成果展示

2022年成果

当前位置: 首页 > 成果展示 > 数字人文成果 > 2022年成果 > 正文

远离主流:媒体注意力与欧美国家极化媒体研究

闫蒲,Ralph Schroeder

北京大学信息管理系,牛津大学互联网研究所

2024-05-09

发布

背景 Background


The rise of partisan news and populism:

• 23% of panellists visited at least one of the partisan news websites such as Breitbart or The Blaze each week (Digital News Report, 2020)

• Trust in news:

• Declined trust of legacy media among the “very left- or right-wing” voter (Newman et al., 2020)

• The link between populist attitudes and consumption of quality newspaper & public broadcasters (Schulz, 2018, 2019)

• Examine some digital platforms as hyperpartisan news as a starting point to understand anti-media and anti-elite content and sentiments in news (Rea, 2020)

Defining hyperpartisan media:

• Hyper-partisan news: news or political information that falls outside of objective, professional, impartial, independent, diverse, and inclusive media.

• Compared to traditional (“mainstream”) media:

• Appeal not to a broad audiences

• Do not cover a diverse range of topics

• Hyperpartisan media & populism:

• Corrupted elites vs the people (Mudd, 2007)

• Media are seen as part of the political elite/establishment (Fawzi, 2009)

image

图1

数据来源 Method and Research Questions


Method

• Datasets: Multi-country surveys (attitudinal) & website visits (behavioural) • Six countries: US, UK, Germany, France, Spain, and Italy • Data collected in Spring 2019

Research questions

• Who visits hyperpartisan media sites? • What is the structure of the audience similarity network of hyperpartisan media domains? • What topics are covered in hyperpartisan websites?

数据结果分析 Findings


Finding #1:

In the regression analysis of hyperpartisan visits, we found that right-wing populist party supporters are more likely than other voters to visit hyperpartisan media websites. Also, Panellists who have higher levels of trust in social media and lower levels of trust in public broadcasters are more likely to consume hyperpartisan media websites. More importantly, the link between distrust in public broadcasters and visits to hyperpartisan media has been reinforced when the panellist support right-wing populist parties.

null

图2

Finding #2:

We conducted community detection on the network of hyperpartisan domains and found the hyperpartisan media websites are clustered by language. Nevertheless, some websites, such as dailykos.com or rt.com, sit across different linguistic communities. The network visualisation of hyperpartisan media can also inform qualitative researchers, by helping them identify hyperpartisan websites that share similar cohorts of audiences

null

图3

Finding #3:

Topic modelling analysis of the news titles of hyperpartisan media suggest that hyperpartisan media agenda indeed covers a less diverse social issues. Politics and religion are common themes in the identified topics.

null

图4

研究结果 Discussions


• Triangulation of webtracking dataset with survey dataset

• The link between populist party support and hyperpartisan media visits

• Media trust of public service media and hyperpartisan media visits

• The importance of professional journalism and public broadcasters: Balancing a polarised and partisan media environment

版权所有©北京大学文科智数化公共平台