Topic

Categories of Chart

Topic Progress:

Click on the numbered tabs to navigate this lesson.

Categories of chart

Before you learn how to use charts as your data visualizations, you should know the various kinds of charts that exist. It’s also important you know what kind of insight can be gotten from each chart. Each kind of chart is unique and is used for a different purpose. For example, bar charts are best used for comparing values of categorical variables while line charts are best used for observing a trend, a change in data per time. 

The next couple of topics will explain each of the categories of charts, how best to use it and what it’s best used for. Understanding what a particular chat is best used for is very imperative in understanding how to perform data visualization.

Bar Charts

Purpose: Bar charts are used for comparing categories. Length denotes high or low values making bar charts ideal for presenting data in ranks.

Grouped Bar Charts

Purpose: The Grouped Bar Charts is an extension of the normal bar chart. It is used to represent and compare different categories of two or more groups. For example, to compare the death rate and birth rate of five countries at the same time, use a grouped bar chart.

The example below compares the number of ICU facilities in public and private hospitals in South Africa during the Covid-19 Outbreak, 2020.

Stacked Bar Chart

Purpose: The stacked bar chart is similar to the Grouped Bar Chart except that instead of multiple bars, you have one bar with internal sub-bars stacked end to end. Each sub-bar represents the groups.

Pie Charts

Purpose: It’s used to compare parts of a whole. They are very effective in showing the size of a category in relation to the whole. But pie charts should be discouraged unless the angles are obvious. This is because angles can be very difficult to compare with the ordinary eye view.

Donut Chart

The Donut Chart is similar to the pie chart except that it has a plain circle inside the bigger circle. It’s also used to compare parts of a whole. In the example below, the sources of drinking water in eThekwini municipality were compared. It’s a comparison of parts of a whole, just like the pie chart.

Line Graph

Purpose: They are used to observe changes over time. Line charts show trends and predict the path of a variable if not interrupted. They are particularly used in showing changes in one or more categories over the same period of time.

Histogram

Purpose: The histogram, also called a frequency distribution graph, is a specialized type of bar graph that resembles a column(bar) graph, but without gaps between the columns. It is used to represent data from the measurement of a continuous variable. Histograms give a rough sense of the density of the underlying distribution of the data.

The below example visualises the salaries of the staff of an organisation. A simple glance at the design shows that a majority of the staff earn between $800 and $1,100. It shows the density of your data, where the bulk of the earning falls in. 

Maps

Purpose: Maps can be used for comparing data values across geographical regions. They are useful for showing concentrations of certain data in different regions. Maps are used for spatial grouping by highlighting regions that belong together under a category. Imagine that you want to visualise the areas a political party won in a country election, using the map of that country will be the most suitable option. Another example could be the mapping of the locations of zoos in Africa, using the Africa map. 

The map below shows the distribution of health workers across Africa. A quick glance indicates that countries with denser colours have a higher number of health workers.

Tables

Purpose: Tables are great for just that—communicating to a mixed audience whose members will each look for their particular row of interest. Tables are usually the easiest and simplest way of presenting election data. 

The table below shows a tabular presentation of the 2015 Nigeria presidential election.

Heat map

Purpose: Heatmap is an advanced table, a step beyond the plane table. One approach for mixing the detail you can include in a table while also making use of visual cues is via a heatmap. A heatmap is a way to visualize data in tabular format, where in place of (or in addition to) the numbers, you leverage coloured cells that convey the relative magnitude of the numbers.

Examine the example below

Scatter Plot

Purpose: Scatterplots can be useful for showing the relationship between two variables because they allow you to plot one variable on a horizontal x‐axis and another on the vertical y‐axis to see whether and what relationship exists. 

For example, to check if a change in GDP affects the death rate, a scatter plot will be the best option for it. The example below compares the sales of ice cream with the temperature of the day.

When a change in one variable affects the other, we say there’s a correlation between the two variables. For example, the example above shows a positive correlation, that is, the higher the temperature, the higher the value of sales.

But it should be noted that correlation is not necessarily causation. A variable can correlate with another but may not necessarily be a factor that affects the change. Hence, correlations should be subjected to further scrutiny.

Slopegraph

Purpose: Slopegraphs can be useful when you have two time periods or points of comparison and want to quickly show relative increases and decreases or differences across various categories between the two data points. 

An example is a comparison of a company’s sales 2016 and in the year 2017,  across several regions as indicated in the diagram here.

More charts

There are more charts to be considered. Some of these charts are not as common as the aforementioned charts. They are useful for a particular purpose. They’re not also present in all visualisation tools but they can be found in Flourish platform, and are very useful.

1. Population Pyramid

Population pyramid is the standard way of showing the age and sex of a population. It’s the best and most suitable way of showing population distribution. Find below the population distribution of Nigeria as a case study, showing the percentage of males and females across different age ranges in Nigeria. 

2. Connected dot plot

The connected dot plot is similar to slopegraph, but in this case, each line is straight and parallel to each other.

It shows differences in values of two variables(only) of groups of categorical variables. In function, it’s similar to the grouped bar chart. Examine the connected dot plot below, showing the gender parity in education, politics and occupation in Nigeria.

3. Bar chart race

The bar chart race is similar to bar chart, except that the ranks are animated. It’s not a static visualization, neither can it be. It’s always interactive and the rank changes with time. An example is shown below:

4. Line Chart Race

The line chart race is similar to a line chart, except that the ranks are animated. It’s not a static visualization, neither can it be. It’s always interactive and the rank changes with time. Find below an example of a line chart race. You may click on the replay button, again and again, to see how the animation plays out.

5. Network Graph

A network graph is a chart that displays the relationship between elements. The elements are called nodes. A network graph helps us to visualize clusters and relationships between the nodes. Click on any of the nodes in the design below and observe the relationships in the network.

6. Parliament Chart

Parliament chart is a visualization that positions dots in a hemicycle shape. It’s useful for visualizing parliament layouts. An example is to visualize the distribution of lawmakers according to their political party.

Knowledge Check! (Categories of a chart)