ArticlePDF Available

Abstract and Figures

Welchen Einfluss haben dialogische Lernsituationen auf den Gebrauch von Evidenz bei adoleszenten Schülerinnen und Schülern? Zusammenfassung. Diese Studie untersucht wie sich argumentativer Diskurs und individuelles Argumentieren in Bezug auf den Gebrauch von Evidenz unterscheiden. In einem 1 × 2 Cross-over Design diskutierten 37 Mittelstufenschülerinnen und –Schüler ein gesellschaftliches Thema mit ihrem Partner, entweder bevor oder nachdem sie einen kurzen Aufsatz zu ihrer eigenen Meinung verfassten. Als Hintergrundinformationen erhielten sie eine Sammlung qualitativ unterschiedlicher Evidenzen zu dem Themenbereich. Die Dialoge und Aufsätze wurden untersucht in Hinblick auf a) die Art der Evidenz und b) auf welche Weise diese genutzt wurde. Die Ergebnisse zeigen, dass sich die Schülerinnen und Schüler in den Aufsätzen häufiger auf die ihnen gemeinsam vorliegenden Evidenzen beziehen (geteilte Evidenz). In den Dialogen nutzen sie Evidenz hingegen häufiger, um den gegenteiligen Standpunkt zu adressieren und zeigen dabei eine klarere Argumentationslinie. Die Ergebnisse weisen auf eine höhere Effizienz der Dialoge im Vergleich zum individuellen Schreiben hin. Gleichzeitig gibt die Studie erste Hinweise für die Gestaltung von Curricula, die Schülerinnen und Schüler dazu anregen Evidenz in ihrer Argumentation einzusetzen.
Content may be subject to copyright.
Jahrgang 30 / Heft 2–3 / 2016
Zeitschrift für
Pädagogische
Psychologie
Themenheft/Special Issue
Learning through Communication
German Journal of Educational Psychology
Herausgeber
Andreas Knapp
Detlef H. Rost
Assoziierte Herausgeber
Tina Hascher
Jörn Sparfeldt
Beirat
Roland Brünken, Martin Brunner,
Joachim C. Brunstein,
Susanne R. Buch, Oliver Dickhäuser,
Roland Grabner, Samuel Greiff,
Michael Grosche, Hans Gruber,
Regina Jucks, Detlev Leutner,
Jens Möller, Jan Retelsdorf,
Tina Seufert, Birgit Spinath,
Nadine Spörer, Robin Stark,
Ricarda Steinmayr, Elsbeth Stern,
Ulrich Trautwein
Gastherausgeber
Regina Jucks
Zeitschrift für Pädagogische Psychologie Jahrgang 30 / Heft 2–3 / 2016
Die Balance zwischen
online und o ine
Isabel Willemse
Onlnesucht
Ein Ratgeber für Eltern, Betroffene und ihr Umfeld
2016. 160 S., 11 Abb., 7 Tab., Kt
€ 19.95 / CHF 26.90
ISBN 978-3-456-85542-4
AUCH ALS E-BOOK
Smartphones, Tablets und Laptops sind
zu unseren ständigen Begleitern ge-
worden, wir verbringen unsere Freizeit
in Sozialen Netzwerken wie Instagram
oder Facebook, mit dem Verschicken von
Bildern oder Textnachrichten und mit
Videogames. Ein Großteil der Jugendli-
chen und Erwachsenen beweist einen
kompetenten und vernünftigen Umgang
mit diesen Gadgets und kann sich prob-
lemlos zwischen digitaler und analoger
Welt hin und her bewegen. Aber es gibt
auch einen kleinen Teil, dem das nicht
gelingt.
Wenn die exzessive Mediennutzung ne-
gative Auswirkungen hat auf das Sozial-
leben und Hobbys, den Beruf oder die
Ausbildung und allenfalls auch die Ge-
sundheit, dann könnte es sich um eine
Onlinesucht handeln. Hierbei handelt
es sich um eine sehr neue Diagnose, die
noch nicht in den offi ziellen Diagnose-
instrumenten vorhanden ist. Nichts-
destotrotz wird sie von Eltern, Betrof-
fenen und ihrem Umfeld erkannt und in
der Beratungspraxis regelmäßig ange-
troffen.
Der Ratgeber wird in einem theoreti-
schen Teil eine allgemeine Einführung
in die Mediennutzung geben, aber vor
allem das Störungsbild genau beschrei-
ben. Hierzu gehören die Diagnosekrite-
rien, Verbreitung, Ursachen und auch
Begleiterkrankungen. Der praktische
Teil enthält viele konkrete Vorschläge
für Bezugspersonen und Betroffene im
Umgang mit Onlinesucht.
www.hogrefe.com
Willemse_210x277.indd 1 14.01.16 16:27
Zeitschrift für Pädagogische Psychologie (2016), 30(2–3), 73–74 © 2016 Hogrefe
Inhalt
Editorial Learning through Communication: How Arguing about Scientifi c
Information Contributes toLearning
Lernen im Diskurs: Wie das Argumentieren über wissenschaftsbezogene
Informationen zum Lernerfolg beiträgt
Regina Jucks, Elisabeth Mayweg-Paus
75
Originalartikel /
Original Articles
Identifying General Cognitive Abilities Involved in Argument
Comprehension and Evaluation
Das Verständnis und die Evaluation von Argumenten:
Zur Rolle allgemeiner kognitiver Fähigkeiten
M. Anne Britt, Kristopher J. Kopp, Amanda M. Durik, Dylan Blaum,
Peter Hastings
79
Who knows? Explaining Impacts on theAssessment of our own Knowledge
and of the Knowledge of Experts
Wer könnte es wissen? Erklären beeinfl usst die Einschätzung unseres
eigenen Wissens sowie unsere Einschätzung des Wissens von Experten
Rainer Bromme, Eva Thomm, Katharina Ratermann
97
Developing Epistemological Understanding in Scientifi c and Social
Domains through Argumentation
Wie das Argumentieren die Entwicklung eines epistemologischen
Verständnisses in wissenschaftlichen und sozialen Bereichen fördern kann
Kalypso Iordanou
109
How Dialogic Settings Infl uence Evidence Use in Adolescent Students
Welchen Einfl uss haben dialogische Lernsituationen auf den Gebrauch
vonEvidenz bei adoleszenten Schülerinnen und Schülern?
Elisabeth Mayweg-Paus, Fabrizio Macagno
121
Better to Agree or Disagree? TheRoleof Critical Questioning
and Elaboration in Argumentative Discourse
Ist Zustimmung oder Dissens besser? Die Rolle des Kritischen
Hinterfragens und Elaborierens im Argumentativen Diskurs
Monja Thiebach, Elisabeth Mayweg-Paus, Regina Jucks
133
Diskussionen / Discussions Making Science Education More Natural – Some Ideas from
the Argumentative Theory of Reasoning
Science education praxisnah – Ideen aus der Argumentative Theory
ofReasoning
Hugo Mercier
151
Inhalt
© 2016 Hogrefe Zeitschrift für Pädagogische Psychologie (2016), 30(2–3), 73–74
New Directions for Research on Argumentation: Insights from theAIR
Framework for Epistemic Cognition
Neue Wege für Forschung über das Argumentieren: Einblicke aus
dem AIR-Framework for Epistemic Cognition
Ravit Golan Duncan, Clark A. Chinn
155
Buchbesprechungen /
Book Reviews
Hasselhorn, M., Ehm, J.-H., Schneider, W. & Schöler, H. (2015). Das Projekt
«Schulreifes Kind». Abschlussbericht der wissenschaftlichen Begleitung
Marlit Annalena Lindner
163
Steins, G. (2014). Sozialpsychologie des Schulalltags. Band I: Grundlagen
und Anwendungen (2. Aufl .)
Steins, G., Bitan, K. & Haep, A. (2014). Sozialpsychologie des Schulalltags.
Band II: Im Klassenzimmer
Jost Stellmacher
165
Neuerscheinungen /
New Books
169
Pädagogische
Psychologie
Zeitschrift für
Ihr Artikel wurde in einer Zeitschrift der Hogrefe AG veröffentlicht.
Dieser e-Sonderdruck wird ausschließlich für den persönlichen Gebrauch
der Autoren zur Verfügung gestellt. Eine Hinterlegung auf einer persönlichen
oder institutionellen Webseite oder einem sog. «Dokumentenserver»
bzw. institutionellen oder disziplinären Repositorium ist nicht gestattet.
Falls Sie den Artikel auf einer persönlichen oder institutionellen Webseite
oder einem sog. Dokumentenserver bzw. institutionellen oder disziplinären
Repositorium hinterlegen wollen, verwenden Sie bitte dazu ein «pre-print»
oder ein «post-print» der Manuskriptfassung nach den Richtlinien
der Publikationsfreigabe für Ihren Artikel bzw. den «Online-Rechte
fürZeitschriftenbeiträge» (http://www.hogrefe.ch/service).
© 2016 Hogrefe Zeitschrift für Pädagogische Psychologie (2016), 30(2–3), 121–132
DOI 10.1024/1010-0652/a000171
Original Article
How Dalogc Settngs In uence
Evdence Use n Adolescent Students
Elisabeth Mayweg-Paus1 & Fabrizio Macagno2
1 Department of Psychology and Sport Studies, University of Münster, Germany
2 Faculdade de Ciências Sociais e Humanas, Universidade Nova de Lisboa, Portugal
Introduction
In recent decades, argumentation has come to be viewed
as a central component of education (Driver, Newton &
Osborne, 2000; Duschl & Osborne, 2002; Kuhn, 1993;
Rapanta, Garcia-Mila & Gilabert, 2013). According to this
view, one of education's most important goals is to provide
learners with the capabilities to assess the available infor-
mation, to select the most relevant and adequate evidence
and to apply this in their judgments and decisions (Os-
borne, Eduran, Simon & Monk, 2001). The use of evidence
for argumentative purposes (namely, to support, assess,
question or refute a claim) is particularly informative, be-
cause it reveals a student's epistemological background
(Sandoval & Millwood, 2005). The interpretation, assess-
ment and justi cation of a piece of evidence, or rather the
critical stance in relation to it, is mirrored by how a student
coordinates and connects it with given claims and other
concurring or con icting evidence (Kuhn 1993; Kuhn, Am-
sel & O'Loughlin, 1988; Kuhn et al., 1995). In contrast, a
student's failure to use evidence in a way that distinguish-
es it from a speci c claim (Jiménez-Aleixandre, Rodríguez
& Duschl, 2000; McNeill, 2011) and is functional in rela-
tion to that claim (McNeill, Lizotte, Krajcik & Marx, 2006)
reveals problems in interpreting such evidence, in distin-
guishing it from an advocated viewpoint and in addressing
it critically. Similarly, a student's epistemological commit-
ment is mirrored by the way she uses such information to
build an argument and defend or attack a viewpoint (Ra-
panta et al., 2013; Sandoval & Millwood, 2005). The qual-
ity of understanding is thus related to the way a student
interprets evidence in order to use it argumentatively
(Sandoval & Millwood, 2005). The dialectical assessment
of a piece of information through arguments, questioning
Abstract: This study examines how evidence is used differently in argumentative discourse compared to individual arguments. Applying a 1 × 2
crossover study design, 37 secondary school students were asked either to discuss a social issue with their partner before individually writing
an essay outlining their opinion or, vice versa, fi rst to discuss and then to write. As background information, they were provided with pieces of
evidence with different levels of quality. Dialogs and essays were analyzed regarding (a) the type of evidence and (b) the way evidence was used.
