Acta Psychologica 246 (2024) 104256

Contents lists available at ScienceDirect

Acta Psychologica
journal homepage: www.elsevier.com/locate/actpsy

Activating argumentation schema to write argumentatively and tactfully

Donghong Liu a, *, Qiong Gan b, Hui Ma a
a
Southeast University, Nanjing, China
b
Central China Normal University, Wuhan, China

A R T I C L E I N F O A B S T R A C T

Keywords: To argue tactfully is a goal in argumentative writing, which entails balanced argumentation schema. Although
CSCL computer-supported collaborative learning (CSCL) has been widely acknowledged as language learning media­
Argumentativeness tion, especially in writing, few studies investigate its effectiveness in activating the balanced argumentation
Tactfulness
schema. This study explores the effectiveness of QQ group discussion, a kind of CSCL mediation most popular in
Argumentation schema
Writing
China, in argumentative writing by means of quasi-experiment and interview. Fifty-six second-year college
students in an English Department participated in this study. The experimental group were asked to have a pre-
writing QQ group discussion on a disputable topic while the control group had an in-class face-to-face discussion
(a regular teaching method for English majors). Content analysis of the essays was made to investigate the use of
Counterargument elements, Qualifier and Concession. The results show that the two groups had no difference in
the use of Counterargument-claim and Rebuttal. However, the experimental group surpassed the control group in
Counterargument-data, Concession and Qualifier, which signifies more argumentativeness and tactfulness. In the
delayed post-test this group still performed better. The interview transcriptions were coded and analyzed by
inductive content analysis with the functions of QQ discussion as the themes. The result not only supported that
of the experiment, but also revealed why and how QQ mediation could help activate the balanced argumentation
schema. It is suggested that CSCL mediation should be promoted in argumentative writing instruction so that the
students could write argumentatively and tactfully.

1. Introduction et al., 2015; Casado-Ledesma et al., 2021) and similar effects have also
been found on argumentative essays (Felton, Garcia-Mila, et al., 2015;
Argumentation has “the potential to promote critical thinking and to Nussbaum, 2008). Despite that, restricted by the course requirement and
foster reflection, deep thinking, and the construction of conceptual syllabus, the focus of the present study is on persuasive writing to improve
knowledge” (Chin & Osborne, 2010: 231). Toulmin’s theory of argu­ argumentativeness and tactfulness. And the students’ inappropriate
mentation (called Toulmin model) is one of the most widely accepted manner to express refutation also motivated us to explore tactfulness in
theories and has been applied to writing instruction for decades. Based persuasive writing. For example, such expressions are frequent in the
on Toulmin model, Wolfe et al. (2009, 2018) advanced “argumentation students’ essays as “even three-year old kid know it”, “absolutely true”,
schema”. The balanced argumentation schema entails counterargument “bullshit” and “insane”.
and facilitates argumentativeness and tactfulness. The fact-based argu­ Dealing constructively with controversies is hard for students. A host
mentation schema just centers on the myside argument that consists of a of studies have investigated the effect of instructional aids in favor of
thesis and supporting evidence. In actual writing the balanced argu­ controversy resolution, such as dialogic discussion in pairs (Felton et al.,
mentation schema elicits either persuasive writing or deliberative writing 2022; Granado-Peinado et al., 2019; Nussbaum, 2008), small group
(Felton, Garcia-Mila, et al., 2015). In persuasive condition the writer can discussion (Li & Zhang, 2022; McDonough et al., 2018, 2019), explicit
argue with the opponents to persuade them while in deliberative condi­ socializing work (Jiang & Zhang, 2020). Some of the studies involve
tion the goal of the writer is to reach consensus. Previous studies have CSCL (computer-supported collaborative learning) as mediation. It has
shown that the argumentative dialogic activities with deliberative been found that collaborative argumentation improves the individual’s
versus persuasive goals promote argumentative skills (Felton, Crowell, competencies when preparing argumentative texts. Dialogue and group

* Corresponding author.
E-mail address: [email protected] (D. Liu).

https://doi.org/10.1016/j.actpsy.2024.104256
Received 17 December 2023; Received in revised form 11 March 2024; Accepted 8 April 2024
Available online 13 April 2024
0001-6918/© 2024 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).
D. Liu et al. Acta Psychologica 246 (2024) 104256

