I remember learning about the difference between facts and opinions in elementary school. We spent days going over different statements and classifying them as one or the other. I think this was supposed to make it easier for us to understand different types of semantic objects, but on the whole it seems like the fact/opinion dichotomy has only made it harder for me to learn things. I actually prefer using an Observations/Theories paradigm, which I think makes it much easier to learn and use knowledge.
Facts and Opinions
Thinking of things as opinions and facts seems to place too little weight on an opinion and too much on facts. Facts about gravity and triboluminescence are presented as edicts handed down from on high, despite the fact that there are still huge questions about how these things work. Opinions, such as personal preferences and boundaries, are often dismissed out of hand as if they don't matter.
Facts, especially in school, are also often treated as being independent of observation. This can lead to issues where people who know a lot of facts have no idea about how to use them in the real world.
Observations and Theories
Observations seem important for both opinions and facts. I observed that apples fall at the same rate as bowling balls, and I also observe that when I eat apples I don't enjoy it. I can then theorize about the nature of gravity, or about my preferences for fruit.
The very terminology of theories makes it easier to remember to also present observations that theory predicts. The fact that objects accelerate at 9.8m/s^2 near the surface of the earth is less usable than a theory tied to observations about bowling balls falling at the same rate as golf balls.
Another nice quality of the observations/theories model is that it nicely splits things into two categories. There are those things in the real world that interact with our senses; and there are those things that are stories we tell about our sensations. Facts, especially have a much more murky relationship with reality. Sometimes a fact is an observation, other times is about a model of observations. This can confuse an issue and make it more difficult to learn a model, or to see the flaws in a model.
Science, Psuedo-Science, and Theories
The criticism that evolution is "just a theory" is often used by creationists to reject it. I think the reason for this is that people will often treat things as being either opinion or facts. If evolution is just a theory, then it isn't a fact. So it's something like an opinion and can be safely ignored.
Creationists are completely right about this, too. Evolution (or gravity or whatever) is just a theory. Intelligent design and young-earth ideas are also just theories. All these theories are just stories that people are sharing to try to explain their observations. The only difference between these stories is in how well they actually predict observations.
Treating all types of theory as the same type of thing makes things a bit harder, since each theory then needs to be evaluated on its own merits. You can't just say: gravity is a fact therefore it is true and unquestionable. If Einstein had done that, we wouldn't have GPS right now.
Another consequence of this view is that looking at a theory isn't enough to tell if you should discard it. You also need to look at what it's being used for. Newtonian mechanics, that thing people are taught in pretty much every high school physics class, is not very accurate in many circumstances. You couldn't use Newtonian mechanics to create GPS. However, it's a great theory for most of what people do all day. Want to calculate how long you'll fall while skydiving? Newtonian mechanics is probably a much better theory to use than general relativity, even though general relativity would be more accurate.
I think this is an important point for things like astrology or young-earth creationism. People who espouse these theories seem to have a much different goal than people who are studying psychology or evolution and geology. Pointing out all the ways that astrology fails to predict the course of a person's life usually won't have any impact on people who like astrology, because in general that's not really what they're using it for. Similarly, pointing out ways that young-earth creationism doesn't mesh with the fossil record won't change a person's mind if what they're really concerned about is their religion.