It is obvious that you should evaluate the relevance of your sources. You evaluate their scientific nature to increase the reliability of your own piece. Indirectly reliability is also determined by the kind of sources you use. If the scholarly nature of your sources is beyond doubt, your paper itself will less likely be doubted.
Think that you will rarely find a source that provides a complete answer to your main questions and sub-questions and that gives a report of the exact same research or problem you are working on.
If you want to use information from websites other than regular scholarly articles and books you should be extra careful and think about the role you give that information in your argument or analysis.
Ask yourself the following questions when reading webpages and be extra careful if the answer is negative most of the times.
These questions do not relate of course to the use of scientific articles or e-books published on websites.
The crux of science lies in the extent to which an author/researcher performs his work objectively and makes it verifiable. In determining the quality and scientific nature of sources you may start from three kinds of checks: