Determining Effective Data Display with Charts Marketing: Analyzing Trends in the Sporting Goods Industry

Information is a source of learning. But unless it is organized, processed, and available to the right people in a format for decision making, it is a burden, not a benefit. William Pollard
LEARNING OBJECTIVES
Level 1

Understand the principles of effective data display Analyze various Excel chart types Determine appropriate uses for different chart types Modify the chart type and the chart source data Specify chart options, including chart and axes titles, legends, and data labels
Level 2

Examine the effectiveness of different chart sub-types Evaluate the stacked and 100% stacked sub-types Explore the Pie of Pie and Bar of Pie sub-types Create various stock charts to display financial data Clarify data with trendlines and moving averages
Level 3

Understand and evaluate radar charts Understand and evaluate bubble charts Compare a bubble chart with a 3-D column chart Create and customize a doughnut chart Explore and customize a dashboard chart CHART TYPES COVERED IN THIS CHAPTER
Area Bubble Column Doughnut Line Pie Radar Stock X Y (Scatter)

160

Chapter 3

Determining Effective Data Display with Charts

CHAPTER INTRODUCTION This chapter teaches you how to use Microsoft Excel charts to provide a visual illustration of quantitative information, giving the viewer an overall picture of a set of data. In many ways, charts are easier to interpret than large amounts of tabular data. This chapter presents some guiding principles you can follow to create effective charts that show quantitative information clearly, precisely, and efficiently. The numerous chart types included in Excel provide great flexibility in presenting data for visual analysis. The chapter begins with the basics of creating and modifying line and column charts while discussing how the choice of chart type can influence a viewers perception of the information presented. Each chart type presents information in different ways that can affect the viewers interpretation of the data. The chapter covers some basic principles to follow when creating charts and provides examples of how charts can be used in specific situations. Coverage of chart sub-types, including different types of stock charts, is also provided. Radar and bubble charts are then used to illustrate some advanced chart techniques. Finally, the last part of the chapter presents techniques for building a management dashboard by combining different chart types within the same chart. CASE SCENARIO Michelle Kim is an analyst in the marketing department at TheZones corporate headquarters in Colorado. Michelle monitors company performance and assists the president and CEO to gather the data necessary for management of the company. Michelle works with counterparts in each of the companys three divisionsEquipment, Footwear, and Apparelto monitor how well the company implements its corporate strategy. Una Baatar, president and chief executive officer of TheZone, has asked Michelle for a report and presentation on the companys performance compared with other industry competitors. The presentation will cover industry trends and the companys overall performance with specific information from each of the divisions. The performance of the companys stock in the market will also be discussed. Una is particularly interested in the Footwear divisions work in bringing the new TZEdge athletic shoe line to market. Michelle will use Excel to create the necessary charts for the report and presentation.

Marketing

161

Chapter 3

Determining Effective Data Display with Charts

LEVEL 1
VISUALIZING DATA If a picture is worth a thousand words, can a chart be worth a thousand data points? Figure 3.1 shows one example from Edward R. Tuftes groundbreaking book, The Visual Display of Quantitative Information. This New York City weather summary for 1980 depicts 2220 numbers. The daily high and low temperatures are shown in relation to the long-range average. The path of the normal temperatures also provides a forecast of expected change over the year; in the middle of February, for instance, New York City residents can look forward to warming at the rate of about 1.5 degrees per week all the way to July, the yearly peak. This distinguished graphic successfully organizes a large collection of numbers, makes comparisons between different parts of the data, and tells a story.
Figure 3.1: Historical chart of New York City weather in 1980

Tuftes book is concerned with the design of statistical graphicsthe combined use of points, lines, numbers, words, shading, and color to present quantitative information. Tufte believes that graphical excellence depends on clarity, precision, and efficiency. Excellent graphics give the viewer the greatest number of ideas in the shortest time with the least ink in the smallest space, according to Tufte. The ink here refers to everything that prints on paper. In the case of an Excel chart, the ink is the combination of data, labels, and chart effects. Tufte presents five data graphics principles related to ink in The Visual Display of Quantitative Information:
162

Above all else show the data. Maximize the data-ink ratio, within reason. Erase non-data-ink, within reason. Erase redundant data ink, within reason. Revise and edit.

Chapter 3

Determining Effective Data Display with Charts

The third and fourth data principles, Erase non-data-ink and Erase redundant data ink, are somewhat related to maximizing the data-ink ratio. Non-data-ink is a part of the chart that decorates more than informs. Redundant data ink is ink that repeats information. For example, the use of a data table and labels on each point in the chart in Figure 3.2 is redundant.
Figure 3.2: An example of chart junk

The fifth and final principle is Revise and edit. Charts can be improved much in the same way that writing is improved by revision and editing. Feedback on a charts usefulness and clarity should be sought wherever possible, and the chart should be adjusted accordingly. The intended message of the chart must be considered. A positive feature in one chart type might become negative if it doesnt support what the chart creator is trying to illustrate. Often, the best way to choose a chart is through trial and error. When a chart is found to be useful, it should be saved as a template or working model for future use.
163

Level 1

The first principle, Above all else show the data, is a reminder not to clutter a chart by adding unnecessary illustration or decoration. The second principle, Maximize the dataink ratio, refers to the portion of the ink that is devoted to displaying the data versus the portion of a graphic that can be removed without losing the data. Everything on a chart needs to have a reason for being there. The excessive use of gridlines and data labels in the chart in Figure 3.2 is an example of what Tufte refers to as chart junk, which results in a chart with a very low data-ink ratioin other words, lots of ink used to display a small amount of data. Chart junk can make charts difficult to read, use, and interpret.

Chapter 3

Determining Effective Data Display with Charts

In business, charts are typically used to summarize information so that it can be presented easily to others. The numerous charting options that Excel offers, however, can sometimes lead to chart junkin this case, the embellishment of charts with a lot of chart effects (ink) that dont tell the viewer anything new. The purpose of formatting is to make a chart easier for the viewer to understand. Too little formatting can leave a chart ambiguous; too much formatting can cause viewers to notice the formatting, not the data. With this in mind, Figure 3.3 shows a much improved chart compared to Figure 3.2.
Figure 3.3: Chart junk removed

Y-axis labels

Y-axis

Chart title

Data point

Data series

X-axis

X-axis labels

Legend

The chart in Figure 3.3 shows consumer purchase data for TheZones three divisions Equipment, Footwear, and Apparelover an eight-year period beginning in the year 2002. As you can see in the figure, the chart contains the following elements:
164

Chart titleThe descriptive text that identifies the charts contents Y-axisThe vertical axis where data values are plotted Y-axis labelsThe labels that identify the data values plotted on the y-axis X-axisThe horizontal axis where categories are plotted X-axis labelsThe labels that identify the categories plotted on the x-axis

Chapter 3

Determining Effective Data Display with Charts

Data seriesThe related data points that are plotted on the chart; each data series on a chart has a unique color or pattern and is identified in the chart legend Data pointsThe points in a data series at which the x-axis and y-axis values intersect LegendA box that identifies the patterns or colors assigned to the data series in a chart As noted earlier, Michelle Kim is a marketing analyst for TheZone and is responsible for monitoring how well the company implements its corporate strategy. She has been analyzing industry trends for the companys three divisions, and has just started to put together a chart showing the sporting goods industrys last eight years of consumer purchase data by equipment, footwear, and apparel. Michelle is using Excel to organize and analyze the various sets of data shown in Figure 3.4, looking for potential problems that could disrupt the companys strategic plans. She will make extensive use of charts to summarize and present the data visually for further analysis.
Figure 3.4: Worksheet of consumer purchase data

3
Level 1
165

The information contained in spreadsheets is stored as tabular data arranged in rows, columns, and worksheets. This allows you to organize and analyze a large amount of data effectively, but can you see what the data means? You turn raw data into information by organizing it into something meaningful. You make information useful and give it a purpose by using it to make decisionsdecisions that you couldnt have made as well without

Chapter 3

Determining Effective Data Display with Charts

the knowledge gained from your analysis. Presenting data visually, in the form of a chart, for example, gives a global view that allows you to effectively convey the meaning hidden in a large data set, compare sets of data, and perhaps uncover a critical relationship or trend that would be hard to reveal with tabular data alone. EFFECTIVE CHARTING IN EXCEL To create a chart in Excel, you first select the data you want to display in the chart, click the Insert tab on the Ribbon, and then click a button in the Charts group or the dialog box launcher in this group if you want to choose from all available chart types and sub-types.
Best Practice Choosing a Chart Location Charts can either be embedded as an object on the same worksheet as the source data for the chart or placed on a separate worksheet referred to as a chart sheet. You can modify the location of a chart at any time using the Move Chart dialog box, which is available on the Chart Tools Design tab, in the Location group. Chart location is a matter of personal preference. Some users find it easier to work with charts that are embedded with the worksheet data because changes to the worksheet are immediately visible in the chart. Other users find that placing the chart on a separate chart sheet makes editing and printing the chart much easier. As a general rule, you should place charts that are developed for presentation and printing on a chart sheet. These charts are typically based on a specific reporting period and the source data usually doesnt change significantly. If the source data is dynamic, seeing the effect of the changes in the chart can be very handy. In this case, the chart should be placed on the worksheet with the data.

Determining the Appropriate Chart Type and Chart Options

Excel provides 11 standard chart types, with 73 sub-types that can be used to present information graphically. In addition, you have the ability to combine these different chart types and sub-types within the same chart. Charts can display a combination of numeric and category data. An Excel chart requires at least one numeric data series and at least one label data series. Excel refers to the label data series as category data. Excel can display up to 255 numeric data series in one chart with up to 32,000 data points in each series, for a limit of 256,000 data points in one chart. The chart type you choose depends on what you are trying to illustrate with the data. Table 3.1 identifies and describes the chart types available in Excel.

166

Chapter 3

Determining Effective Data Display with Charts

Table 3.1: Excel chart types Chart Type Icon Description Column Bar Line Area Pie Doughnut X Y (Scatter) Compares values across categories in a vertical orientation. Values are indicated by the height of the columns. Compares values across categories in a horizontal orientation. Values are indicated by the length of the bars. Displays trends over time or by category. Values are indicated by the height of the lines. Displays trends over time or by category. Values are indicated by the filled areas below the lines. Compares the contribution each value in a single numeric data series makes to the whole, or 100%. Values are indicated by the size of the pie slices. Compares the contribution each value in multiple numeric data series makes to the whole, or 100%. Values are indicated by the size of the doughnut segments. Compares pairs of numeric values on the x- and y-axes, with the data points plotted proportionally to the values on the x-axis; can also be used to display a functional relationship, such as y=mx+b. Values are indicated by the position of the data points. Displays stock price and volume trends over time. Plotted values can include volume, opening price, highest price, lowest price, and closing price. Compares values across categories in a circular orientation. Values are indicated by the distance from a center point. Compares sets of three values. Values are indicated by the size of the bubbles (filled circles). Displays value trends in three dimensions. Values are indicated by areas with colors or patterns on the surface of the chart.

3
Level 1
167

Stock Radar Bubble Surface

Understanding Line and Column Charts

As noted in Table 3.1, a line chart displays trends over time or by category, and a column chart compares values across categories in a vertical orientation. Michelle has already created a line chart based on the consumer purchase data, with markers displayed at each data value (see Figure 3.5). This line chart is useful if the overall trend of the data needs to be emphasized. The vertical y-axis shows the total sales (numeric values) and the horizontal x-axis shows the years (categories). Each line represents the trend in sales of the equipment, footwear, and apparel categories over an eight-year period. In this case, Michelle wants to compare the values in each category more than the overall trend over time, making a column chart the best choice. The two charts shown in Figure 3.5 are based on the same data series; the differences are due to the type of chart chosen. The line chart emphasizes the trend in each category over time, whereas the column chart makes it easier to compare the contribution each category made in a particular year. Note that the column chart in Figure 3.5 is displayed in the default column chart sub-type, the clustered sub-type.

Chapter 3

Determining Effective Data Display with Charts

Figure 3.5: Line chart versus column chart

Line chart emphasizes the trend in each category over time

Column chart emphasizes the contribution that each category made each year

How To Change the Chart Type 1. Select the chart or chart series whose type you want to change. Clicking anywhere in a chart selects the chart. When a chart is selected, three contextual Chart Tools tabsDesign, Layout, and Formatappear on the Ribbon. To change the chart type of a single data series, select that data series either by clicking it in the selected chart, or by selecting it from the Chart Elements dropdown box in the Current Selection group on the Chart Tools Layout or Chart Tools Format tab. 2. In the Type group on the Chart Tools Design tab, click Change Chart Type to open the Change Chart Type dialog box. This dialog box provides the same options available in the Insert Chart dialog box for selecting the chart type. 3. Click the chart type you want to change to, and then click the sub-type (if necessary). 4. Click the OK button.

Comparing Line and X Y (Scatter) Charts

There can be confusion about when to use a line chart and when to use an X Y (Scatter) chart. The data used to plot an X Y (Scatter) chart is different from the data used to create a line chart (or column or bar chart). An X Y (Scatter) chart plots numeric values on both the X- and Y-axes based on the value of the data, whereas a line chart plots numeric values on one axis and category labels equidistantly on the other axis. Consider what happens if the order of the consumer purchase data is switched, as shown in Figure 3.6. The years in the worksheet are listed in descending order, from 2009 to 2002. Figure 3.6 shows this same data plotted as both a line chart and an X Y (Scatter) chart. As you can see in the figure, the line chart plots the data points along the y-axis based on the value (sales volume), but the x-axis is based on position because this axis contains categories (years). The line charts x-axis treats the year data as a category and plots based on position, beginning with 2009. By comparison, the X Y (Scatter) chart plots both the y- and x-axes data as numeric values. So, its x-axis treats the year data as numeric values
168

Chapter 3

Determining Effective Data Display with Charts

and plots them from lowest to highest, based on the year number. Note that the X Y (Scatter) chart is formatted with lines connecting the points; other formats are available for this chart type as well. Also note that the X Y (Scatter) chart automatically generates an x-axis that applies a buffer at both ends, so the years shown on this axis range from 2002 to 2009. An X Y (Scatter) chart should be used instead of a line chart when comparing sets of numerical data on both the x-axis and the y-axis.
Figure 3.6: Line chart versus X Y (Scatter) chart

3
Level 1

The line chart plots the x-axis based on the position of the categories in the data range

The X Y (Scatter) chart plots the x-axis in numeric order based on the values in the data range

Changing the Chart Source Data

After changing the line chart to a column chart, Michelle decides that the chart is too busy and should show data from only the last five years instead of eight. The Source Data dialog box, which is available from the shortcut menu for a selected chart, provides options for modifying the values used to generate a chart. Refer back to Figure 3.4; notice that the categories and numeric data are not in a continuous range of cells. So, Michelle uses the Ctrl key to select both the categories (Equipment, Footwear, and Apparel) and the data for the years 20052009. The data source range changes from A4:I7 to A5:A7, E4:I7.
How To Change the Source Data 1. Select the chart to display the Chart Tools contextual tabs. 2. Click Select Data in the Data group on the Chart Tools Design tab to open the Select Data Source dialog box. See Figure 3.7.

169

Chapter 3

Determining Effective Data Display with Charts

Figure 3.7: Select Data Source dialog box

3. In the Chart data range box, modify the current range of values or select a new range as the source of data for the chart. In the case of a chart with multiple data series, you can change individual data series using the options in the Legend Entries (Series) section of the Select Data Source dialog box. 4. Click the OK button after changing the source data.

Notice the Switch Row/Column button in the Select Data Source dialog box (see Figure 3.7). This option provides settings for displaying the data series by rows or by columns. What would the impact be of changing this option in the current chart? Choosing to display the data by either rows or columns can illustrate different trends and force a different comparison of the data. Refer to Figure 3.8. When plotted by columns, the chart compares the amount of each years sales for each of the three categories (Equipment, Footwear, and Apparel). When plotted by rows, the chart emphasizes the contribution that each category made to each years performance. Michelle is more interested in the amount of each years sales of equipment, footwear, and apparel, so she chooses to plot the data by columns. She doesnt have to open up the Select Data Source dialog box shown in Figure 3.7. She simply clicks the Switch Row/Column button in the Data group on the Chart Tools Design tab. Plotting the data by columns makes it easier to see that sales for the Equipment and Footwear categories appear to be increasing from year to year, whereas sales for the Apparel category are leveling off somewhat.

170

Chapter 3

Determining Effective Data Display with Charts

Figure 3.8: Displaying the data series by columns or by rows

3
Level 1
Compares the amount of each years sales for each category Emphasizes the contribution that each category made to each years performance

Specifying Chart Layout Options

Excel offers many formatting and display options for changing the layout of chart elements, including chart and axes titles, legend, data labels, data table, axes, gridlines, and plot area. Drop-down buttons that lead to options for each of these are found on the Chart Tools Layout tab grouped by Labels, Axes, and Background. Now that Michelle has selected the chart type and determined how to display the source data, she decides to add axes titles that better describe the data on the chart. Many of the Chart Options settings can cause problems with generating excessive ink that obscures the information in the chart. As an illustration of this issue, Figure 3.9 shows the chart that results when too many options are selected, compared to Michelles final chart, which contains only the elements needed to best describe the data, not overwhelm it.
Figure 3.9: Selecting the right number of chart options

Chart with too many options

Final chart

171

Chapter 3 How To

Determining Effective Data Display with Charts

Specify Chart Layout Options 1. Click anywhere in the chart to display the Chart Tools contextual tabs. 2. Click the Chart Tools Layout tab to see available options grouped by Labels, Axes, and Background. See Figure 3.10.
Figure 3.10: Chart Tools Layout tab

Background Axes Labels

This menu provides three groups of drop-down buttons for specifying chart layout options: Labels group Chart Title and Axis TitlesUse the options under these buttons to modify the layout of the chart title or the axes titles. LegendThe options under this button control the display and placement of the chart legend. The legend shows the name of each data series along with the appropriate data marker. You can choose to display the legend in a variety of positions around the plot area. Data LabelsThe options under this button allow you to add and control the display of labels for the data points on the chart. Display options include showing the value of each data point, as well as the series and category to which each data point belongs. Percentages can also be displayed, if appropriate to the chart type. The selections on this tab display labels for each data point, which can become visually overwhelming. You can control the

172

Chapter 3

Determining Effective Data Display with Charts

display of individual labels by selecting them on the plot area or using the Chart Elements drop-down list in the Current Selection group. Data TableUse the options under this button to display a table of values for each data series in a grid below the chart. Axes group AxesUse the options under this button to specify whether the X- and Y-axes are displayed. You can also choose to display the x-axis as a text or a date series axis, instead of automatically choosing a setting based on the chart data. The Date setting shows evenly spaced tick marks across the x-axis, based on the major and minor units of days, months, or years. GridlinesUse the options under this button to modify the display of gridlines in the chart plot area. You can also modify the display of major and minor gridlines for each axis. Background group Plot AreaUse the options under this button to change the formatting of the plot area, the area bounded by the axes. Chart WallUse the options under this button to change the formatting of the chart wall, the two vertical surfaces created in any of the 3-D chart types. Chart FloorUse the options under this button to change the formatting of the chart floor, the horizontal surface created in any of the 3-D chart types. 3-D RotationUse the options under this button to change the viewing angle of a 3-D chart.

