Visualising data - Charts and graph types

Richard May 23rd 2016

HOW TO CHOOSE THE RIGHT CHART OR GRAPH

When it comes to visualising data into graphs and charts there are many different options available to you. The purpose of choosing the right graph is to give your viewers the benefit of seeing a snapshot of the information and gleam some immediate insights without having to traul through all the numbers one by one. The sign of a well crafted graph is one that can offer all the key points from a particular data set in a matter of seconds allowing viewers to digest the key takeaways most important to them.

 

Graph and chart creation has become more of an art in recent years with an explosion in subsets of graph types used to beautify data in all its forms. There are many tools now available to help with complex and intricate graphical data visualisation. Basic charts and graphs however have been is use for hundreds of years (some historical records showing longitudinal information to and from Rome date back to the 1600's) and there are certain methods for displaying data that have stood the test of time. This post is about summing up the most common graph types that can be used to display the majority of data types that will be useful to you in every day cases.

 

Its worth pointing out that structuring your data in a certain way can lead to a skewed perception of the information (something marketeers know all too well). One of the key points to remember is that if your aim is to give a fair, unbiased representation of the data, then choosing the right graph types and scales are hugely important.

 

BAR GRAPHS

Bar graphs are used to show numbers of grouped data that are independent of each. The bars are proportional to each other and the maximum value being displayed. They can be displayed either horizontally or vertically along the y and xaxis. Each axis will either be a category of data or a numerical value.

 

Example of a bar graph

Supposing you wanted to show how many hot drinks were drank in your office over the course of a day by different staff members. Your data might look like this:

 

Bob: 5, David: 2, Jenny: 2, Sarah: 3, Michael: 1

 

Representing this data in a bar graph would give you the following:

Insights from this bar graph

You can see here that Bob is the most prolific of hot beverage drinkers in the office, given the simplicity of the data that's not huge news however with a more complex data set you can imagine how insights can be gleamed much quicker representing it this way.

 

HISTOGRAMS

Histograms are specific types of bar graphs that show the flow of continuous data. They are generally used when categories are ranges of numbers and are particularly useful when you wish to display data collected over a specific time period.

 

Example of a histogram

Now supposing you were interested in which was the most popular time of day for hot drinks to be drunk in your office. A histogram is perfect for this.

Your data:

 

Bob: 08.27, 09.32, 11.52, 13.47, 15.18
David: 09.15, 11.22
Jenny: 09.32, 11.47
Sarah: 09.32, 13.52
Michael: 10.32

 

Insights from this histogram

You can now see at a glance without looking at each of the individual times the most popular time for drinking hot drinks is between 9am and 10am and that this picks up again immediately before and after lunch.

 

PICTOGRAPHS

Pictographs are used to show data in picture format and essentially work the same as grouped bar graphs (where multiple data sets within the same group are represented alongside one another on the same range). Pictures or icons are used to represent portions of the data.

 

Example of a pictograph

Now your interested in viewing the different types of hot drinks your office staff are drinking. This can be done in a number of ways and the data is certainly suitable for a pictograph.

 

Bob: Coffee - 5
David: Coffee - 1, Tea - 1
Jenny: Coffee - 2
Sarah: Coffee - 2, Tea - 1
Michael: Tea - 1

 

Insights of this pictograph

This shows us that coffee is most popular hot drink in the office and that Bob is the biggest consumer of it, also that Michael doesn't share Bobs enthusiasm for the bean, he's more a tea kinda guy.

 

PIE CHARTS

Pie charts are used to show numbers that are directly related to each other and part of the same whole. They are generally the best way to represent percentages and are one of the quickest charts for viewers to understand.

 

Example of a pie chart

We can take the above data and use a pie chart to show the percentage difference in hot beverage choices.

 

Insights from this pie chart

We are now able to view exactly how much more popular coffee is than tea in this office

 

LINE GRAPHS

Line graphs can be used to show how numbers change over a certain time period when data points are connected. They are great for showing trends and analysing the rate at which things change. Points will either be connected through straight lines or can use a curvature to help illustrate rates of change.

 

Example of a line graph

Now, supposing you wanted to look at your companies monthly spend on refreshments over the course of a year to see which are your more expensive months.

 

Insights from this line graph

Here you can see the monthly spend rises steadily throughout the first part of the year hitting its peak in the summer months before drifting down to a sharp fall in December.

 

SCATTERPLOT

Scatterplots are used to display paired data on a y and x axis. They are useful when you want to compare how data points relate to each other on 2 ranges and are great for showing trends there.

 

Example of a scatterplot

We've seen that the kitchen spend goes up in Summer time. To drill down further into the data and spot trends on why this is happening, we could use a scatterplot to see how the weekly consumption of drinks changes depending on the average temperature recorded for that week.

 

Insights from this scatterplot

So now we can see that whilst the amount of coffee being drank goes down during hotter weeks, the bottled water consumption rises accounting for the higher monthly costs during summer.

 

FLOW CHARTS

Flow charts are used to display a process or set of actions that play out in a certain order. Flow charts are widely used in software development to illustrate user/data journeys but they can be used to describe the flow of anything such as: website navigation, workflow management and decision making.

 

They include a direction of travel and traditionally different shapes will be used to represent specific actions.

 

Example of a flow chart

So lastly, you need to decide who's turn it is to make the brews! An important decision like that of course can be made with the help of a flow chart...

 

Author Richard Creative Director

Richard is a designer, whiskey drinker, music lover and Creative Director. After an early career in publishing saw him oversee the production of a range of global titles he turned his eye to digital, managing several design teams in UK agencies before founding Gather with fellow director Paul.