6. Objective data must be externally verifiable to be of merit.

Many times, the data presented to you in support of an argument will be something that you have no way to verify. For example, consider a medium who tells you they can see ghosts. Can other people see these ghosts? Can they be detected in some way other than by seeing them? If the answer is "no" all around, then you cannot accept the data that this medium can see ghosts as either true or false. It is simply of little worth to you. Now, there is a different piece of data here that certainly can confirm: the medium believes they can see ghosts. This is what we call subjective data; information about someone's own perspective or perceptions. This may be of interest for a variety of reasons, but it is not proof of the objective conclusions that they can actually see ghosts or that ghosts exist at all.

A more common situation is in the re-telling of an event from the past. Your friend Joe tells you that when he was a kid, he saw a huge spider in his mom's garden that had red stripes on its knees. You might be able to verify that a species of spider that grows to that size and with those markings does exist, and whether or not they live in the area where he grew up. But you can never verify for sure if he really saw it, or if he saw exactly what he said he saw. You may well respond with something like "Huh, that's cool" or "Isn't that interesting" and you may well think it is both cool and interesting. But this is not data and shouldn't be the basis for any conclusion or decision, such as not purchasing a house in the area because you are arachnophobiac and afraid of encountering any such thing yourself.

These sorts of statements have no merit as objective data, but they may spurn you to do some investigation and find some real data. For example someone's insistence that they can see ghosts may pique your interest in the subject and cause you to want to learn about the paranormal; and someone's claim that they saw a big spider may make you want to find out if such creatures exist and whether they commonly live in an area you are considering buying a home.

There's another place where this tenet is relevant, and that's what we call a "hunch." Hunches are a wonderful example of the human brain at it's best: being able to pick up a tiny thread of relevance from a vast flood of information. But hunches are not data on which to base conclusions. Like the examples above, a hunch provides just a lead, an indicator of something you might want to explore. A private investigator gets a hunch about who the perpetrator might be. But that's not evidence to bring to court. Instead it suggests that he should do some research on his subject, which he hopes will produce real data confirming or denying his guilt in the crime at hand. So when you have a hunch, use it as a reason to do some research. If you feel uncomfortable, like you're not alone in your house, that's not a good enough piece of data for calling the police, or grabbing your gun and unloading the entire clip through the wall into the next room. No, you'll go and look around to see if there is someone there, gathering real data provided by your eyes.

Hunches are of great value and should not be ignored. But neither should they be treated as anything like truth. Decision-making should be postponed until hard data has been collected and a conclusion reached.