discussion can encourage reflection on individuals’ own thoughts and et al., 2011) and to promote self-regulation and problem-solving skills
modification of their own ideas (Kuhn & Crowell, 2011; Reznitskaya & (Yilmaz-Na & Sönmez, 2023). Only a few of these CSCL studies exam­
Wilkinson, 2017). However, few studies investigate such features as ined the participants’ argumentative essays. For example, Yamashita
argumentativeness and tactfulness in an essay. To fill the gap, our study (2021) explored corrective feedback in computer-mediated collabora­
applies CSCL mediation to the investigation of argumentativeness and tive writing and revision contributions. The experimental group
tactfulness. The specific tool is QQ, originally known as Tencent QQ, an received indirect corrective feedback on English articles while the con­
instant messaging software, can provide word and voice chat. It is a free trol group without corrective feedback. The results indicated a signifi­
software for communication and popular among Chinese people. QQ has cant relationship between the presence of corrective feedback and
a function of group discussion which enables group members to discuss accuracy improvement. Jiang and Zhang (2020) compared the learners
certain topics, issue notices, pictures or files. Despite the popularity of who received explicit socializing tasks in pre-writing WeChat talks with
QQ in schools and colleges, research on QQ as mediation in argumen­ those who received implicit tasks. They found that the former got higher
tative writing is still scanty. Just because of its popularity and conve­ scores in their argumentative essays, had lower levels of cognitive load
nience in communication, the efficacy of QQ discussion is worth and higher degree of social presence than the latter. These studies are
exploring for the sake of dissemination. mainly concerned with grammar and wholistic quality of essays. How­
The aim of this study is two-fold: to investigate (a) whether ever, it is still in the dark whether CSCL activities benefit the students in
encouraging students to discuss in QQ group leads to the better acti­ their actual written argumentation work in terms of the argumenta­
vating of argumentation schema (i.e., writing more argumentatively and tiveness and tactfulness.
tactfully); (b) how QQ group discussion as a kind of mediation performs
its function. To address this aim, a pre-test and post-test experimental 1.2. Argumentation schema, argumentativeness and tactfulness
design with a control group was adopted to compare and test the effect
of QQ group discussion, together with interviews. Since previous studies Toulmin model is an influential theory of argumentation. For Toul­
have reached consensus on the facilitating effect of collaborative argu­ min (1958, 2003), argumentation is composed of six elements. Claim is a
mentation, this study compares the effect of two kinds of collaborative conclusion and the final proposition in an argument. Data provides ev­
argumentation (face to face versus CSCL mediation). idence for Claim. Warrant certifies the claim as true and bridges the gap
Before presenting the study, the umbrella terms in this paper must be between Data and Claim. Backing supports Warrant and enables it to be
clarified. “Counterargument” which indicates argumentativeness in­ convincing enough. Rebuttal acknowledges certain conditions under
cludes Counterargument-claim, Counterargument-data and Rebuttal. which Claim does not hold. Rebuttal in fact suggests argumentativeness
“My-side argument” covers Claim, Data, Warrant and Backing. Tact­ which means attacking the opponents’ Data and Claim. Rebuttal and
fulness is embodied in Concession and Qualifier. The Toulmin elements Qualifier anticipate and refute the challenging questions raised by
involved in this study will be defined in Data collection and analysis. people of different opinions. Qualifier indicates the degree of certainty
in Claim. High degree of certainty denotes high verbal aggressiveness
1.1. CSCL activities as mediation in writing which is regarded as destructive and corrosive to relationships (Avtgis &
Rancer, 2004).
Mediation is an important concept in Sociocultural theory (Lantolf, According to Wolfe et al. (2009, 2018), it is an uncritical belief that
2000). Humans get to know and change the world by means of in­ argumentation is only a matter of presenting facts regardless of other
struments, including language which can mediate the relationship be­ aspects such as counterargument or alternative explanation. Counter­
tween humans and the outside world. Mediation includes three kinds: argument elements are necessary for “balanced argumentation schema”
object mediation, other mediation and self mediation (Lantolf & Thorne, since “a complete lack of other-side information is generally problematic
2009). In second language learning, other mediation involves teachers, and demonstrably weakens arguments” (Wolfe, 2012, p.478). The
peer students and the students themselves. The teachers and peer stu­ balanced argumentation schema acknowledges more than one side and
dents offer help explicitly or implicitly by collaborative dialogues, for provides evidence to support various assertions. In Toulmin model,
example pre-writing discussions or activities, which usually benefit the Rebuttal embodies balanced argumentation schema. Moreover, argu­
learners’ Zone of Proximal Development (ZPD). Vygosky (1978) defined mentation schema is “a learned, culturally derived set of expectations
ZPD as “the distance between the actual development level as deter­ and questions about argumentative texts” (Wolfe, 2012, p.480). Despite
mined by independent problem solving and the level of potential its stability in certain context and culture, argumentation schema is
development as determined through problem solving under adult prone to change after powerful learning experience. Both composing
guidance or in collaboration with more capable peers” (p.86). A and comprehending arguments are guided by argumentation schema.
considerable number of studies investigate the benefits of collaborative Once the argumentation schema is evoked, especially the balanced
pre-writing activities in non-CSCL context (Casado-Ledesma et al., 2021; argumentation schema, not only a minimal argument (a Claim and a
Granado-Peinado et al., 2019; Kibler, 2017; Li & Zhang, 2022; McDo­ Datum), but also other elements of the schema such as Warrant, Backing,
nough et al., 2018, 2019). Qualifier and Counterargument may be brought to mind despite the cost
Recent years witness an increase of CSCL studies on collaborative of increased cognitive burdens. In school-based writing, the evoking of
argumentation, with various kinds of softwares employed and facili­ argumentation schema must meet three conditions: demands of an
tating effects reported (Al-hamad et al., 2019; Dwigustini et al., 2021). assignment, expectations about the audience, and the goals of the author
Many CSCL studies have explored the effect of argumentation inter­ (Wolfe, 2012).
vention by measuring the oral argumentation quality or by examining The balanced argumentation schema is conducive to argumentative­
oral argumentation competence and comprehension (Canoz et al., 2022; ness in essay writing. Considerable work has been undertaken con­
Lu & Zhang, 2013; Newell et al., 2011, 2018; Noroozi et al., 2018; cerning argumentativeness over the past decade. Research suggests that
Wecker & Fischer, 2014). This CSCL-related argumentation among argumentativeness can reduce egocentric thinking and increase credi­
students, to be more accurate, debating has been used to elaborate the bility (Rancer, 1998; Wigley et al., 2023). Some studies have found that
content to be learnt, strengthen memory for pieces of information, the students in elementary school were less likely to consider Counter­
deepen understanding of complex ideas, improve inferential ability and argument than adults and accordingly showed weaker argumentative­
academic achievements (Wecker & Fischer, 2014). ness (Crammond, 1998; Leitão, 2003), since balanced argumentation
Collaborative argumentation in most of those studies is used as schema involving higher-order critical thinking is relatively complex in
mediation to foster the acquisition of domain-related knowledge or cognition and thus late-developed. Even at tertiary level it is still hard to
conceptual knowledge (Canoz et al., 2022; Mullen et al., 2019; Newell use this schema in L2 writing, for example Liu and Wan (2020), Li and

2
D. Liu et al. Acta Psychologica 246 (2024) 104256

Zhang (2022), and Qin and Karabacak (2010). (2) Why does (not) QQ group discussion (CSCL mediation) facilitate
More studies focus on the self-perception or belief of argumenta­ argumentativeness and tactfulness? How does it perform its
tiveness. Hsu (2007) investigated American and Taiwanese un­ function of mediation?
dergraduates, and found the Taiwanese students to be lower in
argumentativeness. The English and Chinese versions of the instrument As is believed by many CSCL researchers (Canoz et al., 2022; Noroozi
for Taiwanese respondents were compared and found highly correlated. et al., 2018; Wecker & Fischer, 2014), argumentation among students
However, other studies have different findings and cast some doubt on can promote the acquisition of domain-specific knowledge and under­
Chinese students’ preference for non-argumentative approach. Yeh and standing. Based on this theoretical assumption, we designed this quasi-
Chen (2004) using non-English questionnaire, compared the argu­ experiment (pre-test, post-test, delayed post-test) with a control group to
mentativeness of the students living in China’s mainland, Taiwan, and compare and test the effect of QQ group discussion on activating argu­
Hong Kong. They discovered that the students in mainland scored higher mentation schema. The domain-specific knowledge is argumentation
in argument-approach than those living in Taiwan or Hong Kong. Xie writing knowledge conveyed by the teacher in class, which is likely to
et al.’s (2015) study shows that the Chinese respondents were less form an argumentation schema in the students’ mind.
avoidant of confrontation, more motivated to engage in interpersonal
argumentation and more verbal aggressive than the American re­ 3. Methods
spondents. However, questionnaire can only reveal the respondents’
feel, desire, hope and the like instead of their actual behavior in writing. 3.1. Research context and participants
They also found that the Chinese students were more inclined to take
part in argumentations than the American students. “Compared to US The study was conducted in the English Department of a compre­
respondents, Chinese students had higher approach motivations, were hensive university in China. Altogether 64 students in two classes vol­
more prosocial in their intentions, were more cooperative, felt more unteered to participate in this study. They were in the fall semester of the
persecuted, and experienced more stress” (Xie et al., 2015: 273). Their second school year. In order to keep the participants relatively at the
results seem self-contradictory: on one hand, the Chinese students hoped same proficient level, their performance in“Comprehensive English 1B”, a
to participate in argumentation; on the other hand, they were more primary course for freshmen, was used as a parameter. There was not
sensitive to the interpersonal relationship. However, the results in fact much difference in the performance of those students probably because
expose a dilemma of the Chinese students. Solutions must be found how they were English majors. Finally, 28 students in each class were
to keep balance between the inclination for engagement and the hope selected by excluding those who scored over 90 and below 80 in the final
for harmony. examination of “Comprehensive English 1B”. Although “Comprehensive
Qualifier in Toulmin model indicates a tactful way of persuasion and English” is considered as a course best revealing the students’ profi­
hence contributes to the success of argumentation. However, the ciency, there was some potential limitation of using a single course
research on the tactfulness of expressing claims and refuting opposite performance as a proficiency indicator.
views is rather scarce. Berger (2014) divided Qualifier into three types: The participants were from different provinces of the middle part of
Qualifier in word form, Qualifier in participle form, and Qualifier in China and had already taken eleven courses in the first school
sentence form. Yang (2022) investigated the relationship between Claim year—Comprehensive English 1A, Comprehensive English 1B, English
and writing quality. Based on three types of Qualifiers, Claims were Reading Skills A, English Reading Skills B, English Listening and Speaking
classified. The result showed that the Claim types did not predict the 1A, English Listening and Speaking 1B, Profiles of English Countries, Pro­
holistic writing quality of argumentative essays. However, the division nunciation, English Grammar, Ancient Roman and Greek Myths, and English
of Qualifier goes against Toulmin’s definition. More importantly, tact­ Writing A. Many courses required the students to read several English
fulness is not only revealed in the use of Qualifier but also in expressing articles every week and submit more than one English composition or
refutation. How to express disagreement or rejection decides whether essay in a semester. For example, Comprehensive English courses asked
the argument and Claim can be accepted by the audience. Aggressive the students to write at least three compositions a semester; and English
refutation and claim are likely to turn the argumentation into a quarrel. Writing courses required them to keep a weekly English journal besides
completing writing tasks assigned in class. Moreover, they were taking
2. Research questions and assumptions English Writing B and had already learned exposition (i.e., advancing a
claim and supporting the claim with evidence) when they participated in
Wolfe et al. (2009) believe that argumentation schema is learned and this study. In general, these participants had basic English writing ability
culturally derived, and that argumentation schema can be evoked by and exposition writing knowledge.
demands of an assignment, expectations about the audience, and the The experimental group included 26 females and 2 males; and the
goals of the author. The Chinese college students have already acquired control group consisted of 27 females and 1 male. The participants’ age
Chinese writing conventions and formed L1 essay writing schema. When ranged from 19 to 20. Gender and age factors might influence the
writing Chinese essays the writers are required to provide sufficient experimental results. However, these factors were controlled by select­
arguments to achieve persuasiveness. That is, “My-side argument” is ing the similar number of males and females of the similar age for the
emphasized while “Counterargument” is neglected. However, the stu­ experimental and control groups. Before the experiment, their exposi­
dents may form a new schema (i.e., balanced argumentation schema) tory essays were assessed. The mean score of the experimental group
when they learn English argumentation, as studied by Wolfe et al. was 85.93 and that of the contrast group 86.50. The two sets of scores
(2018) and Liu (2020). How to activate the new schema? If the demand were submitted to t-test but the result showed no significant difference
of an assignment and expectation about the audience are given or fixed, (t = 0.497, p = 0.621). Thus, the two groups were at the same profi­
can the use of CSCL activity (i.e. the goal of the author) effectively help ciency level in English writing. This result also supports the reason­
the students evoke the schema to write more argumentatively? These ableness in the selection of samples based on the scores of Comprehensive
should be under discussion in L2 writing instruction. We accordingly English 1B.
raise our research questions:
3.2. Experiment
(1) Would encouraging college students to discuss in QQ group
(CSCL mediation) facilitate writing more argumentatively and A quasi-experiment was designed since several factors could not be
tactfully? controlled such as group dynamics and individual differences although
most of the factors could be controlled. Both experimental and control

