Scatter charts can be a great way to visually correlate your data.
The exercise and result are in the Scatter Chart file.
Here is some simple hypothetical data tracking interviewing activity.
How is the person doing converting initial interviews to second interviews? One way to show it would be using a scatter chart, also known as an XY chart.
The difference with this chart is that you plot two sets of data against each other. One set of data goes on each axis.
In this example, we would not highlight the month column before firing up the chart wizard. Just highlight the two data columns, then select XY (Scatter) charts from the Chart Wizard.
While in this section of the chart wizard, you can click and hold the Press and Hold button to see a preview of how each scatter chart type would look. The key difference is whether you put a trend line in or not. When you find the one you prefer (in this case we think it looks best without the trend line), then click Next and finish going through the Wizard to complete the chart.
Some pointers on using this type of chart:
We suggest bagging the legend
Create the X and Y axis labels
Consider using the title to spell out the correlation or add in a text box in the body
We are not done yet. Now let's add in a trendline by going to the Chart menu.
In the trendline dialog box, you get a number of choices:
We think the linear trendline is usually the best.
The slope of the trendline can be very meaningful. In this case, it gives you a feel for the conversion rate, turning first interviews into second interviews.
Another thing to consider is how far data points are from the trendline. If they are above or below, what do those data points tell you. What was different during that time frame?
Scatter charts are a great way to visually correlate the data, even giving sharper insights on charts that at face value would appear to state the obvious.
Next we will show how you can use a scatter chart to fix a common chart problem