Tables and graphs are visual ways of presenting and summarising public health data. They are used to share data in a way that the people viewing can quickly see trends or patterns in the data without having to read a lot of text.

Although tables and graphs are often included in reports or documents, they should always be able to be understood and interpreted on their own. They should always include:

• A title that describes ‘what’, ‘where’ and ‘when’ of the data presented. In field epidemiology, we often refer to this as the person, place and time of the data.
• The title should also include a number (based on the order of the tables/graphs in the document). When you refer to the table or graph in the document, you use this number (e.g. Table 1, Figure 1).
• Clear and concise labels that include the units of measurement. In a table, every row and column must be labeled. In a graph, the x (horizontal) and y (vertical) axis must have labels.
• Explanation of all abbreviations and symbols, often in a footnote.
• Explanation of any exclusions from the dataset in a footnote.
• The source of the data, unless this is already clear in the document (e.g. in a methods section).
• The axis of a graph that presents counts (almost always the y axis) should always start at zero and should not include any breaks (i.e. the distance between axis marks should always represent the same value).
• If a table presents data from a group broken into smaller parts, include a total for the group.
• If a graph presents data separated into categories, a key (also called a legend) that explains what each category is and the colour and shape that represents them in the graph.

Tables

A table shows descriptive data presented in rows and columns. An example of a table with all required components is shown below.

Line graph

A line graph is a visual way to show changes, patterns or trends over time.

Time is represented on the x axis and the number or rate is represented on the y axis. By using multiple lines (with different colours or patterns) we can compare multiple categories on the one line graph.

Below is an example line graph with the key required elements.

Bar chart

A visual way of displaying categorical data, where each bar represents a different category, such as age group or sex. It helps to visually show a comparison between different categories.

The length of the bar corresponds with the number in the category it relates to. Bar charts are the only type of graph where the counts or rate is presented on the x axis. The categories are presented on the y axis – that is, the bars will run horizontally. The bars are not touching as they represent separate, distinct categories.

Below is an example bar chart with the key required elements.

Column chart

Columns charts are very similar to bar charts, except the bars are displayed vertically. Each category or value of the variable is represented by a column. The y axis shows the number of cases and the x axis shows the different categories of data that can be presented. There are gaps between the columns because the y axis represents separate, distinct categories. There are four types of column charts (simple, grouped, stacked, 100%) and the type that you choose to use depends on the data and desired emphasis.

Below is an example of a simple column chart with the key required elements.

Grouped column chart

A grouped column chart illustrates data grouped by two or more categories. It is useful to display data in a way that enables easy comparison between sub-sets of a group. Within each group, the bars are touching. The bars of separate groups are not joining. It is also possible to make a grouped bar chart in the same way (but with the bars running horizontally).

Epidemic curve (epicurve)

An epicurve is a very common graph in field epidemiology. It is used routinely in surveillance and outbreak investigations to quickly visualise and understand disease spread over time.

An epicurve uses a type of graph called a histogram, which looks very similar to a column chart but the columns are always next to each other with no space in between. They plot the number of cases on the y (vertical) axis by date/time of onset on the x (horizontal) axis. The x axis usually represents the date/time that a person became sick (their symptom onset date/time). If there are asymptomatic cases (which means they have no date of symptom onset), the case is entered on the date of testing/presentation.

The height of each column represents the number of cases for each time point on the x axis. Each column represents one unit of time. The units you choose will depend on the disease being analysed and the speed at which it spreads (e.g. gastrointestinal illnesses might be displayed in hours, whereas influenza cases might be displayed in days).

Epicurves should have whole numbers on the y axis (not decimals) because it represents counts of people; e.g., In increments of 1, 10, 100, etc (depending on how many cases); not 0.5, 1.5 etc. Below is an example of an epicurve with the key required elements.