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.
Resources
Representing data
Produced by the Learning and Skills Improvement Service (LSIS). It includes a unit on Correlation and scatter graphs (building on knowledge learners already have). To enable students to explore and develop their understanding of what is meant by correlation. Students are given the cards cut-up and place them face up in front of them. They then take it in turns to match a card to any of the others. They should explain/justify why they have made the match. This activity could be used to explore student’s prior knowledge about scatter diagrams and correlation. Students could also be asked to design additional cards of their own and these could be used to further discuss the concept of cause and effect.
Let's compare feet
This resource contains a starter activity in which Key Stage Four which explores data analysis and what conclusions can and cannot be drawn from scatter graphs. This can be used to discuss correlation, trends, continuous and discontinuous data, extrapolation and lines of best fit.
Olympic triathlon
This activity uses seeks to answer the question” Do you think the triathlon will be won by someone who is very strong in one event and average in the other two, or someone who is strong in all three disciplines?” This includes the use of scatter diagrams to explore connects within the data.
This question provides a nice example of using 'real' data and all the complications that this involves but with only three data sets it is manageable and so accessible to students
Solution here
You can use the link below to access some data sets from the 2012 London Olympics if you wished to develop this type of activity further.
http://www.theguardian.com/sport/datablog/2012/aug/06/team-gb-medal-winners-background
Gapminder HIV chart
This is a rich source of large complex data sets. The interactive visualisation present unusual examples of scatter diagrams. This is a peculiarly complex example of a scatter diagram which could be used to discuss why correlation and cause and effect is rarely straight forward in real life.