SOST10031 Chap.7 Data Visualisation and Chart Critique
Data Visualisation and Chart Critique
Data Visualisation and Chart Critique follows timetable week 9, whose seminar focuses on critiquing charts and improving their clarity and accessibility. The chapter starts from a display the module itself uses: the percentage of Andorran survey respondents classified as biased on each of four GSNI dimensions, where one person can appear in several bars and the heights therefore cannot be totalled.
From there it builds a working routine for any chart you meet or make. Decide what a single bar, box or point stands for. Read where each axis begins and ends. Ask whether categories overlap, whether the bin width hides shape, and whether colour is carrying meaning that a label should carry. Finally, hold the title and caption to the evidence on the page.
The routine matters for the portfolio because figures are excluded from the word count yet still have to communicate a finding honestly to a specified reader, alongside your written explanation.
What this chapter covers
- 01
Overlapping categories that cannot be summed
- 02
Choosing a first display for a question
- 03
Axis ranges and truncated scales
- 04
Histogram bin width and hidden shape
- 05
Direct value labels and informative captions
- 06
Accessible colour and readable labels
- 07
Titles that claim no more than the chart shows
Worked example · free
Show why Andorra's GSNI dimension bars cannot be added
- 1Add the four dimension shares: 15.8 + 2.6 + 21.2 + 23.6 = 63.2.
- 2Compare with the share showing at least one bias: 63.2 exceeds 43.13 by 20.07 points, which is impossible unless some respondents are counted in more than one dimension.
- 1Present the shares as separate bars on a 0 to 100 axis, label each value, and state in the caption that respondents may appear in several bars.
Key terms
- Overlapping categories
- Groups that can share members, so their sizes cannot be added to give a total. GSNI dimension shares behave this way.
- Bin width
- The span of values covered by each bar of a histogram. Wider bins smooth away detail, while narrower bins reveal clusters and gaps.
- Truncated axis
- An axis that starts above zero, which can make a small difference between bars look large.
Data Visualisation and Chart Critique FAQ
Why can the GSNI dimension percentages not be added together?
Each dimension share counts respondents classified as biased on that dimension, and one respondent can qualify on several. Adding the shares double counts those people, so the total describes no real group. Only the at-least-one and no-measured-bias figures form a complementary pair that reaches 100%.
What makes a chart misleading even when every number is correct?
Design choices such as a cropped axis, coarse histogram bins, stacked overlapping groups or an overconfident title can distort the impression a reader forms. Comparing the visual ratio with the numerical ratio is a quick way to expose the distortion.
How should a caption be written for a portfolio figure?
State what the axes or bars represent, name the place, measure and period, and point out anything a reader could misread, such as overlapping groups or an earlier observation year. One or two complete sentences usually suffice.
Which display suits a question about how one measure is spread?
A histogram suits a single numerical measure; try more than one bin width before settling. A boxplot summarises the middle half and unusual values compactly, while bar charts serve comparisons of a few groups.
Assessment move
Collect five charts this week: two from the module notebooks and three from news sites. For each one, write the single question it answers, then run the critique routine in order: what one mark represents, where the axes start and stop, whether groups overlap, whether labels and caption name measure, place and period, and whether the title outruns the display.
Note the ratio between the tallest and shortest bar as drawn and as numbers, and flag any mismatch. Redraw one flawed chart on paper with a full scale and direct labels. When you prepare portfolio figures, render the Quarto file, open the PDF at normal size and check that every label is readable and that colour is never the only carrier of meaning.
Finish by rewriting each caption until it explains the figure without the surrounding paragraph.
Working through Data Visualisation and Chart Critique in SOST10031? Sia is AskSia’s AI Computer Science tutor — ask any SOST10031 Data Visualisation and Chart Critique question and get a clear, step-by-step explanation grounded in how SOST10031 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.