While we want you to use scholarly, peer-reviewed sources for this paper, it's a lot easier to find "the good stuff" if you spend some time brainstorming and seeing what's out there on the open web.
Google is great at the very beginning of your search.
Credo is a bunch of encyclopedias smooshed together and will give you an broad overview of a wide range of topics.
Typing your topic into WISE can also be useful in figuring out what's out there!
PsycINFO has some great tools for early searching!
Developing a hypothesis:
1. Start off with a good, just right research question
Way too broad: How does divorce affect families?
Too broad: What factors make divorced fathers more involved in their children's lives?
Just right: Are divorced fathers more likely to be highly involved in their children's lives if families go to group counseling instead of individual counseling?
2. What's your hunch? Remember that you can and should revise your hypothesis as you research and learn more
Way too broad: Oh boy....
Too broad: Hmm... I don't know. Living closely? Getting along with their ex-spouse? Self-identified interest in their children?
Just right: I think that attending family counseling makes divorced fathers are more likely to be highly involved co-parents
3. A good hypothesis is: relational or causal, significant, specific, and gives guidance on what an experiment might look like.
Way too broad: Umm...
Too broad: Divorced fathers who have joint custody and live in close geographic proximity to their ex-spouses are more likely to be 'cohesive coparents'
Just right: Divorced families who participate in family counseling instead of individual counseling are more likely to have fathers who are more involved coparents, regardless of custody status or reason for divorce.