3
Level 1
173

Understanding Area and Pie Charts

An area chart combines the features of a line chart with a bar or column chart by filling in the area below the line, and displaying the trend of values over time or categories. A pie chart is used to display the percentage contribution that each category makes to the whole, or 100%. The sum of the value of the categories must total 100% to use a pie chart. The highlighted numeric data range used in a pie chart is assumed to identify the individual components of some total. Excel calculates the percentage that each cell in the range contributes to the total. Percentages are used every day in business to summarize and compare sets of data. For example, comparing the amount of labor used to create products with different overall prices can be made easier if the labor amount is expressed as a percentage of the total cost. Comparing salary rates and pay increases across various departments in a company is much clearer if the pay increases are expressed in percentages instead of actual amounts. Use a pie chart when its most important to show relative percentages rather than values. Graphing these percentages can allow a viewer to make a quick visual comparison. Earlier, Michelle chose a column chart instead of a line chart to compare the amount of each years sales for each category rather than emphasize the trend in each category over time. How would an area chart display that same data? Figure 3.11 compares a column

Chapter 3

Determining Effective Data Display with Charts

chart with an area chart of the consumer purchase data for all eight years (20022009). As you can see, the area chart does a better job of illustrating the sales trend in the equipment, footwear, and apparel categories over the longer time period (eight years), while still emphasizing the contribution that each category made to a particular years performance. The area chart is somewhat easier to grasp than the column chart, because the large number of columns makes the column chart more difficult to interpret.
Figure 3.11: Column chart versus area chart

Column chart

Area chart

Michelle now wants to examine the contribution that the equipment, footwear, and apparel categories made to the 2009 sales as a percentage. She first creates a new worksheet, named Consumer Purchases 2009, and places the consumer purchase data on this worksheet. She decides to use a 3-D pie chart type to produce a more appealing visual effect compared with a flat pie chart. Because the category labels in cells A5:A7 are separated from the data in cells I5:I7, Michelle must use the Ctrl key to select the two ranges as the data source. An alternative method is to select cells I5:I7 as the data source, create a pie chart using the Pie button in the Charts group on the Insert tab, and then add the range containing the category labels (A5:A7) using the Edit button in the Horizontal (Category) Axis Labels section of the Select Data Source dialog box, shown in Figure 3.12. Care must be taken to not select the total in cell I8 as part of the source data for the pie chart. Doing so would result in a pie chart that has one slice equal to half the pie, and all the other slices together making up the other half of the pie. The chart shown on the left in Figure 3.13 is the default 3-D pie chart produced using the Pie button in the Charts group. To improve the appearance of the default chart, Michelle makes the following modifications to produce the final version of the chart, which is shown on the right in Figure 3.13: Removes the chart legend Includes category names and percentages as data labels on the pie slices Formats the data labels in a larger bold font to make them more visible

174

Chapter 3

Determining Effective Data Display with Charts

Increases the elevation angle of the chart from 15 degrees to 40 degrees to provide more space on the pie slices for displaying the data labels and percentages Enlarges the chart overall by dragging one of its resizing handles on the plot area Explodes the pie slice for the Footwear category, for emphasis, by dragging it away from the other pie slices, because Una is particularly interested in the status of the new TZEdge athletic shoe line
Figure 3.12: Selecting the pie chart source data

3
Level 1

Range for category labels

Range for values

Figure 3.13: Default pie chart and modified pie chart

Default 3-D pie chart produced by the Chart Wizard

Elevation angle increased

Modified 3-D pie chart

Footwear pie slice exploded for emphasis

Data labels and percentages added to each pie slice

175

Chapter 3 How To

Determining Effective Data Display with Charts

Change the Elevation Angle 1. Click the chart to display the Chart Tools contextual tabs. 2. Click the 3-D Rotation button in the Background group on the Chart Tools Layout tab to open the Format Chart Area dialog box. See Figure 3.14.
Figure 3.14: 3-D Rotation in the Format Chart Area dialog box

3. To increase the elevation angle, enter a higher number in the Y: Rotation box or click the up arrow button. To decrease the elevation angle, enter a lower number in the Y: Rotation box or click the large down arrow button. Figure 3.14 shows the 3-D Rotation settings in the Format for the Y: Rotation angle increased to 40 degrees. 4. Click the Close button.

How To Format Data Labels 1. Click any data label on the chart to select the entire series of data labels; or, to select a single data label, click the label again after first selecting the entire series. The second click selects only the data label clicked. 2. With either all the labels for a series or a single label selected, click the Data Labels button in the Labels group on the Chart Tools Layout tab. 3. Select the More Data Labels option to open the Format Data Labels dialog box. See Figure 3.15.

176

Chapter 3

Determining Effective Data Display with Charts

Figure 3.15: Format Data Labels dialog box

3
Level 1
This dialog box provides a variety of formatting options for data labels, organized into categories: Label OptionsThis category provides options for changing the information the label displays as well as the label position. NumberThis category provides options for changing the appearance of numbers in the data labels. Fill, Border Color, Border Styles, Shadow, and 3-D FormatThese categories provide options for changing the appearance of the data labels. AlignmentUse the options in this category to specify the alignment (horizontal or vertical), orientation, position, and direction of the data labels. Click a category in the list on the left of the dialog box to display the available options. 4. After specifying all the necessary formatting options, click the Close button. Best Practice Working with 3-D Charts A 3-D pie chart can be much more visually appealing than a flat pie chart. Most of the chart types have 3-D options, but these should be used with caution. Although the use of 3-D can add visual interest, it might be at the expense of the data being presented. Figure 3.16
177

Chapter 3

Determining Effective Data Display with Charts

compares regular line and column charts with their 3-D equivalents. The 3-D charts are more attractive and interesting, but they are also more difficult to interpret. In the 3-D line chart, the relationship between the apparel and footwear in any particular year is difficult to determine because the lines are no longer on the same axis. The 3-D column chart does a better job of showing this, but it is still not as clear as the regular 2-D line and column charts. Additional care must be taken to ensure that one series isnt hidden behind the other. The data series order can be changed as necessary to adjust this, but if the data points are very close or cross, often this is not enough to prevent one data series from hiding a portion of another. Because 3-D charts are more visually appealing than 2-D charts in many instances, they are a good choice for presentations and reports, provided that the preceding interpretation issues are taken into consideration.
Figure 3.16: Regular charts versus 3-D charts

2-D line chart

3-D line chart

2-D column chart

3-D column chart

178

Chapter 3

Determining Effective Data Display with Charts

Steps To Success: Level 1 In these steps, you need to prepare some charts showing trends in the sporting goods industry. A workbook containing a variety of information on industry-wide trends has already been created. You will begin with data showing the number of people who participate in a variety of sports and activities. Complete the following:
1. Open the workbook named Trends.xlsx located in the Chapter 3 folder, and then

3
Level 1
179

save it as Industry Trends.xlsx. Review the data sets on the four worksheets. You need to determine the most important information in each worksheet and select the best method for displaying that information in a chart. Remember that charts can be used to display single data series, compare multiple data series, show trends, and compare the percentage contribution to the whole.
2. The 2009 Participation worksheet contains participation data for a variety of sports,

ranked in descending order by the values in the Total column. Create one column chart, one bar chart, and one line chart from the data. Place each chart on the same worksheet as the source data. In a cell below each chart, explain what the chart type emphasizes about the data. Select which chart is the best choice for comparing each sports participation level. Place this chart on a separate chart sheet named 2009 Participation Chart as the first worksheet in the Industry Trends workbook, and add appropriate y-axis and chart titles to finalize this chart. Return to the 2009 Participation worksheet and, in a cell below the chart you selected to place on the chart sheet, explain the advantages and any disadvantages of this chart.
3. The 10-Year Participation worksheet contains sports participation data over a 10-year

period, with values shown for every other year, for a variety of sports. Create an appropriate chart to illustrate the participation changes in each sport over the 10-year period. Place this chart on the same worksheet as the source data. Add appropriate titles and a legend to finalize the chart. In a cell below the chart, explain the advantages and any disadvantages of the chart type you chose.
4. The Female Sports Participation worksheet shows the changes in female sports par-

ticipation between 2004 and 2009. Create an appropriate chart that shows the changing trends in the percentage of female participation in each sport. Place this chart on the same worksheet as the source data. Add appropriate titles to finalize the chart. In a cell below the chart, explain the advantages and any disadvantages of the chart type you chose.

Chapter 3

Determining Effective Data Display with Charts

5. The Purchases By Sport worksheet shows the dollar amount of consumer equipment

purchases for 2007, 2008, and 2009. Create an appropriate chart to compare the contribution of each sport and illustrate the changes in consumer purchases from year to year. Place this chart on the same worksheet as the source data. Add appropriate titles and a legend to finalize the chart. In a cell below the chart, explain the advantages and any disadvantages of the chart type you chose.
6. Save and close the Industry Trends.xlsx workbook.