3
D. Liu et al. Acta Psychologica 246 (2024) 104256

groups had the same tasks: collaborative argumentation and argumen­ were asked to express their standpoints on the 4 topics (in favor of,
tative essays. Both groups were required to apply what they had learnt in against or neutral). Roughly equal number of participants were in favor
class. The only difference was the context (virtual vs face-to-face dis­ of and against the view on each topic in each group. Thus, discussion
cussion), with all the other conditions the same for the two groups. groups were formed including both sides of the view.
All the participants had already learned how to write topic sentences When QQ group discussion was implemented for the third time, the
and thesis statements, and how to develop an expository essay. To put it experimental group was asked to write an argumentative essay. The
another way, they had already formed fact-based argumentation schema third disputable topic was a social one:
(or expository schema). According to Wolfe (2012), argumentation
“In the future, there will be a higher proportion of older people than
schema can be changed after powerful learning experience, the partic­
young people in many countries. Is it a positive or negative
ipants learned balanced argumentation schema in class explicitly and
development?”
practiced by doing exercises. They were taught in class to address and
refute an opposing side to increase the persuasiveness of their essay. The experimental group were divided into six groups to have a dis­
Concession was included in the teaching syllabus of argumentation in cussion by QQ without time limit. Each group was made up of four to
this quasi-experiment because it can moderate the rebutting language, five people. The participants were asked to submit a 400-word argu­
enable the opponent to accept the writer’s view and hence make the mentative essay two days later. The writing assignment is displayed in
essay more persuasive. Concession is also an important part of Reser­ the Appendix 1.
vation, a newly defined Toulmin element by Crammond (1998). To be The same teacher also explained the same argumentation writing
specific, the knowledge of writing argumentation that the teacher knowledge to the control group and asked them to submit a 400-word
conveyed in her syllabus included the following elements: argumentative essay two days later. The difference was the face-to-
face discussion in class. They were given 40 min in class to have a dis­
Examining opposing views (Counterargument-claim) cussion on the same topic in a 4-people group (A larger group would
Exploring their reasons (Counterargument-data) make some shy people silent while it is likely to have just one opinion in
Granting the merits of opposing views (Concession) a smaller group or pair work). However, the participants were told to
Refuting opposing views or their reasons (Rebuttal) continue their face-to-face discussion after class if they did not finish it.
Paying attention to the manner when expressing the thesis state­ But all of the groups finished their discussion in class.
ments and conclusion (Qualifier) One week later, the teacher commented on the essays in class. Two
weeks later she asked the two groups to write in 50 min in class another
The participants were asked to do exercises to identify and practice argumentative essay on the topic of “animals in captivity” (see Appendix
using the elements in class after the teacher’s lectures. The purpose was 1). It was a quiz and also served as a delayed post-test in this study. This
to reinforce the declarative knowledge and hence facilitate the forma­ was a social topic less familiar to the college students to align with the
tion of new schema. Take Concession for example. The teacher first one in the experiment.
explained why and how a writer should grant the merits of opposing
views, introduced some useful expressions, and then asked the students 3.3. Interview
to appreciate some samples (i.e., to point out the places of Concession
and comment on them). Interviews were used to complement the results of content analysis in
For Wolfe et al. (2009), three conditions can evoke argumentation the quais-experiment and also to get a better understanding of how pre-
schema: demands of an assignment, expectations about the audience, writing QQ group discussion exerted mediation in argumentative
and the goals of the author. In the present research design, the experi­ writing. Retrospective semi-structured interviews were conducted
mental group and the control group had the same demand of an immediately after the submission of the first essay. Three participants in
assignment and expectation about the audience—the same teacher. The the experimental group volunteered to take the interview which was
difference lies in the goals of the author. The two groups had different conducted in Chinese to ensure that the interviewees could fully express
kinds of pre-writing discussion—QQ group discussion and face-to-face their ideas and opinions. Each interview included the following aspects:
discussion, which might influence their goals and accordingly affect (1) views of the pre-writing QQ group discussion; (2) benefits from the
their argumentation schema evoking. QQ discussion; (3) inadequacies and suggestions for writing instruction.
As is displayed in Table 1, the experimental group was asked to The questions were not predetermined. During the conversation the
discuss in QQ group twice on two disputable topics which were closely interviewees were encouraged to bring up any idea pertaining to the
related to their family life and campus life. Six topics were respectively three aspects.
offered to the two groups to select 4 interesting ones which were used in
this experiment. The two groups showed similar interest. Moreover, they

