Followers Retweet! The Influence of Middle‐Level Gatekeepers on the Spread of Political Information on Twitter. Issue 3 (8th March 2019)
- Record Type:
- Journal Article
- Title:
- Followers Retweet! The Influence of Middle‐Level Gatekeepers on the Spread of Political Information on Twitter. Issue 3 (8th March 2019)
- Main Title:
- Followers Retweet! The Influence of Middle‐Level Gatekeepers on the Spread of Political Information on Twitter
- Authors:
- Hemsley, Jeff
- Abstract:
- Abstract : Twitter allows political candidates to broadcast messages directly to the public, some of which spread virally, potentially reaching new supportive audiences. During the 2014 U.S. gubernatorial election, 74 candidates for State Governor posted 20, 580 tweets, of which 10, 946 were retweeted almost 140, 000 times. By analyzing a collection of tweets posted by gubernatorial candidates that were classified by machine learning into categories of message types, we find that while candidates tend to post tweets that advocate for themselves the most, the public is more likely to retweet attack messages and messages labeled as call‐to‐action. As measured by number of retweets, call‐to‐action tweets tend to reach the broadest audience. We also find that middle‐level gatekeepers, those with between 1, 800 and 26, 000 followers, tend to have the most influence over the flow of political information. Since retweets tend to bring new followers, these findings suggest that politicians wishing to grow their audience may benefit from posting more call‐to‐action and attack messages, and that candidates may wish to find ways to actively enlist the support of middle‐level gatekeepers. Abstract : 摘要: 推 特允许政治候选人将信息直接传递给公众。一些信息散播速度很快, 可能吸引新支持者。2014年美国州长选举期间, 74名候选人发布了20, 580条推文, 其中10, 946条被转推了近140, 000次。通过分析由州长候选人发布的一系列推文, 后者被机器学习技术分类为不同信息类型, 笔者发现: 尽管候选人倾向于发布关于支持自己的推文, 但公众更有可能转推具有政治攻击性的信息, 和被列为"号召行动"的信息。通过转推数量进行衡量, 号召行动的推文似乎能吸引最多的关注。笔者还发现, 拥有1, 800–26, 000名关注者的中等把关人,Abstract : Twitter allows political candidates to broadcast messages directly to the public, some of which spread virally, potentially reaching new supportive audiences. During the 2014 U.S. gubernatorial election, 74 candidates for State Governor posted 20, 580 tweets, of which 10, 946 were retweeted almost 140, 000 times. By analyzing a collection of tweets posted by gubernatorial candidates that were classified by machine learning into categories of message types, we find that while candidates tend to post tweets that advocate for themselves the most, the public is more likely to retweet attack messages and messages labeled as call‐to‐action. As measured by number of retweets, call‐to‐action tweets tend to reach the broadest audience. We also find that middle‐level gatekeepers, those with between 1, 800 and 26, 000 followers, tend to have the most influence over the flow of political information. Since retweets tend to bring new followers, these findings suggest that politicians wishing to grow their audience may benefit from posting more call‐to‐action and attack messages, and that candidates may wish to find ways to actively enlist the support of middle‐level gatekeepers. Abstract : 摘要: 推 特允许政治候选人将信息直接传递给公众。一些信息散播速度很快, 可能吸引新支持者。2014年美国州长选举期间, 74名候选人发布了20, 580条推文, 其中10, 946条被转推了近140, 000次。通过分析由州长候选人发布的一系列推文, 后者被机器学习技术分类为不同信息类型, 笔者发现: 尽管候选人倾向于发布关于支持自己的推文, 但公众更有可能转推具有政治攻击性的信息, 和被列为"号召行动"的信息。通过转推数量进行衡量, 号召行动的推文似乎能吸引最多的关注。笔者还发现, 拥有1, 800–26, 000名关注者的中等把关人, 往往最能影响政治信息的流动。研究结果表明, 既然转推文能带来新的关注者, 那些希望增加关注人数的政客能从发布更多号召行动和政治攻击性推文中获取利益, 研究结果还表明, 候选人希望能找到"积极获取中等把关人支持"的方法。 Abstract : Resumen: Twitter permite a los candidatos políticos transmitir mensajes directamente al público, algunos de los cuales se difunden de forma viral, y potencialmente llegan a nuevas audiencias de apoyo. Durante la elección de gobernador de los EE. UU. de 2014, 74 candidatos para gobernador del estado publicaron 20.580 tweets, de los cuales 10.946 fueron retuiteados casi 140.000 veces. Al analizar una colección de tweets publicados por candidatos a gobernador que se clasificaron por aprendizaje automático en categorías de tipos de mensajes, encontramos que mientras los candidatos tienden a publicar tweets donde se defienden a sí mismos, es más probable que el público retuitee mensajes de ataque y mensajes etiquetados como llamada a la acción. Según lo medido por la cantidad de retweets, los tweets de llamada a la acción tienden a llegar a la audiencia más amplia. También encontramos que los guardianes de nivel medio, aquellos con entre 1.800 y 26.000 seguidores, tienden a tener la mayor influencia sobre el flujo de información política. Dado que los retweets tienden a atraer nuevos seguidores, estos hallazgos sugieren que los políticos que desean aumentar su audiencia pueden beneficiarse de la publicación de más mensajes de llamada a la acción y de ataque, y que los candidatos pueden desear encontrar formas de obtener activamente el apoyo de los guardianes de nivel medio. … (more)
- Is Part Of:
- Policy and internet. Volume 11:Issue 3(2019:Sep.)
- Journal:
- Policy and internet
- Issue:
- Volume 11:Issue 3(2019:Sep.)
- Issue Display:
- Volume 11, Issue 3 (2019)
- Year:
- 2019
- Volume:
- 11
- Issue:
- 3
- Issue Sort Value:
- 2019-0011-0003-0000
- Page Start:
- 280
- Page End:
- 304
- Publication Date:
- 2019-03-08
- Subjects:
- information flow -- political elections -- Twitter -- curation logics -- network gatekeepers -- diffusion
信息流 -- 政治选举 -- 推特 -- 信息内容处理逻辑(curation logics) -- 网络把关人 -- 扩散
flujo de información -- elecciones políticas -- Twitter -- lógicas de curación -- guardianes de redes -- difusión
Internet -- Government policy -- Periodicals
338.926 - Journal URLs:
- http://onlinelibrary.wiley.com/journal/10.1002/(ISSN)1944-2866 ↗
http://www.bepress.com/pso_internet ↗
http://onlinelibrary.wiley.com/ ↗ - DOI:
- 10.1002/poi3.202 ↗
- Languages:
- English
- ISSNs:
- 1944-2866
- Deposit Type:
- Legaldeposit
- View Content:
- Available online (eLD content is only available in our Reading Rooms) ↗
- Physical Locations:
- British Library DSC - 6543.321740
British Library DSC - BLDSS-3PM
British Library HMNTS - ELD Digital store - Ingest File:
- 17061.xml