If you include a third variable, look for a relationship between that variable and the x- and y- variables. Minitab colors bins based on the average value of the third variable for all the observations in that bin. If there is no relationship, the color of the bins will be randomly scattered throughout the binned scatterplot. A pattern in the colors indicates a relationship may exist.
In this example, the research team wants to determine how the carat and color affect the price of diamonds. Minitab uses dark blue to dark red for the bins based on the color of the diamond. Darker shades of red correspond to higher values of the color variable. Darker shades of blue correspond to lower values of the color variable. The price increases as the number of carats increases. However, there is still a large variation in prices for diamonds with the same number of carats. The color of the diamond explains some of this variation. For diamonds with similar number of carats, a higher color number corresponds to a higher price.
The color of a diamond is represented with a letter. Because all the variables must be numeric, the researchers codes the color of the diamond to a numeric scale that ranges from 0-6. The owner gives colorless diamonds a higher number and gives light yellow diamonds a lower number.