5.1 Statistical analysis

There are examples every day in the news or social media where statistics are used to make a claim that seems improbable. When using or interpreting statistics, it is important to take care to check that the information is reliable and presented in a clear and unbiased way. To do this, you need to understand the context of the information, how it was collected and the limits of what it shows.

Figure 18 Analysing statistics involves careful checking of the data

Show description|Hide description

Figure 18 is a collection of reports containing statistics, bar charts and pie charts. A magnifying glass sits on top of the reports.

Figure 18 Analysing statistics involves careful checking of the data

A good starting point is identifying which of the two forms of statistics you are looking at (or wish to use). Descriptive statistics describes and summarises the data and what it shows – for example, a current trend. Inferential statistics uses the data to make predictions about the future (e.g. future trends) or takes the data collected and applies it to a larger population to draw conclusions.

When looking at statistics, the following questions will help you to understand their context and reliability:

• What is the statistical sample used? Usually, larger samples are more reliable than smaller samples.
• Where has the information come from, and is the source reliable?
• Does the visual representation use clear scales and labels? For example, is the same scale used on graphs, particularly when comparing information?
• Have the statistics been independently reviewed? Could the author be biased, and have they mistakenly or deliberately used the statistics to support a particular viewpoint?
• Has a causal link been established if this is being claimed? A correlation between two sets of data does not necessarily mean there is a causal connection between them (The Open University, no date a).