|You are working with the text-only light edition of "H.Lohninger: Teach/Me Data Analysis, Springer-Verlag, Berlin-New York-Tokyo, 1999. ISBN 3-540-14743-8". Click here for further information.
Graphs are most valuable if they are labeled properly. The following
is a short summary of the prerequisites for "good" graphs:
Any graph should be labeled with a short self-explanatory title.
Both the horizontal and the vertical axis should be labeled with variable
names and units. The major values should be clearly labeled with
a meaningful numbering system (e.g. 1-2-5). The dependent variable should
be plotted on the vertical axis and the independent variable on the horizontal
The origin should be zero. If the scaling of any axis prohibits the
inclusion of the origin, a broken axis should indicate the offset from
If the uncertainty of a parameter is known, it should be represented
by an error bar.
Size and Clarity.
The range of the axes should be assigned such that most of the data
(or the region of interest) occupies the whole graph.
Click on the
to see a few examples illustrating bad designs.
Last Update: 2006-Jšn-17