Creating plots of two data series on one graph.
Say you wish to plot two data series on one graph. For instance, you wish to plot two position versus time curves on one graph. To the right are the position and time data for two automobiles.
To plot both of these on the same graph, first follow steps 1-7 from the Graphing Tutorial on the previous page. This will create the first data series to the graph.
While the window below is still open, click again to create the second data series. Then repeat steps 5-7 from the Graphing Tutorial. Continue until all data sets have been added.
Altering the graph's legend.
Let's continue with the above example of the two cars and their data being displayed on the same graph. The default legend is displayed to the right, and consists of data markers along with the terms Series1, Series2, etc. Unfortunately, the terms "Series1" and "Series2" tells us nothing about what is being plotted. Let's learn how to change the labels to read "Car #1" and "Car #2".
First, right-click anywhere on the graph of which you want to alter the legend and select the Source Data... option as shown below. In the window that opens, select the Series tab near the top of the window.
In the new window that pops up, enter the Name of the data series in the text box (see the red circle in the image below). Here we changed "Series1" to "Car #1". To alter the name of another data series, select the appropriate series (see the green circle below) and then enter its name in the text box.
The end result is a more descriptive legend:
Fitting multiple curves on one set of data.
Imagine we have a situation where a robot's position is recorded as it moves forwards and then backwards. As we've done many times before, we can plot this data (see above). However, say we need to determine the robot's velocity as it moves forward and as it moves backwards. Do do this, we need to determine the slope of the data plot when the robot moves forward (positive slope) and the slope when it moves in reverse (negative slope).
Since the above graph is made from only one series of data we can only fit a single trendline to the data. In order to fit two trendlines to the data set, we must create two additional data series from the original data set and plot them on the same graph.
Here's how it is done for our robot example.
Using error bars.
In this example, we measure the mass and volume of various quantities of water. By plotting mass versus volume, we may determine the density of water from the graph's slope. Recall that M = rV. The measured data and the resulting plot is shown below.
Our slope shows tells us that we are within approximately 2% of the accepted density value of 1000kg/m3. However, what we don't know is the quality of our data points. They seem to lie close to the line, but we really should display error bars to be certain.
It was determined by an analysis tool known as standard deviation (see also the Excel tutorial on standard deviation), that the uncertainty of the mass measurement was ± 10kg and the volume measurement had an uncertainty of ± 0.006m3. These uncertainty values will be used error bars on the graph.
Adding error bars in Excel is easy. Here's how:
Copyright © 2000, Clemson University. All Rights Reserved.