Using Charts and Graphs to Represent Data in Xcelsius 2008 Dashboard
In this chapter
- Choosing the Right Components for a Dashboard
- Building on Your Visualizations
- Viewing Grouped Data with Stacked Charts
- Working with XY and Bubble Charts
- Working with Tree Maps
- Issues and Techniques Related to Scaling
- Putting Visual Data Analysis into Focus
- Closing Thoughts
One of the reasons dashboards are popular and successful is that they help you show information and relationships that would otherwise be difficult to see. It should come as no surprise that charts and graphs play a key—or even central—role in the preparation of dashboards.
The goal of this chapter is to introduce you to using charting components in Xcelsius 2008. While many of the features of charting or graphing components are outlined, the emphasis is on how to use charting and graphing components to better convey information. Sometimes, important information is buried in the rows and columns of data. Visualizing data the right way can help to reveal insights.
Choosing the Right Components for a Dashboard
Every dashboard has a story to tell. As you create a dashboard, your choice of components and the way you set your chart attributes can either bring out that story or bury it.
Say that you want to incorporate the following snippet of data, which represents the number of daily visitors to a website, in your dashboard:
Date Visitors Date Visitors 1/1/2009 4667 1/12/2009 4648 1/2/2009 4349 1/13/2009 5154 1/3/2009 3678 1/14/2009 5281 1/4/2009 3094 1/15/2009 5088 1/5/2009 4326 1/16/2009 4709 1/6/2009 4627 1/17/2009 3477 1/7/2009 4615 1/18/2009 3078 1/8/2009 4743 1/19/2009 4617 1/9/2009 4888 1/20/2009 5357 1/10/2009 3321 1/21/2009 5421 1/11/2009 2955 1/22/2009 4902
What is printed here is just a snippet. The full data can easily span a year or more. You can find the full spreadsheet for this example in the file ch05_SampleData.xls.
Your first hurdle is to determine how much data you want to display. Figure 5.1 shows two alternative views of the data. The top-left graph represents a 22-day snippet of data. The lower-right graph represents data over the full range of dates, roughly spanning a full year.
Figure 5.1 Two ways to represent the data on a dashboard.
A quick glance reveals an increasing progression in the data over time, but there is a lot of variation throughout the course of a week. It would be nice to be able to choose an arbitrary point in the timeline and show all the data over, say, a 30-day period. You can do this by using the OFFSET function (see Figure 5.2 or the ch05_SampleDataEnhanced.xls file).
Figure 5.2 OFFSET allows you to choose data from any point in the timeline.
It’s not difficult to turn this into a rudimentary dashboard (see Figure 5.3).
Figure 5.3 A very basic timeline dashboard.
Here are some quick steps to take:
- Launch Xcelsius 2008 and import the ch05_SampleDataEnhanced.xlf file.
- Drag a Column Chart component onto the canvas. If you prefer, you can instead use a Line Chart component or a Combination Chart component.
Map the component to the underlying spreadsheet data.
Click the chart and in the General tab of the chart’s properties panel, click the By Series radio button and then click the + button to add a data series.
Within the added data series, link the Y values to the data to be displayed. If you are following this example, this would be cells H7:H37 of the SourceData worksheet.
If for some reason you want the Y-axis to appear on the right side of the chart, choose Secondary Axis instead of Primary Axis.
Link the category labels to the cells G7:G37.
- Drag a Spinner component onto the canvas and link the data field to the day number (in this example, it is cell H1).
When you open the dashboard in preview mode or export the dashboard, you should be able to move along the timeline by clicking the up and down arrows in the Spinner control or jump to a day number by typing in the number.
There are a number of things you need to fix in the dashboard you’ve created so far. As you use the Spinner control to advance in the timeline, the column chart is a bit jittery between clicks. This is because the data animation feature of the Column Chart component is enabled. In general, data animation is a good thing; however, when you are trying to incrementally advance along a timeline, this feature becomes distracting.
In your Xcelsius workspace, select the Column Chart component. In the Behaviors tab of its properties panel, click the Animations and Effects tab and deselect Enable Data Animation (see Figure 5.4).
Figure 5.4 Turning off data animation can remove the “jitters” from a chart.
Eliminating the jitters is easy enough. Unfortunately, the Spinner component still doesn’t behave exactly as you need it to because you haven’t given it minimum and maximum limits. Unless you specify otherwise, the Spinner component automatically defaults to a minimum of 0 and a maximum of 100. In this example, you should set the minimum to 1 instead of 0. The maximum limit should be a number well above 100. You have a choice of typing your own value or linking to some value on the underlying spreadsheet.
The Column Chart component in Figure 5.4 is set to display 31 vertical bars. They appear a little cramped. You can improve the appearance by setting the marker size for the data series to a value smaller than its default of 17 (see Figure 5.5).
Figure 5.5 Adjusting the vertical bar width in a column chart.