computer Support for Collaborative Learning – CSCL 2003,
COMMUNITY EVENTS – Communication and Interaction (pp. 59-61). Bergen, N: InterMedia.
Mäkitalo, K., Pöysä, J., & Häkkinen, P. (2003, June). Socio-emotional level of grounding in a web-based
conference of small group context. Poster session presented at the International Conference on
Computer Support for Collaborative Learning - CSCL 2003, Bergen, Norway.
O’Donnell, A. M. (1999). Structuring dyadic interaction through scripted cooperation. In A. M.
O’Donnell, & A. King (Eds.), Cognitive perspectives on peer learning (pp. 179-196). Mahwah, NJ:
Lawrence Erlbaum.
Roschelle, J. (1992). Learning by collaborating: Convergent conceptual change. Journal of the Learning
Science, 2(3), 235-276.
Roschelle, J., & Pea, R. (1999). Trajectories from today’s WWW to a powerful educational infrastructure.
Educational Researcher, 28(5), 22-25.
Sawyer, R. K. (2004). Creative teaching: Collaborative discussion as disciplined improvisation.
Educational Researcher, 33(2), 12-20.
Suthers, D., Hundhausen, C. D., & Girardeau, L. E. (2003). Comparing the roles of representations in
face-to-face and online computer supported collaborative learning. Computer & Education, 41, 335-
351.
Weinberger, A. (2003). Scripts for computer-supported collaborative learning. Effects of social and
epistemic cooperation scripts on collaborative knowledge construction. [Dissertation, Ludwig-
Maximilians-Universität]. URL: http://edoc.ub.uni-muenchen.de/archive/00001120/01/Weinberger_
Armin.pdf.
Weinberger, A., Fischer, F., & Mandl, H. (2002). Kodierungssystem für eine Multi-Ebenen-Analyse der
gemeinsamen Wissenskonstruktion [Coding system for a multi-level-analysis of collaborative
knowledge construction]. Unpublished manuscript, Ludwig Maximilian University of Munich,
Institute of Educational Psychology.
Weiner, B. (1985). An attributional theory of achievement motivation and emotion. Psychological
Review, 92, 548-573.
Wosnitza, M., & Volet, S. (2003, August). Significance of social and emotional dimensions in online
learning. Paper presented at the European Association of Research on Learning and Instruction
(EARLI) conference, Padova, Italy.