LEVEL 2
EVALUATING CHART SUB-TYPES EXAMINING SUB-TYPES FOR VARIOUS CHART TYPES As discussed earlier, the chart type you choose depends on what you are trying to illustrate. Charts can compare values across categories (bar and column), display trends over time or category (line and area), and compare the contribution each value in a single data series makes to the whole, or 100% (pie). Some of the chart sub-types available for the basic chart types enable you to sum or stack the values in each category, giving a slightly different perspective on the contribution that each category makes to the sum or the whole. Another option is to create a 100% stacked chart, in which the plotted values are converted into percentages of the total amount within each category. A 100% stacked chart is similar to a pie except that the pieces are in a column instead of a circle. After reviewing the sample charts she has prepared for TheZone, Michelle is pleased with how the visual display of data makes it easier to compare the data across categories. Now she wants to modify the consumer purchases chart, which is currently an area chart, to show the total sales in each year and how much each category contributes to that total. The current chart shows the sales amount for each category, but does not show the total sales amount for each year.
Adding Things Up: Stacked Chart Options

Stacked charts do a good job of illustrating the cumulative effects of data in categories. Line, bar, column, and area charts all have a stacked chart sub-type. Figure 3.17 shows the current consumer purchases area chart compared with three stacked charts of the same data. Michelle decides that the stacked column chart is the one that most clearly shows the totals for each category (in this case, each year). The stacked area and stacked line charts dont clearly show that the lines represent the cumulative total of each year. These charts can give the impression that each line represents the actual values in the data series, and not the cumulative amounts for each category. This could lead to some misinterpretation and require too much explanation of the chart to make the meaning clear.
180

Chapter 3

Determining Effective Data Display with Charts

Figure 3.17: Area chart compared with stacked charts

3
Level 2
Original area chart Stacked area chart Stacked column chart Stacked line chart

How To Change the Chart Sub-Type 1. Click the chart to display the Chart Tools contextual tabs. 2. Click the Change Chart Type button in the Type group on the Chart Tools Design tab to open the Change Chart Type dialog box. The chart type for the current chart is already selected. An alternative method of opening this dialog box is to right-click on the chart and select Change Chart Type. 3. Click the icon for the chart sub-type you want in the Chart sub-type section. 4. Click the OK button. Summing to 100%: Alternatives to Pie Charts

Another option for illustrating the cumulative contribution for each category is to express this contribution as a percentage. As was the case with the stacked sub-type, line, bar,
181

Chapter 3

Determining Effective Data Display with Charts

column, and area charts offer a 100% stacked sub-type as well. Showing the cumulative contribution for each category as a percentage can reduce confusion over whether the line on the chart represents the individual or cumulative contribution to the whole. These charts combine the features of a pie chart with the features of line, column, or area charts. Figure 3.18 compares the original area chart of consumer purchase data with 100% stacked versions of the same data. Although each of these charts clearly shows the cumulative contribution each of the three divisionsApparel, Footwear, and Equipmentmakes to the whole, or 100%, for each year, Michelle chooses to use the 100% stacked column chart. She feels that displaying each year in a column makes the percentage contribution that each of the three divisions made in each year more obvious. In a way, this 100% stacked column chart is the equivalent of eight pie chartsone for each year. Each column presents the same information that would exist in a pie chart for that year; the only difference is the shape.
Figure 3.18: Comparison of 100% stacked charts

Original area chart

100% stacked area chart

100% stacked column chart

100% stacked line chart

182

Chapter 3

Determining Effective Data Display with Charts

Slicing the Pie Too Thin: Summarizing Too Much Detail in Pie Charts

Figure 3.19: Consumer equipment purchases

As illustrated in Figure 3.19, the number of categories makes the pie chart confusing and difficult to interpret. The data is currently sorted alphabetically by category. As a result, the chart is arranged by category. The first category (Archery) starts at the 12:00 oclock position, and the remaining categories are placed in an alphabetical clockwise rotation. The sort order contributes to the disarray and confusion in the chart. Michelle decides to sort the data by the size of the equipment purchases in descending order, so that the data goes from highest to lowest. She creates a new chart, shown in Figure 3.20, using the data sorted in this order. The new chart appears more orderly and less confusing when compared with the original pie chart (see Figure 3.19). Michelle still isnt satisfied with the results, however, because the chart continues to have an excessive number of segments. There are 23 categories shown in the pie chart which, of course, is too great a number, resulting in many small, indistinguishable pie segments.

183

Level 2

Having too many categories can be a problem with pie charts. If you have an excessive number of pie slices, the chart can become very cluttered and confusing. For example, the data in Figure 3.19 shows consumer equipment purchases by sport in both worksheet format and a corresponding pie chart, which Michelle created based on some industry trend data. She wants to analyze which sports generate the most equipment purchases every year, opting to display the data in a pie chart format to show the percent contribution of each sport for the year 2009.

Chapter 3

Determining Effective Data Display with Charts

Figure 3.20: Re-sorted pie chart

Exploring the Pie of Pie and Bar of Pie Chart Sub-Types

How can Michelle modify the pie chart to address the problem of too many segments? One way would be to choose a different sub-type for the chart. Excel offers two interesting pie chart sub-types that can be used to combine many smaller segments of a pie chart into an Other category, which can then be shown as a smaller pie chart or as a single vertical bar chart next to the original pie chart. The two chart sub-types, Pie of Pie and Bar of Pie, are similar, with the exception of the chart type of the second plot in the chart. Michelle wants to emphasize the categories with the largest purchases. Either chart sub-type would do this well, but she chooses the Pie of Pie sub-type, because she feels that the Bar of Pie sub-type could cause confusion over whether the bar represents percentages or values. The resulting chart, shown in Figure 3.21, does decrease the number of pie segments to improve the visual display of data, but still needs formatting enhancements to meet Michelles needs.
Figure 3.21: Pie of Pie sub-type applied to the chart

184

Chapter 3

Determining Effective Data Display with Charts

After adding the chart title Consumer Purchases by Sport and removing the legend from the chart, to make the pie segments larger, Michelle considers how to format the data series to make the chart easier to interpret. The series is currently split by position, with the last eight values displayed in the second, smaller pie chart, or plot. The total of the values in the second plot is represented by the pie segment on the main pie chart to which the smaller plot is connected. There are four options for splitting the data series: PositionSplitting by position assigns a specific number of values to the second plot. In the current chart, the last eight values in the data range are assigned to the second plot. ValueThis option allows you to select a cutoff point that assigns all the values below that point to the second plot. Percent ValueThis option allows you to select a cutoff point by percentage, rather than value, and assign all the percentages below that point to the second plot. CustomWith this option, you can drag individual pie segments between the two charts so you can include exactly the segments you want in the main pie chart and the second plot. When the data series is selected, the Format Data Series dialog box is available by clicking the Format Selection button in the Current Selection group on the Chart Tools Format tab. The dialog box, also available by right-clicking on the data set to open the shortcut menu, provides an option for changing how the series is split, as well as other options affecting the data series display. Michelle decides to make the following changes to the current chart to improve its appearance: Split the series by position, with the last 13 values assigned to the second plot. Emphasize the information in the main pie chart by reducing the overall size of the second plot and increasing the gap width between the two pie charts. Figure 3.22 shows the Format Data Series dialog box settings that will implement these formatting changes.

3
Level 2
185

Chapter 3

Determining Effective Data Display with Charts

Figure 3.22: Changing the format of the Pie of Pie chart

Second plot modified to contain the last 13 values

Gap width increased to 75% Size of second plot reduced to 5%

As an additional enhancement, Michelle adds category name data labels. The resulting chart is shown in Figure 3.23. Notice the label Other, which identifies the pie slice that represents the total of all the slices in the smaller pie plot. You will make some further formatting changes to this chart, which is still too busy, in the steps section at the end of Level 2.
Figure 3.23: Pie of Pie chart with formatting changes

186

Chapter 3 How To

Determining Effective Data Display with Charts

Format Data Series in a Pie of Pie Chart 1. Right-click on either of the pie charts to open the shortcut menu, and then click Format Data Series on the shortcut menu to open the Format Data Series dialog box (see Figure 3.22). 2. Under Series Options, you can select which data points are included in the second pie chart, or plot. The split between the first and second pie chart can be made by the position, value, or percentage of the individual data points on the worksheet. A custom setting allows pie slices to be dragged between the two pie charts. You can also modify the size of the second plot as well as the distance between the main pie chart and the second plot. 3. Use the Fill, Border Color, and Border Styles selections to specify the settings you want for the border and area of the entire data series. Changes to the border style affect all of the pie slice borders. Any changes to the area color or fill effects settings result in the same appearance for all of the pie slices. 4. After making the necessary formatting changes, click the Close button. Note: Any changes to the data label content must be made using the Data Labels button in the Labels group on the Layout tab. You can choose to display the series name, category name, value, and/or percentage. The Layout tab also provides options for the legend display.

3
Level 2
187

From Pies to Doughnuts: Showing Individual Percentages with Multiple Series

Although doughnut charts are not a chart sub-type, they are related to 100% stacked charts and pie charts because of the type of data displayed. One of the limitationsor advantages, depending on your perspectiveof 100% stacked charts is that they show cumulative rather than individual percentages. That is, the value for the second data point is the sum of the first and second values. Pie charts can show individual percentages, but only for one data series. Doughnut charts are a way to show the information contained in a pie chart for more than one series. Michelle wants to show how the percentages of consumer purchases for the three divisions (Apparel, Equipment, and Footwear) have changed from year to year. She has prepared two sample doughnut charts, shown in Figure 3.24, from the same consumer purchase data she has been working with. One of the problems with doughnut charts is that they can easily become confusing if too many data series are charted at once. Michelle decides that even if she charted only two numeric data series, the doughnut charts would be difficult to understand. She thinks that multiple pie chartsone for each year, for examplewould be much easier for decision makers to interpret.

