Exploring the Power of Persuasion in Written Argumentation: A Mixed-Methods Pilot Study (QASA)

Argumentation is a fundamental competence in modern societies, essential for professional, political, and civic life. It is widely recognized as a core skill for academic success and active democratic participation [1-3]. In an era of unlimited digital information access, young people increasingly identify as “digital citizens” and must navigate not only factual argumentation but also misinformation and populist rhetoric. International organizations such as UNESCO emphasize that the ability to exchange perspectives and adapt one’s viewpoint in light of new evidence is central to lifelong learning and civic engagement. Similarly, the OECD identifies argumentation as a key competence for building personal convictions and engaging in democratic decision-making [4,5]. In democratic societies, argumentative competence plays a critical role in fostering perspective-taking and generating solutions to controversial issues. Language and literature instruction can contribute meaningfully to this goal by explicitly teaching both oral and written argumentation using empirically validated approaches [1,6]. International research shows that written argumentation can benefit from prior oral argumentation training [7,8].

Structured debate has proven effective in promoting oral argumen- tative competence and fostering civic engagement [9]. Debate al- lows students to engage with controversial issues in a regulated format that mirrors democratic practices, and its structure closely parallels the organization of written pro–con argumentative es- says. However, little empirical work has examined the relation- ship between debate training and the development of written ar- gumentative competence. Furthermore, there is a need to develop qualitative methods that assess the quality of argumentative texts as accurately as possible so that teachers can give students better feedback.

These research gaps are addressed by the empirical teaching and research project “FDE” (in English: Fair Debating and Written Argumentations/ in German: Fair Debatttieren and Erörtern). As part of this project, two lesson series were implemented at six secondary schools in Germany (both academic and non-academic tracks) between October 2021 and May 2022. These interventions, each comprising six 90-minute sessions, focused on debate (oral argumentation) and writing (written argumentation). As part of an accompanying screening process, students’ written argumentation skills were assessed by pro-and-con argumentation essays (Giera, in press). But how can the quality of these essays—written by ninth- grade students—be comprehensively and thoroughly assessed? This article presents the QASA (Qualitative Argumentation Structure Analysis) sub-study, which focuses on a sample of ninth- grade students from non-academic-track schools (n = 9) and their 18 pro-and-con argumentation essays. QASA describes a method for analyzing argumentation structures in student texts using structural diagrams. This study investigates the extent to which written argumentation competence can be enhanced through debate training in non-academic-track students by assessing the persuasiveness of their argumentative essays, using a five-point rating scale and the QASA method, after a series of lessons on debating. This leads to the central research question of the article:

RQ: To what extent do non-academic-track ninth-grade students demonstrate written argumentation competence in written argumentations after six 90-minute debate sessions, despite having no prior exposure to the pro-and-con argumentation essay genre?

To address this question, Section 2 offers a theoretical overview of international research on the promotion of oral and written argumentation competence. This includes an outline of the debate format and the genre of pro-and-con argumentation, as well as an outline of the link between debate and written argumentation formats. In addition, the key aspects of the QASA procedure are explained.

Section 3 then describes the methodology of the overarching pilot study, “FDE”, with a specific focus on the QASA sub-study and its analysis of the 18 student essays using both quantitative and qualitative methods [10,11].

Section 4 presents the results, which are discussed in Section5. Finally, Section 6 summarizes the findings and provides implications for future research and teaching practice.

Literature Review

Oral Argumentation and Debate in Secondary Education

Oral argumentation is a central mode of reasoning in educational settings, providing students with opportunities to articulate positions, respond to counterarguments, and adapt their reasoning in real time [12]. By engaging in argumentative practices, secondary school students learn how to better structure their reasoning, consider alternative viewpoints, and develop more nuanced understandings of the topics they argue about [13]. Moreover, several studies indicate positive effects of dialogi activities on argumentative writing in various contexts. For example, Casado-Ledesma et al. highlight deliberate discussions as suitable to promote the quality of integration for argumentative texts by Spanish secondary school students [14]. These findings confirm the results of the multi-year study of middle schools by Crowell and Kuhn, who examined the influence of dialogue-based argumentation on the development of cognitive argumentative skills, such as the consideration of counterarguments [15,16]. The investigated dialogue-based approach also seems to be suitable for improving argumentative essay and argumentative knowledge [12].

