@PerryDolia Excellent synopsis! GA.
I would like to add that “data” and “facts” might be better thought of as “observations.” This helps us remember that, due to the unstable nature of perceptions, there could easily be measurement bias.
As @PerryDolia said, we say we “know” something when that knowledge allows us to accurately (or more accurately) predict the behavior of whatever it is we are observing.
But theories don’t observe. People observe and people generate theories to explain their observations. We call the things we observe “data.” A fact is an observation that just about everyone you ask agrees is an accurate observation.
However, all knowledge is provisional. We consider it true if it helps us predict behavior better than we could by chance. Truth is never absolute. Instead it is considered as a probability. In hard sciences, the probability of A leading to B can be expressed as so great that the chance that A does not lead to B is practically unmeasurable.
In social sciences, we only need to be 95% or 99% or 99.9% confident in the relationship between A and B to consider that relationship to be true. Unfortunately, that relationship does not always tell us if there is a causal relationship between A and B.
Observations are easily biased, so it can be hard to measure a level of reliability of an observation. Thus, no matter how great our theories are, they can be greatly mistaken if our observations are inaccurate or if we misunderstand the causal relationship. Since this happens often, people are often mistaken in what they believe they know. Therefore it is a good idea to be skeptical of most observations and theoretical relationships until you have verified them yourself.
For example, people say that race explains discrimination. However, it may be that race is a stand-in for other factors, such as class and income and education, and that when you incorporate these things in your model, race drops out of the model, since it no longer offers any significant additional explanatory power. So what we think we know about race may turn out to be a chimera. Irrelevant.
Knowledge is provisional. Usually better data and better theories will come along (not necessarily in that order) that replaces the old “knowledge” with new, better knowledge. People who hold onto old ideas in the face of theories with better explanatory power—well, there’s a question for you. Why do they do that?
@yarnlady You say you have two sons, but I can not say that has added to my knowledge. If I saw you and your sons, or if you provided birth certificates, or sent pictures and a video, I would be closer to saying I know you have two sons. However, this is the internet, and people are notorious for making up “facts” on the internet. Based on my observations of your behavior on fluther, I’d wager that if you say you have two sons, I’ll be able to observe two sons if I were motivated to verify your assertion. But it’s still a wager until I have more data, and I’m not sure that the information you’ve provided us is sufficient for me to say I know you have two sons.