Corpus-based conceptualization in sociology: possibilities and limits
The problem of the quantitative interpretation of qualitative data is one of the most important in sociological research. Textual analysis has placed emphasis on deep and careful study of texts how personal strategies embodied in the concepts. However, quantitative interpretation has always been problematic. Our paper deals with the corpus-based conceptualization method, which can considered as a method of collecting and organizing data material from linguistic corpora. The corpus-based conceptualization allows us to establish a closer link with the meaning and identify the whole spectrum of meanings. It shows that some sociologists lose essential meanings in the research process because of lack of in-deep immersion in the daily life and speech of the people. We chose the concepts of "altruism" and "mercy" as examples to demonstrate the corpus-based conceptualization and its place in sociological research methodology. Data comes from the Russian National Corpus. The Russian National Corpus consists of 1802 relevant words, with 775 for altruism and 1047 for mercy. Data processing carried out by SPSS 19.0. As the result, we have discussed what difficulties the researcher can meet using this method and have offered SFL and Role and Reference grammar as a way to accurately determine the context. Our suggestions can be used in the preparation of questionnaires, guides, in an analysis of interview transcripts.