In the case you are mentioning, you could display the data in a few ways.
One way is to show a bar graph or scatter plot and the x-axis would just have arbitrary names, e.g. Sample 1, Sample 2, Sample 3, etc., and your data would be shown on the y-axis (remember that the independent variable is on the x-axis, and the dependent variable - your data, is on the y-axis).
If you are trying to show the variance for the same x value (i.e. how different the y values (data) were for one x value), you could do a scatter plot and just enter in the y values but have them all have the same x value (so you'd end up with a vertical plot of data). So for example, let's say you put 5 numbers into a machine (x = 1, 2, 3, 4, and 5), and then the data that came out of that machine was (y = 2, 4, 6, 8, 10), but you did this multiple times, and got (2.5, 3, 4, 6, 8,) the second time. So if you wanted to show how the numbers varied when you entered in x = 2, for the graph you'd only plot (2, 4) and (2, 3), where the first number is x and the second number is y.
You could also do some kind of number line, though I don't know if that would work for this experiment.
Does this make sense? Most graphs do have some kind of x axis. It doesn't have to be quantitative.
If this doesn't make sense, give me some more details on the project and I can give you more specific details.
Stuck? Check out our project guides!
Project Guide: http://www.sciencebuddies.org/science-fair-projects/project_guide_index.shtml
Advanced Project Guide: http://www.sciencebuddies.org/science-fair-projects/competitions_index.shtml