Structured debate differs from informal discussion by its fixed roles, timed stages, and clear decision-focused propositions [6]. These elements require participants to prepare evidence-based claims and anticipate rebuttals, fostering skills essential for both civic engagement and academic discourse.

Research consistently shows that debate training strengthens cognitive and metacognitive processes. Students learn to identify weaknesses in opposing claims, synthesize multiple perspectives, and justify their own stance more effectively [17]. Debate also supports the development of speaking fluency and confidence, especially for learners who may struggle in less structured classroom interactions [18].

The leading model of debate in German schools is the “Jugend debattiert” (Youth Debates) program (Giera, in press). Although the participation of non-academic-track schools in this competition has increased, it remains dominated by academic-track schools, which accounted for roughly two-thirds of participating schools by the end of 2024. Therefore, the FDE project deliberately includes non-academic-track schools.â??â??â??â??â??â??â??

Each debate followed a fixed sequence: opening statements, open exchange, and closing statements. Building on the “Youth debate” curriculum (Giera, in press), the FDE project developed a six- phase teaching unit tailored to heterogeneous learning groups:

1. 1. Trying to Debate: Students are introduced to debate rules and structure (e.g., two pro/two con speakers) through debates on less serious topics (e.g., whether deckchairs should be mandatory in the classroom).
2. 2. Getting to the Point: Students prepare an opening speech to practice clarifying their positions at the beginning of the debate.
3. 3. Referring to Each Other: Students prepare for free discussion, learning to refer and respond to arguments made by the opposing side.
4. 4. Drawing Conclusions: This phase sharpens closing statements.
5. 5. Clarifying Prerequisites: Students identify core aspects and arguments of the debate topic by researching.
6. 6. Debate Practice: Students conduct complete debates, the length of which can be adapted to their performance level.
7.  
8. The aim is for students to engage in simplified debates early on and gradually acquire skills in a cumulative learning process. The curriculum was adapted for inclusive teaching in non-academic- track schools and supplemented by leveled “learning buffet” materials to support diverse learners (Giera, in press).
9.  
10. A key feature of the debate training was the use of a timekeeper to ensure equal speaking time. This helped students reflect on how arguments were received and encouraged perspective-taking. Full debates last 24 minutes; however, in the FDE project, shorter micro-debates (approximately 6 minutes) were used due to limited prior experience and time constraints (Giera, in press). Throughout the process, students practiced forming clear, concise arguments, active listening, and critical self-reflection.

Written Argumentative Competence in Secondary Education

Pro–con argumentative essays are a key genre in secondary edu- cation, requiring students to present balanced reasoning, evaluate multiple perspectives, and conclude with a justified recommenda- tion. Success in this genre depends on rhetorical structure, evi- dence integration, and the ability to consider counterarguments. These skills parallel those practiced in oral debate, but written ar- gumentation additionally demands sustained coherence, audience awareness, and planning strategies [19].

The most widely researched and empirically supported instruc- tional model for developing writing competence is the Self-Reg- ulated Strategy Development (SRSD) approach [20-22]. SRSD encompasses a set of writing strategies and instructional routines designed to build declarative, procedural, and metacognitive knowledge among students in order to enhance overall writing competence [21]. Developed in the early 1980s by Karen Har- ris and Steve Graham, SRSD has since been tested in over 100 school-based studies worldwide. As of January 2025, it has been applied in schools across 19 countries on multiple continents, in- cluding North and South America, Europe, Asia, and Oceania (Gi- era, in press).

The effectiveness of the SRSD approach has also been confirmed in the field of argumentative writing. For example, De La Paz studied the effects of SRSD on students’ ability to write historically reasoned argumentative texts in grade 8 (n = 70) [23]. Students in the intervention group wrote essays that demonstrated stronger content knowledge, improved persuasiveness, more arguments, and greater text length compared to a control group (n = 62). The task closely resembled the pro-and-con argumentation structure used in German dialectical essays.

