Social network as a source of sociolinguistic data: lexicostatistical analysis

The article is devoted to the substantiation of the methodology and comparative lexicostatistical analysis of sociolinguistic data obtained from the social network “VK”. New possibilities of research of verbal Internet communication, including sociolinguistic features of Internet users, are discussed. The primary attention in the paper is paid to the correlation between the age of social network users and lexical features of their texts. Based on the data obtained (about 8 million words), the text corpora of four age groups of social network users aged from 14 to 74 years old have been compiled. The authors have conducted a comparative analysis of the frequency wordlists and, first of all, identified the words that are often used in the texts of users of all four groups. This array generally correlates with the data from the frequency dictionary of the Russian language, although it also has significant differences. The article presents the lists of frequency words typical of each age group; ideographic characteristics of the wordlists (dominant thematic groups of words) and sociolinguistic comments are provided. Conclusions are made about lexical and conceptual differences between texts of different age groups of users, as well as about the productivity of statistical and ideographic analysis of social media texts.