An "everyday" theory may be based on observations, but often have few facts behind them. A scientific theory must be based on facts, observations, and experiments.
Check this site for an excellent explanation.
The following two paragraphs come from Wikipedia:
<<In science a theory is a testable model of the manner of interaction of a set of natural phenomena, capable of predicting future occurrences or observations of the same kind, and capable of being tested through experiment or otherwise verified through empirical observation.>>
<<In common usage, the word theory is often used to signify a conjecture, an opinion, a speculation, or a hypothesis. In this usage, a theory is not necessarily based on facts; in other words, it is not required to be consistent with true descriptions of reality.>>
The above would for example characterize the difference between "I have a theory to explain why the as-yet hypothetical Higgs Boson has not been detected during the recent series of subatomic particle tests conducted at Fermilab", and "I have a theory to explain why my next-door neighbor always mows his lawn on Tuesday evenings". Apart from anything else, there's usually a predictive element in a scientific theory, in which the theory is used to extrapolate from known sets of circumstances to unknown sets of circumstances, and then either gains credibility if it correctly predicts what will happen, otherwise it is either modified or discarded. An everyday use theory is more likely to be a question of "Whenever this happens, so does that". The predictive element is far less relevant, and may be absent altogether.
A "theory" in everyday use is basically an hypothesis, often not more than an intuition, opinion or guess. In science, a guess is called a hypothesis. A (capital "t") theory is sort of a grand explanation of many phenomena. Although nothing in science is ever absolutely proven, a theory is a set of explanations that has so much evidence for it, that for it to be wrong, we would have to radically re-interpret all our data. So, examples of theory would be electromagnetism, plate tectonics, aerodynamic theory, relativity and natural selection. All of these started out as hypotheses, but are now accepted by nearly all scientists because pretty much everything we know about the world supports them. (Which is not to say that they explain everything, or that they won't be modified in the future.)