Chapter 3

Determining Effective Data Display with Charts

Figure 3.24: Two sample doughnut charts

Doughnut chart showing the relative contribution that each year made to the three categories

Doughnut chart showing the relative contribution that each category made to each year

Monitoring a Business with Stock Charts

On a regular basis, Michelle compares TheZones performance in the stock market with that of other competitors in the sporting goods industry. Probably the most important investor data in business can be found in the financial sections of newspapers around the world. The international stock market is the lifeblood of business. Information about the daily variations in a companys value is tracked by everyone from day traders, to investment bankers, to individual investors, and to the companies themselves. Commonly reported stock information also includes the opening and closing prices as well as the highest and lowest prices achieved. The volume of stock sales in a particular period is also commonly reported. Excel line charts can be used to track changes in the prices of stocks at the end of each period, but they are not ideal because each of the values mentioned previously requires a single line representing one data series on the chart. A single line representing the closing prices for the time period would be the easiest to understand, but the range of values that a stock has during the reported period can be as significant as the ending price to many investors. Each of the commonly reported items of stock information can be of importance to investors. Excel can be used to create a form of candlestick plot, which is commonly used in financial stock reporting, to show all of this information in a compact chart. The candlestick plot is a modification of the box-and-whisker plot, which is common in statistics. All of the stock reporting charts available in Excel are somewhat based on the candlestick plot format, taking pieces of it as appropriate to report specific data. Michelle needs to create a chart that summarizes the companys stock prices for the last year. She has a worksheet containing the daily stock price and volume information for the last several years, and now must choose one of the four available stock chart sub-types to
188

Chapter 3

Determining Effective Data Display with Charts

display the data. The different sub-types display varying combinations of the high and low stock prices that are achieved for a day, as well as the days opening and closing stock prices. Some sub-types add the volume or number of shares traded in a day to the chart. All of these chart sub-types require that the data in the worksheet be arranged in a specific order.
Creating a Chart with the High-Low-Close Sub-Type

3
Level 2
189

First, Michelle considers the High-Low-Close chart sub-type. With this sub-type, the data to be plotted must be placed by column in this order: high stock value for the day, low stock value, and, finally, closing stock value, as shown in Figure 3.25. The range of lowest to highest stock prices for each time period is shown by a vertical line. The tip of the arrowhead represents the closing price for the time period.
Figure 3.25: Worksheet and sample High-Low-Close chart

Vertical line represents the range of lowest to highest stock prices for the time period

The tip of the arrowhead represents the closing price for the time period

The data must be in the same order as the name of the chart type: High-Low-Close

Creating a Chart with the Open-High-Low-Close Sub-Type

The Open-High-Low-Close chart sub-type adds information on the opening price of the stock for the time period. This chart sub-type also requires the data to be in a specific order in the worksheet. The order matches the name of the chart sub-type; in this case, the data must be listed in Open-High-Low-Close order to plot the chart properly. After adding the opening stock price to the data series, Michelle creates a chart with the Open-High-Low-Close

Chapter 3

Determining Effective Data Display with Charts

sub-type, shown in Figure 3.26. The range of lowest to highest stock prices is still represented by a vertical line for each time period. A box is included to show the opening and closing prices. The bottom of the box represents the lower of the opening or closing values, while the top of the box represents the higher. Color is used to show if the stock closed at a higher or lower price than its opening price. If the box is white, the stock increased in value for the time period. The white box is referred to as the up bar. If the box is black, the stock decreased in value for the time period. The black box is referred to as the down bar.
Figure 3.26: Sample Open-High-Low-Close chart

A white box indicates an increase in the stock value

The range of lowest to highest stock prices is represented by a vertical line

A black box indicates a decrease in the stock value

The data must be in the same order as the name of the chart type: Open-High-Low-Close

The top and bottom of the box indicate the opening and closing prices

Best Practice Changing the Default Colors for Excel Stock Charts Many analysts dislike the default colors that Excel uses to indicate if a stock closed lower or higher than its opening price. Most businesspeople consider in the black to be positive for a company, but the default Excel stock charts in candlestick format use a black bar to indicate that the stock closed down. One option is to change the color of the up bars to green, and the down bars to red, so that the colors better reflect the positive or negative aspect of the bars.

190

Chapter 3 How To

Determining Effective Data Display with Charts

Adding Volume to Stock Charts

Volume information showing how many shares of stocks were traded in the time period can be added using the Volume-High-Low-Close chart sub-type or the Volume-OpenHigh-Low-Close chart sub-type. These two chart sub-types shift the stock price scale to the right-side axis, or secondary y-axis. Columns are used to indicate the stock volume for each time period. The primary y-axis shows the stock volume on the left, and the columns representing the share volume are compared to it. The secondary y-axis is on the right, and the stock prices shown by the lines/bars/boxes are measured against it. Using the secondary axis in a chart is a common way to include widely differing sets of data from the same time period. These types of stock charts also require the data to be arranged in a specific order in the worksheet. After moving the column containing the volume to the correct location, Michelle creates the chart shown in Figure 3.27.
Figure 3.27: Sample Volume-High-Low-Close chart

Stock volume is shown on the primary y-axis

Stock price is shown on the secondary y-axis

191

Level 2

Format Up and Down Bars in Stock Charts 1. Right-click either an up bar or a down bar in the stock chart to display the shortcut menu. 2. Click either the Format Up Bars option or the Format Down Bars option on the shortcut menu, depending on which type of bar you right-clicked, to display the appropriate dialog box. 3. Select the Fill category, and then specify the color and/or pattern you want to apply to the bars. 4. Click the Close button.

Chapter 3

Determining Effective Data Display with Charts

The default color setting on the column portion of the chart can obscure the High-Low line for time periods with exceptional volume. One option is to format the column data series with a different color that allows the High-Low lines to be displayed, as illustrated in Figure 3.28. Notice that the first volume column is in the same space as the stock price data for that time period. A dark column color would make the stock price information very difficult to interpret.
Figure 3.28: Modified Volume-High-Low-Close chart

Color change makes stock price more visible

What would the impact be if the chart showed all the price and stock information for the previous year, as opposed to showing the information for only April 2009 as it does now? The Volume-Open-High-Low-Close chart sub-type combines all of the available information into one chart: lines for the high-low price range, boxes showing the opening and closing prices, and columns indicating the time period volumes. This information looks great with a small data set, but when all the daily stock information for an entire year is charted, as shown in Figure 3.29, the chart becomes almost illegible. This chart is, essentially, a cluttered line chart superimposed over a column chart. This format wont work at all for the analysis of annual stock information.

192

Chapter 3

Determining Effective Data Display with Charts

Figure 3.29: Volume-Open-High-Low-Close chart showing data for one year

3
Level 2
Stock price information Stock volume information Because this chart shows a years worth of stock price data, the difference between the opening, high, low, and closing prices cant be determined

Adding Trendlines and Moving Averages

Michelle can use a line chart to show the overall trend in the closing price of the stock for the past year, combined with a column chart indicating the stock volume. Or, she could summarize the daily stock prices by month, retaining all the information in the chart. Michelle chooses the first alternative and decides to do the following: Create a line chart from the closing prices for each day and place it directly above the original chart. Add a linear trendline and a 30-day moving average to smooth out the day-to-day variations in the data. Change the source data in the original chart to the stock volume only. Change the chart type of the original chart to column. Remove the title from the new volume chart and make the plot area shorter.

193

Chapter 3

Determining Effective Data Display with Charts

The resulting chart is shown in Figure 3.30.

Figure 3.30: Alternative line and column chart combination

Stock closing prices Linear trendline 30-day moving average

This chart is made up of two charts: a line chart on the top and a column chart on the bottom

Stock volume

This chart shows how movements in stock price are related to changes in volume

This chart is similar to those used by many financial publications to illustrate trends in a companys stock prices and the volume of stock sold in a given period. The purpose of a chart like the one in Figure 3.30 is to show overall trends, not to show the price or volume of a stock for a particular day. Notice that the chart shows that some spikes in the volume of stock sold have occurred at the same time that the stock price has reduced dramatically. This usually indicates that investors have sold off the stock in response to particularly bad news about the companys financial strength or future prospects. The linear trendline shows a definite upward trend in the stock price. Trendlines graphically illustrate trends in the data using a statistical technique known as regression. More advanced uses of trendlines include forecasting forward or backward in the data set, but these topics are best left to those with a strong background in regression to avoid misapplying the tool. The 30-day moving average trendline is used to smooth out the data, making it easier to spot trends. Each point on the 30-day moving average line represents an average of the stock closing prices for the last 30 days. The term moving average is used because the average is calculated each day for the last 30 daysmoving the average along the chart. Varying degrees of smoothing are achieved by changing the length of the moving average. The longer the time period, the less sensitive is the line to day-to-day variability in the price. Generally when a stock price moves above the moving average trendline, it is a good indicator of an upward trend.
194

Chapter 3 How To

Determining Effective Data Display with Charts

Add Trendlines and Moving Averages to a Chart 1. Click the data series to which you want to add a trendline or moving average. Clicking anywhere in a chart displays the Chart Tools tabs. 2. On the Chart Tools Layout tab, in the Analysis group, click the Trendline button to open the Trendline options. 3. Select the Trendline type you want to apply to the data series. The choices include linear, exponential, and linear forecast logarithmic. A two-period moving average selection is also available in this list of options. 4. If you want to change the number of periods in the Moving Average, click More Trendline Options. This opens the Format Trendline dialog box, where you can specify the number of periods over which you want to calculate the average. This dialog box allows you to modify the setting for the various trendlines available in Excel. 5. In the Format Trendline dialog box, set the number of periods and the forecast parameters for the trendline, as necessary. 6. Click the Close button.

