Tip #74: Scatter Plots – useful for understanding the relationships between two variables

May 6, 2009 4190 Comments2009-05-06+11%3A04%3A17Ben+Nash

Suppose, as a Human Resources person, you want to understand the reasons for employee no-shows at work. Because you live in Vermont you suspect that the level of snow fall might have an influence on this but you are not sure. Well, have no fear – we have the tool for you to use – scatter plots. To create a scatter you would simply track the inches of snow fall each day for a month or two and record it on the vertical axis of the chart. Then track the number of no-shows for each day and record this on the horizontal axis. When you have thirty or forty dots on the chart you may be able to see a connection between the number of inches of snow that fell and the number of no-shows. You would look for an increase in the number of no-shows as the snowfall increases, and if you see this, you have what is known as a positive correlation between the two variables (snowfall and no-shows). A perfect positive correlation is given the value of 1 (every time the snowfall increase by x amount the number of no shows increases by y amount). If the opposite were true and if the snow fall increases by x, the no-shows decrease by y you would have a negative correlation. A perfect negative correlation is given the value of -1. If there does not seem to be a connection between to two variables then the correlation would be said to be zero. The closer the number is to 1 or -1, the stronger the correlation, or the stronger the relationship between the variables. The closer the number is to 0, the weaker the correlation. So something that seems to kind of correlate in a positive direction might have a value of 0.67, whereas something with an extremely weak negative correlation might have the value -.21.