Module 2: Using Word, Excel, and PowerPoint for Academic Coursework
Tips for Selecting Chart Types
for Academic Projects
When to Use Certain Types of Charts
Pie Charts
Use the pie chart to present a small number of pieces of data.
There should be at least 3 “slices” and no more than 10.
Use only when the values have a constant sum (e.g., everything
adds up to 100).
The values should have big variations, not equal values.
Column and Bar Charts
Use column or bar charts for discrete, grouped data that has an
order and data that has category names.
The column (vertical) chart should be used for data that have
a natural sequence, such as a time series, for example, 1989,1999,
2000, and so on.
The horizontal (bar) chart should be used for data with no natural
sequence, like Quality of Service, Reputation, Promotions.
Stacked Column Charts
Stacked column charts are a variation of the column chart in which
groups of bars are stacked end to end (instead of side by side).
Stacked column charts are frequently used to compare different sets
of data while showing the components and total of each data set.
Related to the Stacked Column is the Clustered Column, which compares
values across categories.
Example 1: Stacked Column Chart
Example 2: Clustered Column Chart
Formatting Charts
Be sure that every graphical element on your chart is absolutely
necessary. Use this test: If you can remove a title or a line
or a bullet and not affect or change the meaning of the
chart, then that element was not necessary and can be removed.
When you remove “clutter,” your graphic or chart will communicate
more clearly. Compare the examples below.
Before
After
Present one main idea per chart.
Capitalize all important words. Use the same rules that you
would follow for titles in English. See the charts and graphs
in this document as examples.
Use colors and shadings appropriate to the technology that you
will use to present your graphics. If you are going to use the
chart in a printed, black and white research paper, make sure
that the colors are not too dark, and--most important--that the
colors (for example, in your pie chart) will look different from
each other when printed.
If necessary, add labels to help guide readers to important
information. For example, the best solution to the problem of
shading in a pie chart is to label the slices, as shown in Example
3 below.
Example 1: Good
Example 2: Too dark and not enough contrast
Example 3: Slices labelled to clarify elements
Do not use unusual shapes, no matter how interesting, that
make the data difficult to read and understand. For instance,
the cone shape does not add much to our understanding of the chart
below.
Sometimes, you might use the correct chart type, but the result
is that the data is distorted or cannot be read easily. For example,
in Figure 5 below, the number of males using the phone more than
15 minutes per call appears to be less than one. The block representing
females whose calls exceed 15 minutes per call is almost entirely
covered by another column.
In this case, the Chart Type can be changed from a 3-D column
to a clustered column, as shown in Figure 5 below.
Example 1: Data is hidden and distorted
Example 2: Data is clear and easy to read
“Test” your chart on another person. Ask the person to make
a statement based on your chart. Listen carefully. Make adjustments
and test again until other people can clearly understand the key
idea you intend to represent in the chart.
