Data comics outperformed conventional visualisations on both information retrieval and comprehension tasks, with the largest gains in comprehension, particularly multiple-insight questions¶

Claim · All claims
Evidence · 1 study (3 entries) · causal r2 · q3 · i2–i3
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i2 For information retrieval tasks, the correct rate was significantly higher with DC than CV, with a moderate effect size (r = 0.341). → Zirui Shan 2026 q3 i3 For comprehension tasks, DC yielded a significantly higher correct rate than CV with a large effect (r = 0.695), an average increase of 0.417. → Zirui Shan 2026 (2) q3 i3 For multiple-insight comprehension questions, DC showed a large improvement over CV (r = 0.589), an average increase of 0.406. → Zirui Shan 2026 (3)

Evidence¶

Zirui Shan 2026¶

Zirui Shan, Vanessa Echeverria, Yuheng Li, Yi-Shan Tsai, and Roberto Martinez-Maldonado. (2026). Data Comics for Education: Evaluating Effectiveness, Benefits, and the Ethics of AI-Assisted Creation. https://arxiv.org/abs/2607.02361 doi:10.48550/arxiv.2607.02361

q3 · i2 · causal · r2

RQ2 analysis of the within-subjects comparison, splitting the 16 questions per condition by Bloom's taxonomy level. The retrieval contrast reached significance with "a moderate effect size (U= 1601.5,Z= 2.642,p= 0.033,r= 0.341)", an average increase of 0.156.

"For information retrieval tasks, participants achieved a significantly higher correct rate in the DC condition (median = 0.667, IQR = 0.333) compared to conventional visualisations (CV) (median = 0.667, IQR = 0.334), with a moderate effect size (U= 1601.5,Z= 2.642,p= 0.033,r= 0.341)"

Zirui Shan 2026 (2)¶

Zirui Shan, Vanessa Echeverria, Yuheng Li, Yi-Shan Tsai, and Roberto Martinez-Maldonado. (2026). Data Comics for Education: Evaluating Effectiveness, Benefits, and the Ethics of AI-Assisted Creation. https://arxiv.org/abs/2607.02361

q3 · i3 · causal · r2

Same RQ2 analysis, comprehension task level. The article reports DC "showing a large effect size (U= 1727.5,Z= 5.383,p<0.001,r= 0.695)", with data comics increasing the correct rate by a mean of 0.417.

"For comprehension tasks, participants also performed significantly better with DC (median = 0.667, IQR = 0.667) than with CV (median = 0.000, IQR = 0.333), showing a large effect size (U= 1727.5,Z= 5.383,p<0.001,r= 0.695)"

Zirui Shan 2026 (3)¶

Zirui Shan, Vanessa Echeverria, Yuheng Li, Yi-Shan Tsai, and Roberto Martinez-Maldonado. (2026). Data Comics for Education: Evaluating Effectiveness, Benefits, and the Ethics of AI-Assisted Creation. https://arxiv.org/abs/2607.02361

q3 · i3 · causal · r2

RQ2 comprehension subtype analysis. For multiple-insight questions DC "showed an even larger improvement" over CV with a large effect; single-insight questions also favoured DC (r = 0.520, average increase 0.322).

"Formultiple-insightquestions, DC showed an even larger improvement (median = 1.000, IQR = 0.667) compared to CV (median = 0.333, IQR = 0.333), with a large effect (U= 1690.0,Z= 4.564,p <0.001,r= 0.589)"

Discussion¶

Related Claims¶

• University students achieved a significantly higher correct rate answering data questions with data comics than with conventional visualisations — related
• Information retrieval questions were consistently easier than comprehension questions across both visualisation formats — related
• Students perceived data comics as more accessible, more engaging, and easier to retain information from than conventional visualisations — related
• Visualisation literacy positively predicts task accuracy in both conditions, and data comics show no equalising effect for lower-literacy students — related