ABSTRACT
The Social Sciences and, specifically, the sociological research, have progressively assumed the gender factor as one of the strategic keys to understand contemporary phenomena. In fact, as a variable for socio-statistical analysis or as a characterizing trait of individual identity, it is a decisive factor in the interpretation of the deep social transformations, and it inspires the self-reflection of the sociologists about the analytical tools of their discipline. The contribution proposes, through a lexicometric approach, an analysis of the articles published in the last two decades by the oldest journal of Sociology, published by Routledge. The main aim is to highlight the different ways in which gender issues are declined in the international sociological researches presented in the repertoire of the International Review of Sociology and to outline, both on the lexical level and on the topic level, the changes occurred over time.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes on contributors
Isabella Mingo is an Associate Professor of Department of Communication and Social Research at Sapienza University of Rome. Her main research areas include measures and indicators of the social processes (free time, cultural participation, social exclusion, information society), statistical analysis of the textual data.
Mariella Nocenzi is an Adjunct Professor of Department of Communication and Social Research at Sapienza University of Rome. Her main research areas include history of the social theory with specific reference to social diversity, social inequalities, gender and sustainability.
ORCID
Isabella Mingo http://orcid.org/0000-0003-4001-8574
Mariella Nocenzi http://orcid.org/000-0002-2256-4101
Notes
1 This article proposes an extended version of the paper that the Authors proposed at the Conference JADT 2018: 14es Journées internationales d’Analyse statistique des Données Textuelles, Sapienza University of Rome-National Centre for the Researches, Rome, 12th June 2018.
2 See at the International Review of Sociology web site, page ‘Aims and scope’, https://www.tandfonline.com/action/journalInformation?show=aimsScope&journalCode=cirs20.
3 IRaMuTeQ is an open software, distributed under license GNU GPL, based on R statistical software and on Python language. It has now reached version 0.7 alpha 2 and it is still under development (Ratinaud, Citation2009).
4 The lexicalization list includes 110 complex lexicalized entities. Their minimum thresholds of occurrences to consider them in the subsequent analyses are indicated in the following paragraphs.
5 In the analysis by time, it should be pointed out that, as table 1 shows, in the first decade of the considered period the sizes of the sub-corpora are smaller than in the second decade. Furthermore, due to lack of articles, some years in the first decade are not present.
6 The relative frequency of a word in a subtext can be considered as the probability of appearing in n trials (where n is the total number of occurrencies). According to hypergeometric model, the test is carried out under the assumption of a normal distribution, so when the value z is more than |2| we can assume that the word’s presence is characteristic (1-α=95%).
7 Decreasing the frequency threshold to 10 or raising it to 30, ACL results were almost stable.
8 To obtain different partitions, the following IRaMuTeQ parameters have been used: Clustering=Simple on text segment; Number of terminal clusters= 5–15; Minimum frequency of text segment=5; Maximum number of analysed forms=30.000; Svd method=IRLBA (Augmented, Implicitly Restarted Lanczos Bidiagonalization Algorithm).