Table 1
Procedure of the quasi-experiment.
Time Work Specificity

Pre- Participants selection The experimental group and the control group each consisted of 28 participants who were at the same language proficiency level and
experiment & Pre-test writing level
Week 1 Discussion 1 Topic: whether credit cards will bring harm to college students
Experimental group: QQ group discussion
Control group: face-to-face discussion
Week 2 Discussion 2 Topic: whether silence is golden in modern society
Experimental group: QQ group discussion
Control group: face-to-face discussion
Week 3 Discussion 3 and writing Experimental group: after-class QQ group discussion, writing an argumentative essay
Control group: in-class face-to-face group discussion, writing an argumentative essay
Interviews 3 interviewees from the experimental group
Week 4 commenting on the For both groups
essays
Week 5 a delayed post-test Both groups wrote an argumentative essay in class

4
D. Liu et al. Acta Psychologica 246 (2024) 104256

3.4. Data collection and analysis Table 3

Rubric for coding qualifier.
This study focuses on argumentativeness and discretion which are Points Ranks Description
embodied in Counterargument, Qualifier and Concession.
1 Very first acknowledge the opposite view and then claim or
Content analysis was employed to collect data from the essays. The weak reclaim one’s own view and in the meantime use some words
elements identified in this study are as follows: Counterargument-claim, to moderate the tone
Counterargument-data, Refutation, Concession and Qualifier. The rubric 2 Weak Use the words or phrases to express uncertainty such as
for coding is displayed in Table 2. maybe, likely, probably, might, generally speaking; or first
acknowledge the opposite view and then claim or reclaim
Apart from the definitions, we also had linguistic, semantic and one’s own view.
logical considerations when doing identification. For example, to iden­ 3 Neutral Express one’s view in a flat way without using any emphatic
tify counterargument elements some suggestive indicator words and words or moderate words
phrases are conducive, such as “however”, “although”, “despite”, “but”, 4 Strong Make the author prominent in the discourse by emphasizing
his/her discourse responsibility. Such expressions are usually
“even though”.
used as in my opinion/perspective, I hold the view.
A striking difference from the previous studies such as Yang (2022) 5 Very Use some emphatic words, absolutes and superlatives, such as
and Berger (2014), is that Qualifier was assigned five ranks from very strong never, firmly, definitely, very, always, totally, disaster, miserable,
strong to very weak (see Table 3). Qualifier can reveal people’s tactful ultimate.
claiming in their written discourse. The lower the score, the more un­
certainty the claim seems, and the more tactful the writer is. We
only to the essays in the post-test and delayed post-test. Descriptive
examined Qualifiers in their Claims which usually appeared in the
statistics analysis and Mann-Whitney U test were used to analyze the
beginning and end of an essay.
data and to compare the two groups. Mann-Whitney U test (independent
To ensure the validity of the identification of these Toulmin ele­
samples) is used to determine whether two groups are significantly
ments, two of the authors of this paper first discussed the definitions and
different from each other on a certain variable. It can be used in such
typical examples to get familiar with these elements, and then coded a
conditions: first, the sample size is small (<30); second, the normal
sample respectively. The two coders discussed the discrepancies to
distribution is not required; third, the data are continuous instead of
clarify the interpretation of the coding rubric. After reaching agreement,
categorical. This study met the three conditions. Moreover, Mann-
the two coders coded independently the rest compositions—roughly 98
Whitney U test enabled us to do comparison to answer our major
% of the essays. The inter-coder consistency of each element is as fol­
research question: whether QQ group discussion has facilitating effect
lows: Counterargument-claim 98 %, Counterargument-data 94 %,
on argumentativeness and tactfulness.
Rebuttal 96 %, Concession 96 % and Qualifier 88 %. Every discrepancy
The interviews were first recorded and then transcribed. But fillers,
was checked and discussed by the coders until total agreement was
pauses and vocatives were not transcribed precisely because the tran­
reached. Take the example in “Results and discussion” for instance. One
scriptions were used to examine the effect of QQ group discussion rather
of the coders did not find any Rebuttal just because she only noticed
than making conversation analysis. After that, the transcriptions were
“Someone thinks that the old people can take care of baby to liberate the
coded thematically according to categories (e.g. functions, benefits) and
young ladies to go back to their work … In my opinion … I think…” She
sub-categories (e.g. convenience, freedom) in a bottom-up way. For
considered “In my opinion … I think…” as Claim. During the discussion
example, first specific information was found and labeled in each sen­
with the other coder, she realized that the sentences in fact refuted the
tence. When more related items were found, they were compared and
first one although they began with in my opinion and I think, the typical
abstracted, and finally subsumed into a sub-category such as conve­
signifiers of Claim.
nience, freedom or enriching ideas. When more sub-categories were
Since the first two discussions were just discussion without gener­
found, they were compared and subsumed into a category like function
ating any essays, inter-group and intra-group comparisons were done
or benefit. The transcription was double coded by the authors to ensure
inter-coder consistency. The coding procedure was similar to that of the
Table 2 essays. All the differences were resolved via discussion.
Working definitions of Toulmin elements.
Elements Definition Examples 4. Results and discussion
Counterargument- a possible opposing view that Some people contend that
claim can challenge the validity of newspaper remains a typical 4.1. Argumentative essays
Claim media for people to get
information. As can be seen in Table 4, the two groups showed no significant
Counterargument- statistics, facts, personal There are still some people difference in Counterargument-claim and Rebuttal. Both groups could
data experience and so on in who rely on newspapers to
support of Counterargument- get information and to
consider opposite views and refute them. Each essay could address and
claim entertain themselves. They refute more than one opposite view on average. They seemed to have
are used to such enjoyment
of fetching the daily
newspaper, reading it in a Table 4
comfortable chair over a cup Toulmin elements used by the two groups in the 1st essays.
of tea.
Toulmin elements Experimental group Control group Z value
Refutation a statement in which the writer [Concession] Admittedly,
(28) (28)
responds to Counterargument- this view makes a lot of
claim or Counterargument- sense. [Rebuttal] However, Mean S.D. Mean S.D.
data by pointing out the we cannot deny such a fact
possible weakness that an increasing number of Counterargument-claim 1.75 0.75 1.71 0.90 0.46
Concession acknowledging young people, even middle- Counterargument-data 1.61 0.50 1.18 0.82 1.97*
Counterargument-claim or aged people get themselves Rebuttal 1.21 0.88 1.39 0.96 0.61
Counterargument-data informed by the Internet, Qualifier 2.75 0.93 3.71 0.76 3.98***
smart phones rather than Concession 1.36 0.62 0.79 0.79 3.03**
daily newspaper. *
p < 0.05.
Qualifier words or phrases expressing probably, possible, might **
p < 0.01.
possibility in a Claim ***
p < 0.001.