Results showed that in their essays students referred more often to the pieces of evidence provided to them (shared evidence). In contrast, they
used evidence more often to address the opposing viewpoint in dialogs by incorporating it in a more elaborated (clearer) line of reasoning. Find-
ings suggest that dialogues are a more effective tool than individual writing production, and the study provides fi rst hints regarding how to de-
sign curricula that will encourage students to use evidence in a more sophisticated way in their argumentation.
Keywords: argumentative dialogue, individual argument, argumentative function, argumentative structure, evidence use
Welchen Einfl uss haben dialogische Lernsituationen auf den Gebrauch von Evidenz bei adoleszenten Schülerinnen und Schülern?
Zusammenfassung: Diese Studie untersucht wie sich argumentativer Diskurs und individuelles Argumentieren in Bezug auf den Gebrauch von
Evidenz unterscheiden. In einem 1 × 2 Cross-over Design diskutierten 37 Mittelstufenschülerinnen und –Schüler ein gesellschaftliches Thema
mit ihrem Partner, entweder bevor oder nachdem sie einen kurzen Aufsatz zu ihrer eigenen Meinung verfassten. Als Hintergrundinformationen
erhielten sie eine Sammlung qualitativ unterschiedlicher Evidenzen zu dem Themenbereich. Die Dialoge und Aufsätze wurden untersucht in
Hinblick auf a) die Art der Evidenz und b) auf welche Weise diese genutzt wurde. Die Ergebnisse zeigen, dass sich die Schülerinnen und Schüler
in den Aufsätzen häufi ger auf die ihnen gemeinsam vorliegenden Evidenzen beziehen (geteilte Evidenz). In den Dialogen nutzen sie Evidenz
hingegen häufi ger, um den gegenteiligen Standpunkt zu adressieren und zeigen dabei eine klarere Argumentationslinie. Die Ergebnisse weisen
auf eine höhere Effi zienz der Dialoge im Vergleich zum individuellen Schreiben hin. Gleichzeitig gibt die Studie erste Hinweise für die Gestaltung
von Curricula, die Schülerinnen und Schüler dazu anregen Evidenz in ihrer Argumentation einzusetzen.
Schlüsselwörter: argumentativer Diskurs, Evidenz, argumentative Funktion
Author's personal copy (e-offprint)
122 E. Mayweg-Paus & F. Macagno, Infl uence of a Dialogic Setting
Zeitschrift für Pädagogische Psychologie (2016), 30(2–3), 121–132 © 2016 Hogrefe
and attacks is related crucially to its understanding (Nuss-
baum & Edwards, 2011). The crucial issue is understand-
ing how to improve students' use of evidence in their ex-
plicit reasoning in order to design strategies that will help
them to adequately integrate evidence in their line of rea-
soning. Our study addresses this question by testing
whether and how the argumentative mode (dialogic vs.
monologic) a ects the use of evidence from a functional
and structural perspective.
Dialogic argumentation
Dialogic argumentation is viewed as a process in which
two or more people engage in a debate characterized by
opposing claims (Kuhn & Udell, 2003) that they need to
support with arguments and question critically (Walton,
1989). In an argumentative dialogue, a participant is sub-
ject to the interlocutor's scrutiny of her own position. This
scrutiny and the e orts to challenge the opposing view can
be assumed to impact directly on the quality of argumen-
tation. In a dialogic argumentative setting, a student needs
to be more critical regarding not only her own position but
also the opposing one. She should achieve this by drawing
on more evidence and elaborating it further in order to ad-
dress the challenges to her own position in a relevant way
(Walton & Macagno, 2007). In dialogues, students need to
analyze the reasons for preferring one point of view or one
piece of evidence over another. Therefore, they are en-
couraged to take a critical stance toward the presented evi-
dence (Osborne et al., 2001). They should engage directly
in understanding and assessing their interlocutor's rea-
sons, thereby elaborating more and in greater depth on
their own and the other's point of view.
Argumentative dialogues are considered to be an e ec-
tive educational strategy for developing reasoning skills
(Koballa, 1992; Kuhn, 1992; Kuhn & Crowell, 2011). In-
deed, argumentative dialogues have been shown to have a
positive e ect on the ability to generate counterarguments
and rebuttals (Kuhn, Goh, Iordanou & Shaen eld, 2008).
Kuhn and Moore (2015) have shown that students who
were taught within a 2-year dialogue-focused curriculum
(see also Kuhn, 2015) tended to back up their claims with
more evidence than students belonging to a nondialogic
group. Moreover, the same intervention revealed that stu-
dents engaging in a dialogic setting were more likely to
draw on evidence from their own personal knowledge (in-
stead of from a list provided to them) and to address and
weaken an opposing claim instead of using evidence only
to support their own view.
This study raises two crucial questions: (a) How can the
use of evidence serve as an indicator of quality of argu-
mentation? (b) How does the dialogic setting impact on
the quality of evidence use? Although, as indicated above,
many studies have investigated the use of evidence in edu-
cational contexts, it is not entirely clear what type of evi-
dence they used and how they used it. To address these is-
sues, we designed a study to cast light on the relationship
between the various types of uses of evidence and their
argumentative quality and to test the di erences between
a dialogic and a monological setting. To do this, we needed
to draw some basic distinctions on both a content level
(namely between various types of evidence) and a func-
tional level (namely between various types of argumenta-
tive uses of evidence).
The  rst distinctions concern what type of evidence is
taken into account. Some studies refer solely to scienti c
evidence (such as empirical data from experiments and
surveys, see Iordanou & Constantinou, 2014, 2015; Sand-
oval, 2003; Sandoval & Millwood, 2005); others also in-
clude evidence drawn from single cases (nonscienti c evi-
dence, see Kuhn & Moore, 2015). Therefore, we decided to
provide students with a broader range of  ve subcatego-
ries of evidence. To determine which evidence is used, we
not only considered these distinct types but also analyzed
where the evidence came from. In particular, we distin-
guished between evidence that we provided as input on
cards and evidence stemming from students' personal ex-
periences or knowledge. This enabled us to test not only
the e ect of the argumentative mode on the quality of evi-
dence per se but also how it might trigger di erent sources
of evidence.
The second distinction concerns how evidence is used
for argumentative purposes. Our basic assumption is that a
critical understanding of evidence can be revealed through
its argumentative uses, or more speci cally through a dis-
tinction between more and less elaborated (or sophisticat-
ed) uses to support a viewpoint or rebut an incompatible or
di erent one. Hence, we distinguished the types of evi-
dence use on the basis of the structure of an argument and
the possible functions of a piece of evidence within that
argument (Pollock, 1974; Toulmin, 1958; Walton, 2006).
Such functions were then compared and assessed in terms
of which uses were more or less sophisticated. This served
as the basis for developing a speci c coding scheme.
Function and structure of evidence use
From an argumentative point of view, evidence serves two
fundamental functions (or purposes): It can be used to
support one's own viewpoint or to weaken a position that is
incompatible or di erent from the one defended (the in-
terlocutor's perspective). From a structural point of view,
evidence can support or weaken a position in di erent
ways. These can be distinguished by drawing on Toulmin's
Author's personal copy (e-offprint)
E. Mayweg-Paus & F. Macagno, Infl uence of a Dialogic Setting 123
© 2016 Hogrefe Zeitschrift für Pädagogische Psychologie (2016), 30(2–3), 121–132
(1958) argument structure. According to Toulmin, an argu-
ment can be represented as an interconnected set of (a) a
claim (C), (b) data (D), (c) a warrant connecting claim and
data (“since W”) and (d) backings (B) substantiating the
warrants. The structure of Toulmin's model as applied in
our research is depicted in Figure1.
This reading of Toulmin's scheme aims to distinguish
between two distinct argumentative uses of evidence that
relate to the structure of an argument. The data are con-
sidered as the premise (or the premises) that are linked di-
rectly to the claim, providing the factual and relevant rea-
son to draw the conclusion (the claim). The backing
instead, is conceived as the support that can be given to an
element grounding the claim. This is usually the warrant,
but can also be the data. In this sense, the backing is not
essential for the logical and material support of the claim.
It reinforces the elements supporting it.
In this interpretation of Toulmin's model, the conclu-
sion is grounded on premises (data and warrant) that are
an interpretation of the available information. This casts
light on which elements thereof are relevant to the conclu-
sion. For example, in order to support the claim that
“smoking should be banned,” an arguer may use the avail-
able evidence that “major studies indicate a relation be-
tween smoking cigarettes and lung disease.” However, this
piece of information needs to be interpreted in order to be
used to support the intended conclusion, thereby casting
light on which aspect thereof is relevant in this case. There-
fore, the arguer will point out that because smoking a ects
health (or more speci cally lung health), it should be for-
bidden. This data, namely the interpretation used as a
premise, allows the triggering of a commonly shared war-
rant from negative consequences (whatever a ects health
should be forbidden). The piece of evidence is then used to
back up the data. In this view, evidence is used in a rele-
vant argumentative way to support the desired conclusion
after being interpreted accordingly. Otherwise, it can be
used simply as a direct support of a claim, but no relation
of relevance to the warrant is brought to light. For exam-
ple, the arguer may simply say that “smoking should be
banned, because major studies indicate a relation between
smoking cigarettes and lung health.” This uninterpreted
piece of evidence leaves the interlocutor with the burden
of retrieving the intended relation to the conclusion, but
also the possibility of rejecting the argument directly (a
“relation” means that no evidence of an actual cause has
been found). Whereas a piece of data can be supported by
several backings (thus saving the argument from one of
the possible criticisms of the evidence cited), when evi-
dence is simply cited as a data, the whole argument is ex-
posed to default if such evidence is rebutted by a con ict-
ing interpretation (which applies even more particularly
for neutral pieces of evidence). Uninterpreted evidence
thus makes the process of argument reconstruction much
more complex and less clear, and grants the interlocutor
the possibility of interpreting it strategically (Macagno,
2008, 2012).
For these reasons, we distinguished between two levels
of evidence use di erentiated according to their argumen-
tative relevance:
1. First-order evidence: Evidence used to support or chal-
lenge a claim directly in an argument (functioning as
Toulmin's data).
2. Second-order evidence: Evidence used to support or
challenge the validity of how data ( rst-order evidence)
is used to support a claim in an argument (functioning
as Toulmin's backing).
Apart from the dimensions of validity and acceptability, an
argument can be evaluated by considering the two dialec-
tical criteria of clarity (see Aristotle, trans. 1995) and e ec-
tiveness (considered as diminishing the possibilities of
possible attacks) (Macagno & Walton, 2014). Hence, the
arguments leaving less room (and less of a burden) of in-
terpretation to the interlocutor should be considered as
both clearer and more e ective. Based on this, we can
claim that the most sophisticated use of evidence corre-
sponds to its more elaborated use as a backing, which pre-
supposes its interpretation and its consequent reasoned
and critical use in a way that is signi cant for the conclu-
sion (Sandoval & Millwood, 2005).
