Preparing
Graphs1
Author: J. M. McCormick
Last Update: August 12, 2008
General Considerations2
When we prepare a graph the
independent variable is always on the “x-axis”, and the dependent variable is
always on the “y-axis”. We indicate which variable is which by saying “as a
function of” or “versus”, with the dependent variable coming first, and the
independent variable coming second. So if someone says, “volume was plotted as
a function of mass” or “the volume is plotted versus mass,” it means that mass
was on the x-axis and volume was on the y-axis. Watch what you say/write as
there is only one correct usage!
All graphs must have a title and
axis labels. The title should tell the reader what is in the graph. Avoid the
obvious use of “vs.” or “versus” in the title. The axis labels have two parts:
the first is the name of the parameter, and the second is the unit. The axis
labels can have one of three formats: parameter name (unit); parameter name,
unit; parameter name/unit. For example, if the parameter was temperature and it
was measured in Kelvin, then the axis label could be Temperature (K), or
Temperature, K or Temperature/K.
Devise a scale for each axis so
that the tick mark labels end in a “0” or a “5”. If this is not possible, use a
scale so that the last digit in the tick mark labels is an even number. Tick
mark labels with the last digit being an odd number, other than “5”, are almost
never used. The axes do not need to start at zero. For example, if all the x
values occur between 400 and 600 nm, a graph of these data could start at 400
nm. The key to preparing good graphs is selecting a scale that shows all of the
data and minimizes large regions of blank space. The one exception is when
you need to extrapolate back to a certain value, but the data are not
necessarily close to that value. In this case,
you would want to have the value to which you will extrapolate shown on the graph, even though there may be some blank
space.
The number of significant figures in the tick
marks is usually less than that in the original data. For example, if our
temperature axis has values between 200 and 320 K with each individual point
measured to the nearest 0.01 K, we would not write the tick marks as 200.00,
220.00, etc., but rather 200, 220, etc. (see Fig. 1 and Fig. 2).
Hand-Drawn Graphs
For hand-drawn graphs in the
notebook choose a scale so that the graph fills most, if not all of the page.
Use a small, dark dot for each data point. Draw a circle around the point to
help the reader locate the actual data point. The radius of the circle usually
approximates the uncertainty in the point unless this gives a circle that is too
large. With hand-drawn graphs, one usually does a linear regression “by eye”,
which means that a ruler is used to put a line through the data such that all
points lie as close as possible to the line. The slope and intercept can then
be determined from the line itself using points on the line at the extremes of
the graph to maximize the precision.
Figure 1. A properly executed hand-drawn graph.
Spreadsheet Graphs
With a spreadsheet it is much
easier to prepare graphs, but it is also much easier to produce a poor
quality graph. To assure that your graphs are correctly prepared (e. g., look
good and are easy to understand by the reader), follow these standard
procedures:
-
-Most of the graphs that you will
prepare for this class are called “XY Scatter” plots in Excel. The
other formats will be used sparingly because they are generally not useful in
chemistry.
-
-
-When creating the graph, make the
chart occupy a new sheet; do not paste it onto the work sheet containing the
data.
-
-
-Make all lines (the axes, the
tick marks, the box around the plot area and any data lines) black with medium
intensity (third choice from the top under “Weight” in the format boxes).
-
-
-The plot area should have no
fill color (not the default setting).
-
-
-Turn off all gridlines (not
the default setting), unless told otherwise by your instructor.
-
-
-Legends showing what a given
symbol means are useful, but not always needed (e. g., in a formal laboratory
report this information is given
in the caption that accompanies the figure). Be sure that your selection of lines and legend titles clearly
distinguish between multiple data sets and fits.
-
-
-Watch your tick marks. The
spreadsheet automatically selects the spacing, which may not be appropriate for
your graph (see General Considerations).
-
-
- Be careful with the use of color
to distinguish different data sets on the same graph. Yellow and light blue do
not show up very well when printed either on color or black and white printers.
Do not distinguish different data sets by color if you do not have a color
printer. In general, it is best to dispense with color entirely and make all
lines and symbols black (or at least a uniform dark color).
-
-
-Data are always shown as symbols
and fits to the data are shown as lines or curves. Do not connect the
data points with lines. This will confuse the reader as to whether these lines
represent a fit, or not.
-
-
-Sometimes a fit is not required,
or cannot be done, but you still want to show a trend in the data. In this case
a smooth line that passes through the data as an “aid to the eye” is used, and
is so indicated in the text accompanying the graph.
-
-
-A 16-point font for axis labels and graph titles and a 14-point
font for the tick mark labels are better choices for general work than the
default settings.
-
-
-Depending on the final size of
the graph, you might want to increase the size of the symbols to 9 point.
-
-
-Most spreadsheets automatically
select the number of decimal places on the tick labels from the data, but not
always correctly. For example, if your data were between 50 and 100, you would
want tick labels of 50, 60, 70, 80, 90, and 100. But if the data in the
spreadsheet are set to two decimal places, most spreadsheets would make the
labels 50.00, 60.00, 70.00, 80.00, 90.00 and 100.00. Although this is not
necessarily incorrect, the convention is to show the minimum number of decimal
places on the tick labels.
Remember that the default settings in a spreadsheet may not
fit your needs. Therefore, it is critical that you know how to change the
settings (see the Guide to Excel).
Figure 2. A graph that was properly prepared using
a spreadsheet.
References
- 1. Click here to obtain this file in PDF format (suitable for printing).
-
-
2. Dodd, J. S., Ed. The ACS Style Guide: a Manual
for Authors and Editors; The American Chemical Society: Washington , DC,
1986.
Return to Data Analysis Page
Return to Truman Chemistry Lab
Page