When reliability and usefulness are to be evaluated, it is wise to ask some questions about the document.
Source of information
Where did you get the document from? What databases or search engines have been used? Do the databases have scientific content? How does the ranking in the database work?
Who is the publisher of the document? If it is a journal, is it indexed in any databases? One should keep in mind, though, that even if a publisher has a good reputation, a single work does not necessarily live up to high standards.
Who is the author? What does the author's background look like? Is the author affiliated with any institution or organization? If that is the case: what values or aims are pursued by the institution or organization? What has she/he written before, what experiences and education does she/he possess? Has the author been cited in other sources or bibliographies?
What is the purpose of the document? Are the contents factual or aimed at creating a public opinion? Do the contents seem reliable and based on thorough research, or is it ill-founded and possible to put into question? Is the view of the author objective and unpartial? Is the language free from strong emotional expressions and preconceived conceptions? Are the results and conclusions reasonable? Is there anything controversial?
Date of publication
When was the source published? Is the source up to date enough for your purposes? Some subject domains, e.g. the natural sciences, tend to demand relatively new sources, while some fields within e.g. the humanities may use older sources.
Structure in scientific documents - IMRaD
In many research areas, scientific documents are usually structured according to IMRaD - Introduction, Method, Results and Discussion. This structure is reflected in the disposition and the headings. Read more about IMRaD.