Discursive Constructions of Domestic Violence Among Iranian Instagram Users

Abstract

Domestic violence (DV) is a widespread social phenomenon, adversely impacting public mental and physical health. The abatement of such issue necessitates a priori social awareness and a posteriori social support. With that in mind, the present study aimed at disclosing dominant discursive constructions of DV among Iranian Instagram users. Driven by Fairclough’s (1992) notion of “moments of crisis,” critical discourse analysis was employed as the theoretical framework to elucidate the results obtained from 1,028 comments on nine DV-related posts. Our primary objective was to uncover Iranian Instagram commentators’ perception of the causes and consequences of DV as well as the perpetrator and victim’s roles through probing into the embedded social values and assumptions underpinning such discursive constructions represented within comments. Meticulous inspection of the data yielded six major themes, namely, (a) blaming the responsible, (b) offering solutions to the victims, (c) portraying the male perpetrators, (d) overgeneralizing the transgression, (e) disregarding men’s rights, and (f) portraying the female perpetrators. The results indicated that female and male commentators held dissimilar perspectives with regard to DV and that though the conceptualizations of DV were not destitute of prejudice, both perpetrators and victims were encouraged to take requisite steps to minimize the adverse impact of DV on their families and relations as well as on social and public health. The discursive themes identified in the current work may serve as the stepping stone toward the mitigation of DV within society.

Domestic violence risk prediction in Iran using a machine learning approach by analyzing Persian textual content in social media

“…Although there have been studies in Iran to identify the impact of this violence in various cities and across the country [ [30] , [31] , [32] , [33] , [34] , [35] ], to qualitatively understand the conditions of the victims [ 36 , 37 ], to determine if various treatments are effective [ [38] , [39] , [40] ], to identify why this phenomenon occurs [ [41] , [42] , [43] ], and to consider ethical considerations in research related to this topic, no study in the field of DV prediction using machine learning has been conducted. A similar study extracted the main topics of discussion in the Persian language in the field of DV, or in general, in the Instagram and Twitter social networks [ 20 , 44 ] which aids in monitoring these networks. However, none intended to classify all data into two distinct categories, critical and uncritical, using a model of machine learning algorithms.…”

“…Data that lacked a name, tweet, caption, bio, or photo were deemed bots and were removed from the dataset. Furthermore, the minimum number of data collected was not reported as a criterion in the research background; however, the minimum sample size in psychological studies of social media using machine learning based on tweets was 2000 tweets [ 23 ], whereas 1028 comments related to DV were examined in the Persian sample study [ 20 ]. The first global study on DV was also carried out on Facebook, with 8,856 posts and 28,873 text comments totaling approximately 30,000 data samples [ 24 ].…”

“…According to the study by Ebadi-Jalal et al, the structure of DV-related conversations among 1028 comments on the related posts on Instagram was identified and categorized. The findings of this article have finally categorized the main topics related to these conversations under the following headings: blaming responsibility, proposing solutions for victims, portraying male criminals, overgeneralizing the offense, ignoring women's rights, and portraying female perpetrators [ 20 ].…”