5
D. Liu et al. Acta Psychologica 246 (2024) 104256

mastered such fundamental elements of argumentation as referring to Table 6

opposite views and refuting them. However, the experimental group Toulmin Elements Used by the experimental group in the two essays.
used more Counterargument-data and Concessions. More Toulmin elements 1st essay (28) 2nd essay (28) Z value
Counterargument-data indicate that the writers could place themselves
Mean S.D. Mean S.D.
in the opponents’ position and explore why they think so. It can be seen
that the experimental group override the control group in critical Counterargument-claim 1.75 0.75 1.14 0.45 3.04*
Counterargument-data 1.61 0.50 1.21 0.50 2.40*
thinking and hence in argumentativeness. Furthermore, the experi­ Rebuttal 1.21 0.88 1.29 0.71 0.53
mental group tended to acknowledge the reasonable aspects of the Qualifier 2.75 0.93 2.79 0.79 0.18
opposite views or their supporting evidence before refuting them. Concession 1.36 0.62 0.75 0.75 2.99**
Moreover, the experimental group scored significantly lower than the *
p < 0.05.
control group in Qualifier, which suggests less certainty and more **
p < 0.01.
moderate tone in expressing Claims. Higher frequency of concession,
together with moderate tone revealed in Qualifier, demonstrates that the
experimental group were better at considering the persuasiveness and Table 7
acceptability of their own views and thus writing more tactfully. Toulmin elements used by the control group in the two essays.
As for the control group, they paid more attention to Toulmin elements 1st essay (28) 2nd essay (28) Z value
Counterargument-claim and Rebuttal than to Counterargument-data.
Mean S.D. Mean S.D.
Take the following paragraph for example. The writer only put for­
Counterargument-claim 1.71 0.90 1.39 0.69 1.62
ward the Counterargument-claim without explaining why or offering
Counterargument-data 1.18 0.82 0.89 0.57 1.71
any evidence (i.e. Counterargument-data). The Rebuttal mentions three Rebuttal 1.39 0.96 1.14 0.65 1.50
aspects: the old people “don’t like” and “don’t have the obligation” to look Qualifier 3.71 0.76 3.11 0.31 2.95**
after their grandchildren, and even would “burden young people”. In fact, Concession 0.79 0.79 0.71 0.81 0.41
in the whole essay the writer did not offer enough Subclaims or Data to **
p < 0.01.
support them, but only concentrated on putting forward a series of
Counterargument-claims and refuting them one by one. This example
group, diminish in use occurred in every Toulmin element whereas the
illustrates the deficiency of the control group in evoking the argumen­
frequency of Qualifier displayed significant decline. The descending
tation schema.
tendency in the two groups revealed the impact of time restriction.
[COUNTERARGUMENT-CLAIM] Someone thinks that the old However, this fact cannot be covered: although time factor exerted in­
people can take care of baby to liberate the young ladies to go back to fluence on the argumentativeness and tactfulness in both groups, and
their work. [REBUTTAL] In my opinion, maybe some old people even more greatly to the experimental group, the effect of the experi­
like taking care of babies, but there are a lot of old people don’t like ment still lingered, illustrated by the experimental group’s higher fre­
those sorts of things. After all, they don’t have the obligation to do quencies of Counterargument-data and Qualifier in Table 5.
that. And I think it will burden young people to take care of old
people and babies simultaneously.
4.2. Interviews
The results of comparison in the second essays are displayed in
Table 5. The two groups still show significant differences in The interviews were designed to answer the second research ques­
Counterargument-data and Qualifier except Concession. Although the tion. The results are displayed according to the three aspects: opinions to
second essay was a timed writing and implemented two weeks after the the pre-writing QQ group discussion, benefits from such discussion, in­
first one, the experimental group could still use their argumentation adequacies and suggestions.
schema better than the control group. It might be a positive “aftereffect”.
However, the experimental group used fewer Concession this time than A. Why QQ group discussion could activate argumentation schema
in their first essays. It is speculated that the time factor causes every effectively
schema to compete fiercely to be processed in the working memory and
as a result the less powerful schema is usually eliminated. Concession is a. Convenient for recording and reading
not the indispensable element in argumentation although it reveals a
good manner and conducive to persuasion. Therefore, with time re­ The interviewees reported that QQ discussion was convenient for
striction, the less competitive Concession was used less frequently so as them to read and check what they had said just because when discussing
to guarantee the use of major elements. they typed their words and the words were recorded automatically. Even
Intra-group comparisons were also made to the two groups, as is if they sent vocal messages to the group, they could be saved as well.
shown in Table 6 and Table 7. The frequencies of Counterargument- They considered this function better than face-to-face discussion.
claim, Counterargument-data and Concession in the experimental
1st interviewee: “The chat saved in QQ provided us with reference when
group decreased significantly in the delayed post-test. For the control
we rearranged the text later. It is more convenient.”
2nd interviewee: “Exactly. To discuss in QQ group is easy and efficient.
Table 5 Everyone in our group was active. In addition, I could do some research
Toulmin elements used by the two groups in the 2nd essays. online and express my views by using my phone.”
Toulmin elements Experimental group Control group Z value
(28) (28)
3rd interviewee: “It is convenient to record what we discussed in QQ
group. In our previous face-to-face discussion, we did not record what we
Mean S.D. Mean S.D.
had talked about until halfway in the discussion. But by QQ, we can write
Counterargument-claim 1.14 0.45 1.39 0.69 1.47 down each point. I mean, we don’t have to recall the points that have been
Counterargument-data 1.21 0.50 0.89 0.57 2.16*
mentioned before. That might waste a lot of time.”
Rebuttal 1.29 0.71 1.14 0.65 0.87
Qualifier 2.79 0.79 3.11 0.31 2.29*
Concession 0.75 0.75 0.71 0.81 0.30
*
p < 0.05. b. Freedom from the restriction of time and space

