Scatter diagrams and correlation

Students need to make data handling decisions that are informed by the context they are working with and Science is an ideal opportunity to do this. Encourage students to discuss the benefits and disadvantages of different techniques. Use a range of subject specific contexts for these discussions. Provide exemplars where the data is biased or does not answer the question posed and ask students to critique them. Encourage students to continually revisit the question they are trying to answer and reflect on whether what they are doing will inform this question.

Some students may think that correlation implies causation. Students need to be confident about why correlation does not necessarily imply causality in real life situations. Science lessons are an ideal opportunity to do this. Show examples of scatter diagrams that imply causality but where this is due to a third factor. For example life expectancy and number of TV sets. Remind them that the line of best fit is used to summarise the data and also to make predictions. Some students may think that a line of best fit always has to pass through the origin

Some pupils may not see a trend because they are looking at the detail instead of the bigger picture. Help them to tell the story of the graph by using living graph activities. Print a large copy of a scatter diagram representing scientific data. Have cards that describe key points on the chart or graph. Cut these up and students have to work in teams to place them on the correct place on the graph.

In Science courses, the most likely place where scatter graphs will be used to explore the (possible) correlation between two variables in the Health, disease and the development of medicine topic of the Biology course. The example using Olympic Triathlon activity below, whilst perhaps not exactly matched to the course, provides a nice example of the kind of data/examples that may appear.