Plot Window
The plot window allows you to plot the values in two table
columns against each other.
You can display it using the Plot (
)
button when the chosen table is selected in the
Control Window's Table List.
On the plotting surface a marker is plotted for each row in the
selected Row Subsets
at a position determined by the
values in the table columns selected to provide the X and Y values.
A marker will only be plotted if both the X and Y values are not blank.
If more than one subset is being plotted, they will be drawn using
different markers. A key on the right hand side indicates the marker
being used for each subset. The marker types can be changed using the
Marker Types menu.
You can zoom in and out of the plot by dragging with the left mouse
button down and right (zoom in) or up and left (zoom out) - this takes
a little practice but is easy to use after a couple of goes.
If you get lost you can push the
Rescale button (
)
to return the scaling to normal.
Below the plot there are two sets of controls for selecting the
table column which will provide the X and Y axis values.
Each one consists of two parts:
-
Column selector
- A selection box from which column names in the main table can be
selected. Only columns which can be plotted from (i.e. scalar numeric
ones) will be displayed in this selector.
-
Log checkbox
- This checkbox can be clicked to toggle whether the axis in question
is to be plotted logarithmically or not. If it is logarithmic,
any negative values are simply ignored (not plotted).
-
Flip checkbox
- This checkbox can be clicked to toggle whether the axis in question
is to be plotted reversed. Normally (unticked) X axis increases
left to right and Y axis increases bottom to top. When ticked,
X axis increases right to left, and Y axis increases top to bottom.
To the right is a set of checkboxes headed
Row Subsets.
Click on these to choose which of the table's
defined Row Subsets should be plotted on this graph.
Different subsets are plotted
using different markers, so you can see where different groups of results
lie in relation to each other.
You can alternatively use the Subsets To Plot item on the
Subsets menu.
The subsets are plotted in order of which was most recently selected.
This makes a difference on a crowded plot or where some points are
members of multiple subsets, since the most recently plotted symbol
will appear on top. If points from one subset are being hidden behind
those from another, you can deselect and reselect that subset and
they'll be shown on top.
The following extra buttons are available on the toolbar:
-
Export as EPS
- Pops up a dialogue which will print the current plot as an EPS file.
In general this is a faithful and high quality rendering of what
is displayed in the plot window. However, if plotting is being done
using the transparent markers, it won't come out right since
transparency cannot be represented in PostScript; the markers will
be rendered as if they were opaque.
Currently, if there are many points being plotted, this can result
in a rather large output file.
-
Export as GIF
- Pops up a dialogue which will output the current plot to a GIF file.
The output file is just the same as the plotted image that you see.
Resize the plotting window before the export to control the size
of the output GIF.
-
Rescale
- Rescales the axes of the current plot so that it contains all
the data points in the currently selected subsets.
By default the plot will be scaled like this, but it it may have changed
because of changes in the subset selection or from zooming in or out.
-
Replot
- Redraws the current plot. It is usually not necessary to
use this button, since if you change any of the plot characteristics
with the controls in this window the plot will be redrawn
automatically. However if you have changed the data, e.g. by
editing cells in the Data Window,
the plot is not automatically redrawn (since this is potentially an
expensive operation and you may not require it).
Clicking this button redraws the plot taking account of any changes
to the table data.
-
Grid
- Toggles whether a grid is drawn over the plotting surface or not.
-
Draw Subset Region
- Allows you to draw a region on the screen defining a new
Row Subset. When you have finished
drawing it, click this button again to indicate you're done.
See Appendix A.3.6.1 for more details.
-
Subset From Visible
- Defines a new Row Subset
consisting of only the points which
are currently visible on the plotting surface.
See Appendix A.3.6.1 for more details.
The Marker Types menu allows you to select a
set of markers which will be used for plotting.
Some of these sets are marked "Transparent" - for these,
instead of pixels on the plot blocking out ones already plotted,
the more markers that are plotted at a given screen position,
the darker in colour it will appear. This can be useful if
you have very many points to plot, since you can see by the
colour of pixels on the plot how many points are there in
crowded regions. Unfortunately transparent points are not rendered
properly when exported to PostScript files (they come out opaque),
but they still work when exported to GIF format.
The marker type set used initially depends on how many rows there
are in the table (large dots for few rows, small ones for many).
The Regression menu provides facilities for
calculating and plotting linear regression lines for some or
all of the subsets on display. The following options appear
on the menu:
-
Plot Regression For Subsets...
- Presents you with a checkbox menu which allows you to select
which subsets regression lines will be displayed for.
Each subset whose box you select will have a line plotted on the
graph representing its least-squares fit regression line
in the same colour as the points for that subset.
-
Display Regression Coefficients
- Displays a window giving the gradient, intercept and
product moment correlation coefficient for each regression line
which has been plotted.
Note the regression lines plotted are those calculated from all the
visible points in the subset in question - any points
off the edge of the graph are disregarded. Thus, zooming in and
out will change the correlation coefficients and line geometry.
The linear correlation functionality is experimental in this version
of the program, and will be improved in future.