I finished writing my internship report for a master degree. Now I am writing its abstract. Is it possible to mention a statistic study on the abstract and reference it in the bibliography or is this something bad to do ?
If this is just an internal document that will never be seen outside your own institution, it doesn't really matter.
For a document that will be distributed outside your institution, the abstract should be fully comprehensible even if that's all you see. It's quite common for databases to include only the abstract or for it to be much easier for people to see the abstract than the rest of the document. (For example, on ArXiv, each paper's "home page" includes the abstract but you'd have to download the whole paper to see the references.) For these reasons, an abstract that says something like
We build on the statistical analysis of .
isn't very useful. This isn't so bad with a long reference style:
We build on the statistical analysis of Smith and Jones (2005).
is better, especially if that is a well-known paper in the field but not everything builds on well-known papers and the reference is potentially vague if Smith and Jones wrote multiple papers together that year.
The best way to do this is to give a full citation in the abstract itself:
We build on the statistical analysis of Smith and Jones (J. Interest. Res., 3(1):41–59, 2005).
Some journals forbid citations of any kind in the abstract but that seems to force people into the suboptimal "Smith and Jones" style.