A database is a lot like a filing cabinet. In the cabinet, you have folders that contain individual pieces of paper. Like this a database is constructed of records that describe a larger work, such as a scholarly journal (the folder in our analogy) and records that describe the unique articles in that journal (the paper).
In each record, there will be individual fields with discrete pieces of information, such as a field to list the title of the article or the author's name. Some databases will offer a way to search those individual fields. This can be a very powerful search that can help narrow down an overwhelming list of results.
Many databases offer truncation searching. This means that you can search the root of a word and any of its derivatives simply by using the truncation symbol provided by the database. Each database uses their own pre-determined symbols, so check the help provided to see what that might be.
You probably remember all this from using Wexis, but just as a refresher...
This is really the meat of any system that organizes information.
The Skinny on Finding Books
In libraries, most librarians use the Library of Congress subject heading system to describe the topic of a book. The idea is that you would be able to find all the books that deal with that subject by doing a subject heading search. And because we all use the same system (and honestly, we collaboratively created the records that describe books so most of use describe the same books the same way) it is a pretty uniform system and works pretty well.
In all, I'd suggest doing a keyword search for books, find a good one on your topic, then use the subject headings to find more.
The Lowdown on Finding Journal Articles
Each database vendor uses their own system to describe their material so there is no uniformity. Most will have some sytem of thesaurus or controlled vocabularly, but it varies from system to system. So, you have two choices. 1.) Figure out the system used by the database that you will be using or 2.) Do full text searching for articles on your topic, then use any functionality you can to find more.
In all, I'd suggest doing a keyword search for articles, find a good one on your topic, then use whatever descriptors are provided to find more.