臉書粉絲專頁互動與選舉結果之相關性研究

NCU Institutional Repository > 管理學院 > 資訊管理研究所 > 博碩士論文 > Item 987654321/81251

jsp.display-item.identifier=請使用永久網址來引用或連結此文件: http://ir.lib.ncu.edu.tw/handle/987654321/81251

题名:	臉書粉絲專頁互動與選舉結果之相關性研究
作者:	黃琮閔;HUANG, CONG-MIN
贡献者:	資訊管理學系
关键词:	社群媒體;政治參與;選舉預測;Social media;Facebook;Political participation;Election forecast
日期:	2019-07-12
上传时间:	2019-09-03 15:40:31 (UTC+8)
出版者:	國立中央大學
摘要:	社群媒體對選舉的影響扮演越來越重要的角色，本研究旨在提出一個用於解釋得票率所受影響之整合模型，並分析不同屬性之線上聲量指標與選舉結果的相關性，且依據實証結果詮釋社群媒體活躍程度與選舉結果的關係。本研究以2018年台灣直轄市長選舉19位候選人為研究對象，分析其1,137篇Facebook貼文之互動資料，並以縱慣性研究的概念為基礎，在指標設計中加入時間跨度，使指標能夠反映隨時間變化的互動量，一共設計了42個指標。研究結果顯示「所有貼文按讚數總和」、「平均粉專按讚成長」、「平均粉專追蹤成長」、「最大單篇貼文按讚數」以及「平均每篇貼文按讚數」是對得票率最具影響力之前五大指標，且整體上按讚的影響力大於分享大於留言。此結果可供未來學者與候選人設計社群媒體策略時之重要參考。;The impact of social media on elections is playing an increasingly important role. Our research proposes an integrated model extracting the key impact factor of vote shares and analyzing the correlation between online popularity and election results. This research resolves the relationship between social media engagement and election outcome based on empirical results. We take 19 candidates from the mayor election of Taiwan in 2018 as the research object, analyzes social media data from 1,137 Facebook posts, and adds time span to the index design based on the concept of longitudinal research. With this measurement, we adopt 42 indicators which can reflect time-varying interactions. A total of 42 indicators were designed. Our research shows that “Sum of all posts likes”, “Average growth of fan page likes”, “Average growth of fan page follows”, “Max single post likes” and "Average post likes” are the top five indicators that have the most influence on vote shares, and the overall influence of likes is greater than shares and greater than the comments. Our research provides a new integrated model which employ the longitudinal measurement and extract top five social media impact factor of elections which are indispensable groundworks for future researches.
显示于类别:	[資訊管理研究所] 博碩士論文

文件中的档案:

档案	描述	大小	格式	浏览次数
index.html		0Kb	HTML	165	检视/开启

在NCUIR中所有的数据项都受到原著作权保护.

社群 sharing

数据加载中.....