Social media are becoming increasingly important tools of political communication which makes them the subject-matter of study of political sciences – not only in borderline disciplines such as political marketing or political communication, but also in political sciences and international relations. Existing expert publications mainly focus on surveying traditional media and usually do not reflect the reality of the continuously growing influence of social media; however, they tend to pay attention predominantly to the distinctiveness of the media space. The purpose of the article is to contribute to the discussion about methodology distinctions and even counter positions of such surveys through comparison with methods of surveying traditional media. Every contemporary researcher must cope with the largest amount of – relatively easily available – sources in the history of mankind. Online environment is characterised by high degree of innovativeness related to the process of continuous transformation; compared to traditional media, the content of social media is unstable and it is necessary to put much higher demands on the context of online contents and relevancy of information sources. The study points out methodology risks that should be reflected mainly in the preparatory stage of surveys, e.g. when determining the selection sample, construction of categories of the content, or media, which are to be surveyed, as well as the setting of coding system. Due to the abundance of various content types, it is getting increasingly difficult to distinguish the relevant from the trivial. It requires much more intensive preparation of the researcher who must be aware of different environments and understand the methodology of differences surveyed by the study.