From UVA Writing Center

If a claim states what you think and reasons state why you think that, evidence shows how you know something to be the case. Here is the example from reasons again: If you went to the doctor and the doctor said "I need to give you a shot in the stomach with this foot-long syringe," you would most likely want to know why. If the doctor says "Because this is the only way to cure the mutant strain of malaria you contracted in Myanmar," then the doctor has given you a reason that may convince you to be poked. But you may want to verify the doctor's claims with some real-world evidence, so you may ask the doctor how the vaccine works. If the doctor then went on to explain how the vaccine contained the special anti-mutant-malarial antibodies that devoured mutant malaria, then the doctor has given you evidence to support the reason. Together, that reason and evidence provide support for the overall claim let me impale you with this syringe.

Recall that reasons exist in the mind of the writer and rest on evidence, whereas evidence is found outside of the writer in the world and rests on itself.

Contents 1 Primary and Reported Evidence 2 Four Maxims of Quality Evidence 2.1 Is the Reporting of the Evidence Accurate? 2.2 Is the Reporting of the Evidence Precise? 2.3 Is the Evidence Presented Representative? 2.4 Does the Evidence Come from Reliable Sources? 3 Trusting Evidence

Primary and Reported Evidence

In an essay all evidence is reported evidence. It's just not possible to present, say, the murder weapon with your paper; you'll have to describe it. This being the case, the accuracy, precision and closeness to the source of the evidence is of prime importance. And that's where the following Four Maxims of Quality Evidence becomes of use:

Four Maxims of Quality Evidence

Is the Reporting of the Evidence Accurate?

In other words, are you representing the evidence as well as you can. The facts have to be right.

• Can the evidence be looked up and independently verified?
• Is that evidence not contestable? In other words, can it not be reasonably challenged (you can challenge that the Loch Ness Monster exists, but challenging the evidence that eagles fly would be quite a chore).
  • Corollary: It is also impossible to prove a negative: One would have a hard time proving the Loch Ness Monster doesn't exist, if there's no evidence one way or the other. One fallacy of logic is to challenge your audience to accept a claim or prove it wrong. The burden of proof rests with the presenter, so it is up to the writer to present the evidence of Nessie, not up to the audience to disprove the claim.
• Is the Evidence Comprised of Facts? Don't confuse reasons with evidence; evidence rests on itself.

Is the Reporting of the Evidence Precise?

In other words, your facts may be straight, but are the facts appropriate to support the argument you're trying to make? If you're making the argument that there are far too many werewolves in Charlottesville eating undergrads on weekends, a factual discussion about the number of eaten undergrads being between 12 and 112 may be accurate, but not nearly precise enough to be useful.

• How closely should the evidence match audience expectations? A five-page paper about a supercollider does not need to present a full theoretical breakdown of sub-particle physics.

Is the Evidence Presented Representative?

Depending on what you are arguing, you may need a large amount of data, or a specific amount. If you're discussing how many cattle have been eaten by werewolves in Albermarle County, your data will be fairly specific. If you want to talk about werewolf attacks in general, you'll need a much larger data set.

• How strong does your evidence need to be? An argument that audiences will strongly resist will require stronger evidence.
• Does your evidence match the reason you are presenting?

Does the Evidence Come from Reliable Sources?

Your source for a piece of evidence needs to be:

• Current: You wouldn't get new glasses based on your prescription from when you were 12 years old (unless you're 12 years old).
• Reputable in the field you're discussing: You wouldn't use a plumber as a source for an essay on the lower intestine.
• A disinterested party: You wouldn't necessarily quote a fast food joint's press release about nutritional content unless you're holding that release up to scrutiny.
• Close to the source of the evidence: A magazine article quoting a journal quoting an experiment is not as good as the published results of the experiment itself.

Trusting Evidence

Wherever you get your evidence from, know that it's already being shaped by the source of that evidence. Back to those werewolves: Reports of the werewolves of Charlottesville will vary widely according to who is presenting those accounts. The local newspaper will describe the problem in words, while the local TV news may show some images of carnage, the four-year-old may describe the giant doggie who barked funny, and the old coot storyteller may frame the account inside the history of Charlottesville supernatural oddities since Thomas Jefferson.

This is know as the Roshomon Effect, which comes from the 1950 Akira Kurosawa film Roshomon. The film presents six different accounts of a rape and murder in feudal Japan, and each account differs from the others; the Woodcutter even contradicts his own first account and gives another.

So when judging evidence, here are a few things to be skeptical of:

Memories: Memories are notoriously unreliable and often change according to circumstances. An eyewitness's recollection of an event is only circumstantial; there needs to be corroborating evidence that verifies that account.

Anecdotes: Anecdotes are interesting, but not as evidence. They're basically memories put in narrative form so they make good stories, and as far as evidence goes, they're as useful as the paper they're written on.

Authorities: An authority on a subject can be useful as far as evidence goes, but can also be selectively used to make it seem like the authority's opinion supports the writer's. To make an authority's account useful, you also need to show the evidence that the authority has gathered - otherwise you run the risk of just restating your point, but in someone else's words.

Quantitative Data: Mark Twain made popular the phrase "There are three kinds of lies: lies, damn lies and statistics." Stephen Colbert has said "The problem with evidence is that it doesn't always agree with your facts." Statistical data looks great because it's hard to argue with numbers. But when statistical, quantitative data is gathered, the factors used to assemble the data must also be taken into account. For instance, in April, 2006, U.S. Secretary of the Interior Gale Norton announced that for the first time since the U.S. Fish and Wildlife Service started measuring wetlands, the number of wetlands had finally increased. The figure was a net gain of 191,800 acres since 1998. However, in a footnote on pg. 76 of the report, it is noted that man-made ponds like golf course water hazards were also included in the acreage. This hadn't been done before, and when golf course hazards aren't counted, the overall acreage actually continued to decrease.

Visual Evidence: Photography and film can seem to be especially powerful evidence. However, not only can images be doctored, like the covers of so many checkout-lane magazines, but the context in which they occur affects the interpretation of the image. Because of this, you want to be sure to tell your reader the context of any images you present. One example of how photographs can lead or mislead interpretation occurred in New Orleans during the aftermath of Hurricane Katrina (2005).  The topic of "looting" was all over the news, and the topic gained a particularly racial overtone.  In the below photos, note how in the caption for the top photo of the young black man says he was "looting a grocery store" in New Orleans, while the caption for the bottom photo of the young white man says he was "finding bread and soda from a local grocery store."  Neither caption refers to the race of either young man, but the context with the photos suggests very different race-based interpretations of the same behavior.