We live in a time of immense information overload. One of the best ways to make sense of, and reveal patterns in, this excess of data is to visualise it.
By shaping this ubiquitous resource we can reveal new insights that would otherwise be lost in a sea of numbers.
Data journalism involves the use of key information sets and data to inform a story. It lets you tell a story in a way that is easy for the audience to comprehend.
Data is more important today due to its abundance and the fact that we have at our disposal the tools and techniques required to analyse it and produce visualisations quickly and efficiently.
Graphs make comparing data easier, but it is essential to choose a type of graph that is appropriate for the data, rather than one that is aesthetically pleasing.
Three commonly used types of graphs are bar charts, line charts and pie charts. Bar charts are easy to understand and are highly recognisable. They are particularly effective with categorical, numerical data as they allow for quick comparisons of information and can reveal highs and lows at a glance. Line charts connect individual numeric data points, usually over a period of time, and pie charts show relative proportions or percentages of information.
Data visualisation is a useful means of representing large or complex data sets in a way that is easily comprehendible by the audience. By providing the audience with tools to analyse and make comparisons with the data, the amount of time required to understand the content is reduced.
Data visualisation is sometimes referred to as “functional art”, however there’s a fine line between successful data visualisation and superfluous infographics that don’t present the data in a meaningful way, as seen in the below image. It is also important to consider what information is included in the visualisation so as not to overwhelm the audience with unnecessary information.
There are four main types of data; nominal and ordinal, which are categorical, and interval and ratio, which are numerical.
Nominal data consists of names categories that are inherently unordered. This data can be counted to calculate percentages but cannot be used to calculate averages. When there’s only two categories the data is referred to as dichotomous.
Ordinal data refers to the order of the values being measured. No category on an ordinal scale has a truly mathematical value, but they can be used to calculate percentages.
Interval data is numeric data that doesn’t have a meaningful zero point, in that zero doesn’t indicate the absence of the value being measured, e.g. 0:00am isn’t the absence of time and 0°C isn’t the absence of temperature.
Conversely, ratio data is numeric data in which the zero point is meaningful and does indicate an absence of the value, e.g. $0 and 0ml. This is evidenced in the chart below.
We live in an extremely data oriented time, in which the amount of data we create and share is growing exponentially, as shown in the screen grab below. As a result, data visualisation strategies are vital to make sense of and organise this data.
Data can be defined as values of quantitative or qualitative variables belonging to a set of items. Data itself has no meaning until it is interpreted and visualised. Effective visualisation helps users in analysing data and evidence, aiding in accessibility, understandability and usability.
I think the most notable idea proposed in this lecture pod was that not all information visualisations are based on data, but all data visualisations are information visualisations, e.g. most process visualisations are information visualisations but not data visualisations, as evidenced in the infographic below which is essentially an illustrated list, as it lacks numerical data.