2D Scatterplots - Advanced Tab

Graphical Analytic Techniques

The Advanced tab of the 2D Scatterplots dialog contains a variety of options for creating 2D scatterplots.

Variables. Click the Variables button to display the standard variable selection dialog, in which you select the variable(s) to be plotted. When you select more than one variable for one of the univariate graphs types (i.e., Regular, Frequency, or Voronoi), then a sequence of graphs is produced (one for each of the variables). The selection that you make will then be displayed in the area of the dialog below the Variables button. If you select marked subsets, then the method of categorization selected will apply to all scatterplots.

Graph type. Select the type of 2D scatterplot to be plotted from the Graph type list. Click the type of the plot link listed below for a brief description of that type of graph.

Fit. You can fit an equation to the points in the line plots by selecting one of the predefined functions in this box.

Statistics. You can include a variety of statistics as footnotes in the graph by selecting one or more of the check boxes under Statistics.

R square. Select this check box to include the R-square for the plotted variables.

Correlation and p. Select this check box to include the correlation for the plotted variables and its corresponding p-value.

Regression equation. Select the Regression equation check box to include the regression equation in the plot.

Ellipse. Use the options under Ellipse to superimpose an ellipse on the data in the scatterplot. You can display one of two types of ellipses (Normal or Range, see below), or leave it off (select Off).

Normal. Select this option button to produce an ellipse based on the assumption that the two variables follow the bivariate normal distribution. The orientation of the ellipse is determined by the sign of the linear correlation between the two variables (the longer axis of the ellipse is superimposed on the regression line). The ellipse shows the prediction interval for a single new observation, given the parameter estimates for the bivariate distribution computed from the data, and the given N. Note that if the number of observations in the scatterplot is small, then the prediction interval may be very large, exceeding the area shown in the graph for the default scaling of the axes. Thus, in some cases (with small N) you may not see the prediction interval ellipse on the default graph (change the scaling to show larger intervals for the two variables in the plot). For additional information see, for example, Tracy, Young, and Mason (1992), or Montgomery 1996; see also the description of the prediction interval ellipse.

Range. Select this option button to produce a fixed size ellipse such that the length of its horizontal and vertical projection onto the x- and y-axes (respectively) is equal to the mean ± (Range * I) where the mean and range refer to the X or Y variable, and I is the current value of the coefficient field.

Coefficient. Specify the coefficient that controls the ellipses described above in this box.

Regression bands. Use the options under Regression bands (applicable when Linear or Polynomial is selected as the Fit) to display Confidence or Prediction bands around the fitted (regression) line. You can enter the probability value (Level), which represents the probability that the "true" fitted line (in the population) falls between the bands. The standard error for the fitted line (which represents the predicted values, given the respective linear or polynomial fit) is computed based on the polynomial regression model (it is assumed that the data and their polynomial transformations are normally distributed, e.g., see Neter, Wasserman, & Kutner, 1985, p. 246).

Mark Selected Subsets. Click this button to display the Specify Multiple Subsets dialog, in which you can specify selection conditions that will subset (i.e., categorize) the cases in one plot. For more information about this option, see Mark Selected Subsets.