Data Visualisation, Presentation and Communication
Choosing the right chart
The chart must match the data and the message. Bar charts compare categories; column charts do the same with vertical bars. Line charts show trends over time. Scatter plots reveal relationships between two numeric variables. Pie charts show parts of a whole and only work with a small number of categories summing to 100%. Histograms show the distribution of continuous data. Heat maps show intensity across two dimensions. Choosing a pie chart for ten categories, or a line chart for unordered categories, is a classic error.
Design principles
Good visualisations have a clear title, labelled axes with units, a legend where needed, a stated data source, and consistent scales. Avoid chartjunk - unnecessary 3D effects, gradients and gridlines that distract. A truncated axis that does not start at zero exaggerates differences and can mislead; this is a key trap in questions asking why a graph is misleading. Colour should be accessible to colour-blind viewers and used meaningfully rather than decoratively.
Communicating findings
Audiences differ. A technical audience wants method, sample size and limitations; an executive audience wants the headline finding and recommended action first. Structure presentations around the decision being supported, not the order in which analysis was performed. State uncertainty and limitations honestly - omitting them is a form of misrepresentation.
Worked example
A report claims sales "soared" using a bar chart whose vertical axis starts at $80,000 rather than zero, making a 3% rise look dramatic. The fix is to start the axis at zero or use a line chart with a clearly labelled scale, and to state the actual percentage change in the caption.