Speakers draw from four primary evidence types — statistics, examples, testimony, and analogies — each of which serves a distinct rhetorical function and carries distinct risks. Statistics establish scale and significance but risk numbing the audience if not contextualized; specific examples make abstractions concrete but risk overgeneralization from a single case; testimony lends credibility through expertise or firsthand experience but depends entirely on source quality; and analogies make unfamiliar ideas accessible but can mislead if the comparison breaks down under scrutiny. Skilled speakers combine evidence types strategically, using a statistic to establish the scope of a problem, then an example to give it a human face, then expert testimony to validate the solution.
Take a single claim and support it four different ways — once with each evidence type. Evaluate which combination would be most convincing for a specific audience. Analyze speeches that rely too heavily on one type and identify where the argument becomes monotonous or unconvincing.
From your prior study of evidence in speeches, you know that claims without support are merely assertions, and that evidence is what transforms assertion into argument. This topic sharpens that knowledge into something more actionable: not just that you need evidence, but that evidence comes in distinct types, each with its own rhetorical strengths, weaknesses, and failure modes, and that the combination of evidence types is itself a rhetorical decision.
Statistics communicate scale and significance. When you tell an audience that 1 in 5 adults experiences a mental health condition each year, you are doing something no anecdote can do: establishing the scope of a problem in a way that is hard to dismiss as an outlier. But statistics abstract away from experience, and audiences often process numerical data passively rather than viscerally. The most common failure with statistics is presenting them without context — "17,000 cases per year" means nothing unless your audience knows the baseline. Always anchor statistics to something the audience can intuitively grasp: "17,000 cases per year — that's roughly every resident of this city."
Examples do the opposite work: they make abstractions concrete and emotionally resonant. A single vivid case study can move an audience more than a page of data, because human cognition is wired to engage with specific stories. The corresponding risk is the hasty generalization — if you support a broad claim with a single example, skeptical listeners will ask whether the example is representative. The solution is to treat examples as illustrations of a pattern that you've already established with statistics, rather than as primary proof. Together, a statistic and an example are more powerful than either alone: the statistic establishes that the pattern is real and widespread; the example gives the pattern a face.
Testimony lends credibility through authority or experience. Expert testimony (a scientist, a practitioner, a recognized authority) establishes that your claim is endorsed by someone whose knowledge exceeds the audience's. Eyewitness or personal testimony establishes authenticity and emotional truth — "this is what it actually felt like." Both depend entirely on the credibility of the source, which means you must introduce credentials explicitly rather than assuming the audience will recognize them. The failure mode for testimony is argument from authority — citing credentials as a substitute for reasoning rather than as evidence supporting a reasoned argument.
Analogies are often underestimated as evidence, but they are among the most powerful explanatory tools available. A good analogy works by mapping an unfamiliar structure onto a familiar one, allowing your audience to transfer existing understanding rather than build new understanding from scratch. "The immune system works like an army" lets audiences use military intuitions to reason about biological concepts they don't otherwise have. The risk is that every analogy breaks down somewhere — no two things are identical — and a sophisticated audience may probe where the mapping fails. The speaker's responsibility is to use the analogy for what it illuminates, and to know in advance where the comparison's limits lie. The strategic art of evidence is in combining all four types: statistics for scale, examples for human resonance, testimony for authority, and analogy for explanation — each compensating for the others' weaknesses.