The Influence of Political YouTube Content Consumption on Affective Polarization in Korea : How YouTube Shapes Emotional Divides = 한국에서 정치 유튜브 콘텐츠 소비가 정서적 양극화에 미치는 영향: 유튜브는 어떻게 감정의 분화를 만들어 가는가

다국어 초록 (Multilingual Abstract)

This study investigates how the consumption of political YouTube channels’ content influences affective polarization among South Korean adults. The research explores two central questions. The first concerns whether such consumption predicts affective polarization, and the second examines how specific mechanisms connect viewing behavior to polarization outcomes. To answer these questions, the study adopts a convergent mixed-methods design that integrates a quantitative survey of 100 participants with qualitative in-depth interviews with 10 viewers.
The quantitative results revealed a statistically significant positive relationship between consumption time and affective polarization (r=.338, p<.001). Hierarchical regression analysis demonstrated that consumption time of political YouTube channels' content significantly predicted affective polarization (β=.338, p<.001), and this relationship remained robust even after controlling for political orientation (β=.352, p<.001), explaining 14.3% of the variance in polarization.
The qualitative analysis identified three core mechanisms through which the consumption of political YouTube channels' content intensifies affective polarization. First, algorithmic reinforcement narrowed users’ informational environments, reducing exposure to diverse perspectives and contributing to the formation of echo chambers. Second, multimedia storytelling amplified emotional reactions and deepened viewers’ affective involvement with political messages. Third, as viewers developed stronger trust in particular channels or creators, they became more likely to accept partisan claims as credible, which increased the persuasive impact of the content. Interview participants described how repeated exposure to ideologically consistent material gradually solidified their political views and heightened their negative emotions toward opposing groups.
Integration of quantitative and qualitative findings reveals that consumption of political YouTube channels' content contributes to affective polarization through structural features of the platform and content format rather than viewers' pre-existing ideological positions. These findings underscore the need for platform-level interventions, media literacy education, and policy measures to mitigate polarization in digital political communication.

국문 초록 (Abstract)

본 연구는 정치 유튜브 채널 콘텐츠 소비가 한국 성인의 정서적 양극화에 미치는 영향을 분석한다. 연구는 두 가지 핵심 질문을 중심으로 진행된다. 첫 번째는 이러한 콘텐츠 소비가 정서적 양극화를 예측하는지에 관한 것이며, 두 번째는 시청 행동이 어떤 구체적 메커니즘을 통해 양극화 결과로 이어지는지를 살펴보는 것이다. 이를 위해 본 연구는 정량적 설문조사 100명과 심층 인터뷰 10명을 결합한 수렴적 혼합방법 연구 설계를 채택하였다.
양적 분석 결과, 소비 시간과 정서적 양극화 사이에 통계적으로 유의한 정적 관계가 나타났다(r=.338, p<.001). 위계적 회귀분석 결과, 정치 유튜브 채널 콘텐츠 소비 시간은 정서적 양극화를 유의하게 예측하였으며(β=.338, p<.001), 이 관계는 정치 성향을 통제한 후에도 유지되었다(β=.352, p<.001). 이는 양극화 분산의 14.3%를 설명하였다.
질적 분석 결과, 정치 유튜브 채널 콘텐츠 소비가 정서적 양극화를 심화시키는 세 가지 주요 메커니즘이 드러났다. 첫째, 알고리즘 강화는 이용자의 정보 환경을 좁혀 다양한 관점에 대한 노출을 줄이고 에코 챔버 형성을 촉진했다. 둘째, 멀티미디어 스토리텔링은 정서적 반응을 강화해 정치적 메시지에 대한 감정적 몰입을 높였다. 셋째, 특정 채널이나 크리에이터에 대한 신뢰가 커질수록 이용자들은 당파적 주장을 더 쉽게 수용하며 콘텐츠의 설득력이 강화되었다. 인터뷰 참여자들은 이념적으로 일관된 콘텐츠에 반복적으로 노출되면서 정치적 견해가 굳어지고 반대 진영에 대한 부정적 감정이 심화되는 경험을 설명했다.
양적 및 질적 결과의 통합은 정치 유튜브 채널 콘텐츠의 소비가 시청자의 기존 이념적 입장보다는 플랫폼과 콘텐츠 형식의 구조적 특징을 통해 정서적 양극화에 기여함을 보여준다. 이러한 연구결과는 디지털 정치 커뮤니케이션에서 양극화를 완화하기 위한 플랫폼 수준의 개입, 미디어 리터러시 교육, 정책적 조치의 필요성을 강조한다.

