Politicians of all parties have used statistics to back up their points over the past few weeks and months or to pull apart their opponents’ arguments. But how can we work out whether to believe the figures and what do they really mean?

Statisticians, journalists and scientists have launched Making Sense of Statistics, a guide that provides a few questions you can ask and outlines the pitfalls to look out for when weighing up claims that use statistics. As more and more data and statistics are wheeled out on the eve of the General Election, we have to take a step back and decide what it all means.

Here is an interesting take on the science behind statistics that all Media Editors should read. No doubt it would get in the way of many a good headline:

http://www.senseaboutscience.org.uk/PDF/MSofStatistics.pdf

Section 1: If a statistic is the answer, what was the question?
Statistics are the product of conscious choices: what to count, how to count it and how to express the results. To understand them we need to consider what choices were made when the study was designed, ask how big the sample was, how it was chosen and, in making projections or forecasts, what assumptions were used.

Section 2: Common pitfalls
There is more than one type of average; each one can give a different answer and we need to find out why a particular one was used. To make a story more dramatic, people regularly use the most extreme number from a range of possible values, that is, an outlier – a possible but not very likely value.

Section 3: How sure are we?
Statisticians check if a result is consistent with
chance or if it is ‘statistically significant’. Even if a result is statistically significant it doesn’t mean it is practically significant or of importance to society. Confidence intervals give the scale of potential uncertainties in counting, measuring or observing data. Just because there has been a run of events – deaths at a hospital, accidents on a road, draws in football matches – it doesn’t necessarily mean that something beyond chance has caused it.

Section 4: Percentages and risk – Knowing the absolute and relative changes
To understand the importance of any increase or decrease we need to know both the absolute and relative change. To know if a change in risk matters to an individual we need to know what the risk was to begin with.