|
|
I am pleased to announce that two new graphics
extension commands for SPSS Statistics are now available on the SPSS Community
website (www.ibm.com/developerworks/spssdevcentral).
They are in both the Extension Commands and the Graphics Tools Collections.
Both include a dialog box interface and standard SPSS syntax.
STATS REGRESS PLOTS is
a procedure helpful in studying bivariate relationships or regression residuals.
It produces a set of small plots of one y variables against
a set of x variables. Scatters are used for scale variables, and bar, line,
or boxplots for categorical ones. Options for scatters include coloring,
sizing, shaping, and labeling points by other variables, various fit lines,
and control over the size. If the y variable is a regression residual,
you might scale the points by the leverage or distance values saved from
the regression. For single scatters, you can have boxplot borders
in both dimensions.
STATS SUBGROUP PLOTS is
a procedure for graphically comparing subgroups of a dataset across a set
of variables.
It produces one or more rows of small plots for each group
in a partition of the dataset. Each plot shows the distribution in the
group overlaid on the distribution in the entire sample. The type of plot
depends on the variable's measurement level: histograms or kernel densities
for scale variables and bar charts for categorical variables.
Both commands require the Python Essentials also available
from the SPSS Community and at least Statistics version 18. Both
work mainly by generating appropriate GPL syntax.
I think these are pretty cool. I hope that you will
find them useful.
Jon Peck (no "h")
Senior Software Engineer, IBM
[hidden email]
new phone: 720-342-5621
|
Locked