Similarly, Mason et al. examined the impact of SRSD in a quasi- experimental study involving 33 students in grades 7 and 8 [24]. The focus was on improving the quality of short argumentative texts (“quick writes,” n = 279). The findings showed that students who received SRSD training in addition to regular instruction produced longer and more persuasive texts than those in the comparison groups. Another study on fifth- and sixth-grade students (n = 564) found comparable results: the intervention group (n = 326) produced longer and more detailed argumentative texts than the control group (n = 266) [25].

Within the FDE project, the SRSD approach was adapted for the first time in Germany to teach the writing of dialectical essays as part of a classroom unit. Mirroring the structure of debate training, the study investigated how students’ argumentative writing competence developed over the course of the project.

Integrated Development of Oral and Written Argumenta- tive Competence

In recent years, international research has increasingly turned its attention not only to the isolated development of oral and written argumentative skills but also to their interconnected promotion. For example, Shi et al. (p. 109) describe dialogue as a “pathway to argumentative writing [3].” This perspective is grounded in a sociocultural approach, which holds that “[…] that writing development is influenced by participating in situations that afford opportunities to appropriate semiotic tools and social practices” (p. 1347) [26].

Several studies have explored the effect of argumentative discus- sions on students’ critical thinking and writing skills. Reznitskaya et al., for example, examined the impact of collaborative discus- sions (small-group dialogues designed to foster critical reason- ing) on students' reflective judgment [8]. These sessions support- ed students in explaining discussed issues, linking ideas to prior knowledge, drawing conclusions, and developing new insights based on arguments and counterarguments. They also enabled stu- dents to present sufficient evidence and expand their knowledge base. These findings confirmed earlier research on the influence of group discussions on written argumentation [7].

In terms of debate as a specific form of oral argumentation, international research shows that only a few studies have directly investigated its relationship with written argumentation in school settings. In the quasi-experimental field, the studies by Malloy et al. in the U.S. and el Majidi et al. in the Netherlands are particularly noteworthy. Malloy and colleagues observed enhanced quality of argumentative texts among 40 fifth-grade students following several rounds of debate training. Similarly, el Majidi et al. studied

147 eighth-grade students who participated in ten structured debates and later demonstrated stronger argumentative structure and persuasiveness in their written texts compared to the control group [9]. Against this background, the present study aims to provide insights into argumentative writing by focusing attention on structural elements of pro-and-con argumentation essays. The QASA sub-study of FDE uses a small sample to show which of these structural elements of written argumentation will improve in quality after debate training (see Section 2.1). In the following Section, an outline of FDE and the QASA methodology is given.

Outline of Qasa Methodology

First, the central text type analyzed in the QASA procedure—the pro–con argumentative essay—is defined, and then, the QASA method itself is explained in detail. The pro–con argumentative essay is a text genre that requires students to develop coherent, supported pro and con arguments regarding a controversial issue, integrating both a thesis and its antithesis, and concluding with a well-reasoned summary [19]. The focus on this genre in the current study is justified by its curricular inclusion in lower- secondary education across different school tracks [28]. Moreover, this text type is a mandatory component of the intermediate-level German language examination at the end of grade 10 in Berlin and Brandenburg. From a (writing) instructional perspective, the pro– con argumentative essay supports the development of some of the skills necessary for written argumentation—such as stance-taking, qualifying, and conceding—within academic writing [28].

The QASA method has been adapted for use with pro–con argumentative essays, as depicted in Figure 1. It builds on corpus- based text analysis techniques proposed by Stede, tailored to create graphical representations of argument structures [29]. Comparable methods include automated argument-mining approaches, such as those developed by Stab and Gurevych, which identify argumentative discourse components and relations in persuasive essays. These systems also segment text into Argumentative Discourse Units (ADUs) and classify their relational connections. Below, the key terminology required to understand QASA is introduced, followed by detailed methodological steps (see Fig. 2).

