If everyone can change the Wikipedia, can it be trusted? Do any studies exists that show it is less trustworthy than classical encyclopedias?
You are not supposed to trust it - Wikipedia itself tells you that it is not a reliable source.
Instead, you are supposed to trust the sources it cites in the articles. This will mostly be secondary sources, e.g. newspaper or magazine articles, books or studies about a topic, provided that these meet Wikipedia's requirements for reliable sources.
The latter is also supposed to maintain overall quality of Wikipedia articles and prevent the inclusion of outright wrong information, though - anyone who adds or deletes or changes content of an article is supposed to back this up by sources, at least when their change is challenged.
This can't be guaranteed to work in all cases, though - and this is a difference to classical encyclopedias, where a number of people might be tasked with verifying the included information.
Wikipedia seems to do quite well, though, they have an article about the reliability of Wikipedia - again, you are not supposed to trust this article, but the sources cited in it.
Please note that I have added several links to Wikipedia in this answer - but not all of these links go to articles, some refer to pages in the Wikipedia: namespace. The latter is where Wikipedia's own policies are defined. You are supposed to follow those, no trust needed, or you may be sanctioned by users with the required privileges.
Wikipedia has an entire article on this subject, including several studies, most famously Giles, J. (2005). "Internet encyclopaedias go head to head: Jimmy Wales' Wikipedia comes close to Britannica in terms of the accuracy of its science entries" published in Nature.
Generally, Wikipedia seems of roughly equal reliability as conventional encyclopedias.