6
D. Liu et al. Acta Psychologica 246 (2024) 104256

The first and third interviewees mentioned that QQ group discussion 2nd interviewee: “For example, if I had something to say on a point is­
could free them from time and space. They had more time to think, to sued by other members, I could directly ‘quote’ it (QQ’s built-in quotation
arrange his thought and organize his language. Whenever some ideas function). It is very clear, and I like it.”
popped up in their mind they could deliver them to the discussion group
3rd interviewee: “As such, I could have a clear understanding of the
at any time, without waiting for other people to stop or worrying about
whole discussion framework and logic. This is in part because we had
being interrupted. Compared with face-to-face discussion, QQ discus­
more time for review after we typed in our words.”
sion did not impel them to find some time available to everybody to sit
together. Moreover, there was no embarrassment of failure to refute due
to blank mind.
b. Enriching my-side argument
1st interviewee: “In QQ discussion, I had more time to organize my own
language. I mean, more time for thinking. Then I typed into QQ. I think
By reading, rereading and understanding other people’s views and
this is a process to rearrange my own ideas. But face-to-face discussion is
ideas, the interviewees generated new ideas to support their own views.
different. When refuting other views, I usually failed to speak at the
Even in the process of refutation, new ideas emerged in their mind. As
moment. Instead, in QQ I was allowed for a relatively long time to
the third interviewee put it, just because they had more time to read and
conceive ideas and clarify the logic.”
think, they had more evidence to defend their own views and refute the
3nd interviewee: “QQ discussion is not limited by time and space. opposite side.
Traditionally, face-to-face discussion requires everyone to sit together and
1st interviewee: “Maybe when I made my own argument, I would get
discuss something in the limited time. So, sometimes a flash of inspirations
some points. Also, I could have some new ones when refuting others’
we have after the discussion cannot be shared instantly. However, we
views.”
could send our ideas to the QQ group at any time even when the discussion
was over without interrupting the others.” 2nd interviewee: “I could review others’ arguments as well as my own in
QQ, so that I could think over and over again, for example, whether I had
missed something from others or what new points I could put forth based
on others’ views. That helped me have a better understanding of one’s
c. Avoiding direct confrontation
statement through an in-depth analysis of the opinion and the logic behind
it. As such, my refutation would be more effective.”
All of the interviewees agreed that QQ discussion could make the
participants avoid direct confrontation so that they had more freedom to 3rd interviewee: “If it takes us long time to understand a point of view,
express ideas fully. They did not have to consider facework and even that means we think about it for a long while, which perhaps stimulates
possible hostility caused by disagreement. By QQ discussion, embar­ other inspirations.”
rassment of divergence and even friction can be decreased. They can
focus on the clarity of idea expressing.
1st interviewee: “In QQ group chatting, we have more freedom to express c. Promoting cognitive engagement
our opinions. In other words, we would like to discuss with each other
despite the disunity of ideas that could lead to friction.” By cognitive engagement, it is meant sustained attention, mental
st effort and self-regulation (Helme & Clarke, 2001). All of the in­
2 interviewee: “I think, when communicating with each other face to
terviewees reported that their group members were active in the dis­
face, we may care about others’ feelings and speak in a euphemistic way.
cussion like them. The first and second interviewees provided more
Hence, the discussion will be less intense.”
details: they took the role of Cons and were actively attending to the
3st interviewee: “As for expressing views, we could say what we think. Prons’ ideas and refuting them. The other students in the groups were
For example, in face-to-face talking, sometimes we may feel embarrassed also highly cognitively engaged since they “incessantly” read and
to refute one’s view, especially when the others all agree with it.” rebutted the Cons. The active cognitive engagement is the practice of
critical thinking and argumentativeness in writing.
1st interviewee: “In the discussion that day, I was the only one to hold a
B. How QQ group discussion helped to activate argumentation schema negative view, and what I needed was to constantly refute others’ views
from different aspects. Thanks to the special function of QQ grouping, I
The benefits that the interviewees profited from the QQ discussion think my refutation was targeted.”
corroborate the results of content analysis of the experimental group.
2nd interviewee: “A girl and I were against a viewpoint, but others might
support it. They focused on what points we had posted in QQ, and con­
a. Fully understanding peers’ thoughts
tradicted us step by step in the whole discussion.”
As the first interviewee said, they realized that people had a variety 3rd interviewee: “I think many students played an active role in the QQ
of thoughts and views on the same social occurrence. This is the first step discussion, like expressing their minds in the group as was evidenced by
of critical thinking. They could know other people’s ideas clearly and our last online discussion.”
even grasp the whole framework of the group discussion just because
they could read the ideas for a longer time owing to the special function
of QQ group discussion mentioned above. That might contribute to their
C. Inadequacies and suggestions
writing more argumentatively and in a more polite manner.
1st interviewee: “I learned about our team members’ ideas in a visual QQ group discussion can be improved. The interviewees pointed out
way. Also, I found that their thoughts diverged in all directions as a saying the inadequacies and put forward suggestions. They thought some group
goes that ‘A thousand readers make a thousand Hamlets’.” members were not active enough and they expected more contribution
of ideas and more interactions in the group. Even typing was not
considered as an efficient way and Tencent conference was suggested.
However, Tencent conference is like face-to-face talk which debilitates

7
D. Liu et al. Acta Psychologica 246 (2024) 104256

the students to challenge and refute. Despite the imperfectness of their moderate their tone by means of Qualifier and Concession.
suggestions, the interviewees revealed their positive attitude and their Finally, QQ mediation facilitates the activation of balanced argu­
eagerness to improve it. mentation schema. Researchers attribute the scarcity of Counterargument
in the students’ essays to the failure of identifying alternative views
1st interviewee: “We need to encourage more group members to talk.
because of overwhelming cognitive load (Coirier et al., 1999; Li &
Indeed, only a few were very actively involved in the discussion. And as
Zhang, 2022), or lacking awareness of the role of Counterargument in
for me, I like to communicate with and listen to others. That is why I hope
argumentation enhancement (Nussbaum & Kardash, 2005). Cognitive
more students can join us.”
load and the lack of awareness might result from inactive argumentation
2nd interviewee: “We may send vocal messages instead of typing, or turn schemas. Wolfe et al. (2009) maintain that a possible cause of poor
to Tencent Conference.” argumentative writing is “that the writer has a deficient argument
schema… if one has a minimalist argument schema that has slots for
3rd interviewee: “The disadvantage is that typing takes some time. We
only a claim supported by a single reason, then the important sub-goals
need to set time limit and improve efficiency.”
of including backing for reasons and rebutting other-side information
Generally speaking, the interviewees thought that the advantages of will be absent” (p.184–185). But our study cannot totally support this
QQ group discussion overshadow its weakness. The result not only view since both groups cpuld address Counterargument-claims and
corroborates that of the content analysis, but also explains why QQ rebut them. The generation of Counterargument-data and the manner of
group discussion facilitated writing argumentatively and tactfully, and rebuttal make a difference. It is proved in the study that if the argu­
how it performed the functions. mentation schema is not strong enough it cannot be evoked by the face-
to-face discussion in class. But QQ discussion can evoke and strengthen
5. General discussion the schema.
The control group produced fewer Counterargument elements than
The research results have displayed that the experimental group did the experimental group in the first post-test essays. The insufficiency of
better in Counterargument-data, Concession and Qualifier. QQ group the face-to-face discussion might also be a cause. Although their group
discussion can mediate pre-writing thinking and hence enhance argu­ discussion could be continued after class, no group did so and even some
mentativeness and tactfulness. The reasons are to be discussed in this groups stopped the discussion before the class was over. Just as the in­
section. terviewees said, the students were likely to avoid discussion because of
Chin and Osborne (2010) contend that epistemic activities such as “the disunity of ideas that could lead to friction”, “care about others’ feelings
discussion are powerful mechanisms by which students can not only and speak in a euphemistic way”, or “feel embarrassed to refute one’s view”.
collaboratively construct new meanings, but also be stimulated to have As a consequence, their thought is restricted to one side, failing to know
some extended cognitive engagement. While students are reasoning fully alternative views and supporting evidence. However, their second
about the advantages and disadvantages, Pros and Cons, causes and essays reveal much progress owing to the explicit input. The teacher’s
consequences of alternative views, they are actually exposed to a greater comments on the first essay in class is explicit input as she emphasized
variety of ideas than thinking and writing alone. QQ group discussion what was a good argumentative essay with their sample essays. Teach­
can mediate the students’ thinking better than face-to-face discussion er’s explicit input may strengthen the argumentation schema in the
owing to its unique technical functions. students’ mind. Nevertheless, the effect of explicit input is still inferior
Firstly, the effective mediation is owed to the convenience of QQ to the practice in QQ discussion that can promote the formation of
group discussion, which allows more freedom in time and space. Using balanced argumentation schema, evoke and reinforce it. The second essay
mobile phones or other mobile devices they could look up for some of the experimental group serve as the proof.
references or sources to enrich the my-side argument and refute the In addition, this study reveals that even at modern times the desire
opposing views effectively. Our study corroborates Al-Hamad et al. for harmony still characterizes Chinese people and continues influ­
(2019) and Chen et al. (2019). Moreover, as Burston (2015) puts it, encing their communication and arguing behaviors. It cannot be
mobile assisted language learning works smoothly in this digital era due concluded solely by means of questionnaires that the Chinese students
to the young people’s new life style and the demand for fast and effective are more argumentative and verbal aggressive. Actual communication,
learning solutions. both verbal and written, exposes the fact. Despite the prevailing value of
Secondly, QQ discussion as socializing activity downgrades the harmony and face, it is still likely for the teachers to take some effective
embarrassment which occurs frequently in social communication. The measures to help the students be argumentative with appropriate
students can express their ideas more bravely even to those who hold manners (e.g. using more Concessions and moderate Qualifiers).
different views. They engage in the discussion cognitively and per­ The research results have implication for teaching. QQ discussion as
formed actively due to human’s competitive nature and the lower de­ a kind of effective CSCL mediation should be used frequently in argu­
gree of apprehension in social communication as can be seen in the mentative writing instruction to reinforce the schema and intensify
interviews. This study has solved the problem found by Xie et al. (2015), practice, as the students in the delayed post-test still had some trace of
i.e., the contradiction between hope for participation in argumentation the balanced argumentation schema in their mind despite the attrition in
and hope for harmony. QQ group discussion can avoid the negative ef­ memory and time restriction in writing. In addition, as the interviewees
fects such as being “persecuted” and stressful, and exert positive effects suggested, QQ group discussion can be implemented in a more efficient
like being “prosocial” and “cooperative”. way. A more heated discussion will prompt the students to think more
Thirdly, owing to the function of recording in QQ discussion, the actively. Some students tend to be silent and only take in other people’s
students can scroll up and down to check their own words and their peer ideas. In view of that, roles can be assigned to those students who are
students’ ideas. They can fully understand their peers’ ideas, especially silent and passive in the group so as to motivate them to express what
alternative ideas, and finally elicit more thinking and reflection. This they think and to benefit the other team members. To achieve the goal,
result corroborates Jiang and Zhang (2020) who contend that the more a teachers should also adjust their roles. They are not only organizers and
learner borrows relevant knowledge from the peers in discussion ac­ coordinators in the discussion, but also mediators who arouse the in­
tivities, the higher opportunity he/she would solve the similar problem. terest of the students to participate in the discussion and encourage them
Moreover, just because the students understand the Counterargument to express ideas, challenge and refute different views in an appropriate
well, they make claims or draw their conclusions in a cautious way, way.
avoiding aggressive and arrogant manners. All of that explains why the
experimental group could produce more Counterargument elements and

