I truly enjoy creating readable, user-friendly and attractive charts for presenting data and analysis results. I like using my cartographic skills (I was a geography major) to create information-rich charts with appropriate fonts, color schemes and icons. Most often, I create bar or column charts and pie charts with single-instance data for my projects. Sometimes though, I need to show multiple types of data or a change over time and bar, column and pie charts just don’t cut it.

Unfortunately, Excel help menus (we use the 2003 version) don’t provide a lot of help in creating more complex charts or information on how to format your data to produce a desired chart. I therefore turn to external sources and websites of Excel experts. A favorite site of mine comes from a Microsoft Most Valuable Professional: http://peltiertech.com/Excel/Charts/index.html. This site features examples of charts made from complex data and tutorials on how to create them yourself.

For the Improve Group evaluation of Performing Arts Workshop’s Artists-in-Schools program I needed to display how teachers’ responses to survey questions changed over the course of the program period. I wanted to be able to put multiple questions on the same chart, show a change over time and show change for three different respondent types. To create the chart at the end of this page, I used multiple tutorials:
• http://peltiertech.com/Excel/Charts/ColoredVerticalBand.html, which taught me how to make floating bar/column charts,
• http://peltiertech.com/Excel/ChartsHowTo/ClusterStack.html, which showed me how to cluster columns, and
• http://peltiertech.com/Excel/ChartsHowTo/ChartType.html, which showed me how to have multiple chart types for data series.

The following chart displays classroom teachers’ responses to a question on how comfortable they feel using the artistic disciplines in their regular curriculum using a 4-point scale from Very uncomfortable to Very comfortable. Teachers completed the survey twice, once at the beginning of the Artists-in-Schools program (pre-test) and again at the end of the program (post-test). The columns on this graph represent the range between the pre-test average scores and the post-test average scores; the square dots represent the average post-test scores for each artistic discipline. Each type of respondent (comparison, etc.) is color-coded consistently across all other charts made for this project. Along with a brief explanation of the data, readers can clearly see where each respondent type started out, where they ended up and how their growth compares to each other.

Once I had created a chart that was formatted and styled in a way that I liked, I saved it as a new Chart Type so that I could apply it easily to the numerous other charts I needed to create. A tutorial for this step can be found here http://peltiertech.com/Excel/ChartsHowTo/CreateCustomTypes.html.