Evidence use can be divided into four categories that
combine the argumentative purpose of the use of a piece
of evidence (functional level) with its structural role (struc-
tural level). A claim can be supported with  rst- or second-
order evidence (evidence used as data or as a backing).
Similarly, a position can be attacked with  rst- or second-
order evidence, namely, by using evidence as a backing to
support a contrary conclusion (counter-claim), or as an un-
derminer, attacking the foundations of the argument in
support of the incompatible view (Macagno, Mayweg-Paus
& Kuhn, 2014; Mayweg-Paus, Macagno & Kuhn, 2015). An
underminer can weaken either the interpretation of the
Figure 1. Toulmin's structure of argument analysis.
Author's personal copy (e-offprint)
124 E. Mayweg-Paus & F. Macagno, Infl uence of a Dialogic Setting
Zeitschrift für Pädagogische Psychologie (2016), 30(2–3), 121–132 © 2016 Hogrefe
evidence used by the interlocutor (namely the data) or the
evidence itself. Whereas the  rst type of refutation can be
considered as corresponding to the direct support of a
claim, underminers mirror the interpretation and the as-
sessment of the pieces of evidence. For this reason, where-
as  rst-order evidence employed both to support a claim
and attack an argument reveals a lower quality of evidence
use,  rst-order evidence used as a backing or an under-
miner can be considered as a sign of higher quality use.
Rationale of this study
To identify the speci c e ect of dialogic interaction on the
use of evidence in argumentation, we compared adoles-
cents’ arguments on a social issue in a peer-peer argumen-
tative dialogue and in an individual essay. In both cases,
students engaged in an argumentative activity (because
they needed to provide reasons supporting a claim) de-
rived from an explicit alternative or doubt (in dialogues) or
an implicit one (in essays). All participants received a set of
cards with short pieces of evidence of di erent types, and
they were told that they could use these if they wanted to.
The number of pieces of evidence supporting a position,
the opposing position, and a neutral position on the topic
were equal.
The aim of this study was twofold: From a content point
of view, our goal was to assess the di erences between the
two argumentative modes relative to the following dimen-
sions: (a) the use of the di erent types of evidence provid-
ed in the cards, and (b) the source of the evidence that the
students used in general (whether it was drawn from the
cards or from their personal experience or knowledge).
From a functional and structural point of view, we intend-
ed to test the impact of the dialogic setting on the critical
evaluation of evidence and points of view. Therefore, we
developed a coding system designed to capture how evi-
dence was used from a functional-structural point of view
and what counted as a more elaborated or sophisticated
use (Clark, Sampson, Weinberger & Erkens, 2007).
Method
Participants and design
Thirty-seven students (54.1 % female) from two biology
classes (10th graders) at a public high school in New Jersey
participated in the study. Their mean age was 15.62 years
(SD= .92). A total of 48.6 % of the participants were His-
panic, 27 % were Black, 18.9 % were White, 2.7 % were
Asian, and 2.7 % were American Indian. The two classes
were comparable regarding their grade point average
(GPA) on a 4-point scale (Class A: M= 2.27, SD= .61, Class
B; M= 2. 1, SD= .82), F(1, 35) = .522, p= .4 0, ns, as well as on
their biology performance (in percentage) in particular
(Class A: M= 74.15 %, SD= 13.88 %, Class B: M= 77.52 %
SD= 8.62 %), F(1, 35) =.757, p= .30, ns. Note that students
needed to achieve at least 50 % to pass the biology class.
The experimental manipulation was performed in a 1 × 2
crossover design with students writing an individual essay
either before or after engaging in mutual discussion with a
peer partner.
Materials and procedure
The topic chosen was “banning cigarette sales in the US.”
Students are likely to have di erent views on this topic,
and, at the same time, they can be expected to be open to
considering other possible perspectives on it and to engag-
ing in deeper discussions. Students were given a set of
cards containing 15 distinct pieces of evidence selected ac-
cording to  ve categories depending on the type of source
on which they were based: anecdotal-single case, popular-
ity, authority/expert source, descriptive statistics and labo -
ratory evidence. These  ve categories were ranked along a
dimension ranging from low to high quality. The de nition
of quality was based on two criteria. The  rst was purely
theoretical, grounded on relevant works in argumentation
(Walton, 2006). The second was statistical: We asked a
group of 10 experts (academic sta at a German and a Por-
tuguese University (6 male, M= 33.80 years old, M= 7.50
years working in academic context) to rank the  ve types
according to how helpful they would be in an academic
discussion. The experts' ranking (from the strongest to the
weakest) can be summarized as follows:
1. Laboratory evidence (80 % of the sample. 10 %: statis-
tics; 10 %: authority);
2. Descriptive statistics (80 % of the sample; 10 %: author-
ity; 10 %: anecdotal);
3. Authority (80 % of the sample; 20 %: laboratory evi-
dence);
4. Popularity (70 % of the sample; 30 % anecdotal);
5. Anecdotal (60 % of the sample; 30 %: popularity; 10 %:
statistics).
For each category, students received one piece of evidence
in favor of banning cigarette sales, one against and one
that was neutral. Six pieces of evidence were taken from a
list already used in the work of Kuhn and Moore (2015).
Four other pieces were slightly modi ed versions of evi-
dence used by a random sample of 20 laypeople inter-
viewed in New York City's Grand Central Station. The
Author's personal copy (e-offprint)
E. Mayweg-Paus & F. Macagno, Infl uence of a Dialogic Setting 125
© 2016 Hogrefe Zeitschrift für Pädagogische Psychologie (2016), 30(2–3), 121–132
sample covered a wide range of professions (such as dance
teachers, physicians, police o cers, etc.) and age groups
(mean age was 38.32 years, SD= 14.90; max: 65, min: 18
years). The remaining  ve pieces were developed on the
basis of an Internet search. Table 1 gives an overview of the
15 pieces of evidence.
The intervention took place within a 90-min school
class. First, students were provided with two questions
asking for their view on the topic (“Should cigarette sales
be banned in the U.S.?” and “How sure are you of your
opinion?”). This information enabled us to match students
to opposing-site pairs. Then, the cards containing the 15
di erent pieces of evidence were distributed to them. Stu-
dents were asked to read the information on the cards
carefully and rank the cards according to their strength. In
the next step, students in Class A were instructed to write
an essay outlining their view on the topic. Students in Class
B were randomly assigned to opposing-site pairs, and were
asked to discuss the topic verbally. In the last phase, stu-
dents in Class B were instructed to write an essay and stu-
dents in Class A were randomly assigned to opposing-site
pairs and asked to discuss the topic verbally. All face-to-
face dialogs were audiotaped.
Dependent measures
Coding of dialog quality
The  rst step in the coding procedure was to identify the
relevant pieces of evidence. Therefore, the dialogs and the
essays were  rst divided into idea units (Asterhan &
Schwarz 2009; Jucks & Paus 2013). Idea units were further
divided into on-task (addressing the argumentation task)
and o -task units. Then, we classi ed the on-task units
into evidence units (namely units that reported or referred
to evidence) and nonevidence units. Interrater agreement
(calculated on roughly 50 % of both dialogues and essays)
was good (in dialogs: Krippendor 's α = .94; in essays:
Table 1. Evidence list – Classifi cation system.
Evidence type Purpose Example
Anecdotal (single case)
Reference to a specifi c case
Con Georg Harrison, a musician for the Beatles, was a smoker and died of lung cancer in the age of 58.
Neutral A man called John Stasser used to smoke, but he was not addicted to cigarettes.
He simply enjoyed smoking.
Pro A woman named Helen Reichert lives in NYC; she is 108 years old and has been smoking half
apack of cigarettes every day for over 80 years.
Authority
Reference to a source such
as scientists, government,
institutes
Con The government emphasizes that smoking is one of the factors enhancing the probability of
getting heart attacks.
Neutral Dr. F. R. Moore, president of a major cancer center, claims that smoking, combined with other
causes, may lead to lung cancer, but there are many other factors.
Pro Scientists have shown that a strong genetic infl uence of people living long and healthy lives
regardless whether or not they smoke.
Popularity (popular opinion)
Reference to the majority/
common sense
Con A survey showed a large majority of people support banning cigarettes sales.
Neutral A survey showed a large majority of people think smoking is an important issue.
Pro A survey showed a large majority of people think everybody has the right to decide whether they
want to smoke.
Laboratory evidence
(authority)
Reference to a study by
describing the design/fi ndings
Con Several major studies indicate a relation between smoking cigarettes and lung disease.
Neutral Laboratory research shows that people use smoking to help them cope when facing highly
stressful situations.
Pro Studies show the nicotine in cigarettes causes fast-acting chemical reactions in your brain that
relieve anxiety and nervousness.
Statistics: descriptive
(induction)
Reference to numbers
and data
Con Each year, an estimated 443,000 people die from smoking or exposure to secondhand smoke,
andanother 8.6 million have to live with a serious illness caused by smoking.
Neutral An estimated 17 million Americans try to quit smoking each year, and about 8 % of them succeed.
Pro Thousands of farmers in the U.S. make their living from farming tobacco leaves, and the tobacco
industry contributes an average of $ 16.5 billion to the economy each year
Author's personal copy (e-offprint)
126 E. Mayweg-Paus & F. Macagno, Infl uence of a Dialogic Setting
Zeitschrift für Pädagogische Psychologie (2016), 30(2–3), 121–132 © 2016 Hogrefe
Krippendor 's α= .94). In a next step, we coded the on-
task units of both essays and dialogs by taking into account
two distinct aspects of evidence use: the kind of evidence
and the structure and function of evidence use.
Type of evidence used
We took into account the  ve di erent evidence categories
from the cards plus the category of personal evidence con-
taining evidence not drawn from the cards but from per-
sonal experience or knowledge (personal experience =
95 % and personal knowledge from hearsay= 5 %). Inter-
rater agreement on roughly 50 % of the dialogs and the
essays was good for both the classi cation of the students'
evidence in the  ve categories (Krippendor 's α= .85–94),
and the detection of personal evidence in the dialogs
(Krippendor 's α= .80) and in the essays (Krippendor 's
α= .79). Table 2 presents an overview of these results.
Structure and function of evidence use
We grounded our analysis on two interrelated dimensions
of the use of a piece of evidence (Clark et al., 2007): its
function and its structure. For the functional coding, we
referred to a coding scheme developed by Kuhn and Moore
(2015) that di erentiates between evidence advanced to
support a claim and evidence used to weaken a claim.
Within these two generic goals of evidence use, a further
distinction was the way a piece of evidence can be used to
support or weaken a claim ( rst- and second-order evi-
dence). We referred to this distinction as concerning the
“structure” of evidence use, and we based our di erentia-
tions on Toulmin's (1958) argument pattern. The  nal cod-
ing scheme consisted of four distinct categories (see also
Table 3 for examples and interrater reliabilities): (a) sup-
port premises, (b) support claim, (c) weaken claim and (d)
weaken evidence. Support premises is the code for an evi-
dence unit used as a backing in Toulmin's pattern, namely
as second-order evidence. Support claim is the use of evi-
dence to support a viewpoint directly, namely as  rst-or-
der evidence. Weaken claim represents the code for the
second-order evidence units used to counter an argument,
namely to attack the conclusion directly or the data on
which it is based. Weaken evidence is the code for  rst-order
Table 2. What kind of evidence was used?
