ACCA Strategic Professional · Advanced Performance Management · Data science and analytics
Brightwell Co's sales dashboard uses a bar chart whose vertical axis starts at 95 rather than 0, so a rise in customer satisfaction from 96 to 98 appears to be a fourfold increase in bar height. What is the principal problem with this design?
The truncated axis exaggerates the difference between values and can mislead users. Because bar length encodes magnitude, starting at 95 makes a small two-point improvement look like a dramatic increase. Bar charts can show scores, and hiding data until audit is unnecessary.
- AThe truncated axis exaggerates the difference and may mislead usersCorrect
- BThe chart should show only one bar to avoid comparison
- CBar charts cannot display percentages or scores
- DThe data should be hidden until the year-end audit
Explanation
Bars encode value by length, so starting the axis at 95 makes heights 1 and 3 units, visually a threefold-plus difference for a 2% rise. This distorts perception. Bar charts can show scores, and comparison is their purpose.
Did you get it right without looking?
One question tells you little. A timed set on Data science and analytics shows your real accuracy, how long you take and where you lose marks.
More Data science and analytics questions
- Harbor Logistics uses an algorithm trained on five years of historical manager ratings to recommend annual pay rises. Analysis shows female …
- Zentra Retail's finance director wants a single-screen display for regional managers showing a handful of key performance indicators, update…
- Brindle Airlines has a model that forecasts seat demand per flight. It now adds an optimisation engine that, given forecast demand, fleet co…
- Marlow Telecom wants to use customer call-centre recordings, collected for complaint handling, to build a model that predicts staff bonuses …
- Marlowe Foods links part of its regional managers' bonuses to a dashboard of ten KPIs. Managers complain that the dashboard is cluttered and…
- Orchid Logistics built a model to predict late deliveries. It achieves 99% accuracy on the training data but only 62% on new data it has not…