8
D. Liu et al. Acta Psychologica 246 (2024) 104256

6. Conclusion 1. Introduce the topic and present your position in the introductory first
paragraph.
By conducting a quasi-experiment, we have found that the experi­ 2. Include both FOR and AGAINST to develop the supporting
mental group performed better than the control group in paragraphs.
Counterargument-data, Concession and Qualifier. They used a measured 3. Finally, restate your position in the concluding paragraph.
tone to refute and claim. In the delayed post-test, the experimental
group still showed higher frequencies in Counterargument-data and
Qualifier. QQ mediation can engage the students in the discussion in References
flexible time and place, and allow them enough time to take in and
digest information, and even think critically. The in-depth involvement Al-Hamad, R. F., Al-Jamal, D. A. H., & Bataineh, R. F. (2019). The effect of Mall
instruction on teens’ writing performance. Digital Education Review, 35, 289–298.
and engagement help evoke and reinforce the balanced argumentation https://doi.org/10.1344/der.2019.35.289-298
schema. As a result, the experimental group could write more argu­ Avtgis, T. A., & Rancer, A. S. (2004). Personalization of conflict across cultures: A
mentatively and tactfully. comparison among the United States, New Zealand, and Australia. Journal of
Intercultural Communication Research, 33, 109–118.
CSCL mediation has been proved effective to activate and reinforce Berger, R. E. (2014). A scientific approach to writing for engineers and scientists. Wiley.
the balanced argumentation schema. In order to help the students write Burston, J. (2015). Twenty years of MALL project implementation: A meta-analysis of
argumentatively and refute tactfully, teachers are encouraged to use learning outcomes. ReCALL, 27(1), 4–20. https://doi.org/10.1017/
S0958344014000159
CSCL mediation in writing instruction. They can act as mediators as well Canoz, G. M., Ucar, S., & Demircioglu, T. (2022). Investigate the effect of argumentation-
as organizers and coordinators in the discussion to spur the students to promoted interactive simulation applications on students’ argumentation levels,
express ideas, challenge and refute different views tactfully. academic achievements, and entrepreneurship skills in science classes. Thinking Skills
and Creativity, 45, 101–106.
This study is a preliminary one aimed at exploring how and why QQ
Casado-Ledesma, L., Cuevas, I., Van den Bergh, H., Rijlaarsdam, G., Mateos, M.,
mediation facilitates argumentativeness and tactfulness. It cannot reach Granado-Peinado, M., & Martín, E. (2021). Teaching argumentative synthesis
every aspect of this topic. The major inadequacy of this study lies in the writing through deliberative dialogues: Instructional practices in secondary
method. It was the written products of the participants that was education. Instructional Science, 49(4), 515–559. https://doi.org/10.1007/s11251-
021-09548-3
compared. Accordingly, group dynamics and individual differences of Chen, C., Liu, H., & Huang, H. (2019). Effects of a mobile game-based English vocabulary
the participants such as motivation, belief, personality, willingness of learning app on learners’ perceptions and learning performance: A case study of
communication and learning style were neglected. Further studies will Taiwanese EFL learners. ReCALL, 31(2), 170–188. https://doi.org/10.1017/
S0958344018000228
be done to investigate the individual differences in QQ discussion. Chin, C., & Osborne, J. (2010). Supporting argumentation through Students’ questions:
Case studies in science classrooms. The Journal of the Learning Sciences, 19(2),
Funding 230–284. https://doi.org/10.1080/10508400903530036
Coirier, P., Andriessen, J., & Chanquoy, L. (1999). From planning to translating: The
specificity of argumentative writing. In J. Andriessen, & P. Coirier (Eds.), Foundation
This work was supported by the National Social Science Funds (Grant of Argumentative Text Processing (pp. 1–28). Amsterdam: Amsterdam University
No. 21FYYB 016). Press.
Crammond, J. G. (1998). The uses and complexity of argument structures in expert and
student persuasive writing. Written Communication, 15, 230–268. https://doi.org/
CRediT authorship contribution statement 10.1177/0741088398015002004
Dwigustini, R., Sari, N., & Susilawati., and Nisa, B.. (2021). Fostering students’ writing
skills by the integration of Mall application. JOLLT Journal of Languages and Language
Donghong Liu: Writing – review & editing, Writing – original draft,
Teaching, 9(1), 34–43. https://doi.org/10.33394/jollt.v%vi%i.3264
Resources, Project administration, Methodology, Funding acquisition, Felton, M., Crowell, A., Garcia-Mila, M., & Villarroel, C. (2022). Capturing deliberative
Formal analysis, Conceptualization. Qiong Gan: Investigation. Hui Ma: argument: An analytic coding scheme for studying argumentative dialogue and its
Investigation. benefits for learning. Learning, Culture and Social Interaction. https://doi.org/
10.1016/j.lcsi.2019.100350
Felton, M., Crowell, A., & Liu, T. (2015). Arguing to agree: Mitigating my-side bias
Declaration of competing interest through consensus seeking dialogue. Written Communication, 32(3), 317–331.
https://doi.org/10.1177/0741088315590788
Felton, M., Garcia-Mila, M., Villarroel, C., & Gilabert, S. (2015). Arguing collaboratively:
The authors declare no potential conflicts of interest with respect to Argumentative discourse types and their potential for knowledge building. British
the research, authorship, and/or publication of this article. Journal of Educational Psychology, 85(3), 372–386. https://doi.org/10.1111/
bjep.12078
Granado-Peinado, M., Mateos, M., & Martín, E.y. C., .I. (2019). Teaching to write
Data availability collaborative argumentative synthesis in higher education. Reading and Writing, 32
(1), 2037–2058. https://doi.org/10.1007/s11145-019-09939-6
Data will be made available on request. Helme, S., & Clarke, D. (2001). Identifying cognitive engagement in the mathematics
classroom. Mathematics Education Research Journal, 13(2), 133–153. https://doi.org/
10.1007/BF03217103
Appendix 1 Hsu, C. F. (2007). A cross-cultural comparison of communication orientations between
Americans and Taiwanese. Communication Quarterly, 55(3), 359–374.
Jiang, D., & Zhang, L. J. (2020). Collaborating with ‘familiar’ strangers in mobile-assisted
Writing Assignment of the first essay:
environments: The effect of socializing activities on learning EFL writing. Computers
Write an argumentative essay on old age in about 400 words, you & Education, 150, 1–14. https://doi.org/10.1016/j.compedu.2020.103841
should: Kibler, A. (2017). Peer interaction and learning in multilingual settings from a
sociocultural perspective: Theoretical insights. International Multilingual Research
Journal, 11, 199–203. https://doi.org/10.1080/19313152.2017.1328970
1. Introduce the topic and present your position in the introductory first Kuhn, D., & Crowell, A. (2011). Dialogic argumentation as a vehicle for developing
paragraph. young adolescents’ thinking. Psychological Science, 22(4), 545–552.
2. Include both FOR and AGAINST to develop the supporting Lantolf, J. P. (2000). Second language learning as a mediated process. Language Teaching,
33, 79–96.
paragraphs. Lantolf, J. P., & Thorne, S. L. (2009). Sociocultural theory and the genesis of second language
3. Finally, restate your position in the concluding paragraph. development. New York: Oxford University Press.
Leitão, S. (2003). Evaluating and selecting counterarguments: Studies of children’s
rhetorical awareness. Written Communication, 20(3), 269–306.
Writing Assignment of the second essay: Li, H. H., & Zhang, L. J. (2022). Investigating effects of small-group student talk on the
Write an argumentative essay on animals in captivity in at least 250 quality of argument in Chinese Tertiary English as a foreign language learners’
words, you should: argumentative writing. Frontiers in Psychology, 13, 1–14. https://doi.org/10.3389/
fpsyg.2022.868045

