I hate to be a bit redundant, since other experts have nicely answered this question, but I thought I'd add my ten cents.
As it relates to research, an abstract can be thought of as a snapshot of what is to come in a journal article. A snapshot doesn't give us all the details, numbers, and specifics involved, but gives us a clue as to the goal, outcome, and theories connected to a given piece of research. It is a nice way to quickly get a sense for whether a piece of research is going to be usable in our own research. If writing an abstract, it is important to be concise but give enough information that that an informed researcher or student can decide whether the work applies to their own research.