NIJAMSS peer-reviewed article

From Data to Decisions: A Practical Framework for Accurate, Ethical and Accessible Graphical Presentation in Statistical Analysis

Daniel A. Otwori

Abstract: Graphical presentation is a central component of statistical analysis because it converts numerical information into visual evidence that can reveal distribution, comparison, trend, association and uncertainty. However, graphs can also mislead when the chart type, scale, denominator, colour, annotation or level of aggregation is inappropriate. This applied methodological review examines the selection, design and interpretation of common statistical graphics, including bar charts, dot plots, histograms, box plots, line charts, scatter plots, maps and uncertainty displays. Evidence from statistical, cognitive and visual-communication literature is synthesised to explain why visual effectiveness depends on matching the graph to the research question and data structure rather than on decoration. The review shows that bar and dot charts support categorical comparison; histograms, density plots and box plots reveal numerical distributions; line charts display ordered change over time; scatter plots examine relationships; and confidence-interval plots communicate estimate precision. Pie charts and three-dimensional effects are often less effective because angle, area and perspective are harder to compare accurately. The paper proposes a Graph Selection-Design-Interpretation Framework built around seven principles: question-data alignment, truthful scales, denominator transparency, visible uncertainty, accessibility, reproducibility and ethical communication. Three technical tables provide guidance on chart selection, visual-integrity auditing and public-health dashboard design. The paper concludes that graphical presentation is not a cosmetic final step but an analytical method that shapes interpretation and decision-making. High-quality graphs should be accurate, simple, accessible and reproducible, while showing enough context to prevent false certainty or misleading comparison.

Keywords: Biostatistics; Data visualisation; Graph design; Public health; Scientific communication; Statistical graphics; Visual integrity

Journal

Nexus International Journal of Applied Science, Medicine and Social Sciences

Publication date

May 2026

Volume / Issue

Vol. 1, Issue 1

Paper number

002

Article type

Applied Methodological Review

Published paper ID

NIJAMSS-V1S1-MAY2026-002

Publisher

Nexus Academic Press

Access

Open Access

Full text PDF: https://www.nijamss.org/archives-download/nijamss-vol-1-series-1-may-2026-paper-002.pdf