Kind of evidence Example from use in dialog Example from use in essay
From cards
(α= .85 – .94)
“I am against banning because there are people who are
like 100 years old, and sometimes even more, and they
have smoked for 80 years and they are still alive.
“I think cigarette sales shouldn't be banned. There is
awoman who was 108 years old and who has been
smoking half a pack of cigarettes for over 80 years.
Personal
(α= .80 (dialog),
α= .79 (essay))
“(…) my grandfather has lung cancer, so that'sbecause
Iam like against, just because ofsmoking (…)
“I believe, cigarettes sales should be banned. I person-
ally know a young boy who has a hole in his throat since
birth. That hole was due to his mother's addiction before
and while she was pregnant with him (…)
Table 3. Our coding scheme.
Category Description Example
Support claim
(α= .82 (essay),
α= .79 (dialog))
Evidence (italics) is used to support directly the generic
viewpoint, but it is not related to an argument.
“My fi rst statement is a survey that showed a large ma-
jority of people support banning cigarettes sales. And a
Beatles musician George Harrison was a smoker and died
of lung cancer at 58. I strongly believe cigarettes sales
should be banned.
Support premises
(α= .75 (essay),
α= .73 (dialog))
Evidence is used to back up the premises of an
argument(bold), giving strength to it. It is related
indirectly to the student's claim, because it supports
aline of reasoning.
“There are many reasons to why it should be banned, not
only because its bad for you but it can be addicting to
people, like an estimated 17 million Americans try to quit
smoking each year and about 8 % of the succeed.
Weaken claim
(α= .81 (essay),
α= .82 (dialog))
Evidence is used to weaken the opposing claim
by providing a counter-reason not to accept it.
“A lot of people believe that smoking can lead to a short
life and causes various sicknesses, but not always.
Scientist have shown a strong genetic infl uence of
people living a long healthy life.”
Weaken evidence
(α= .85 (essay),
α= .79 (dialog))
Evidence is used to directly weaken the evidence that
supports the opposing view either directly or indirectly
(underlined).
Although Helen Reichert is 108 and has been smoking
80years that doesn't mean nothing. Not every one
person is alike. Just like she lived to be 108,
George Harrison only lived to be 58.
Author's personal copy (e-offprint)
E. Mayweg-Paus & F. Macagno, Infl uence of a Dialogic Setting 127
© 2016 Hogrefe Zeitschrift für Pädagogische Psychologie (2016), 30(2–3), 121–132
evidence used for attacking directly the interlocutor's evi-
dence used as a backing. In this latter case, the interlocu-
tor's argument is undermined by countering the evidential
grounds on which it stands.
Results
To take random e ects into account and increase the gen-
eralizability of results, we employed linear mixed-e ect
models. These provide several advantages (see Barr, Levy,
Scheepers & Tily 2013; or Baayen, Davidson & Bates 2008,
for details). To determine the potential e ects of the argu-
mentative mode (dialog and essay) on the use of evidence
and on the function and structure of evidence use in the
following analyses, we classi ed the argumentative mode,
the class and the intercept as the  xed e ects and the sub-
jects and the dyads as random e ects. In particular, we
used the factor mode to compare aspects of evidence use
of students in both classes in either dialog or essay. We
used the factor class to compare aspects of evidence use in
both modes between the two classes, whereas we viewed
the interaction between the two factors as illustrating an
e ect of the order of intervention.
The analysis showed that dialogs were longer (number
of on-task units) (M= 20.35, SD= 12.30) than essays (M=
10.76; SD= 4.03), t(71) = –4.01, p< .001, d= 1.05. An aver-
age of 3.73 (SD= 2.28) on-task units in the dialogs and 2.53
(SD= 1.73) on-task units in the essays referred to evidence,
with no di erence between the two modes, t(71) = –1.46,
p= .15, ns. The following analyses were based on the pro-
portions of each category in the distinct three coding sys-
tems on the total number of on-task units referring to evi-
dence. To make the data suitable for further analyses, we
then performed a logit-transformation of these propor-
tions and used these transformed values.
Evidence Use. In the essays, 81.9 % of the evidence was
drawn from the cards and the remaining 18.1 % was per-
sonal evidence. Moreover, 26.6 % of the card evidence
(representing 81.9 % of the total evidence) was anecdotal,
12.6 % was from authority, 1.9 % was from popularity,
9.0 % was laboratory and 31.9 % was descriptive statistics.
In the dialogs, 67.8 % of the evidence was drawn from the
evidence cards and the remaining 32.2 % was personal ev-
idence. Here, 23.7 % of the card evidence was anecdotal,
12.5 % was from authority, 3.0 % was from popularity,
5.5 % was laboratory and 23.1 % was descriptive statistics
(see also Figure 1).
We then examined whether students' performance in
dialogs and essays di ered with regard to the use of evi-
dence from the cards. The analysis revealed that students
used more evidence from the cards in the essays (M= 2.83,
SD= 2.70) than in the dialogs (M= 1.38, SD= 3.10,  xed
coe cient: -0.75), t(71) = 2.30, p= .02, d= 0.49. Moreover,
there was no speci c in uence of either the class, t(71)=
1.03, p= .31, ns, or the order of intervention, t(70)= –0.82,
p= .41, ns. Furthermore, the amount of evidence used in
essays and dialogs did not di er for any of the  ve prede-
ned subcategories (p> .2, ns.).
The function and structure of evidence use
A total of 26.7 % of the evidence in the essays and 30.9 %
in the dialogs was second-order evidence used to support
the premises of an argument, whereas 60.3 % of the evi-
dence in the essays and 29.5 % in the dialogs corresponded
Figure 2. Means and standard deviations for the fi ve types of evidence presented on the cards as well as personal evidence in essays and dialogs
(“What kind of evidence was used?”).
0
5
10
15
20
25
30
35
Anecdotal Authority Popularity Laboratory Statistical Personal
Essays
Dialogs
Author's personal copy (e-offprint)
128 E. Mayweg-Paus & F. Macagno, Infl uence of a Dialogic Setting
Zeitschrift für Pädagogische Psychologie (2016), 30(2–3), 121–132 © 2016 Hogrefe
to  rst-order evidence used to support a claim directly. In
addition, 12.2 % of the evidence in the essays and 27.2 % in
the dialogs was used to weaken the other's claim, and
1.0 % in the essays and 12.4 % in the dialogs was used to
weaken the other's evidence (see also Figure2).
We then examined whether students' performance in
essays and dialogs di ered in terms of the purpose of evi-
dence use as represented by the four coding categories.
The analysis showed that students gave evidence to sup-
port the premises of their arguments to an almost equal
extent in both dialogs and essays, t(71) = 0.41, p= .68, ns.
Furthermore, there was no speci c in uence of the class, t
(71) = 0.54, p= .54, ns, or the order of intervention, t(70) =
–1.27, p= .21, ns ( xed coe cient: –1.19). However, in their
essays, students used  rst-order evidence more often to
support their claim directly (M= 0.84, SD= 3.97) than in
the dialogs (M= –1.74 SD= 3.28), t(71) = 2.86, p= .01, d=
0.71, ( xed coe cient: –2.99). Again, there was no speci c
in uence of either the class, t(71) = 0.83, p= .41, ns, or the
order of intervention, t (70) = –0.81, p= .42, ns.
We further analyzed whether there were any di erences
in the students' use of evidence to attack the opposing site.
Students used more  rst-order evidence to weaken the op-
posing claims in dialogs (M= –2.07, SD= 2.75) than in es-
says (M= –3.48, SD= 2.74), t(71) = –2.66, p= .01, d= 0.51,
( xed coe cient: –0.13). However, there was no speci c
in uence of either class, t(71) = –0.55, p= .59, ns, or the or-
der of intervention, t(70) = 1.42, p= .1 6, ns. A further analy-
sis showed that students used second-order evidence
more often to attack the evidence supporting the opposing
claim in the dialogs (M= –3.31, SD= 2.74) than in the essays
(M= –4 .49, SD= 0.64,  xed coe cient: –1.14), t(71) = –2.74,
p = .01, d = 0.59. As in the former analyses, there were
again no di erences between the classes, t(71) = –1.48, p=
.14, ns, or the order of intervention, t (70) = 1.29, p= .20, ns.
Table 4 shows a prototypical essay (D08_1A), coded
with the categories of the coding scheme. The student's
claim was put forward in Move 2, after supporting it di-
Figure 3. Means and standard deviations afor the four categories representing functional and structural aspects of evidence use (“How was evi-
dence used?”).
0
10
20
30
40
50
60
70
Support claim Support premises Weaken claim Weaken evidence
Essays
Dialogs
Table 4. Example of an essay.
Move Contribution Code
1 Each year, an estimated 443,000 people die
from smoking or exposure to secondhand
smoke and another 8.6 million have to live
a serious illness caused by smoking.
Supp_claim
2 If I was invited to join a discussion about ban-
ning cigarette sales in the US, I would tell them
that I believe cigarettes should be banned and
that I am one hundred percent certain of my
point of view.
Other
3 I personally know a young boy who has a hole in
his throat since birth. That hole was due to his
mother's cigarette addiction before and while
she was pregnant with him.
Supp_ claim
4 Commercials on cigarettes give information on
what cigarettes are made out of.
Supp_ claim
5 Studies show that the methane found in ciga-
rettes is also found in cat urine and dog poop.
Supp_ claim
6 Studies also show the nicotine in cigarettes
causes fast-acting chemical reactions in your
brain that relieve anxiety and nervousness.
Supp_ claim
7 My grandmother can smoke up to a pack of
cigarettes a day.
Supp_ claim
8 And reading these facts makes me even more
nervous of her health in the future.
Other
Author's personal copy (e-offprint)
E. Mayweg-Paus & F. Macagno, Infl uence of a Dialogic Setting 129
© 2016 Hogrefe Zeitschrift für Pädagogische Psychologie (2016), 30(2–3), 121–132
rectly in Move 1. Moves 3 to 7 were all intended to support
the claim directly (“I believe cigarettes should be
banned”), without interpreting it or providing a link be-
tween it and the various pieces of evidence. In particular,
the student used statistical evidence in Move 1, personal
evidence in Moves 3 and 7, scienti c evidence in Move 6
and mixed type of evidence in Moves 4 and 5 (scienti c
evidence known through personal experience or hearsay).
In all these moves, the student used evidence as  rst-order
evidence without showing why such information could be
relevant to the desired conclusion.
Table 5 reports how the coding scheme captured the dif-
ferent uses of evidence in a dialogue. After the claim was
put forth and defended with arguments by A, B used Turn
2 to advance a contrary opinion backed by evidence. Its
conclusion (smoking causes death) was then attacked in
Turn 3 with an argument supported by second-order evi-
dence. In Turns 4 and 5, the students used  rst-order evi-
dence to support the claims without any interpretation.
However, in Turns 6 to 10, they opened up a meta-dialogic
discussion on the relevance of the problem of addiction,
using second-order evidence pro and contra the relation-
ship between cancer (and mortality) and smoking. Finally,
in Turn 11, Student B advanced another argument leading
to the choice of banning cigarette sales, which was then
quali ed (undermined) by Student A, who pointed out that
the dangers resulting from addiction do not a ect every-
one.
