Effective argumentative writing requires more than grammatical accuracy; it necessitates understanding writing as a social and communicative act. Argumentation, while common, poses significant challenges for EFL/ESL learners, demanding the strategic use of rhetorical structures, logical reasoning, and linguistic features to connect with readers. Metadiscourse, encompassing linguistic elements that build rapport and signal writer stance, is crucial in argumentative writing. This study addresses a gap in research by exploring interactional metadiscourse in the context of a gender-sensitive topic, particularly considering cultural factors influencing Saudi women's assertiveness in expressing opinions. The research questions focus on how Saudi EFL male and female students use interactional metadiscourse markers in their writing about car driving, specifically exploring variations based on gender and stance taken on this gender-sensitive topic.

Existing research on metadiscourse markers in argumentative writing often focuses on factors like proficiency, nativeness, learning context, and topic. Studies have shown varied results regarding gender differences, with some indicating male writers use more boosters and females more hedges, while others present contrasting findings. Research on Arab writers' use of metadiscourse has primarily focused on professional writers or postgraduates, with limited investigation into undergraduate EFL students. A key gap identified is the lack of research exploring metadiscourse use in relation to gender-sensitive topics and stance variation, especially within a cultural context like Saudi Arabia, where societal expectations have traditionally limited female assertiveness.

This mixed-methods study employed both quantitative and qualitative approaches. 144 undergraduate English translation students (59 males, 85 females) at King Saud University participated. Data were collected via an argumentative essay writing task on the topic of driving ability differences between men and women. The essays were categorized into four sub-corpora based on gender and stance (male/female for male/female driving). AntConc software was used for quantitative analysis of metadiscourse markers, based on Hyland's (2005) interactional metadiscourse model. Qualitative analysis involved examining the context of marker usage. Independent Samples t-tests and one-way ANOVAs were conducted to compare groups, ensuring assumptions of parametric statistics were met. Inter-rater reliability was assessed using Cohen's kappa (0.647).

Quantitative analysis revealed significant gender differences in metadiscourse marker usage. Female writers used significantly more attitudinal lexis, boosters, hedges, and self-mentions than males. Normalised frequencies confirmed these findings. ANOVA results indicated significant differences between the four groups (males/females for male/female driving) in the use of boosters, attitude markers, hedges, and self-mentions. Post-hoc tests revealed specific group differences. For instance, females arguing for male drivers used significantly more hedges than other groups; females supporting female drivers used significantly more self-mentions than males supporting male drivers. Qualitative analysis provided contextual support for these findings, illustrating differences in how specific words were used to express stance and attitude. For example, 'all' was used differently by males (to emphasize universal male superiority) and females (to generalize female driving behavior).

The findings challenge previous research by showing that female writers used a greater number of interactional metadiscourse markers than males. This might reflect the changing social dynamics in Saudi Arabia, where women are increasingly empowered. The significant use of self-mentions, hedges, boosters, and attitude markers suggests a more personalized writing style among females. The use of boosters by females, particularly those supporting female drivers, shows assertiveness in their arguments. The higher usage of hedges by females arguing for men’s driving could be a reflection of their cultural background and a less assertive style of arguing. The use of hedges doesn’t necessarily indicate less confidence but rather reflects sensitivity toward the audience. The stance taken (supporting same gender vs. opposite gender) significantly influenced metadiscourse use. The study highlights that choosing a gender-sensitive topic and stance interacts to shape how writers use metadiscourse resources.

This study demonstrates that gender and stance significantly influence the use of interactional metadiscourse markers in argumentative writing on a gender-sensitive topic. Female writers, particularly those supporting their own gender, exhibited more assertive metadiscourse patterns. The findings have pedagogical implications for EFL instruction, emphasizing the importance of explicit teaching of metadiscourse strategies. Future research should investigate larger corpora, diverse settings, and other gender-sensitive topics to further explore these dynamics.

The study's limitations include the relatively small corpus size, particularly for certain subgroups, and the single institutional setting. The use of a single gender-sensitive topic and the fact that essays were written under exam conditions could also affect the generalizability of findings. Future research should address these limitations by employing larger and more diverse datasets.