The idea of a graph is to give a quick and interpretable presentation of a set of experimentally acquired data that show trends. The graph should be able to stand on its own It should be a presentation in itself.

1) The x-axis represents what you do; the y-axis is the measured result of what you did..

From the first graph you might read: "For 10 gallons of applied water I can then expect to get 30 bushels of pumpkins" whereas the second graph makes no sense because it reads: "For 30 pumpkins I can then expect to get 10 gallons of applied water"

2) Associated with the label of each axis one must also indicate the UNITS of measure used. If you say a car goes at a speed of 5, this has absolutely no meaning since it could be 5 miles per hour, 5 kilometers per hour, 5 cm per sec, or 5 oak trees passed per day.

Notice how useless these graphs would be if they just said 'pumpkins' and 'applied water' with no units. In fact, these units could use improving: we would like to know if this is gallons for the whole year, per week, per day, or what???? Is it bushels per plant, per acre, per week or what??? (These are things that can be defined in the legend).

3) Except in unusual cases, start the axis at the origin and increment the axis in a consistent way using round numbers. (One exception would be where you are making a log plot where there is no operational definition for an origin.) In this graph you obviously suspect that zero plant mass would give you zero photosynthesis, but the way the graph is plotted gives you little chance of seeing that relationship.

4) Make certain your experimental points are obvious (i.e., clear, large and distinct). An X, circle, triangle or a big fat blotch can be used. The above graphs are good in that they emphasize the actual data points. Also, if a point represents an average of more than one piece of datum, then show the standard deviation using error bars.

5) Respect the fact that the line drawn on a graph among distinct data points is your personal interpretation of what the data points are indicating. For example, if you have a good reason to believe that the experiment you have done has data that should fall along a straight line, then you have the right to draw a straight line through (but generally not beyond) the data points.

6) Give a clear FIGURE LEGEND indicating what the graph is all about. Remember, a reader of your graph initially knows nothing about it, and the first place one normally looks for information is the title/legend. Make clear what the graph is about and include all information that influences the results (such as concentrations, temperatures, type of organism, etc.). This is often the most difficult aspect of graph making. See #2 and "legend examples" under "Basics of Tables" for more information.

7) You should plot curves that are useful to compare on the same graph. On the other hand, if there is no useful comparison, then DO NOT plot various curves on the same graph.

(By the way, what are the implications of the above left graph if the axes were switched?)

8) If you are using a computer to do your plot, do not expect the computer to do all of your work. You still have to supply the program with the correct information and you have to decide if all of the above rules have been obeyed.

[ ] ==> What I have controlled is on the horizontal (x) axis, and what interests me is on the vertical (y) axis.

[ ] ==> The units on the axes are appropriate, evenly spaced, and start from zero.

[ ] ==> The units are properly defined.

[ ] ==> The experimental points are obvious and distinct.

[ ] ==> IF appropriate, a best fit line is drawn and the equation shown. Otherwise trends are shown by drawing lines between points.

Title/Legend MUST include:

[ ] ==> Organism and type of biological sample that was used.

[ ] ==> The amount of sample used and how was it pretreated?

[ ] ==> Basic description of how the experiment was done. Basic method and instrument(s). (Or direct the reader to where they can find this information)

[ ] ==> Description of the conditions the sample was exposed to? (pH, chemical concentrations, temperature, ...)

[ ] ==> Indicate any data manipulations- for example do the points represent averages, were standard deviations determined, etc.

[ ] ==> useless and repeated information is avoided.

[ ] ==> I've shown my graph to someone else ( ) and they understand it.

[ ] ==> What I have controlled is in the left most column, and what was been determined is in the other columns.

[ ] ==> The units are properly defined in the title or at the tops of the columns.

[ ] ==> The information is ordered from top to bottom in a logical way.

[ ] ==> Not all of the entries are strictly quantitative; otherwise the data would be better graphed.

[ ] ==> Where numbers are used, the number of digits shown reflects the experimental accuracy.

Title/Legend MUST include:

[ ] ==> Organism and type of biological sample that was used.

[ ] ==> The amount of sample used and how was it pretreated?

[ ] ==> Basic description of how the experiment was done. Basic method and instrument(s). (Or direct the reader to where they can find this information)

[ ] ==> Description of the conditions the sample was exposed to? (pH, chemical concentrations, temperature, ...)

[ ] ==> Indicate any data manipulations- for example do the points represent averages, were standard deviations determined, etc.

[ ] ==> useless and repeated information is avoided.

[ ] ==> I've shown my table to someone else ( ) and they understand it.