Социология: 4М. № 36

Социология: 4М. № 36

по-русски    in english    

Sociology: methodology, methods, mathematical modeling

No. 36, 2013

 

Contents

Practices of formalized data collection and analysis

 Kuchenkova A.V., Tatarova G.G. A Strategy for Using Logical-Combinatorial Methods in Procedures of Typological Analysis

The article proves the necessity of applying two well-known logical-combinatorial methods (qualitative comparative analysis (QCA) and DSM-method) to solve problems of typological analysis. We propose a strategy for their integrated use in a common research situation which is characterized by the need of secondary conceptualization of large-scale studies data. The results of testing this strategy are analyzed.

Keywords: sociological contextuality, typology, typological analysis, linguistic and logical structure of typological analysis, logical-combinatorial method, DSM method, QCA-method

Ponarin E.D., Lisowski A.V., Zelikova Y.A. Poisson Models for the Dependent Variables - Can We Predict the Scores of Football Matches?

This paper describes the general case of the Poisson regression and points out its distinction from the case of log-linear models for contingency tables. The Poisson model is applied to scoring in the Russian national football league. Its difference from the OLS regression in practical application is discussed.

Keywords: Poisson regression, log-linear models, football

Tovar García E.D. Educational Trajectories of Schoolchildren in Yaroslavl Oblast: Discriminant Analysis in Comparison with Logistic Regression

The article studies educational trajectories of schoolchildren in Yaroslavl Oblast. Conclusions point out that schoolchild’s educational achievements, educational plans of his/her friends and the level of education of his/her father are key predictors of a decision about continuing education. Thanks to this information it is possible to know which schoolchildren are at risk of not continuing their studies. In the course of the research comparative advantages of the logistic regression and the discriminant analysis in the case of binary dependent variables were examined. With the necessary prerequisites for the use of methods fulfilled, both strategies work well classifying schoolchildren.

Keywords: Educational trajectory, educational achievement, family resources, human capital, logistic regression, discriminant analysis, Yaroslavl Oblast

Unformalized data: collection and analysis methods

Divisenko K.S. The Experience of Using Structural-Semantic Analysis for an Autobiographical Narrative

The article presents a structural semantic method of narrative analysis based on A.-J. Greimas’ semiological theory. The religious search and conversion described in the autobiographical text are used to illustrate this method. In the end the author presents some conclusions on possible applications of this method for reconstruction of the religious life-world.

Keywords: J. Greimas, autobiography, actant, actantial model, life-world (Lebenswelt), narrative, religion, representation of religious experience, semiology, structural-semantic method.

Online research methods

Koltsova O.Y., Maslinsky K.A. Identifying the Thematic Structure of the Russian Blogosphere: Automatic Text Analysis Methods

The article considers methodological results of the research of the Russian language. Automatic methods of texts analysis and the corresponding software are adapted and tested in relation to the final research goals (detection of topical structure of the blogosphere, description of its evolution and discussion communities formation). The methods of dividing texts into topical groups are classified into two large types – cluster analysis and topic modeling, a software representing each type is chosen and applied. Experiments are run on two datasets, each consisting of 104 blog posts. The choice of topic modeling is justified, the full technological research cycle from data collection to sociological analysis is described.

Keywords: Internet, blogs, methods of sociological research, text clustering, topic modeling, big data

Network analysis

Safonova M.A., Wiener B.E. Co-citations’ Network Analysis of Ethnological Publications in Russian Periodicals: Preliminary Results

The authors of the paper are interested in elucidation of a cognitive structure for Russian ethnology. It means we want to uncover which research areas are found within this discipline. The authors use network analysis of Russian scholastic periodicals, published in 2005-2009, as a research tool for an investigation of co-citations. The study found that for the given task an algorithm of van Dongen was more effective than an algorithm of Newman.

Keywords: cognitive structure of science, network analysis, Russian ethnology.

Analytical reviews

Maloshonok N.G. On the Student Engagement Measurement: Basic Methods and their Restrictions

The paper describes a concept of “student engagement”, which is widely applied among international researchers, and examines how the concept can be used in assessment of excellence of universities. It considers basic approaches to operationalization and measurement of student engagement and proposes a new classification of methods for investigating this phenomenon.

Keywords: student engagement, university excellence, process and outcome indicators, learning process



© 1998-2024. Институт социологии РАН (http://www.isras.ru)