Misleading charts and samples
See how a cut axis changes a graph, choose a chart for a statement, judge whether a survey sample is representative, and see that the same mean can hide very different spreads.
- France: grade 4
- Germany: grade 8
- Syria: grade 5
- Palestine: grade 5
- Ukraine: grade 6
Guide
This studio is about honest statistics: how a chart or a number can give a wrong impression, and how to check it.
A value axis that does not start at 0 (the break mark shows it) makes small changes look huge. Always read the numbers on the axis and work out the real change, for example as a percentage.
Choose a chart for the statement you want to make: a pie chart for parts of one whole, a line graph for a development over time, a bar chart to compare groups.
A sample is representative when it is a small copy of the whole group. Asking only people at the football pitch, or only friends, gives a biased result. A random draw from a list of everyone is better.
An average alone can mislead: two groups can have the same mean and a very different range.