Overall considerations Choosing a chart type
Overview
Picking the right chart type is important for communicating the insights in your data.
When choosing a chart type, consider both the type of data you have and the specific relationships or trends you want to highlight.
If you need further help, or have any questions, contact digitalcontent@ons.gov.uk (opens in a new tab)
Common relationships in data visualisations
There are eight relationships that data visualisations commonly convey. Consider the important trend or comparison you want to convey to the user and use the chart type that best illustrates it.
Some chart types can illustrate different relationships depending on the data used. For example, a bar chart might show magnitude, parts of a whole, distribution, or change over time.
Correlation
The relationship between two variables.
Deviation
The difference between a value and an average or another value.
Distribution
How data values are spread for a variable.
Geography
The pattern of data across different locations or areas.
Magnitude
The size of values.
Parts of a Whole
The relative sizes of components within a whole.
Ranking
The position of data within a hierarchy or scale.
Time
How a value changes over time.
Choose simple and familiar charts
Use simple charts that will be familiar to users whenever possible.
Complex or unfamiliar charts can be difficult to interpret and may obscure the relationships you are trying to display.
Charts that show multiple relationships
Some chart types can display more than one relationship at the same time.
For example, an area chart can show both the parts of a whole and how they change over time. A connected-dot plot can illustrate both the magnitude of two variables and their deviation.
Choose a chart type that best illustrates the most important relationship you want to convey.
Charts showing multiple relationships are often less clear than charts focused on a single relationship.
If necessary, use multiple charts to present different relationships, as two simple charts can be more effective than one complex chart.
Consistent baselines
When comparing data, it is often easier to interpret charts where components start from a consistent baseline.
For example, it can be difficult to compare individual components in stacked charts if they do not share a common starting point.
Consider using charts like line charts, where users can easily compare trends.
Example of a chart without a consistent baseline for categories
In this example, it is hard to compare the relative size of components and data may be misinterpreted.
Example of a line chart with a consistent baseline for categories
In this example, it is easier to compare the relative size of components and see the trend described in the chart title.
Exceptions to using consistent baselines
Using a consistent baseline may not always the best choice.
A stacked chart may be more effective when dealing with large value differences, when precise comparison is less important, or when you want to show the size of a total.
Start with the big picture
Where you do need to use more a complex visualisation, it can be helpful to provide a simple overview of the data first. This helps build the users’ understanding of the data before diving into more detailed or complex charts.
For example, if comparing the breakdown of housing types for different regions, first present the national breakdown before looking into regional differences.
Stock or flow data in change over time charts
Time series data can be categorised as either stock data or flow data.
Understanding whether you have stock or flow data can guide your choice of chart type. For example, a line chart might show total public debt (stock data), while a column chart shows the monthly budget deficit (flow data).
Stock Data
This data shows the value of something at a specific point in time, for example, total wind turbines in the UK. It remains constant until updated.
Stock data is typically better suited to line or area charts.
Flow Data
This data measures occurrences or rates of change over a specific time interval, for example, number of wind turbines built in the UK last year. It resets to zero at the start of each time interval.
Flow data is often best represented with a column chart, where each column represents a specific time period.