Let me echo others. Students should choose a topic that interests them. They will engage more readily with the material, likely spend more time researching and writing, and deliver a more-polished paper to their instructor, who may have delight in reading their final work.
Reference resources, such as encyclopedias, Wikipedia, or CQ Researcher, can also be valuable to students, since they offer basic background information that their instructor may already expect them to know beforehand. Reference articles, like a trusted friend, can identify specific aspects of a topic which might be even more interesting than the general topic. A bibliography or list of references at the end of a reference article can also launch them into their research immediately. At any rate, students should aim for a specific topic, especially if their paper can only be a few pages long.
Creating a concept map, a spider graph, or other brainstorming activities can guide students to discover what interests them the most, as well as make a better decision about how they want to specifically narrow down their topic.
My supervisor has identified three ways in which a researcher can narrow down their topic:
Three Ways to Narrow a Topic
A person interested in the obesity epidemic can narrow down that huge topic by selecting a population group, such as children, teenagers, Generation X, middle-aged, whites, Hispanics, lower income, etc. Likewise, they could also limit even further to a geographic location: rural, urban, a specific state (Idaho, Utah, South Carolina, etc.), or a region (Western U.S., the Midwest, the South, etc.). Last, a time limitation can shrink the scope of the research quickly: last 3-5 years, the 90s, the 19th century, and so on. True, there probably was no obesity epidemic in the 19th century, unless it occurred among the nouveau riche, fat cats, or the robber barons as many called them.
If you are interested in a multimedia screen cast that discusses this same idea of narrowing down a topic, take a look at a Prezi presentation I created not long ago: Narrowing A Topic.