3
Level 2
195

Michelle is pleased with the results, but still wants to explore how to show annual stock data using a Volume-Open-High-Low-Close stock chart. She copies the worksheet and begins to consider how to summarize the data. The stock volume is simple; she needs to add up the stock traded during the month and can use a SUM function to do so. MAX and MIN functions can be used to take care of the highest and lowest stock price for each month. The opening and closing prices for each time period present more of a problem. How do you select a value that corresponds to the first and last day of the month from a range of values? Michelle decides that the quickest way to do this is to simply select these cells in the worksheet. She creates a summary table of the monthly stock prices and volumes, shown in Figure 3.31.
Figure 3.31: Worksheet with summary table

Summary table of monthly stock data for the past year

Annual volume, opening, high, low, and closing price data

Chapter 3

Determining Effective Data Display with Charts

The resulting Volume-Open-High-Low-Close stock chart is shown in Figure 3.32. Notice that the primary y-axis scale showing stock volumes has been increased to keep the stock volume columns from interfering with the stock price information.
Figure 3.32: Annual Volume-Open-High-Low-Close stock chart

The volume scale has been modified to keep graphs from interfering with each other

Annual volume, opening, high, low, and closing price data summarized by month

Steps To Success: Level 2 In these steps, you will work with the same consumer equipment purchasing data used to complete the Pie of Pie chart that Michelle worked on earlier. This data and the Pie of Pie chart are contained in the Excel workbook named Purchase.xlsx. Youll use this data to create a Bar of Pie chart. This same workbook contains detailed purchasing data broken out by age and gender for the following: footwear, camping, fitness, and golf. Youll use this data to create charts that illustrate the relationship between age and gender. The goal is to apply Tuftes principles to effectively present this quantitative information. Complete the following:
1. Open the workbook named Purchase.xlsx in the Chapter 3 folder, and then save it as

Consumer Purchases.xlsx. Review the data sets on the six worksheets. You need to determine how the chart sub-types covered in Level 2 can be used to illustrate the most important information in each worksheet.
2. The Pie of Pie worksheet contains the consumer purchase data organized by sport that

Michelle used to create the Pie of Pie chart earlier. Use this data to create a Pie of Pie chart for the 2009 data. Place this chart on the same worksheet as the source data. Format the chart to look like the chart shown in Figure 3.23. Include the chart title 2009 Consumer Purchases by Sport on your chart.

196

Chapter 3

Determining Effective Data Display with Charts

3. Use the data found on the Purchases by Sport worksheet to create a Bar of Pie chart

that illustrates the percentage contribution by each sport to the total purchase amount. Create one Bar of Pie chart for each year. Place the three charts on the same worksheet as the source data. Include a chart title for each chart that is similar to the chart title you included in the Pie of Pie chart in Step 2.
4. The Footwear Purchases worksheet contains percentage data for footwear purchases

3
Level 2
197

organized by age groups and gender. The data also includes a breakout of the age and gender of the U.S. population by percentage. Create a clustered column chart, a stacked column chart, and a 100% stacked column chart to compare the purchases by different age groups each year with the U.S. population. Place the three charts on the same worksheet as the source data. Include appropriate titles and a legend in each chart. In a cell below each chart, explain what you think the chart illustrates.
5. The Camping Purchases worksheet contains percentage data for camping purchases

organized by age groups and gender. The data also includes a breakout of the age and gender of the U.S. population by percentage. Create an area chart, a stacked area chart, and a 100% stacked area chart to compare the purchases by different age groups each year with the U.S. population. Place the three charts on the same worksheet as the source data. Include appropriate titles and a legend in each chart. In a cell below each chart, explain what you think the chart illustrates.
6. The Fitness Purchases worksheet contains percentage data for fitness purchases orga-

nized by age groups and gender. The data also includes a breakout of the age and gender of the U.S. population by percentage. Create a line chart, a stacked line chart, and a 100% stacked line chart to compare the purchases by different age groups each year with the U.S. population. Place the three charts on the same worksheet as the source data. Include appropriate titles and a legend in each chart. In a cell below each chart, explain what you think the chart illustrates.
7. The Golf Purchases worksheet contains percentage data for golf purchases organized

by age groups and gender. The data also includes a breakout of the age and gender of the U.S. population by percentage. Create a doughnut chart and one other chart of your choice to compare the purchases by gender each year with the U.S. population. Place the two charts on the same worksheet as the source data. Include appropriate titles and a legend in each chart. In a cell below each chart, explain what you think the chart illustrates.
8. Save and close the Consumer Purchases.xlsx workbook.

Chapter 3

Determining Effective Data Display with Charts

LEVEL 3
EXPLORING MORE ADVANCED CHART TYPES EVALUATING THE EFFECTIVENESS OF RADAR, BUBBLE, AND DASHBOARD CHARTS Michelle is very pleased with the progress that she has made using Excel to chart and visually analyze data. However, after further reviewing both the Pie of Pie and Bar of Pie charts showing consumer purchase data, she doesnt feel that these charts adequately convey the information. She decides to look into radar charts as an alternative way to chart the consumer purchase data. Another data set she is working with has three data points for each category. Charting on three axes can be very challenging. Some options for handling this type of data include a 3-D column chart or a bubble chart. In addition, Michelle wants to set up a management dashboard that shows a visual summary of various performance data she uses on a regular basis.
Understanding Radar Charts

Excel has many advanced chart features and types that can be used to present data in a variety of ways. One of the more advanced chart types is the radar chart. Radar charts are named for their resemblance to the plots on radar screens as they scan a 360-degree circle. Values radiate from the center of the chart in a way that can be compared to radar screen plots, showing the distance of an object from the radar in the center. The categories are represented by lines that radiate out from the center. Although radar charts somewhat resemble pie charts, each segment in a radar chart is always the same size as the other segments. The number of categories determines the number of segments in the chart. The radar chart is also similar in structure to a spiderweb, with each web strand that roughly forms a circle representing the data values. The area under the curves in a radar chart can be shaded, similar to an area chart. Radar charts are like area charts in that they both allow you to compare the contributions of different categories. Whereas area charts emphasize the amount of the contribution, radar charts emphasize the relative comparison of the categories. Radar charts combine the comparison features across categories that area charts provide, with the comparison of the relative contribution to the whole that pie charts provide. As she continues to analyze the data set of consumer equipment purchases by sport, Michelle wants to show which sports generate the most equipment purchases every year. The Pie of Pie chart created earlier shows the percentage contribution of each sport to the industry. Now, Michelle decides to try a radar chart to show the numerical value that each sport contributed to the industry. The worksheet data is sorted alphabetically by category with data for the three-year period 20072009. Michelle selects the 2009 data and produces the chart shown in Figure 3.33. The bold values next to each web strand represent the values of the data.

198

Chapter 3

Determining Effective Data Display with Charts

Figure 3.33: Consumer equipment purchase data plotted as a radar chart

3
Level 3
The circular lines that radiate out from the center represent the axis values for this chartin this case, the $4,000 value The straight lines that radiate out from the center represent categoriesin this case, the Archery category

This chart isnt very informative. The large range of values in the data set results in large jumps in the chart, as evidenced by the wide variances in the shaded areas representing the data values. Viewing the chart, you can pick out only the largest contributors to industry purchases, such as Exercise and Fishing Tackle, with little idea of the values contributed by the rest. Sorting the data in descending order by purchases in 2009 would reorganize the information presented in the radar chart. The resulting chart would only do a slightly better job of showing the dollar amount contributed by each sport category and allowing you to see the relative contribution from each sport category. One of the problems is the range of values present in the data. There are too many insignificant data points that add more clutter than information to the chart. One option is to reduce the range of data charted to those categories that are more significant. Michelle is most interested in those categories that have over $300 million in consumer purchases in 2009. There are 15 equipment categories that meet the criteria, as shown in the radar chart in Figure 3.34. Note that the shaded area moves in a circular fashion around the chart, going up and down as necessary to indicate the data values. For example, the Exercise category data value is $4,992.21, so the shaded area for this category appears very close to the circular line representing $5,000. By contrast, the Fishing Tackle category data value is $2,265.21, so the shaded area moves down for this category and appears close to the circular line representing $2,000. How could you show the categories with the smaller values? If needed, a second chart could be created with the smaller values. Care would have
199

Chapter 3

Determining Effective Data Display with Charts

to be taken that the difference in values for the two charts is obvious or a problem with interpreting the data could result. A note describing this would have to be added to the second chart.
Figure 3.34: Revised radar chart limited to purchases over $300 million

The Exercise category has the largest amount of purchases in 2009

Radar charts can display more than one series, but displaying too much information on the chart can cause problems. For example, Figure 3.35 shows the consumer purchase information from three years in the chart on the left, which is labeled as a filled radar chart. The series overlap in many values, making it very difficult to determine the values represented by the chart. Changing the chart type from a filled radar to a nonfilled (standard) radar chart, shown on the right in Figure 3.35, doesnt help much because the values are so close.
Figure 3.35: Radar charts with multiple data series

In this case, the data for the three years is so close that the 2007 series hides the other two years of data

The standard radar chart uses lines to represent the data

Filled radar chart

Standard radar chart

200

Chapter 3

Determining Effective Data Display with Charts

Radar charts offer a powerful method of displaying and comparing data across categories. As with all of the charts in Excel, the intended message of the chart must be considered. A positive feature in one chart type might become negative if it doesnt support what the chart creator is trying to illustrate. Often, the best way to choose a chart is through trial and error.
Understanding Bubble Charts

3
Level 3
201

Bubble charts allow three-dimensional data to be plotted in 2-D on two axes. A bubble chart is similar to an X Y (Scatter) chart. The x- and y-axes are numeric rather than having one axis containing categories, as in bar and column charts. The third dimension is also numeric and is represented by varying the size of the data point that intersects with the xand y-axisthe larger the bubble, the larger the value. The varying sized data points, or bubbles, give this chart its name.
Plotting 3-D Data in Two Axes: Bubble Charts vs. 3-D Column Charts

Michelle has been working with some market share data for the Footwear division and wants to display the relationship between the number of styles offered, sales, and market share. A 3-D column chart has three axes and should, therefore, work for this data. Figure 3.36 shows the 3-D column chart; in this chart, the plot area was rotated to better show each data series. The great difference in scale for the number of styles, sales, and market share make this chart less than useful.
Figure 3.36: 3-D column chart of Footwear division market share

The magnitude of the Sales figures makes this chart difficult to interpret

Michelle decides to try a bubble chart to display this set of data. She also removes the legend from the right side of the chart and adds a title. The resulting chart is shown in Figure 3.37.