Discussion
The study reported here was designed to investigate how a
dialogic setting in uences middle school students' evi-
dence use in argumentation. The crossover design, in
which each participant's argumentative behavior in a dia-
logic and a monologic mode is compared directly within
the person, allowed us to control for individual di erences
as well as e ects of order. The experimental setting em-
ployed in this study helped to render the di erences result-
ing exclusively from the argumentative setting more
salient. The data were analyzed by selecting three funda-
mental criteria: the argumentative purpose of evidence
use (function), the structure thereof (argumentative struc-
ture) and the content of the pieces of evidence.
From a functional point of view, we detected a funda-
mental di erence between the two settings regarding the
consideration of the opposing position. Students used
more evidence to support their own claim than to weaken
the opposing one – regardless of whether they were writing
an essay (87 %) or discussing with a peer partner (60 %).
However, whereas the essays included almost exclusively
rst-order evidence in support of the own view, the use of
evidence in the dialogs was more balanced. Here, students
tended to use second-order evidence to weaken the oppos-
Table 5. Example of a dialog.
Turn Dialog
Partner
Contribution Code
1 A “Ok, I feel like it shouldn't be like,
people should keep selling them be-
cause farmers live from tobacco, farm-
ing tobacco. So without tobacco they
can't, they can't live without tobacco.
They don’t get money, they don't get
paid. They don't have no money to pay
their bills if they don't farm tobacco.
Other
2B
“But at the same time people are dying
because of that like last year, an esti-
mated 443,000 people die from smoking
or exposure to secondhand smoking.
Supp_ claim
3 A “Yeah but scientists showed that also
for people, uhm, with cancer, uhm, for
help take away their stress and it gives
them the desire to eat”
Weak_
claim
4 B “Yeah, but how do you feel about
people trying to quit smoking, like each
year about 17 million people try to quit
smoking, but only about 8 % of them
succeed.
Supp_ claim
5 A “But people think that everyone has
the right to smoke or not.
Supp_ claim
6 A “Yeah, but it doesn't matter if you
smoke or not because there are people
that are like 100 years old and they
have been smoking for 80 years and
they are still alive, it doesn't matter if
you smoke or not. It depends on how
you are…
Weak_ev
7 B “But at the same time remember
Georg Harrison, he used to be smoking
and died of lung cancer at only 58.
Weak_ev
8A
“Yeah but that wasn't only the smoking,
he was eating…
Other
9 B “Not at all. Other
10 A “But he got lung cancer because he
didn't know though.
Other
11 B “You never know but at the same time
the government is still like what it
does, it enhances the probability of
getting heart attack, it is telling us
that's danger and in my opinion it
should be banned.
Weak_
claim
12 A “Yeah, but it's like, there are people
that don't get addicted to cigarettes
they only enjoy smoking them. It's not
everybody that is smoking cigarettes.
Weak_
claim
Author's personal copy (e-offprint)
130 E. Mayweg-Paus & F. Macagno, Infl uence of a Dialogic Setting
Zeitschrift für Pädagogische Psychologie (2016), 30(2–3), 121–132 © 2016 Hogrefe
ing view, which is in line with Walton's (1989) dual goals of
argumentative discourse. These results replicate  ndings
on more practiced students reported by Kuhn and Moore
(2015). Apparently, although being taught dialogic argu-
mentation as a 2-year-long intervention (compared with a
nondialogic curriculum) clearly improved students' ability
to use evidence to support their own claim, the proportions
of evidence use addressing the opposing view remained
almost the same between dialogs and essays. This shows
how much the setting itself, namely the argumentative
mode, in uences the way evidence is used.
From a structural perspective, we found a crucial di er-
ence in the kind of argumentative use of evidence to sup-
port or weaken a viewpoint. Students used evidence in dif-
ferent ways: They cited it to support their claim directly
( rst-order evidence), or they interpreted it to back up
their arguments or weaken the grounds of the opposing
one (second-order evidence). In the essays, when support-
ing their view, students used  rst-order evidence more of-
ten, that is, a less elaborated argumentative strategy. In the
dialogic setting, when referring to the opposing view in
order to weaken it, students used second-order evidence
more frequently to weaken the argument put forward by
the interlocutor. However, this second  nding has to be
interpreted with caution, because the essays generally in-
cluded less evidence weakening the opposing claim (this
also holds for  rst-order evidence used to attack the op-
posing side's evidence directly). Nevertheless, these  nd-
ings indicate that in dialogs, arguers engage more deeply
and authentically in a critical analysis of the positions and
the evidence at their disposal. Students seem to be more
critical – by referring more frequently to the opposing view
–and clearer– by showing a more elaborated line of rea-
soning. This  nding brings to light another important is-
sue. In the dialogic setting, students used second-order
evidence more frequently, but they also challenged it more
often. The use of second-order evidence seems to lead the
interlocutor to the choice of trying to weaken or rebut it. As
a matter of fact, this option is easier and more strategic
than the alternative, namely supporting alternative data
warranting a contrary position with a di erent backing
(Macagno & Walton, 2014). The simple second-order evi-
dence undercutter leaves the burden of persuasion with
the speaker, and the attacker simply has to continue to re-
but the backings without any need to provide an alterna-
tive line of argument.
From a content point of view, the students drew almost
exclusively on evidence from the evidence cards in their
essays (on average 82 % of references to evidence were of
this type). In the dialogs, in contrast, results were very dif-
ferent. Participants were comparably more likely to draw
on evidence from their own personal knowledge or experi-
ences. Here, 68 % of evidence references came from the
cards, whereas 32 % came from personal knowledge. In
our view, this di erence might be attributed to the stu-
dents' (perceived) expectations as well as their prior expe-
riences associated with the setting. Essay writing is a well-
established educational method in the school context, and
students are used to being evaluated on the basis of their
performance. Therefore, they will tend to opt for the piec-
es of evidence provided by the teacher. In contrast, peer
dialogs represent a much more natural setting. Not sur-
prisingly, students will then tend to ground their line of ar-
gumentation much more on evidence taken from their
personal experience. Moreover, the presence of the peer
partner increases this e ect. When evidence was taken
from personal knowledge or experience in our study, the
further reply of the partner was a ected by this argumen-
tative choice.1 However, this interdependency of the di-
alog partners raises a crucial issue: The dynamics of every
single dialog depend very strongly on the input of the par-
ticular interlocutors (such as prior knowledge on the topic
or prior experience in argumentation). Thus, the students'
behavior was not determined solely by the mode, but also
by the speci c structure of the dyad they were assigned to.
In this study, the sample size of the within design as well as
the statistical procedure chosen allowed us to control these
potential e ects. Nevertheless, future research should ad-
dress this point by incorporating individual measures and
investigating their impact on dialogic dynamics. In this
context, it is necessary to consider the in uence of other
relevant aspects such as emotions and/or motivations.
No di erence was found regarding the  ve qualitative
categories from the cards. Students showed a strong ten-
dency to use descriptive statistics as evidence in their ar-
guments (32 % in essays and 23 % in dialogs). Then, in the
argumentative setting, they referred preferably and more
or less equally (27 % in essays and 24 % in dialogs) to anec-
dotal evidence (the weakest category according to the ex-
perts' ratings), whereas laboratory evidence (actually rat-
ed by experts to be the strongest evidence) was used only
very rarely (9 % in essays and 6 % in dialogs).
These descriptives point strongly to the need to develop
students' argumentative skills, particular those used for dis-
tinguishing, evaluating and using di erent types of evidence
in a sophisticated way. The present  ndings can improve our
understanding of how curricula based on the dialogic ap-
proach (Kuhn, Hemberger & Khait, 2014) can contribute to
1 Our results confi rm what Kuhn and Moore (2015) observed for students with 2 years of practice in dialogical argumentation. However, whereas
the amount of shared evidence (or evidence from a list provided to the students) they reported in essays was almost the same as in our fi ndings,
it was much less in dialogs (around 20 %). This might be attributed to their higher experience with arguing in a dialogical setting.
Author's personal copy (e-offprint)
E. Mayweg-Paus & F. Macagno, Infl uence of a Dialogic Setting 131
© 2016 Hogrefe Zeitschrift für Pädagogische Psychologie (2016), 30(2–3), 121–132
the development of argumentation in the classroom. Where-
as the dialogic setting itself exerts a positive in uence on the
functional and structural dimensions of evidence use, this
does not apply directly to the understanding of evidence
quality. Beyond further practice –which could already be
proven to be of high relevance for the development of argu-
ment skills (Crowell & Kuhn, 2014)– additional support is
required to help students use “strong” evidence in an elabo-
rated way. One approach to help students develop a more
complete and precise understanding of di erent types of
evidence could be metacognitive prompting (see Jucks,
Schulte-Löbbert & Bromme, 2007; Thiebach, Mayweg-Paus
& Jucks, 2016).
Metacognitive prompts are questions or hints that en-
courage metacognitive activities during the learning pro-
cess. Prompts are designed to overcome super cial pro-
cessing (King, 1992). In particular, re ection prompts
(Davis, 2003) target learners' re ection on their own un-
derstanding and learning, and provide them with possibili-
ties regarding how to use di erent prompts. Prompts can
be implemented in di erent ways. Within the new tech-
nologies, they are often used in computer-supported set-
tings (Chen, Wei, Wu & Udden, 2009; Stadtler & Bromme,
2007). When using evidence for an argumentative pur-
pose, students could be provided with re ection prompts
asking them to think about how convincing the evidence
they intend to use might be (e. g. how easily [or not] it
could be rebutted by the opposing side). Additionally, one
could make transcripts of former (successful) dialogs
available that can be used for re ective activities. Enhanc-
ing such meta-level awareness has already been shown to
be a very e cient approach for developing argumentative
strategy use (see Kuhn et al., 2008).
From a theoretical point of view, this study focused
mostly on the function and the structure of the use of evi-
dence by developing and applying a coding scheme in
which these two dimensions are combined. We addressed
the problem of the content of the pieces of evidence sepa-
rately by classifying these pieces of evidence into classes
ranked by experts and analyzing the possible di erences
in their use. A basic link that needs to be investigated is the
relationship between the functional-structural dimension
and the content. A piece of evidence can be used as a
premise or as a backing to either support a viewpoint or at-
tack an argument. However, is the speci c piece of evi-
dence adequate for its intended function? Sandoval and
Millwood (2005) and McNeill (2011) have tackled this is-
sue. The  rst authors introduced the coding categories of
su ciency and conceptual quality to capture whether suf-
cient relevant data are cited to justify a claim, and wheth-
er they can back it up reasonably. The second author pro-
posed the criteria of su ciency and appropriateness to
analyze the type and quality of the relationship between
evidence and claim. However, such coding schemes fail to
distinguish the functional-structural from the content di-
mensions, and the criteria for assessing the appropriate-
ness and the conceptual quality of evidence use are left
unspeci ed. Drawing on advances in argumentation theo-
ry, it is possible to develop the functional-structural coding
scheme to assess the relevance of a piece of evidence for
the intended conclusion (Macagno, 2008; Macagno &
Walton, 2014). Evidence can be evaluated as preferentially
pro or contra a given conclusion, and its degree of support
can be ranked as weak or strong. In this fashion, it would
be possible to examine not only whether or not students
have taken the con icting view into account and have in-
terpreted the piece of evidence, but also whether they have
understood its meaning and its possible argumentative
consequences.
