University of Sydney · FACULTY OF STATISTICS

QBUS5001 Chap.1 Data Visualisation, Communication and Ethics

- one subject, every graph, every model, every mark
6 Chapters4-page Bible
Our own words - no uploaded lecturer files
Updated for this semester
Chapter 1 of 10 · QBUS5001

Data Visualisation, Communication and Ethics

This chapter turns chart design into a statistical decision. It links variable type and business question to an appropriate display, then checks axes, scales, denominators, annotations and missing context. The central habit is to make a visual claim reproducible and ethically bounded: show what the data support, state what could change the conclusion and avoid upgrading association or timing into causation.

A categorical comparison usually needs position or length on a shared baseline, while a numerical distribution needs a display that preserves shape, centre, spread and unusual observations. Time on the horizontal axis supports a trend story only when intervals and measurement rules remain comparable. A bivariate chart must reveal which variable is placed on each axis and whether colour or size introduces a third variable.

Ethical communication begins before styling: define the population, denominator and time window, disclose missing or excluded records, and test whether a truncated scale or selective range reverses the visual impression. Reproducibility means another analyst could rebuild the chart from the same data and choices. The final caption should state the supported pattern, not a stronger causal story that the design cannot identify.

In this chapter

What this chapter covers

  • 01

    Selecting charts from the number and type of variables

  • 02

    Using position, colour and size as visual channels

  • 03

    Checking axes, units, denominators and aggregation

  • 04

    Identifying misleading scales and omitted context

  • 05

    Documenting exclusions and reproducible steps

  • 06

    Separating observed patterns from causal claims

Worked example · free

Repair a misleading store comparison

Q [4 marks]. Store A has 480 returns from 24,000 orders and Store B has 300 returns from 10,000 orders. Compare quality and propose a chart. The mark allocation used here is not a University assessment scheme.
Chart choice follows the variable structureArrows connect one or two variables to suitable displays and then to a communication check. Questionwhat varies? One variablebar or histogram Two variablesscatter or groups Communication checkscale, labels, contextethics, reproducibility
The arrows move from the number and type of variables to a suitable chart, then to the scale, context, ethics and reproducibility checks that make the display defensible.
  • +1Compute comparable return rates: Store A is 2.0% and Store B is 3.0%.
  • +1Explain that raw counts answer workload, while rates answer return propensity under unequal exposure.
  • +1Use a common zero-based percentage axis and label both numerator and denominator.
  • +1Conclude that Store B has the higher observed rate and request comparable product-mix context before attributing cause.
Store A has the larger count, but Store B has the higher return rate. A rate chart with order volumes printed beside each store supports the quality comparison.
Sia tip — Ask what the denominator represents before choosing a chart or writing better and worse.
Glossary

Key terms

Visual Channel
A graphical property such as position, colour or size used to encode a variable.
Truncated Axis
An axis that begins away from zero and may visually amplify a small absolute difference.
Reproducible Analysis
An analysis whose documented data and steps allow another analyst to obtain the same result.
FAQ

Data Visualisation, Communication and Ethics FAQ

How do I choose between a bar chart and histogram?

Use a bar chart for categorical counts or proportions and a histogram for the distribution of one numerical variable. Histogram bars represent adjacent numerical intervals, so their order and width carry meaning that ordinary category bars do not.

When is a zero baseline important?

A zero baseline is especially important for bar lengths because readers compare their magnitude directly. For a line chart, a restricted range can reveal variation, but the scale and any break must be explicit so the visual effect does not overstate the business change.

What makes a visualisation ethical?

An ethical display preserves the relevant denominator, scale, units, exclusions and uncertainty, and it uses causal wording only when the design supports it. It also gives enough source and method detail for another reader to audit how the picture was produced.

Study strategy

Exam move

Practise chart selection from short business prompts before opening Excel. For each prompt, name the variables, their measurement types, the comparison or trend of interest and the denominator. Sketch the axes, then write the strongest conclusion the design supports and one alternative explanation. Finish by checking scale, labels, source, missing values and any excluded observation.

Build a small audit routine around every visual. First, identify the decision the reader is expected to make and decide whether the chart actually displays the relevant comparison. Second, trace each visual channel back to a variable and check that legends, units and category order make the mapping unambiguous. Third, inspect the raw range so an axis choice cannot hide variation or magnify noise.

Fourth, reproduce one summary directly from the data, such as a median or group count, to catch filtering errors. Then rewrite the caption twice: once as a purely descriptive claim and once as a cautious decision implication. Comparing those versions trains the boundary between evidence and persuasion.

Keep one misleading redesign beside the corrected version so the ethical cost of a scale or denominator choice remains visible during revision. For a final audit, compare the chart with a plain table of the same values. If the visual suggests a stronger ranking, change or discontinuity than the table supports, inspect the scale, binning and category order.

Record one sentence about uncertainty or data quality so the communication remains useful without pretending the sample is a complete view of the market.

Working through Data Visualisation, Communication and Ethics in QBUS5001? Sia is AskSia’s AI Statistics tutor — ask any QBUS5001 Data Visualisation, Communication and Ethics question and get a clear, step-by-step explanation grounded in how QBUS5001 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.

A+Everything unlocked
Unlocks this Bible + all 202 of your subjects at University of Sydney, plus 1,000+ more Bibles across the AskSia library.
Sia - your QBUS5001 tutor, unlimited, worked the way the exam marks it
The full 4-page Bible + practice bank with worked solutions
Chrome extension - sync your LMS so Sia knows your deadlines
Bilingual EN / Chinese on every Bible and every Sia answer
$0.99 Trial
30-day money-back · cancel in one tap · how it works
QBUS5001 · Foundation in Data Analytics for Business - independent study guide on the AskSia Library. More University of Sydney subjects · Microeconomics across all universities