Graphs - 2D Graphs - Scatterplots

Ribbon bar. Select the Graphs tab. In the More group, click 2D and from the menu, select Scatterplots to display the 2D Scatterplots dialog box.

Classic menus. From the Graphs - 2D Graphs submenu, select Scatterplots to display the 2D Scatterplots dialog box.

Two-dimensional scatterplots visualize a relation (correlation) between two variables X and Y (e.g., weight and height). Individual data points are represented in two-dimensional space, where axes represent the variables (X on the horizontal axis and Y on the vertical axis). The two coordinates (X and Y) that determine the location of each point correspond to its specific values on the two variables.

Numerous types of X-Y scatterplots are available: Regular, Double-Y, Multiple, Frequency, Bubble, Voronoi, and Quantile. Regular scatterplots are simple scatterplots of one set of X-Y variables. Double-Y plots involve one X variable and two Y variables. One Y variable is plotted against the left y-axis scale and one is plotted against a separate right y-axis scale. Different point markers are used for the two plots. Multiple scatterplots have one X variable and multiple Y variables. All Y variables are fit to a single y-axis scale, but each variable is represented on the graph by a different point marker. Frequency plots are regular X-Y scatterplots in which the size of point markers is proportional to the number of superimposed points at that set of X-Y coordinates. Bubble plots are regular X-Y scatterplots in which the size of point markers reflects the value of a third (weighting) variable. In Voronoi plots, the space between each plotted point marker on a regular scatterplot is divided by straight lines that define the areas surrounding each point such that any set of X-Y coordinates within the defined space is closer to its center point than to any other point. In Quantile scatterplots, the quantiles of the X variable are plotted against the quantiles of the Y variable. If the resulting data points fall on a straight line, the two variables follow the same distribution.

STATISTICA provides a variety of fit types for the graph: Linear, Polynomial, Logarithmic, Exponential, Distance Weighted LS, Neg Expon Weighted LS, Spline, and Lowess. Note that the selected function is applied to each set of X-Y variables (e.g., a Double-Y plot has two fit lines applied, one for each y-variable. Options to mark selected subsets of the data are available for Regular, Quantile, and Voronoi type plots.

See also, Conceptual Overviews - 2D Scatterplots.