References
Aristotle. (1995). Topics. In J. Barnes (Ed.), The works of Aristotle.
Princeton, NJ: Princeton University Press.
Asterhan, C. S. C., & Schwarz, B. B. (2009). Argumentation and ex-
planation in conceptual change: Indications from protocol anal-
yses of peer-to-peer dialog. Cognitive Science, 33, 374–400. doi:
10.1111/j.1551-6709.2009.01017.x
Barr, D. J., Levy, R., Scheepers, C., & Tily, H. J. (2013). Random ef-
fects structure for confi rmatory hypothesis testing: Keep it
maximal. Journal of Memory and Language, 68, 255–278. doi:
10.1016/j.jml.2012.11.001
Baayen, R. H., Davidson, D. J., & Bates, D. M. (2008). Mixed-effects
modeling with crossed random effects for subjects and items.
Journal of Memory and Language, 59, 390–412. doi: 10.1016/
j.jml.2007.12.005
Chen, N.-S., Wei, C.-W., Wu, K.-T., & Udden, L. (2009). Effect of high
level prompts and peer assessment on online learners' refl ec-
tion levels. Computers & Education, 52, 283–291. doi: 10.1016/
j.compedu.2008.08.07
Clark, D., Sampson, V., Weinberger, A., & Erkens, G. (2007). Analytic
framework for assessing dialogic argumentation in online
learning environments. Educational Psychology Review, 19,
343–374. doi: 10.1007/ s10648–007–9050–7
Crowell, A., & Kuhn, D. (2014). Developing dialogic argumentation
skills: A three-year intervention study. Journal of Cognition and
Development, 15 (2), 1–46. doi: 10.1080 /15248372.2012.725187
Davis, E. A. (2003). Prompting middle school science students for
productive refl ection: Generic and directed prompts. Journal of
the Learning Sciences, 12 (1), 91–142. doi: 10.1207/s15327809
jls1201_4
Driver, R., Newton, P., & Osborne, J. (2000). Establishing the norms
of scientifi c argumentation in classrooms. Science Education,
84, 287–312. doi: 10.1002/(SICI)1098–237X(200005)84:3<287::
AID-SCE1>3.0.CO;2-A
Duschl, R. A., & Osborne, J. (2002). Supporting and promoting argu-
mentation discourse in science education. Studies in Science
Education, 38 (1), 39–72. doi: 10.1080/03057260208560187
Iordanou, K., & Constantinou, C. P. (2014). Developing pre-service
teachers' evidence-based argumentation skills on socio-scien-
tifi c issues. Learning and Instruction, 34, 42–57.
Author's personal copy (e-offprint)
132 E. Mayweg-Paus & F. Macagno, Infl uence of a Dialogic Setting
Zeitschrift für Pädagogische Psychologie (2016), 30(2–3), 121–132 © 2016 Hogrefe
Iordanou, K., & Constantinou, C. P. (2015). Supporting use of evi-
dence in argumentation through practice in argumentation and
refl ection in the context of SOCRATES learning environment.
Science Education, 99, 282–311. doi: 10.1002/sce.21152
Jimenez-Aleixandre, M. P., Rodriguez, A. B., & Duschl, R. A. (2000).
Doing the lesson or doing science: Argument in high school ge-
netics. Science Education, 84, 757–792.
Jucks, R., & Paus, E. (2013). Different words for the same con-
cept: Learning collaboratively from multiple documents. Cog-
nition and Instruction, 31(2), 1–28. doi: 10.1080/07370008.
2013.769993
Jucks, R., Schulte-Löbbert, P., & Bromme, R. (2007). Supporting
experts' written knowledge communication through refl ective
prompts on the use of specialist concepts. Journal of Psycholo-
gy, 215, 237–247. doi: 10.1027/0044-3409.215.4.237
King, A. (1992). Comparison of self-questioning, summarizing, and
notetaking-review as strategies for learning from lectures.
American Educational Research Journal, 29, 303–323, 303–323.
doi: 10.3102/00028312029002303
Koballa, T. R. (1992). Persuasion and attitude change in science
education. Journal of Research in Science Teaching, 29 (1),
63–80. doi: 10.1002/tea.3660290107
Kuhn, D. (1993). Science as argument: Implications for teaching
and learning scientifi c thinking. Science Education, 77,
319–337.
Kuhn, D. (2015). Thinking together and alone. Educational Re-
searcher, 44(1), 46–53. doi: 10.3102/0013189X15569530
Kuhn, D., Amsel, E., & O'Loughlin, M. (1988). The development of
scientifi c thinking skills. Orlando, FL: Academic Press.
Kuhn, D., & Crowell, A. (2011). Dialogic argumentation as a vehicle
for developing young adolescents' thinking. Psychological Sci-
ence, 22, 545–552. doi: 10.1177/0956797611402512
Kuhn, D., Garcia-Mila, M., Zohar, A., Andersen, C., White, S. H.,
Klahr, D., & Carver, S. M. (1995). Strategies of knowledge acqui-
sition. Monographs of the Society for Research in Child Develop-
ment, 60(4), i. doi: 10.2307/1166059
Kuhn, D., Goh, W., Iordanou, K., & Shaenfi eld, D. (2008). Arguing on
the computer: A microgenetic study of developing argument
skills in a computer-supported environment. Child Develop-
ment, 79, 1310–1328. doi: 10.1111/j.1467-8624.2008.01190.x
Kuhn, D., Hemberger, L., & Khait, V. (2014). Argue with me. New York,
NY: Wessex.
Kuhn, D., & Moore, W. (2015). Argumentation as a core curriculum.
Learning: Research and Practice, 1 (1), 66–78. doi: 10.1080/
23735082.2015.994254.
Kuhn, D., & Udell, W. (2003). The development of argument skills.
Child Development, 74, 1245–1260. doi: 10.1111/1467-8624.
00605
Kuhn, D. (1992). Thinking as argument. Harvard Educational Re-
view, 62, 155–179. doi: 10.17763/haer.62.2.9r424r0113t670l1
Macagno, F. & Walton, D. (2014). Emotive language in argumenta-
tion. New York, NY: Cambridge University Press.
Macagno, F. (2008). Dialectical relevance and dialogical context in
Walton's pragmatic theory. Informal Logic, 28, 102–128.
Macagno, F. (2012). Reconstructing and assessing the conditions
of meaningfulness: An argumentative approach to presupposi-
tion. In H. Ribeiro (Ed.), Inside arguments: Logic and the study of
argumentation (pp. 247–268). Newcastle upon Tyne, UK: Cam-
bridge Scholars Publishing.
Macagno, F., Mayweg-Paus, E., & Kuhn, D. (2014). Argumentation
theory in education studies: Coding and improving students'
argumentative strategies. Topoi: An International Review of Phi-
losophy. doi: 10.1007/s11245-014-9271-6.
Mayweg-Paus, E., Macagno, F., & Kuhn, D. (2015). Developing argu-
mentation strategies in electronic dialogs: Is modeling effec-
tive?Discourse Processes. doi: 10.1080/0163853X.2015.1040323
McNeill, K. L. (2011). Elementary students' views of explanation,
argumentation, and evidence, and their abilities to construct
arguments over the school year. Journal of Research in Science
Teaching, 48, 793–823.
McNeill, K. L., Lizotte, D. J., Krajcik, J., & Marx, R. W. (2006). Support-
ing students' construction of scientifi c explanations by fading
scaffolds in instructional materials. The Journal of the Learning
Sciences, 15, 153–191. doi: 10.1207/s15327809jls1502_1
Nussbaum, E. M., & Edwards, O. V. (2011). Critical questions and
argument stratagems: A framework for enhancing and analyz-
ing students' reasoning practices. Journal of the Learning Sci-
ences, 20, 443–488.
Osborne, J. F., Erduran, S., Simon, S., & Monk, M. (2001). Enhancing
the quality of argument in school science. School Science Re-
view, 82 (301), 63–70.
Pollock, J. L. (1974). Knowledge and justifi cation. Princeton, NJ:
Princeton University Press.
Rapanta, C., Garcia-Mila, M., & Gilabert, S. (2013). What is meant
by argumentative competence? An integrative review of meth-
ods of analysis and assessment in education. Review of Educa-
tional Research, 83, 483–520. doi: 10.3102/0034654313487606
Sandoval, W. A. (2003, July). The inquiry paradox: Why doing sci-
ence doesn't necessarily change ideas about science. In Pro-
ceedings of the sixth international computer-based learning in
science conference (pp. 825–834).
Sandoval, W., & Millwood, K. (2005). The quality of students' use of
evidence in written scientifi c explanations. Cognition and In-
struction, 23 (1), 23–55. doi: 10.1207/s1532690xci2301_2
Stadtler, M., & Bromme, R. (2007). Dealing with multiple docu-
ments on the WWW: The role of metacognition in the formation
of documents models. International Journal of Computer Sup-
ported Collaborative Learning, 2, 191–210. doi: 10.1007/s11412-
007-9015-3
Thiebach, M., Mayweg-Paus, E., & Jucks, R. (2016). Better to agree
or disagree? - The role of critical questioning and elaboration in
argumentative discourse. Zeitschrift für Pädagogische Psychol-
ogie, 30(2–3), 133–149.
Toulmin, S. (1958). The uses of argument. Cambridge, UK: Cam-
bridge University Press. doi: 10.1017/S0031819100037220
Walton, D. N. (1989). Dialogue theory for critical thinking. Argumen-
tation, 3, 169–184. doi: 10.1007/bf00128147
Walton, D. N. (2006). Fundamentals of critical argumentation. New
York, NY: Cambridge University Press.
Walton, D. N. & Macagno, F. (2007). The fallaciousness of threats:
Character and ad baculum. Argumentation, 21(1), 63–81.
Manuscript submitted: 13.3.2015
Manuscript accepted: 2.9.2015
Confl ict of interest: No
Elisabeth Mayweg-Paus
Westfälische Wilhelms-Universität Münster
Institute of Psychology for Education
Fliednerstraße 21
48149 Münster
Germany
e.mayweg@uni-muenster.de
Author's personal copy (e-offprint)
... In such contexts, a person is subject to others' scrutiny of their own position, which, in turn, enhances one's need to be more critical not only toward one's own position but also the opposing position. As research shows, the dialogic nature of the interaction directly affects how people handling evidence: In a collaborative setting, students seem to use evidence more often to address opposing viewpoints in an elaborated way, whereas in an individual setting they are more likely to stick to the information given to them (shared evidence) instead of integrating new information (Kuhn and Moore, 2015;Mayweg-Paus and Macagno, 2016). The potential of collaborative engagement for helping people deal with scientific online information efficiently lies in specific communicative moves (such as exchanging multiple perspectives) that can elicit (deeper) cognitive processing of information and a critical reflection of sources. ...