목차 (Table of Contents)

• ABSTRACT i
• 국문 초록 iii
• PREFACE v
• TABLE OF CONTENTS vii
• LIST OF TABLES xi
• LIST OF FIGURES xii
• CHAPTER 1: INTRODUCTION 1
• 1.1 Research Background and Problem Statement 1
• 1.2 Research Purpose and Questions 2
• 1.3 Significance of the Study 3
• 1.3.1 Theoretical Contributions 3
• 1.3.2 Methodological Contributions 3
• 1.3.3 Practical Contributions 4
• 1.4 Scope and Structure 4
• 1.4.1 Research Sample and Period 4
• 1.4.2 Methodology Overview 5
• 1.4.3 Delimitations and Limitations 6
• 1.4.4 Thesis Structure 6
• CHAPTER 2: LITERATURE REVIEW 6
• 2.1 Theoretical Foundations of Affective Polarization 7
• 2.2 Digital Media and Partisan Content Effects 8
• 2.3 YouTube as Political Information Platform 9
• 2.4 YouTube and Social Media in South Korean Context 9
• 2.5 Theoretical Framework 10
• 2.6 Research Gaps and Study Positioning 11
• CHAPTER 3: RESEARCH DESIGN AND METHODOLOGY 12
• 3.1 Mixed-Methods Research Framework 13
• 3.1.1 Philosophical Foundations 13
• 3.1.2 Convergent Design Rationale 13
• 3.1.3 Integration Strategy 14
• 3.2 Quantitative Component: Survey Analysis 14
• 3.2.1 Sampling and Recruitment 14
• 3.2.2 Survey Instrument Development 15
• 3.2.3 Affective Polarization Measurement 16
• 3.2.4 Statistical Analysis Approach 17
• 3.3 Qualitative Component: Semi-Structured Interviews 18
• 3.3.1 Sampling Framework for Qualitative Analysis 18
• 3.3.2 Interview Protocol and Data Collection 19
• 3.3.3 Thematic Coding and Analysis 20
• 3.3.4 Conservative-Progressive Asymmetries 21
• 3.4 Analytical Methods and Validity 22
• 3.5 Ethical Considerations 22
• CHAPTER 4: RESULTS 23
• 4.1 Quantitative Analysis Results 23
• 4.1.1 Sample Characteristics 24
• 4.1.2 Measurement Properties and Descriptive Statistics 25
• 4.1.3 Bivariate Relationships 27
• 4.1.4 Regression Analysis 28
• 4.2 Qualitative Findings: Mechanisms of Polarization 33
• 4.2.1 Algorithm Reinforcement 34
• 4.2.2 Confirmation Bias Reinforcement 36
• 4.2.3 Emotional Amplification 38
• 4.2.4 Outgroup Derogation 39
• 4.2.5 Ingroup Solidarity 40
• 4.2.6 Meta-Cognitive Awareness 42
• 4.3 Mixed-Methods Integration: Convergent Analysis 44
• 4.3.1 Convergence Matrix: Mapping Themes to Quantitative Items 44
• 4.3.2 Viewing Time as Amplification Mechanism 47
• 4.3.3 Ideology as Moderator: Parallel Pathways to Polarization 49
• 4.3.4 Meta-Cognitive Awareness: The Knowledge-Behavior Gap 51
• CHAPTER 5: DISCUSSION 54
• 5.1 Overview of Key Findings 54
• 5.1.1 Quantitative Findings: Viewing Time as Primary Predictor 54
• 5.1.2 Qualitative Findings: Six Mechanisms 55
• 5.1.3 Mixed-Methods Integration 56
• 5.2 Theoretical Implications 56
• 5.2.1 Polarization as Structurally-Induced Phenomenon 56
• 5.2.2 Algorithmic Systems as Parallel Reality Constructors 57
• 5.2.3 The Parallel Pathways Framework 57
• 5.2.4 Affective vs. Ideological Polarization: Entanglement and Temporal Sequencing 58
• 5.2.5 The Meta-Cognitive Paradox: Agency and Algorithmic Determinism 59
• 5.3 Methodological Contributions 61
• 5.3.1 Value of Convergent Mixed-Methods Design 61
• 5.3.2 The Convergence Matrix Innovation 61
• 5.3.3 Cross-Cultural Measurement Considerations 62
• 5.4 Practical Implications 62
• 5.4.1 Platform-Level Interventions 62
• 5.4.2 Policy and Regulatory Approaches 63
• 5.4.3 Educational Interventions 63
• 5.4.4 Civil Society Initiatives 64
• 5.5 Limitations and Future Directions 64
• 5.5.1 Design Limitations 64
• 5.5.2 Measurement Limitations 65
• 5.5.3 Sampling Limitations 65
• 5.5.4 Contextual Limitations 66
• 5.5.5 Future Research Priorities 66
• CHAPTER 6. CONCLUSION 66
• 6.1 Contributions 67
• 6.2 Implications and Applications 68
• 6.2.1 Platform Design Implications 68
• 6.2.2 Media Literacy Education 68
• 6.2.3 Policy and Regulatory Approaches 69
• 6.3 Limitations, Future Directions, and Concluding Reflections 69
• 6.3.1 Study Limitations 69
• 6.3.2 Future Research Priorities 70
• 6.3.3 Concluding Reflections 70
• REFERENCES 72
• APPENDICES 76