Statistical methods have been used in linguistics for a long time. However, recently, information technologies have boosted the development of statistical tools, which are now more actively used for applied tasks, including processing and presentation of text data. The purpose of the work is to describe a number of statistical metrics used in lexicographic studies, involving a frequency dictionary of the Russian language, text corpora and databases that present information about lexical collocability. These measures are implemented to differentiate vocabulary on different grounds, highlighting key words and phrases characteristic of texts of a certain style or topic. The paper also provides a brief historical overview of the application of quantitative methods to text analysis.