The analysis begins with the segmentation of text into Elementary Discourse Units (EDUs)—typically, full simple sentences or clauses serving coherent communicative functions. A subset of these constitutes Argumentative Discourse Units (ADUs), which contribute directly to argumentation. An ADU may encompass multiple EDUs when necessary to form a complete argumentative contribution. Content zones are conventional text parts that play specific roles in achieving the text's purpose [11]. For pro–con essays, content zones include the thesis, pro arguments, and con arguments. Each ADU is assigned to one content zone within a predefined inventory.

The analytical output is a graphical argument structure tree (see Fig. 2), which highlights a text’s core argumentative elements while omitting non-argumentative coherence devices (see Section 4). Explanatory or organizational sentences not part of the argument can be excluded from this representation. Following Stede’s methodology, the QASA annotation process involves the following steps [29]:

1. Segmentation—The text is partitioned into EDUs and subsequently into ADUs.

2. Labeling—Each EDU or ADU is assigned a content zone label, identifying the central thesis (ZTH), if apparent, and categorizing ADUs as pro or con arguments.

3. Linking—ADUs (theses, arguments, examples) are connected via relational links to form a complete structural representation of the argumentation.

Research Design and Methodology

This section summarizes key elements of the research design of the FDE study in order to clarify the context of the QASA sub-study described in Section 2.4. For a more detailed account of the project’s design and findings, see Giera [31].

   The Comprehensive Project “Fair Debating and Written Argumentation” (FDE)

The FDE project is the first in the German-speaking world to empirically investigate the integrated development of oral and written argumentation skills in a quasi-experimental field study. It included a longitudinal intervention with 357 ninth-grade students from six schools in the German state of Brandenburg (three academic-track and three non-academic-track schools). The study was conducted between October 2021 and May 2022. A total of 357 students from six schools participated, divided equally between academic-track and non-academic-track schools. Its aim was to assess and promote the (written) language skills of students with and without special language needs in grade 9 through two instructional units—one on debating and one on pro-and-con argumentation essay writing (see Sections 2.1 and 2.2).

The instructional design was informed by the structured debates mentioned in Section 2.1 and the Self-Regulated Strategy Devel- opment (SRSD) framework for teaching writing [21,22]. Each unit comprised six 90-minute sessions, implemented weekly during regular language lessons. Random assignment within classes fol- lowed an alternating-number procedure, creating two treatment groups: DE (first debate, then essay writing) and ED first essay writing, then debate). This enabled analysis of possible sequence effects on students’ argumentation skills. The present sub-study focuses exclusively on the DE group from non-academic-track schools, as these students had no prior exposure to the pro–con argumentative essay genre.

The DE sequence began with six debate training sessions based on a scaffolded format, moving from short “micro-debates” to full debates, each adhering to a structured sequence: opening statements, open exchange, and closing remarks (see Section 2.1). The strategies emphasized evidence-based reasoning, rebuttal practice, and time management through the use of a designated timekeeper [6].

Following a brief holiday break, the same students participated in six SRSD-based lessons on writing pro–con argumentative essays. Instruction covered planning (e.g., generating and organizing arguments), drafting (maintaining a balanced structure), and revising (strengthening coherence and persuasiveness). The lessons explicitly drew connections between debate phases and corresponding essay sections (introduction, body, conclusion).

To test the effectiveness of different intervention sequences, the students in each class were randomly assigned (by alternating numbers on the class roster: 1,2,1,2, etc.) to two treatment groups: DE (debate first, then essay writing) and ED (essay writing first, then debate). This enabled comparison of the effects of different instructional orders (see Table 1 in Section 3.2).

The focus on grade 9 reflects the curricular importance of written argumentation for achieving the intermediate school-leaving certificate at the end of grade 10 in Germany (see Section 2.2). To evaluate the research question, the teaching units were developed in cooperation with teachers who train students and schoolteachers, implemented in classrooms, and then evaluated.

To track the development of writing competence over time, a total of 1,024 student essays were collected across four measurement points. Students composed their essays digitally using the Gorilla. Sc software [32]. Each writing task was based on authentic, standardized exam prompts used in grade 10, with different topics used at each time point to prevent retest effects. 

