Please use this identifier to cite or link to this item:
http://hdl.handle.net/11375/32042
Title: | An exploration of political Facebook posts and their public's reactions |
Authors: | Morrone, Paola |
Keywords: | political social media;political engagement;citizen engagement;machine learning;sentiment analysis |
Publication Date: | 2019 |
Abstract: | This research explored how and to what extent do political Facebook posts engage their followers; to what extent does the Facebook post’s content and features influence the follower’s response; and to what extent can a model predict the followers’ responses to a particular post. The research was based on all the Facebook posts published on the Liberal MP’s official Facebook pages including their associated follower responses over the course of one year. The data was classified using quantitative methods; post and response messages were subjected to content analysis to identify topics and sentiment; and predictive modeling was used to predict engagement levels and response sentiment. The results showed that the Facebook messages posted by the Liberal members of parliament over the lapse of one non-elections year, neglected, for the most part, to engage citizens in either voicing their opinions, sharing the messages or sharing their emotions regarding the topics at hand. Although the correlations between each individual Facebook post features and the followers’ response were weak, the study found that shorter messages received more positive follower engagement; posts with higher positive connotation received more positive comments as well. Similarly, videos and links receive more positive engagement. The results also showed that it is possible to predict follower engagement and response sentiment within a 30% error with the most significant predictors being the post message length, the time of publication, and the post sentiment. The study recommends further research using other social media venues and different political parties as well as incorporating demographic data as predictors. |
URI: | http://hdl.handle.net/11375/32042 |
Appears in Collections: | Master of Communications Management |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
Morrone_Paola_2019_MCM.pdf | 1 MB | Adobe PDF | View/Open |
This item is licensed under a Creative Commons License