Most people (especially in Western societies) require some type of evidence, usually in the form of facts and statistics, before they will accept someone else’s claims or position.¹⁴ Facts represent documented occurrences, including actual events, dates, times, people, and places. Facts are truly facts only when they have been independently verified by people other than the source. For example, we accept as true that Abraham Lincoln was the sixteenth president of the United States because this fact has been independently verified by eyewitnesses, journalists, historians, and so forth. Listeners are not likely to accept your statements as factual unless you back them up with credible evidence.

Use Statistics Selectively

Statistics are quantified evidence that summarizes, compares, and predicts things, from batting averages to birthrates. Statistics can clarify complex information and help make abstract concepts or ideas concrete for listeners.

Although audience members may want you to offer some statistics in support of your assertions, they don’t want an endless parade of them. Rather than overwhelm the audience with numbers, choose a few statistics that will make your message more compelling. For example, to avoid bombarding the audience with sales figures, the late Steve Jobs, co-founder and former CEO of Apple Computer, Inc., described iTunes’s early success this way: “We’re selling over five million songs a day now. Isn’t that unbelievable? That’s 58 songs every second of every minute of every hour of every day.”¹⁵

Use Statistics Accurately

Statistics add precision to speech claims, if you know what the numbers actually mean and use the terms that describe them accurately. Following are some statistical terms commonly used in speeches that include statistics.

Use Frequencies to Indicate Counts

A frequency is simply a count of the number of times something occurs:

“On the midterm exam there were 8 As, 15 Bs, 7 Cs, 2 Ds, and 1 F.”

Frequencies can indicate size, describe trends, or help listeners understand comparisons between two or more categories:

Use Percentages to Express Proportion

As informative as frequencies can be, the similarity or difference in magnitude between things may be more meaningfully indicated in percentages. A percentage is the quantified portion of a whole. Describing the number of males and females in the 2013 Colorado population in percentages, rather than in counts (as in the previous section), shows more clearly and quickly the relationship between the two amounts: 50.2 percent male and 49.8 percent female.

Percentages also help audience members easily grasp comparisons between things, such as the unemployment rate in several states:

In May 2013, Nevada had the highest rate of unemployment, at 9.5 percent. At 3.8 percent, Nebraska had the lowest rate.¹⁹

Because audience members cannot take the time to pause and reflect on the figures as they would with written text, consider how you can help listeners interpret the numbers, as in this example:

As you can see, Nevada’s unemployment rate is two and one-half times greater than that of Nebraska.

(See Chapter 16 for more on the differences between oral and written language.)

Use Types of Averages Accurately

An average describes information according to its typical characteristics. Usually we think of the average as the sum of the scores divided by the number of scores. This is the mean, the arithmetic (or computed) average. But there are two other kinds of averages—the median and the mode. As a matter of accuracy, in your speeches you should distinguish among these three kinds of averages.

Consider a teacher whose nine students scored 5, 19, 22, 23, 24, 26, 28, 28, and 30, with 30 points being the highest possible grade. The following illustrates how she would calculate the three types of averages:

The following speaker, claiming that a rival organization misrepresented the “average” tax rate, illustrates how the inaccurate use of the different types of averages can distort reality:

The following example shows how the Tax Foundation’s methodology can overstate the tax burdens of the typical family. Suppose four families with incomes of $50,000 each pay $2,500 in taxes (5 percent of their income) while one wealthy family with income of $300,000 pays $90,000 in taxes (30 percent of its income). Total income among these five families is $500,000, and the total amount paid in taxes is $100,000. Thus, 20 percent of the total income of the five families goes to pay taxes. But it would be highly misleading to conclude that 20 percent is the typical tax burden for families in this group.²⁰

Present Statistics Ethically

Offering listeners inaccurate statistics is unethical. Following are steps you can take to reduce the likelihood of using false or misleading statistics.

Use Only Reliable Sources

Include statistics from the most authoritative source you can locate, and evaluate the methods used to generate the data. The more information that is available about how the statistics came about, including how and why it was collected and what researchers hoped to learn from it, the more reliable the source is likely to be.

Present Statistics in Context

Statistics are meaningful only within a proper context. To help audience members accurately interpret statistical information, inform listeners of when the data were collected, the method used to collect the data, and the scope of the research:

These figures represent data collected during 2014 from questionnaires distributed to all public and private schools in the United States with students in at least one of grades 9–12 in the fifty states and the District of Columbia.

Avoid Confusing Statistics with “Absolute Truth”

Even the most recent data will change the next time the data are collected. Nor are statistics necessarily any more accurate than the human who collected them. Offer the data as they appropriately represent your point, but refrain from declaring that these data are definitive.

Orally Refer to Your Sources

Clearly identify the source of your information and provide enough context (including approximate date of publication) to accurately interpret it. For guidelines on orally citing your sources, see Chapter 10, “Citing Sources in Your Speech.”

Avoid Cherry-Picking

To cherry-pick is to selectively present only those statistics that buttress your point of view while ignoring competing data.²¹ Cherry-picking is a popular tool of politicians and policymakers, who are often accused of selectively referring to only those statistics that boost their arguments and policies.

When you find yourself searching for statistics to confirm an opinion or a belief you already hold, you are probably engaging in cherry-picking. Choosing from among the mean, median, and mode of a distribution the one average that makes the best case for your point, when in fact one of the other averages is the better indicator of what your data represent, is an instance of cherry-picking. Researching statistical support material is not a trip through a buffet line to select what looks good and discard what doesn’t. You must locate as much information as possible that is pertinent to your particular point, and then present it in context or not at all.

Use Visual Aids Whenever Possible

When your speech relies heavily on statistical information, use appropriate tables, graphs, and charts to display the statistics (see Chapters 20–22).

Win Acceptance of Your Supporting Materials

Audience members will accept your examples, narratives, testimony, facts, and statistics as legitimate only if they believe that they are derived from sources that are credible. It is up to you to establish your sources’ trustworthiness and reliability. You can do this by alerting listeners to the sources’ qualifications to report on the information. For detailed guidelines on how to research and cite supporting material, see Chapter 9, “Finding Credible Print and Online Materials,” and Chapter 10, “Citing Sources in Your Speech.”