Term of Award

Fall 2018

Degree Name

Master of Science in Applied Engineering (M.S.A.E.)

Document Type and Release Option

Thesis (restricted to Georgia Southern)

Copyright Statement / License for Reuse

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.

Department

Department of Information Technology

Committee Chair

Cheryl Aasheim

Committee Member 1

Jeffrey Kaleta

Committee Member 2

Adrian Gardiner

Abstract

Political bias in the media is not only relevant because it has been extensively discussed, it is important to study because it can have significant influence on political knowledge, attitudes and the beliefs of the people. Although several studies that investigate political bias have been published, there is need for an approach to detect this bias using modern computing method called text analytics. The results presented in this study are the outcomes of the analysis carried out on the news articles extracted from three mainstream media in the U.S using a newly developed web scraping tool. LIWC was then used to analyze the data and several text analytics techniques are employed through Python to formulate other key variables important to our study. With our results, we compare the potential difference in left versus right media bias between CNN and Fox News. Also, we compare differences between fact reporting versus opinion or persuasion. These findings correspond with the existing studies in media bias (Otero, 2018). We discovered that there are differences in media bias and these are defined by linguistic processes, psychological processes, and other language variables. The dataset provided in the study can serve as a basis for further research on media bias. We also discovered that discriminant analysis is a useful statistical tool for text analysis research as it considers all relevant language attributes expressed in the news to formulate its model. Finally, we described important language attributes to be examined when analyzing news media for bias.

OCLC Number

1085541971

Research Data and Supplementary Material

Yes

Download

Share

COinS