9
D. Liu et al. Acta Psychologica 246 (2024) 104256

Liu, D. (2020). Alterations in the Chinese EFL learners’ argumentation schemas: A Toulmin, S. E. (1958). The uses of argument. Cambridge: Cambridge University Press.
longitudinal study based on Toulmin model. Modern Foreign Languages, 43(6), Toulmin, S. E. (2003). The uses of argument (2nd ed.). Cambridge: Cambridge University
793–805. Press.
Liu, D., & Wan, F. (2020). What makes proficient writers’ essays more persuasive? A Vygosky, L. S. (1978). Mind in Society: The development of higher mental process.
Toulmin perspective. International Journal of TESOL Studies, 2(1), 1–13. https://doi. Cambridge, MA: Harvard University Press.
org/10.46451/ijts.2020.06.01 Wecker, C., & Fischer, F. (2014). Where is the evidence? A meta-analysis on the role of
Lu, J., & Zhang, Z. (2013). Scaffolding argumentation in intact class: Integrating argumentation for the acquisition of domain-specific knowledge in computer-
technology and pedagogy. Computers & Education, 69, 189–198. supported collaborative learning. Computers & Education, 75, 218–228.
McDonough, K., De Vleeschauwer, J., & Crawford, W. (2018). Comparing the quality of Wigley, C. J., Rancer, A. S., & Lin, Y. (2023). Initial report of the Argumentativeness
collaborative writing, collaborative prewriting, and individual texts in a Thai EFL Index of Behavioral Indicators for measuring trait argumentativeness.
context. System, 74, 109–120. https://doi.org/10.1016/j.system.2018.02.010 Communication Quarterly, 71(5), 483–498. https://doi.org/10.1080/
McDonough, K., De Vleeschauwer, J., & Crawford, W. J. (2019). Exploring the benefits of 01463373.2023.2228864
collaborative prewriting in a Thai EFL context. Language Teaching Research, 23, Wolfe, C. R. (2012). Individual differences in the “myside bias” in reasoning and written
685–701. https://doi.org/10.1177/1362168818773525 argumentation. Written Communication, 29(4), 477–501. https://doi.org/10.1177/
Mullen, S., Richter, T., Schmid, S., & Berthold, K. (2019). How to improve argumentation 0741088312457909
comprehension in university students: Experimental test of a training approach. Wolfe, C. R., Britt, M. A., & Butler, J. A. (2009). Argumentation schema and the myside
Instructional Science, 47, 215–237. https://doi.org/10.1007/s11251-018-9471-3 bias in written argumentation. Written Communication, 26(2), 183–209. https://doi.
Newell, G. E., Beach, R., Smith, J., & VanDerHeide, J. (2011). Teaching and learning org/10.1177/0741088309333019
argumentative reading and writing: A review of research. Reading Research Quarterly, Wolfe, C. R., Gao, H., Wu, M., & Albrecht, M. (2018). Most any reason is better than
46, 273–304. none: Consequences of implausible reasons and warrants in brief written arguments.
Newell, G. E., Bloome, D., Kim, M., & Goff, M. (2018). Shifting epistemologies during Written Communication, 35(3), 255–285. https://doi.org/10.1177/
instructional conversations about “good” argumentative writing in a high school 0741088318767370
English language arts classroom. Reading and Writing, September 20, 1–24. Xie, Y., Hample, D., & Wang, X. (2015). A cross-cultural analysis of argument
Noroozi, O., Kirschner, P. A., Biemans, H. J. A., & Mulder, M. (2018). Promoting predispositions in China: Argumentativeness, verbal aggressiveness, argument
argumentation competence: Extending from first- to second-order scaffolding frames, and personalization of conflict. Argumentation, 29, 265–284. https://doi.org/
through adaptive fading. Educational Psychology Review, 30, 153–176. 10.1007/s10503-015-9352-8
Nussbaum, E. M. (2008). Using argumentation vee diagrams (AVDs) for promoting Yamashita, T. (2021). Corrective feedback in computer-mediated collaborative writing
argument-counterargument integration in reflective writing. Journal of Educational and revision contributions. Language Learning & Technology, 25(2), 75–93.
Psychology, 100(3), 549–565. Yang, R. (2022). An empirical study of claims and qualifiers in ESL students’
Nussbaum, E. M., & Kardash, C. A. M. (2005). The effects of goal instructions and text on argumentative writing based on Toulmin model. Asian-Pacific Journal of Second and
the generation of counterarguments during writing. Journal of Educational Foreign Language Education, 7(6), 1–17.
Psychology, 97, 157–169. Yeh, J. H., & Chen, L. (2004). Cultural values and argumentative orientations for Chinese
Qin, J., & Karabacak, E. (2010). The analysis of Toulmin elements in Chinese EFL people in Taiwan, Hong Kong, and Mainland China, 51–64. Intercultural
university argumentative writing. System, 38(3), 444–456. https://doi.org/10.1016/ communication: A global reader.
j.system.2010.06.012 Yilmaz-Na, E., & Sönmez, E. (2023). Unfolding the potential of computer-assisted
Rancer, A. S. (1998). Argumentativeness. In J. C. McCroskey, J. A. Daly, M. M. Martin, & argument mapping practices for promoting self-regulation of learning and problem-
M. J. Beatty (Eds.), Communication and personality: Trait perspectives (pp. 149–170). solving skills of pre-service teachers and their relationship. Computers & Education,
Hampton Press. 193, 1–18.
Reznitskaya, A., & Wilkinson, I. A. (2017). The most reasonable answer: Helping students
build better arguments together. Harvard Education Press.

10