Chapter 3

Determining Effective Data Display with Charts

Figure 3.37: Bubble chart of Footwear division market share

The market share of each shoe style is represented by the size of the bubbles

The bubble chart displays the number of styles on the x-axis and the sales on the y-axis; the market share is represented by the size of each bubble on the chart. Now, the chart needs some descriptions to clarify the data on the chart. Michelle runs into a slight problem when attempting to add the shoe style categories to the chart. Adding the series name as a data label to the chart results in the problem shown in Figure 3.38. The series name has no meaning because all the data is in one series.
Figure 3.38: Adding a series name to the bubble chart

In this chart, the market share for each shoe is in the same series

The shoe category names can be assigned to the bubble chart by selecting the Index tab of the Data Source dialog box and then adding each shoe type as an individual series. Adding the series requires the individual entry of the name, x-value, y-value, and bubble size. Michelle adds the x- and y-axis titles and selects the series name as a data label to create the final chart shown in Figure 3.39.

202

Chapter 3

Determining Effective Data Display with Charts

Figure 3.39: Final bubble chart showing Footwear division market share versus sales versus number of styles

3
Each shoe style is entered as an individual series

Solving Bubble Chart Problems

Bubble charts can be temperamental to work with. The source data must be selected without including labels. Figure 3.40 shows what happens if the row and column labels are selected along with the source data. Excel is confused by the addition of what appears to be category data and attempts to assign numerical values to it, thereby rendering the chart useless.
Figure 3.40: Range selection problems with bubble charts

Excel attempts to plot the Styles and Sales data as a series, resulting in bubbles that are plotted for these values

Best Practice Adding Category Labels to Bubble Charts Without Using Multiple Series There is a way to trick Excel into displaying category labels without having to go to the trouble of entering individual name, x-value, y-value, and bubble size values for each item: use custom formatting to hide the displayed value for the range of cells associated with the bubble sizein this case, market share. The value data label displays the cell contents for the range of cells identified as the data source for the bubble size, as shown in Figure 3.41. The chart data range for this figure is (B4:C7, E4:E7).
203

Level 3

Chapter 3

Determining Effective Data Display with Charts

Figure 3.41: Bubble chart tricks

Cell E4 contains the formula "=D4" No matter what value is entered into these cells, the category names will be displayed because of the custom cell format

The value data label shows this custom cell format for the market share bubble size

To create a custom cell format that accomplishes this trick, display the Number tab of the Format Cells dialog box; then enter the category text within quotation marks, as shown in Figure 3.42. The category name becomes the displayed contents of the cell. The actual cell contents are hidden behind the text string and are used for any calculations.
Figure 3.42: Custom cell formats

No matter what value is entered into the cell, the category name will be displayed because of the custom cell format

204

Chapter 3

Determining Effective Data Display with Charts

Creating a Management Dashboard

Michelle has been reviewing several business intelligence software packages featuring dashboards that display performance indicators in a fashion similar to the instrument panel in a cars dashboard. These dashboards usually feature a set of charts that summarize several sets of data graphically. The most intriguing ones feature charts that resemble actual dashboards. Michelle prefers the flexibility that using Excel gives her, but would like to add a dashboard worksheet to provide a quick, visual summary of performance indicators such as customer satisfaction, customer returns, on-time deliveries, monthly sales, and other accounting measures. She decides the first step is to create a chart that duplicates the gauges found on many of the software packages she has been reviewing. The chart will illustrate the average customer purchases per year for TheZones top customers.
Building a Dashboard Chart

3
Level 3
205

Michelle builds the dashboard chart shown in Figure 3.43. This chart is a doughnut chart with some of the data segments hidden to create the 180-degree circular band. Doughnut charts assign the relative size of each segment according to the segments contribution to the total. For example, in a two-segment doughnut chart, each with the same value, each segment takes up one-half of the doughnut. Three equal segments each take up one-third, four segments each take up one-fourth, and so on. Because a doughnut chart is really a circle, Michelle can use the properties of a circle to fit the values into her dashboard chart.
Figure 3.43: Final dashboard chart

The user can enter the lower and upper limits on the worksheet and the appearance of the gauge will change to match

Value indicator

Defining the Normal Operating Range

The chart shows three segments in the top arch: the leftmost section for values below the normal operating range limit, the normal operating range section, and the rightmost section for values above the upper limit. A fourth segment of the doughnut chart takes up one-half of the chart and is hidden at the lower part of the chart. Because there are

Chapter 3

Determining Effective Data Display with Charts

360 degrees in a circle, the easiest thing to do is to convert all values into degrees. The dashboard chart needs to display only one-half of the circle, so the lower hidden segment will always have a value of 180 degrees. Michelle wants to be able to dynamically adjust the normal operating range limits on the chart by typing values into the worksheet that define the lower and upper limits. The worksheet in Figure 3.44 is used to accomplish this.
Figure 3.44: Dashboard chart setup

These cell formulas are used to build the final dashboard chart

These three sections will be hidden by changing the color to match the background

Michelle enters the lower limit, upper limit, and maximum (max) value of the gauge into the cells in row 10 of the worksheet. The max value will always occur at the 180-degree position of this chart, so it is divided by 180 degrees, in cell D10, to get a conversion value to use for the other values. The chart source data for the first gauge segment is created in cell B3 by multiplying the lower limit value in cell A10 by cell D10, converting the lower limit into degrees. The second chart segment is calculated in cell B4 by multiplying the upper limit value in cell B10 by the conversion factor in cell D10 and then subtracting the value for the first segment in cell B3. The third segment is whatever remains in the 180-degree arc; this is calculated in cell B5 by subtracting the value for the first two segments from the value of the lower hidden segment in cell B6.
Creating the Value Indicator

The value indicator is the portion of the doughnut chart that is below the main arch of the dashboard chart (see Figure 3.43). The value indicator only requires three segments for its doughnut chart instead of the four needed for the gauge. The first segment is calculated in cell C3 by multiplying the value entered into cell A13 by the angular conversion factor in cell D10. The value of the second segment is calculated in cell C4 by subtracting the value of the third and final segment in cell C5 from the first segment in cell C3. The third segment is the portion of the doughnut that will be hidden at the bottom of the chart and

206

Chapter 3

Determining Effective Data Display with Charts

is equal in value to the normal operating range fourth segment that is to be hidden in the same way.
Creating the Doughnut Chart

Michelle uses the Insert tab on the Ribbon to create the doughnut chart shown in Figure 3.44. She uses the Ctrl key to select the range B3:B6, C3:C5. The default doughnut chart starts the angle of the first slice at the 12 oclock position, or 0 degrees. Michelle right-clicks the data series and opens the Format Data Series dialog box. The angle of the first segment setting is located on the Options tab. Michelle sets the angle to 270 degrees to have the chart start the gauge at the 9 oclock position. This hides the bottom half of the doughnut chart. Michelle hides the lower segments of the chart by selecting each segment and opening the Format Data Point dialog box, shown in Figure 3.45. In the Fill category, she sets the Fill option to Solid fill, and in the Border Color category she sets the Border Color to No line, making these segments seem to disappear. The second segment of the value doughnut can also be hidden in this manner.
Figure 3.45: Format Data Point dialog box

3
Level 3
207

Adding Digital Values to the Chart

The lower and upper normal operating range limits and the plotted value can be added to the chart to improve the information that the chart provides. The Shapes button in the Insert group on the Layout tab contains a variety of shapes that can be added to the chart.

Chapter 3

Determining Effective Data Display with Charts

The plotted value digital display is an oval object found on the Shapes palette. Michelle selects the chart, clicks the Oval button on the Shapes palette, and then draws the oval in the middle of the chart (see Figure 3.43). To display the value in the oval, she clicks the oval on the chart, types = in the Formula Bar, and then selects the cell containing the value she wants to link to this shape. For the plotted value, this is cell A13. Michelle then formats the text with bold and a larger font size to improve clarity. She creates text boxes, again using the Shapes palette, to display the upper and lower limits in the dashboard chart. These can be linked to cell values in the same way that the oval shape was. With the dashboard chart complete, Michelle tests it by entering different numbers for the normal operating range lower and upper limits, the max gauge value, and the value to be plotted. She ensures that the chart responds properly to these changes. Michelle is pleased with her work on this dashboard style chart. She will use this chart as a template for the other dashboard style charts she is planning to use to monitor performance indicators or metrics for the company. Steps To Success: Level 3 In these steps, you will create charts to illustrate and analyze consumer purchase data for the Footwear division of TheZone. Complete the following:
1. Open the workbook named Footwear.xlsx in the Chapter 3 folder, and then save it

as Footwear Analysis.xlsx.
2. Use the Footwear Purchases by Price worksheet in the workbook to produce charts

that compare the average price for each category of footwear. Create a radar chart and an appropriate version of a bar chart to see which one best illustrates the information. Place the two charts on the same worksheet as the source data. Include appropriate titles and a legend in each chart. Determine which chart does the best job of illustrating the relative contribution of each category to the purchase amount, and which one is best for emphasizing the year-to-year trend for each category. In a cell below each chart, explain what you think the chart illustrates.
3. Michelle has been working on sport participation data by gender and equipment sales.