To explain the research design, it is important to mention that all research data were collected anonymously. All students were anonymized using five-digit IDs, generated by them. One limitation of the pilot study is that only a few students entered the same code for screening at all measurement points. For the present study, the development of persuasiveness in the argumentation essays was only considered for those students who had entered the same anonymized code at two measurement points (see Table 1). Students’ written argumentation skills were assessed using both holistic and analytical ratings based on the IMOSS coding scheme on a five-point scale [10]. This rating method also formed the basis for the quantitative analysis of the 18 student texts included in the QASA sub-study, which is described in Section 3.2.

Methodological Approach of the QASA Sub-study Building on the FDE project, the QASA sub-study focuses specifically on a sample from the DE treatment group from non- academic-track schools. Nine students were selected whose data met two criteria: (i) authorship of essays in both the pretest (tâ?ÂÃÃÃÃÃÃÂ??????ÂÃÃÃÃÃÂ?????ÂÃÃÃÃÂ????ÂÃÃÃÂ???ÂÃÃÂ??ÂÃÂ?) and intermediate test (tâ??), and (ii) clear identification via anonymized student IDs. Each student produced one essay per time point, yielding a corpus of 18 texts. Figure 1 illustrates how the QASA sub-study was derived from the broader FDE project.

Figure 1: Derivation of the Sub-Study from the FDE Research and Teaching Project

To address the central research question (see Section 1) regarding the development of persuasive writing in pro-and-con argumentation essays before and after debate training, the essays from all non- academic-track students in the DE group were analyzed at time points t1 and t2 (see Figure 2). It was crucial that each essay could be definitively matched to an anonymized student ID in order to track individual development over time. 

To address the central research question (see Section 1) regarding the development of persuasive writing in pro-and-con argumentation essays before and after debate training, the essays from all non- academic-track students in the DE group were analyzed at time points t1 and t2 (see Figure 2). It was crucial that each essay could be definitively matched to an anonymized student ID in order to track individual development over time.

For the QASA sub-study, only the DE group’s phase following debate training was analyzed, as these students had no prior knowledge of pro-and-con argumentation essays due to curricular timing (see Section 2.2). This sub-study employed a mixed- methods design.

Table 1: Timeline of the FDE Project for Intervention and Control Groups

Quantitative Analysis

Persuasive coherence was assessed using the IMOSS coding procedure, adapted for argumentative writing, on a 5-point scale [10]. Rating was conducted using a double-blind peer review. IMOSS was chosen due to its high reliability, having been validated in the large-scale study DESI, involving nearly 20,000 student texts. Rater agreement was 70.9% for pragmatic items, with internal consistency at Cronbach’s α = .91 and an overall reliability coefficient of α = .90 [33]. The system has been successfully applied to argumentative texts written by ninth-grade students, making it suitable for the present analysis. The results are presented in Table 2 (Section 4).

Qualitative Analysis

A qualitative argumentation structure analysis (QASA) was conducted following the model by Stede, from which the sub- study takes its name [11]. This method is explained in more detail in Section 2.4. Figure 2 shows an example of a structure tree representing a pro-and-con argumentative essay written by a 9th- grade student.

The student was asked to write an argumentative essay on whether digital e-books should be introduced at his school instead of analog books. The text sections are assigned to the structure tree. These are structural components typical of a controversial argument: the central thesis, pros and cons, outline notes, and summary. This text is a very short example, but it illustrates what structure trees look like. The thesis is marked with a round gray node at the root of the tree. Pro arguments are shown as round nodes, and con arguments as square nodes. Three arrow types indicate relationships:

Sharp arrow: supportive relation;

Round arrow: counterargument;

Dashed arrow: example.

This analysis enables insights into the complexity and persuasiveness of students’ texts. As shown in Tables 2 and 3 (Section 4), the QASA method allows for detailed comparisons of the number and type of arguments (ADUs), as well as their relational structure, before and after the intervention. 

Figure 2: Example of Argumentation Structure Tree