Possibly one of the most important steps in conducting a meta-analysis is developing a question.
Your question should reflect your interests and can be as broad or specific as desired.
The question may need to be modified slightly throughout the process due to a lack of sufficient articles or an overwhelming number of relevant articles.
During this process, it is also important to verify that your work in this area will be original. Use the various search engines to determine whether or not there is a pre-existing meta-analysis in your area of interest. If there is not, you can begin to create search terms designed to gather all the relevant articles (Step 3).
However, if someone has already meta-analyzed the field of interest, you should consider several factors before abandoning your topic in despair.
How old is the existing meta-analysis, and how many studies have been published in the field since its publication date?
It may be time for an update!
Do you agree with the existing meta-analysis' exclusion criteria?
If they were too exclusive/limiting you can argue the case that they did not accurately summarize the field.
Are there multiple moderators they did not examine that you would be interested in?
Your focus could be in a different direction.
If none of these produce a valid reason to redo the existing meta, it's time to give up and return to the top of this page with a new question.
