MODULE 4
Data and Research Visualisation
Interpret research findings, select appropriate visual formats and produce diagrams, infographics, charts or data narratives that communicate results clearly.
INTERPRET · VISUALISE · COMMUNICATEBEFORE YOU START
Come prepared for Module 4
- Bring the findings or dataset you want to visualise. Even rough numbers or a summary table is enough.
- Decide who your audience is (academic/technical vs general) before choosing a chart type.
- If practising bibliometric mapping, export a reference list (BibTeX/RIS/CSV) from Module 1's Publish or Perish exercise to import into VOSviewer or Bibliometrix.
Accuracy over aesthetics: A visually appealing chart that misrepresents the data (distorted axes, cherry-picked scale) is worse than a plain, honest one. Always sense-check what a chart implies before publishing it.
Scan to open this module on your phone
drshahizan.github.io/bdc/module-4-data-visualisation.html
Trainer's Tip, Dr. Shahizan
I choose the chart type before I choose the colours. Decide what comparison or relationship you want the reader to see first, then pick a chart that shows exactly that. AI tools are fast at making pretty charts and slow at knowing what you are trying to say.
EXERCISE 4.1
Chart-Type Decision Helper ~10 min
Describe what your data needs to show. The helper recommends suitable chart types and explains why, so you can brief an AI tool or design software with confidence.
WORKED EXAMPLE
See a completed recommendation
Sample input: Goal: Trend over time · Variables: 2 variables · Audience: General audience
EXERCISE 4.2
Spot the misleading chart ~10 min
The same data can look completely different depending on how the axes are drawn. Both charts below plot the exact same two numbers, Tool A at 82 and Tool B at 78, a real difference of only 4 points.
The y-axis starts at 75, not 0. That makes Tool B's bar look less than half of Tool A's, when the real difference is only 4 points.
The y-axis runs the full 0 to 100 scale. Tool A and Tool B now look as close as they actually are.
- Does the y-axis start at zero? If not, is the break clearly marked and justified?
- Is the scale consistent across every bar, point or panel in the chart?
- Are the categories being compared actually comparable (same time period, same population, same unit)?
- Would the chart's headline takeaway change if you redrew it with an honest axis?
Apply this to your own charts: Run every chart you plan to present, including one an AI tool generated for you, through these four questions before you publish or present it.
EXERCISE 4.3
Recreate a chart from your own data ~15 min
Use the Chart-Type Decision Helper above with real numbers instead of a hypothetical scenario.
- Pull a small dataset you already have, for example the number of matching papers per year from your Module 1 Publish or Perish search, or your own study's results.
- Fill in the table below with your actual numbers.
- Decide your goal (comparison, trend, proportion, relationship, distribution) and run it through the Chart-Type Decision Helper above.
- Build the recommended chart in a tool of your choice (Excel, Canva, or a bibliometric tool from below), and check it against the four questions from Exercise 4.1.
| Year | Count |
|---|---|
| e.g., 2021 | e.g., 12 |
| e.g., 2022 | e.g., 19 |
| e.g., 2023 | e.g., 31 |
EXERCISE 4.4
Generate and fact-check a Studio infographic ~15 min
Gemini Notebook Studio can turn your uploaded sources into a polished infographic in seconds. The exercise is not making the infographic, it is checking it.
- Upload 3 to 5 verified sources on your topic into Gemini Notebook, then open Studio and generate an infographic. See Gemini Notebook Studio Prompts for ready-made generation prompts.
- List every number, statistic or claim shown in the infographic.
- Check each one against the actual uploaded sources. Mark it Correct, Wrong or Not traceable to a source.
- Fix or remove anything Wrong or Not traceable before you would consider using the infographic anywhere real.
| Claim or number in the infographic | Check result |
|---|---|
| e.g., "68% of respondents reported improved efficiency" | e.g., Correct, matches Table 2 in Source A |
A pretty graphic is not evidence: An infographic is only as trustworthy as the numbers behind it. Never present or submit an AI-generated visual you have not personally verified against the source data.
RECOMMENDED TOOLS
Tools for bibliometric mapping and scientific communication
Once you know the chart type you need, these tools help you build bibliometric network maps, topic overviews and polished scientific visuals.
VOSviewer
Constructs and visualises co-authorship, citation and keyword co-occurrence networks.
Visit VOSviewer →Bibliometrix
R-based toolkit for comprehensive bibliometric analysis and science mapping.
Visit Bibliometrix →Gephi
Open-source platform for visualising and analysing complex networks.
Visit Gephi →Open Knowledge Maps
Open platform that creates visual overviews of research topics from scholarly literature.
Visit Open Knowledge Maps →BioRender
Creates professional scientific illustrations for biology, medicine and the life sciences.
Visit BioRender →Canva
Creates presentations, posters, infographics and other visual research communication materials.
Visit Canva →RELATED TOOLS & PROMPTS
Build the visual once you know the chart type
AI Tools for Visualisation
Curated tools for charts, diagrams, infographics and data storytelling.
Explore Visualisation Tools → 24 STUDIO PROMPTSGemini Notebook Studio Prompts
Generate a source-grounded infographic or slide deck from your verified sources.
Explore Studio Prompts → CUSTOM STUDIO OUTPUTBuild a Gemini Notebook Studio Prompt
Create a concise brief for the exact visual output you need.
Build a Studio Prompt →WHY USE AI HERE
Traditional vs AI-assisted visualisation
| Task | Traditional approach | AI-assisted approach |
|---|---|---|
| Choosing a chart type | Guess based on habit or what's familiar | Chart-Type Decision Helper recommends types based on your actual goal |
| Mapping citation networks | Manually tally co-citations and co-authorships | VOSviewer or CiteSpace generate the full network automatically |
| Getting a topic overview | Read through papers and sketch a summary by hand | Open Knowledge Maps generates a visual topic overview from the literature |
| Producing a polished figure | Rebuild formatting manually in a general design tool | Canva or BioRender offer research-ready templates for posters and figures |
The trade-off: AI speeds up chart selection and network mapping; you still own the accuracy and honesty of the final visual.
RECAP
Key takeaways and terms
Key Takeaways
- Match the chart type to your goal (comparison, trend, proportion, relationship, distribution, process or geography), rather than choosing a chart first and forcing the data to fit it.
- Bibliometric tools (VOSviewer, CiteSpace, Gephi, Bibliometrix) reveal patterns across a whole literature, not a single study.
- Keep axes honest and labels clear. A chart should clarify data, not distort it.
- Studio prompts can generate a source-grounded visual directly from your verified papers.
- Co-citation network
- A map showing which papers are frequently cited together, indicating related themes.
- Bibliometric analysis
- The quantitative study of publication and citation patterns.
- Choropleth map
- A map shading regions by a numeric value to show geographic patterns.
- Data storytelling
- Presenting data with a narrative and clear takeaway, not just raw numbers.
SELF-CHECK
Test your understanding
5 MODULES · 1 DAY
See how Module 4 fits into the full programme.
Review timings for every module, breaks and the closing wrap-up.
View Full Schedule