TY - GEN

T1 - Public Opinion Dynamics in Online Discussions

T2 - 13th International Conference on Social Computing and Social Media, SCSM 2021, held as part of the 23rd International Conference, HCI International 2021

AU - Bodrunova, Svetlana S.

AU - Blekanov, Ivan S.

AU - Maksimov, Alexey

N1 - Publisher Copyright: © 2021, Springer Nature Switzerland AG.

PY - 2021

Y1 - 2021

N2 - Objectives: Social media have become a place where the bulk of grassroots political discussion takes place. Today, the growing body of research is dedicated to cumulative patterns on online deliberation, the predecessors of which were the concept of the spiral of silence, the silent majority hypothesis, and influencer studies. However, when applied to the dissonant, disruptive, and discontinued online discussions of today where gatewatching is much less predictable and cumulation of support is often accompanied by communicative aggression, these concepts need to be reconsidered and re-tested. Also, the current public communication online is much more multi-level than before; even within one platform, several communication layers may be defined, and their inter-relations in terms of public opinion aggregation remain under-studied. Research goal. This paper aims at discovering patterns of cumulative deliberation in online communication. We first discuss the umbrella concept of cumulative deliberation. Then, we test the dynamics of the discussion on Belarusian oppositional YouTube in terms of impact of cross-account commenting on growth of commenting within the cross-account community and the overall discussion. Method and sampling. We have collected the data by YouTube crawling. The data include all user comments of 2018 for six salient Belarusian oppositional accounts. To define the cross-account commenters, we used Gephi-based web graph reconstruction. We manually coded user posts for interactivity, aggression, and criticism. Dependencies in the dynamics of the discussion were tested by correlational and cluster analysis. Results. We have discovered that users diverged into two mutually exclusive modes of expression, namely aggressive-dialogical and (self-)critical. Of the cross-account commenters, several users demonstrated ‘cumulative’ behavior and personal opinion bubbles. While there were nearly no dependencies discovered in the dynamics of user posting, criticism and self-criticism show capacity of spurring/diminishing the dynamics of public communication online, even if commenting on the whole is cumulative, not dialogical.

AB - Objectives: Social media have become a place where the bulk of grassroots political discussion takes place. Today, the growing body of research is dedicated to cumulative patterns on online deliberation, the predecessors of which were the concept of the spiral of silence, the silent majority hypothesis, and influencer studies. However, when applied to the dissonant, disruptive, and discontinued online discussions of today where gatewatching is much less predictable and cumulation of support is often accompanied by communicative aggression, these concepts need to be reconsidered and re-tested. Also, the current public communication online is much more multi-level than before; even within one platform, several communication layers may be defined, and their inter-relations in terms of public opinion aggregation remain under-studied. Research goal. This paper aims at discovering patterns of cumulative deliberation in online communication. We first discuss the umbrella concept of cumulative deliberation. Then, we test the dynamics of the discussion on Belarusian oppositional YouTube in terms of impact of cross-account commenting on growth of commenting within the cross-account community and the overall discussion. Method and sampling. We have collected the data by YouTube crawling. The data include all user comments of 2018 for six salient Belarusian oppositional accounts. To define the cross-account commenters, we used Gephi-based web graph reconstruction. We manually coded user posts for interactivity, aggression, and criticism. Dependencies in the dynamics of the discussion were tested by correlational and cluster analysis. Results. We have discovered that users diverged into two mutually exclusive modes of expression, namely aggressive-dialogical and (self-)critical. Of the cross-account commenters, several users demonstrated ‘cumulative’ behavior and personal opinion bubbles. While there were nearly no dependencies discovered in the dynamics of user posting, criticism and self-criticism show capacity of spurring/diminishing the dynamics of public communication online, even if commenting on the whole is cumulative, not dialogical.

KW - Belarus

KW - Communication

KW - Cumulative deliberation

KW - Deliberation

KW - Granger test

KW - Networked discussions

KW - Political protest

KW - YouTube

UR - http://www.scopus.com/inward/record.url?scp=85112227101&partnerID=8YFLogxK

UR - https://www.mendeley.com/catalogue/f3aebdde-c318-3bbb-896f-53963e8dce2e/

U2 - 10.1007/978-3-030-77626-8_14

DO - 10.1007/978-3-030-77626-8_14

M3 - Conference contribution

AN - SCOPUS:85112227101

SN - 9783030776251

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 205

EP - 220

BT - Social Computing and Social Media

A2 - Meiselwitz, Gabriele

PB - Springer Nature

Y2 - 24 July 2021 through 29 July 2021

ER -