A scatter plot is utilized to depict the relationship between two variables and can analyze whether a cause-and-effect relationship exists. When using a scatter plot, commonly the variables being plotted on the y and x axes are related to each other. Furthermore, useful inferences and conclusions can be drawn when the y variable is dependent upon the x variable.
In comparison to other statistical figures, the points plotted on a scatter plot are not joined by a straight or curved line. However, in some instances the points may form a straight line. When this results, this means that there is a strong correlation between the two variables.
Using scatterplots, both positive and negative relationships can be observed. In some cases the points being plotted may just be randomly scattered and this is indicative of two variables which are not correlated at all. In addition to depicting whether a linear relationship exists between two variables, a scatter plot can also indicate whether outliers exist within a data set.
For example: Is there a correlation between age and the time spent watching TV? In this case, the time (in hours) is the dependent variable and age (in years) is the independent variable. Age is the independent variable because it changes on a yearly basis and this is not influenced by other factors. Time is the dependent variable in this case. Even though time stands alone, the amount of time spent watching TV will change during different stages of an individual’s life.
Figure 1. This scatter plot examines whether a relationship exists between the age of individuals and the time they spend watching TV. Results for individuals up to the age of 18 were recorded and the total sample size (n) = 80.
As can be observed from Figure 1, there is no correlation between the two variables being analyzed. Scatter plots provide a simple way of visually testing the correlation between different factors and thus, being able to use them properly is a valuable skill.