Purpose
A scatter diagram is used to identify the relationship between two different characteristics. The data generated will be used to confirm a supposition that two variables are related. This tool is often used as the follow up to a Cause and Effect (Fishbone ) Diagram to find out if the connection between causes and the effect are factual, rather than just a consensus.

How does it work ?

• Provides both a visual and statistical means to test the strength of a potential relationship.
• Supplies the data to confirm a hypothesis that two variables are related.

Method

1) To test a theory, collect 50 - 100 samples of data that you think may be related, then construct a data sheet.

Theory: There may be a relationship between coating thickness and dip time.

2) Draw the axis lines of the diagram, using the vertical axis for the dependent variable, or the 'effect', and the horizontal axis for the independent variable, or cause. Then plot the data.

3) Interpret the results. The results shown in the example identify that dip time has an effect on coating depth, but only up to a point. From 2 minutes to to 3.75 minutes, the coating depth increases steadily. At 4 minutes there is a rapid increase in depth, but after this period of time, the application of the coating becomes erratic with very little increase in depth.

This result is described as a positive correlation as it shows that an increase in time can affect the coating depth.

4) Other types of results.

Possible Positive Correlation If the time is increased, the coating depth increases somewhat, but there are probably other influencing factors involved.	No Correlation There is no demonstrated connection between time and depth of coating.