From Data to Discourse: Interpretive Shifts in the Use of Visualizations in Educational Research Articles

Introduction. Visual elements, including tables, figures, and diagrams, are a routine feature of contemporary research articles. Previous scholarship has described their formal properties and has documented the prevalence of descriptive commentary, yet the cross sectional dynamics of visual meaning within the IMRaD structure remain insufficiently specified. In particular, it is still unclear how the rhetorical and epistemic status of the same visual changes when it is taken up in Discussion after initial presentation in Results. The aim of the study is to conduct a discursive-rhetorical and epistemological analysis of the functions of data visualization in the structural components of original research articles.
Materials and Methods. The study examined a purpose-built corpus of 50 peer reviewed education research articles published between 2018 and 2024. All verbal references to visual elements were manually coded using an analytic scheme covering integration type, interpretation depth index, and interpretative function. The coding also captured causal relation patterns, argumentative frames, interpretative shifts between Results and Discussion, and recurrent phrase frames. Coding reliability was assessed through double coding of a subset of the corpus and calculation of intercoder agreement. Quantitative analysis combined descriptive statistics with chi square tests of association.
Results. The corpus contained 576 unique visual elements and 551 verbal references. Visuals were concentrated in Results and Methods. Referential and descriptive integration predominated, whereas interpretative integration was uncommon. Interpretation depth was heavily weighted toward the lowest levels, and higher order interpretation and theoretical generalization accounted for a small minority of cases. Argumentative framing was dominated by Support, and interpretative shifts between Results and Discussion were rare, occurring in 2.4 percent of references. Phrase frame analysis likewise showed a strong reliance on referential and descriptive formulae.
Discussion and Conclusion. In education research articles, visuals function primarily as instruments of data presentation rather than as resources for interpretation and theorization. The scarcity of interpretative shifts suggests a structural constraint on the extent to which visuals are recontextualized as argumentative evidence across sections of the article. Although visuals are ubiquitous, their epistemic potential is activated only intermittently. These findings motivate more explicit pedagogical attention to visual literacy in research writing and invite editorial reflection on how multimodal evidence is expected to contribute to knowledge claims within the IMRaD genre.