She built a worksheet by combining data from the 2009 overall and 2009 womens participation in sports data with the equipment purchases by sport data. Use the data in the Participants and Equipment worksheet to create a bubble chart that shows the number of women and the number of men participating in sports, as well as the amount of equipment sales that each sport generates. Place the chart on the same worksheet as the source data. Include appropriate titles and category labels in the chart.
4. Save and close the Footwear Analysis.xlsx workbook.
208

Chapter 3

Determining Effective Data Display with Charts

CHAPTER SUMMARY This chapter presented many of the ways that charts can be used to illustrate quantitative information. Charts offer the opportunity to add visual analysis to problem solving. Tuftes guiding principles on the creation of graphics can be applied in Excel to create excellent charts that present quantitative information clearly, precisely, and efficiently. In Level 1, you learned how the choice of chart type can influence a viewers perception of the information presented. Differences between the main chart typesline, column, bar, area, pie, and X Y (Scatter)were examined, as these charts were used to summarize and present a variety of data. Many of the chart types can be used on the same types of data, resulting in different interpretations of that data. Level 2 moved into coverage of chart sub-types, such as stacked and 100% stacked, for line, column, and area charts. These sub-types further summarize the data being presented, allowing a comparison of the contribution of a particular data series to a greater whole. The Pie of Pie and Bar of Pie chart sub-types were introduced to show how particular elements of pie charts can be summarized in another format. The different sub-types of stock charts were covered in some detail, showing the strengths and weaknesses of the various ways that Excel presents stock price and volume information. You also learned how trendlines and moving averages can help to clarify the data in stock charts. In Level 3, the discussion was expanded to include more advanced chart types, such as radar and bubble charts. You also saw how to combine chart types within the same chart to build a management dashboard. This management dashboard consists of speedometerlike gauges that can be used to monitor various business metrics at a glance, allowing the viewer to see if any particular areas of a business require more investigation. Table 3.2 summarizes the chart types and sub-types presented in this chapter.
Table 3.2: Summary of Excel chart types and sub-types Chart Type Column Clustered Column Stacked Column 100% Stacked Column 3-D Column Bar Clustered Bar Stacked Bar 100% Stacked Bar Compares values across categories Compares the contribution of each value to a total across categories Compares the percentage each value contributes to a total across categories Compares values across categories Compares the contribution of each value to a total across categories Compares the percentage each value contributes to a total across categories Compares values across categories and across series Description

3
Chapter Exercises
209

Chapter 3

Determining Effective Data Display with Charts

Table 3.2: Summary of Excel chart types and sub-types (cont.) Chart Type Line Line Stacked Line 100% Stacked Line 3-D Line Pie Pie 3-D Pie Pie of Pie Exploded Pie Bar of Pie X Y (Scatter) Displays the contribution of each value to a total Displays the contribution of each value to a total with a 3-D visual effect Displays a main pie chart with user-defined values extracted and combined into a second pie plot Displays the contribution of each value to a total value while emphasizing individual values Displays a main pie chart with user-defined values extracted and combined into a stacked bar as the second plot Compares pairs of values; display options include Scatter with data points connected by lines; Scatter with data points connected by smoothed lines without markers; and Scatter with data points connected by lines without markers Displays the trend of values over time or categories Displays the trend of the contribution of each value over time or categories Displays the trend of the percentage each value contributes over time or categories Displays the contribution of each value to a total (similar to a pie chart), but can contain multiple series Displays the contribution of each value to a total value while emphasizing individual values (similar to an exploded pie chart), but can contain multiple series Displays changes in values relative to a center point Compares sets of three values; similar to a scatter chart with the third value displayed as the size of a bubble marker Requires three series of values in this order Requires four series of values in this order Requires four series of values in this order Requires five series of values in this order Displays a trend over time or categories Displays the trend of the contribution of each value over time or categories Displays the trend of the percentage each value contributes over time or categories Displays a line with a 3-D visual effect Description

Area Area Stacked Area 100% Stacked Area Doughnut Doughnut Exploded Doughnut

Radar Bubble Stock High-Low-Close Open-High-Low-Close Volume-High-Low-Close Volume-Open-High-Low-Close

210

Chapter 3

Determining Effective Data Display with Charts

CONCEPTUAL REVIEW
1. List and describe the significance of each of Tuftes five data graphics principles. 2. What are the three steps involved in using the Insert tab to create a chart? 3. Give an example of a low data-ink ratio in a chart. 4. How do you change the chart type of an existing chart? 5. What charting limits does Excel have in terms of data points and series? 6. How many standard chart types and sub-types are available in Excel? 7. What are the differences and similarities between a line chart and an X Y (Scatter) chart?

3
Chapter Exercises
211

When should you use each one?

8. Explain the difference between the data points in a line chart and an X Y (Scatter)

chart.
9. What are the differences between a bar chart and a column chart? Give an example of

when you would use each one.

10. What are the differences between a column chart and an area chart? Give an example

of when you would use each one.

11. How do pie charts differ from doughnut charts? 12. When should you use a stacked line, column, or area chart? How do the stacked charts

differ from regular charts?

13. When should you use a 100% stacked line, column, or area chart? How do the 100%

stacked charts differ from stacked charts?

14. What chart sub-types are available for the stock chart in Excel? Explain how you inter-

pret the data markers on each of the sub-types.

15. How does a radar chart differ from other charts? Give an example of when you would

consider using a radar chart.

16. Define a bubble chart and explain what type of data is suitable for plotting on a bubble

chart.
17. What is a dashboard chart used for?

Chapter 3

Determining Effective Data Display with Charts

CASE PROBLEMS Level 1Illustrating Travel Data for the Indiana Department of Tourism As a travel industry consultant, you have been working for the state of Indiana Department of Tourism. The department manager has asked you to present a report on where Indiana residents travel to and where Indiana visitors come from. The modes of transportation must also be presented. You will also include some information from a survey that your firm has recently completed. The travel survey is located in the workbook named Indiana.xlsx. Complete the following:
Marketing

1. Open the workbook named Indiana.xlsx in the Chapter 3 folder, and then save it as

Indiana Travel.xlsx.
2. Create a bar chart to illustrate the differences among the travel characteristics of the

average Indiana resident, the average U.S. resident, and the average Indiana visitor.
3. Create a bar chart to illustrate a state-by-state comparison of the number of residents

from the rest of the nation that visit Indiana to the number of Indiana residents that visit other states. Rank the information in this chart by the states with the most visitors to Indiana.
4. The Indiana Department of Tourism is very interested in learning more about how

people travel in Indiana. They want to use this information to better target advertising to specific demographic groups and methods of travel. Use the appropriate chart or series of charts to illustrate how people travel; the purpose of their travel; and their income levels, age, and gender. This information should be presented for people who live within the state and those visiting the state with a comparison between the two groups.
5. Save and close the Indiana Travel.xlsx workbook.

Level 2Analyzing Stock Performance for Universal Investments As an analyst at the Universal Investments financial company, you regularly monitor the performance of your clients investments. You are preparing for a meeting with one of your investors to discuss how their stocks have performed over the last year and changes they should make to their portfolio. Your client has asked for information on four companies: Intel, AMD, Time Warner, and Wal-Mart. You will find daily stock price and volume information contained in the Invest.xlsx workbook. Complete the following:
1. Open the workbook named Invest.xlsx located in the Chapter 3 folder, and then save

Finance

it as Investment Performance.xlsx.
212

Chapter 3

Determining Effective Data Display with Charts

2. Create a chart for each company using the appropriate stock chart sub-type. Be certain

to use each chart sub-type only once.

3. Summarize the data to clearly show some of the data markers in the charts. 4. Place each chart on a new sheet in the workbook. 5. Use the appropriate chart to compare the change in stock price for each company over

3
Chapter Exercises
Sales
213

the last year.

6. Using the information on the Portfolio worksheet, prepare the appropriate chart to

display the relative contribution of each stock to the value of the portfolio.
7. Save and close the Investment Performance.xlsx workbook.

Level 3Illustrating Patterns in Gas Prices for CKG Auto You have been assigned to CKG Autos sales department as an analyst. The marketing team for the new Safari Wildebeest Sport Utility Vehicle has become concerned that the rising cost of gas will negatively affect vehicle sales. They are considering introducing limited rebates at times when gas prices are especially high. The company might also increase its marketing of the Safari Meer Cat, the new compact, sporty SUV the company introduced late last year. Your supervisor has asked you to put together a report and presentation showing the price per gallon of gas by region and large metropolitan areas. The company is interested to see if there are any patterns in the data that can be used to better time the rebate promotions. The Safari.xlsx file contains data on gas prices for various regions and metropolitan areas around the country. Complete the following:
1. Open the workbook named Safari.xlsx located in the Chapter 3 folder, and then save

it as Safari Analysis.xlsx.
2. Create a series of charts that compare gas prices by time of the year and by location.

The time can be in quarters, months, or week of the month, as you deem appropriate. The location should be charted separately by region, by state, and by city.
3. For each locationregion, state, and citydetermine if a particular location has higher

gas prices than the others. Add a comment near each chart to explain your findings.
4. Determine if a particular time of year has higher gas prices than the other times of the

year. Add a comment near each chart to explain your findings.

5. Create a bubble chart to illustrate the number of weeks that each citys gas price is

above the gas price for the entire United States; the number of weeks that it is below the gas price for the entire United States; and the average gas price in dollars per gallon for the year. Add appropriate titles, labels, and a legend to the chart.
6. Save and close the Safari Analysis.xlsx workbook.