... Based on research approaches to collaboration and argumentation, collaborative argumentation settings tend to encourage people to engage with evidence more reflectively and in a more differentiated way (e.g., Chin and Osborne, 2008;Kuhn and Moore, 2015;Mayweg-Paus and Macagno, 2016), which may help students critically question the evidence presented in the conflicting information (e.g., Mayweg-Paus and Macagno, 2016;. However, individual reflection settings may also have advantages, as they are much more easy to manage and may help students reflect on their sourcing competencies (Mullins et al., 2011;Zimmermann and Mayweg-Paus, 2021). ...
... Based on research approaches to collaboration and argumentation, collaborative argumentation settings tend to encourage people to engage with evidence more reflectively and in a more differentiated way (e.g., Chin and Osborne, 2008;Kuhn and Moore, 2015;Mayweg-Paus and Macagno, 2016), which may help students critically question the evidence presented in the conflicting information (e.g., Mayweg-Paus and Macagno, 2016;. However, individual reflection settings may also have advantages, as they are much more easy to manage and may help students reflect on their sourcing competencies (Mullins et al., 2011;Zimmermann and Mayweg-Paus, 2021). ...
Article
Full-text available
In view of the COVID-19 pandemic, students had to cope with the challenging situation of handling a vast amount of potentially conflicting online information while staying informed. Reading conflicting scientific information has been shown to require cognitive effort for one to integrate it successfully, but reading such information during a crisis–such as the COVID-19 pandemic–may cause additional emotional stress, as students also had to cope with critical aspects of the pandemic (e.g., physical distancing and uncertainty). Different studies have indicated that in crises, stress can be relieved by seeking online social support (as a coping strategy). Similarly, working together (as collaborative learning) can also help people more critically discuss information on a cognitive level. Based on the approaches of online collaborative learning and online social support seeking, we were interested in whether an individual vs. collaborative communication setting would lead to any differences in students’ cognitive as well as emotional engagement with conflicting information about COVID-19. In a 2 × 2 mixed design, N = 109 education science students were exposed to two conflicting texts regarding COVID-19 testing that contained current scientific information. The online experiment was conducted in Germany in April 2020, which was the beginning of lockdown in that country. After reading the two texts, participants were asked to reflect on their engagement with the conflicting information either individually (individual group, n = 49) or via chat collaboratively (collaboration group, n = 60 in 30 dyads). With respect to participants’ written reflections (content-analyzed regarding cognitive as well as emotional engagement), participants in the collaborative group, compared to those in the individual group, more often discussed the pandemic in general and less often engaged emotionally when discussing the evidence from texts. All participants reported higher perceived information overload, lower self-efficacy in sourcing information about COVID-19, and higher active coping strategies after the reflection task compared to before reading the information, with no significant differences between the collaborative and individual groups. We discuss these findings regarding any opportunities and challenges that arise in online collaboration between students for cognitive and emotional engagement when handling conflicting information about COVID-19.
... Accordingly, an important feature of this type of consensus building is that individuals do not accept contributions of their partners as they are. In this context, efficient communication comprises strategies that directly address and challenge the argumentative structure and content (e.g., scientific evidence) of the other's contributions (Mayweg-Paus and Macagno, 2016;Mayweg-Paus et al., 2016a). In particular, critical questioning seems to be a strong argumentative strategy given its capacity to address deeper grounds of disagreement, bringing into light background knowledge and knowledge beliefs that might otherwise escape attention. ...
... Collaboratively achieving epistemic aims in dialogic argument depends substantially on the discourse partners' efforts to deeply elaborate on and challenge their partner's knowledge and arguments (e.g., Kuhn and Udell, 2003). In this context, the dialogic character (or two-way mode) of argumentation can support OESI through (a) enhancing the quality of argumentation and the use of evidence (Crowell and Kuhn, 2014;Mayweg-Paus and Macagno, 2016) and (b) the evaluation and reconciliation of diverging claims (Nussbaum and Edwards, 2011;Felton et al., 2015a). In an argumentative dialogue, a person is subject to the interlocutor's scrutiny of her own position, which enhances her need to be more critical not only toward her own position but also the opposing one. ...
Article
Full-text available
Many urgent problems that societies currently face-from climate change to a global pandemic-require citizens to engage with scientific information as members of democratic societies as well as to solve problems in their personal lives. Most often, to solve their epistemic aims (aims directed at achieving knowledge and understanding) regarding such socio-scientific issues, individuals search for information online, where there exists a multitude of possibly relevant and highly interconnected sources of different perspectives, sometimes providing conflicting information. The paper provides a review of the literature aimed at identifying (a) constraints and affordances that scientific knowledge and the online information environment entail and (b) individuals' cognitive and motivational processes that have been found to hinder, or conversely, support practices of engagement (such as critical information evaluation or two-sided dialogue). Doing this, a conceptual framework for understanding and fostering what we call online engagement with scientific information is introduced, which is conceived as consisting of individual engagement (engaging on one's own in the search, selection, evaluation, and integration of information) and dialogic engagement (engaging in discourse with others to interpret, articulate and critically examine scientific information). In turn, this paper identifies individual and contextual conditions for individuals' goal-directed and effortful online engagement with scientific information.
... Percentages of students who employed these strategies at least five times also favored the intervention condition. Again, these findings are consistent with prior studies showing the effectiveness of the intervention in enhancing use of evidence in argumentation (Iordanou & Constantinou, 2015;Kuhn & Moore, 2015;Mayweg-Paus & Macagno, 2016). In addition, we show here that gains observed in interpersonal dialogs in prior studies transfer to and are identifiable in individually constructed dialogs. ...
... This finding was particularly encouraging given the accumulation of research indicating a strong tendency across age groups to substitute mechanism explanation for empirical evidence when supporting a causal claim (Ahn, Kalish, Medin, & Gelman, 1995;Brem & Rips, 2000;Kuhn, 1991;Kuhn & Pearsall, 1998Ross, Lepper, & Hubbard, 1975). Our present results favoring the younger group suggest that skills in using evidence are highly amenable to intervention, and that extended practice of coordinating claims with evidence afforded by peer dialogs, in which participants more frequently than in individual writing draw on evidence to weaken an opposing side (Kuhn & Moore, 2015;Macagno, 2016;Mayweg-Paus & Macagno, 2016), constitutes a productive path to promoting construction of evidence-based counterarguments in both individual and dialogic arguments. Our results also show that skills with evidence use to justify claims might not necessarily develop with age, as 7th and 9th grade controls showed comparable performance in this regard. ...
Article
Full-text available
This study investigates the possibility that solitary dialog, in which individuals construct in writing a hypothetical dialogic argument, may more fully reveal individual skill achievement in argument than do conventional argumentive essays. A sample of 54 11-12-year-old Chinese students individually composed such written dialogs, subsequent to their participation in a four-month dialog-based argument curriculum that previously reported gains in both dialogic and essay assessments. Also partaking in the constructed dialog task reported on here were two non-intervention control groups from the same school; one the same age (n=50) as and the other two years older (n=52) than the intervention group. As well as outperforming their agemates, the intervention group’s performance on the constructed dialog task showed they had achieved skill equal to that of the older group in counterargument and were superior to them in using evidence to justify claims. The possibility is considered that the my-side bias reported in typical argumentive essays is due to limited understanding of the purpose of essay writing, rather than lack of skills per se.
... Such skills do show improvement, however, with engagement and practice (Kuhn & Crowell, 2011;Hemberger et al., 2017;Iordanou, 2010;Iordanou & Constantinou, 2015;Rapanta, 2019;Reznitskaya & Wilkinson, 2017;Ryu & Sandoval, 2012;Shi, 2019). In general, progress in dialogic argument precedes that in individual argument (Kuhn, 2019;Kuhn & Moore, 2015;Mayweg-Paus & Macagno, 2016;Shi, 2019). Initially, the essays of novices consist largely of supporting arguments for a favored position. ...
Preprint
Full-text available
Schooling traditionally affords students more experience in learning and practicing procedures than in identifying what a situation calls for. When asked to choose appropriate numerical data to support their causal claims, college students perform surprisingly poorly. In one case we describe, almost all chose limited, inconclusive data as sufficient evidence, despite having available the more comprehensive data needed to support their claim, and despite their established competence to employ such data for this purpose. Our objective in highlighting this weakness is to make a case that choosing one's evidence warrants the status of an important metacognitive intellectual skill and educational objective, one central to but that extends well beyond the domains of scientific and mathematical reasoning and hence warrants greater attention both in and beyond the science curriculum. People may choose evidence to justify their assertions in an ill-considered way, with potential adverse effects in both private and public communication.
... Such skills do show improvement, however, with engagement and practice (Kuhn & Crowell, 2011;Hemberger et al., 2017;Iordanou, 2010;Iordanou & Constantinou, 2015;Rapanta, 2019;Reznitskaya & Wilkinson, 2017;Ryu & Sandoval, 2012;Shi, 2019). In general, progress in dialogic argument precedes that in individual argument (Kuhn, 2019;Kuhn & Moore, 2015;Mayweg-Paus & Macagno, 2016;Shi, 2019). Initially, the essays of novices consist largely of supporting arguments for a favored position. ...
Article
Full-text available
Schooling traditionally affords students more experience in learning and practicing procedures than in identifying what a situation calls for. When asked to choose appropriate numerical data to support their causal claims, college students perform surprisingly poorly. In one case we describe, almost all chose limited, inconclusive data as sufficient evidence, despite having available the more comprehensive data needed to support their claim and despite their established competence to employ such data for this purpose. Our objective in highlighting this weakness is to make a case that choosing one’s evidence warrants the status of an important metacognitive intellectual skill and educational objective, one central to but that extends well beyond the domains of scientific and mathematical reasoning and hence warrants greater attention both in and beyond the science curriculum. People may choose evidence to justify their assertions in an ill-considered way, with potential adverse effects in both private and public communication.
Chapter
Full-text available
Investigación idiográfica con interés en analizar el uso dado a la evidencia en producciones argumentativas de estudiantes de décimo grado, que participan de un diseño pedagógico sustentado en la argumentación como herramienta de mediación para el desarrollo del pensamiento crítico-reflexivo. Se realizó análisis discursivo a 112 producciones argumentativas de 28 casos escogidos por muestreo intencional. Examinando aspectos dialécticos propios de la estructura del argumento (tipo y naturaleza) y, aspectos pragmáticos de la argumentación en situación de uso (función). Los resultados muestran que, se privilegia el uso de pseudoevidencias para sustentar los puntos de vista. Se incluye progresivamente el uso evidencias genuinas. Existe mayor presencia de uso de evidencia como apoyo al propio punto de vista y como recurso para debilitar la posición de otros. Estos hallazgos además de dar cuenta de la sofisticación en el uso de evidencias y del desarrollo del raciocinio argumentativo, también, contribuyen a repensar las condiciones de los ambientes de aprendizaje en los que se hace cada vez más relevante la introducción de acciones pedagógicas que fomenten tanto la producción argumentativa y la vigilancia epistémica en la construcción del conocimiento académico, aspectos que la literatura reporta como competencias centrales del proceso de enseñanza-aprendizaje en contextos formales.
