This is a fairly tricky question as Abstract Algebra is one of those things that makes a lot more sense one you've spent some time studying some of it's subject areas.
I will however have a stab at it...
Abstract algebra normally follows the same pattern of taking a set, $Z$ say, and attributing some properties to the elements of those sets. We then seek to prove certain things about those sets.
We do in fact use sets of these sorts all the time, since number systems are formed in exactly this way. For example, the set of integers $\mathbb{Z}$ is simply a set of numbers with the properties of the numbers being that they can be added and subtracted from each other, with an identity element (an element such that adding it to other elements just gives you what you started with) zero etc.
The goal of the 'abstract' part of abstract algebra is due to the fact that rather than just studying very specific cases, we instead look for common properties across many types of sets and mathematical objects and study the properties that they all have in common. This is similar to a the fact that while there are many hundreds of breeds of dogs, they all have very similar physiologies and so a vet can study the common aspects of all dogs and in doing so she can then perform operations on any of them, even though there are other aspects in which they all differ.
So this why in abstract algebra we don't specifically study each individual set such as $\mathbb{C}$ and$\mathbb{R}$, but instead we notice that they all satisfy certain properties in common. We then define a Field (as one example) as being a set $\mathbb{F}$ with these properties in common (you'll study Fields when you start your course). Then anything we prove about Fields can be applied to all of the sets which also share the properties of a Field.
There are many other types of objects such as as Groups and Rings which are also studied because they have properties that we know exist for many specific cases.
Therefore, in conclusion Abstract ALgebra is the process of noticing similar properties in different mathematical objects and then specifically seeing what we can prove about objects with these properties in order to be able to say things about every object with those given properties.