Writing an abstract is an important part of academic research. Having decided that your title sounds interesting, potential readers will often decide whether to read (or cite) your paper based on the abstract. There are a couple of templates you can follow:
A fantastic template is this annotated Nature abstract: http://www.cbs.umn.edu/sites/default/files/public/downloads/Annotated_Nature_abstract.pdf. As with all templates, it’ll require a bit of tweaking to make it fit, but the general instructions (basic instruction, background, problem, here we show, …) equally apply to the social sciences.
Here’s another template you could adapt: (1) What was the focus in the research to date? In one sentence. (2) What is distinctive about your papers? What is your contribution in terms of theory or empirical research? Include your research question here. (3) Describe the data and methods used. Do me a favour and avoid the temptation to describe your data as unique or original. (4) Spell out your main findings: what have you found out that we didn’t know before? What are the key conclusions you draw from the research?
Make sure your key results are mentioned. It is very frustrating to read an abstract that just mentions the question without the results. I understand the rationale behind this: the reader should be tempted to read the paper. More often, however, the reader will put the paper away and read something else (worse still, the paper is behind a pay wall, and the reader has to decide whether to buy the article…). A final tip: don’t say that hypotheses will be generated, implications will be discussed, or conclusions will be drawn: that’s the job of a paper.