Article
Based on educational-psychological backgrounds on learning through communication and current requirements regarding the development of media skills this article exam- ines the use of online forums in academic teaching from a theoretical and a practical perspective. First, the potential and the specific characteristics of asynchronous and text-based communication in forums are outlined. Then, didactical methods to support frequent participation and transactive communication are introduced. The main focus lays on the role of moderation processes by university teachers or peers as well as on the potentials of using cooperation scripts as instructional guidance for argumentation and exchange processes. Against this background concrete didactic scenarios as well as recommendations on how they might be realized in virtual environments are de- duced and illustrated. Subsequently, suggestions are made on how the effectiveness and quality of these approaches can be proved and secured by using the indicators of participation and knowledge and skill acquisition. Finally, practical implications and fu- ture research perspectives are discussed.
Article
Full-text available
The definition and the assessment of the quality of argumentative texts has become an increasingly crucial issue in education, classroom discourse, and argumentation theory. The different methods developed and used in the literature are all characterized by specific perspectives that fail to capture the complexity of the subject matter, which remains ill-defined and not systematically investigated. This paper addresses this problem by building on the four main dimensions of argument quality resulting from the definition of argument and the literature in classroom discourse: dialogicity, accountability, relevance, and textuality (DART). We use and develop the insights from the literature in education and argumentation by integrating the frameworks that capture both the textual and the argumentative nature of argumentative texts. This theoretical background will be used to propose a method for translating the DART dimensions into specific and clear proxies and evaluation criteria.
Article
Full-text available
In the domain of medical science, factual evidence is usually considered as the criterion on which to base decisions and construct hypotheses. Evidence-based medicine is the translation of this approach into the field of patient care, and it means providing only the type of care that is based on evidence that proves its effectiveness and appropriateness. However, while the literature has focused on the types and force of evidence used to establish the recommendation and treatment guidelines, the problem of how evidence is used in doctor–patient interactions to motivate, or refuse, treatment or habit change has been almost completely neglected. In this specific context, characterized by the epistemic unbalance between the interlocutors and a specific conversational goal (making a decision shared by the patient), the scientific categories of evidence are often irrelevant. The goal of this paper is to address this challenge by analyzing the role evidence plays in doctor–patient interactions in diabetes care. After introducing an analytical distinction between the epistemic and the pragmatic aspect of evidence, we will propose a classification of the types and functions of evidence in chronic care communication, and illustrate its possible uses through its application to our corpus of diabetes-care consultations. From our qualitative analyses, it is possible to observe how in this communicative context a crucial role is played by the conflicts of evidence, in which providers correct patients’ uses of evidence leading them to using less defeasible levels of evidence.
Article
Full-text available
This article explores the concepts of critical questions (from D. N. Walton, 199670. Walton , D. N. 1996 . Argument schemes for presumptive reasoning , Mahwah, NJ : Erlbaum . View all references) and integrative and refutational argument stratagems as an approach for teaching argumentation and critical thinking. A study was conducted for 6 months in 3 sections of a 7th-grade social studies classroom in which 30 students discussed and wrote about current events. One section served as a comparison group. Over time the experimental group made more arguments that integrated both sides of each issue. Collectively, the experimental group also successfully constructed salient critical questions, particularly in regard to weighing values and designing practical creative solutions. In-depth analysis of 1 student showed how conceptual structures and argument practices improved incrementally over time and how the appropriation of stratagems may have been facilitated by the dialectical nature of the intervention (e.g., using critical questions and stratagems successfully in discourse). The theoretical and practical importance of Walton's dialogue theory, and the critical question approach to argumentation, are discussed.
Data
Full-text available
The study presented here examines how interacting with a more capable interlocutor influences use of argumentation strategies in electronic discourse. To address this question, 54 young adolescents participating in an intervention centered on electronic peer dialogs were randomly assigned to either an experimental or control condition. In both conditions, pairs who held the same position on a social issue engaged in a series of electronic dialogs with pairs who held an opposing position. In the experimental condition, in some dialogs, unbeknownst to them (because dialog took place electronically), the opponent was a more capable (“expert”) adult. Dialogs in the control condition were only with peers. Argumentation strategies of the experimental group who argued with the “expert” showed immediate strategy improvements in their subsequent peer dialogs, improvement absent in the control group (Cohen’s d ¼ 1.12). In particular, the experimental group showed greater use of counterargument in general and advanced forms of counterargument (undermining) that challenges the deeper premises or reasoning on which an argument is based. Implications with respect to mechanisms of change in the development of argumentation skills are considered.
Article
Full-text available
The purpose of this study was to determine whether providing students with continuous written instructional support or fading written instructional support (scaffolds) better prepares students to construct scientific explanations when they are no longer provided with support. This article investigated the influence of scaffolding on 331 seventh-grade students' writing of scientific explanations during an 8-week, project-based chemistry unit in which the construction of scientific explanations is a key learning goal. The unit makes an instructional model for explanation explicit to students through a focal lesson and reinforces that model through subsequent written support for each investigation. Students received 1 of 2 treatments in terms of the type of written support: continuous, involving detailed support for every investigation, or faded, involving less support over time. The analyses showed significant learning gains for students for all components of scientific explanation (i.e., claim, evidence, and reasoning). However, on posttest items lacking scaffolds, the faded group gave stronger explanations in terms of their reasoning compared to the continuous group. Fading written scaffolds better equipped students to write explanations when they were not provided with support.
Article
Full-text available
The study presented here examines how interacting with a more capable interlocutor influences use of argumentation strategies in electronic discourse. To address this question, 54 young adolescents participating in an intervention centered on electronic peer dialogs were randomly assigned to either an experimental or control condition. In both conditions, pairs who held the same position on a social issue engaged in a series of electronic dialogs with pairs who held an opposing position. In the experimental condition, in some dialogs, unbeknownst to them (since dialog took place electronically) the opponent was a more capable (“expert”) adult. Dialogs in the control condition were only with peers. Argumentation strategies of the experimental group who argued with the “expert” showed immediate strategy improvements in their subsequent peer dialogs, improvement absent in the control group (Cohen's d = 1.12). In particular the experimental group showed greater use of counterargument in general as well as an advanced form of counterargument (undermining) that challenges the deeper premises or reasoning on which an argument is based. Implications with respect to mechanisms of change in the development of argumentation skills are considered.
Book
Full-text available
This book analyzes the uses of emotive language and redefinitions from pragmatic, dialectical, epistemic, and rhetorical perspectives, investigating the relationship between emotions, persuasion, and meaning, and focusing on the implicit dimension of the use of a word and its dialectical effects. It offers a method for evaluating the persuasive and manipulative uses of emotive language in ordinary and political discourse. Through the analysis of political speeches (including President Obama's Nobel Peace Prize address) and legal arguments, the book offers a systematic study of emotive language in argumentation, rhetoric, communication, political science, and public speaking.
Article
Ist Zustimmung oder Dissens besser? Die Rolle des Kritischen Hinterfragens und Elaborierens im Argumentativen Diskurs Zusammenfassung. Beim Umgang mit wissenschaftsbezogenen Informationen im Internet spielt der argumentative Austausch mit anderen eine immer zentralere Rolle. Als besonders förderlich für den Wissenserwerb durch argumentativen Diskurs haben sich das kritische Hinterfragen und Elaborieren der Information des Partners erwiesen. In der vorliegenden Studie wurde untersucht, wie diese lernförderlichen Kommunikationsaktivitäten im argumentativen Diskurs über wissenschaftsbezogene Informationen gefördert werden können. In einem 1 × 2 Zwischengruppendesign wurde entweder ein Unterschieds- oder ein Gemeinsamkeitenfokus erzeugt, indem Dyaden instruiert wurden, während der Diskussion eines wissenschaftsbezogenen Themas insbesondere auf Unterschiede oder auf Gemeinsamkeiten in den Sichtweisen und Argumenten zu achten. Die Ergebnisse zeigen, dass ein Fokus auf Unterschiede sich förderlich auf die Qualität des argumentativen Diskurses sowie auf das individuelle kritische Denken auswirkte. Die intrinsische Motivation hinsichtlich der Zusammenarbeit mit dem Partner und der Aufgaben war hingegen bei einem Gemeinsamkeitenfokus höher. Implikationen für die Verbesserung der Qualität argumentativer Diskurse und die Förderung des individuellen kritischen Denkens werden diskutiert.
Article
Linear mixed-effects models (LMEMs) have become increasingly prominent in psycholin-guistics and related areas. However, many researchers do not seem to appreciate how random effects structures affect the generalizability of an analysis. Here, we argue that researchers using LMEMs for confirmatory hypothesis testing should minimally adhere to the standards that have been in place for many decades. Through theoretical arguments and Monte Carlo simulation, we show that LMEMs generalize best when they include the maximal random effects structure justified by the design. The generalization performance of LMEMs including data-driven random effects structures strongly depends upon modeling criteria and sample size, yielding reasonable results on moderately-sized samples when conservative criteria are used, but with little or no power advantage over maximal models. Finally, random-intercepts-only LMEMs used on within-subjects and/or within-items data from populations where subjects and/or items vary in their sensitivity to experimental manipulations always generalize worse than separate F 1 and F 2 tests, and in many cases, even worse than F 1 alone. Maximal LMEMs should be the 'gold standard' for confirmatory hypothesis testing in psycholinguistics and beyond.
Article
Basing its arguments in current perspectives on the nature of the scientific enterprise, which see argument and argumentative practice as a core activity of scientists, this article develops the case for the inclusion and central role of argument in science education. Beginning with a review of the nature of argument, it discusses the function and purpose of dialogic argument in the social construction of scientific knowledge and the interpretation of empirical data. The case is then advanced that any education about science, rather than education in science, must give the role of argument a high priority if it is to give a fair account of the social practice of science, and develop a knowledge and understanding of the evaluative criteria used to establish scientific theories. Such knowledge is essential to enhance the public understanding of science and improve scientific literacy. The existing literature, and work that has attempted to use argument within science education, is reviewed to show that classroom practice does provide the opportunity to develop young people's ability to construct argument. Furthermore, the case is advanced that the lack of opportunities for the practice of argument within science classrooms, and lack of teacher's pedagogical skills in organizing argumentative discourse within the classroom are significant impediments to progress in the field.
Article
In attempting to define intelligence in real-world contexts, psychologists have focused primarily on the kinds of thinking that people do in work-related environments. In this article, however, Deanna Kuhn describes another form of thinking that should be central to efforts to describe real-world intelligence: thinking as argument. It is in argument, the author maintains, that we find the most significant way in which higher order thinking and reasoning figure in the lives of most people. Kuhn describes her research, which examines the extent to which a process of reasoned argument underlies the beliefs people hold and the opinions they espouse about important social issues. Her results indicate that argumentive reasoning ability does not differ systematically as a function of sex or age (from adolescence through the late sixties), but is strongly related to education level. Kuhn believes that social contexts, such as the classroom, are the most promising arena for practicing and developing argumentive th...
Book
Presenting the basic tools for the identification, analysis, and evaluation of common arguments for beginners, this book informs by using examples of arguments in dialogues, both in the text itself and in the exercises. (Examples of controversial legal, political, and ethical arguments are analyzed.) Illustrating the most common kinds of arguments, the book also explains how to evaluate each kind by critical questioning. Douglas Walton demonstrates the reasonable nature of arguments under the right dialogue conditions by using critical questions to evaluate them.