4. Electronic Theses and Dissertations (ETDs) - Faculties submissions

Permanent URI for this communityhttps://hdl.handle.net/10539/37773

For queries relating to content and technical issues, please contact IR specialists via this email address : openscholarship.library@wits.ac.za, Tel: 011 717 4652 or 011 717 1954

Browse

Filters

20241

Settings

Author: search.filters.author.Ngwenya, Nonhlanhla NomusaSubject: search.filters.subject.Bias

Search Results

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    A computational study of media bias in South African online political news reporting over the period 2021 - 2023
    (University of the Witwatersrand, Johannesburg, 2024) Ngwenya, Nonhlanhla Nomusa; Alence, Rod
    The study examined the presence of tonality bias in South African political news reporting over the period 2021 until mid-2023. The study employed the methods of the Lexicoder Sentiment Dictionary, a lexical-based method, and Latent Semantic Scaling, a semi-supervised machine learning method. Sentiment was utilised as a proxy for tonality. Online commercial media publishers were contrasted against the state-owned news publisher to ascertain how online news reporting contributed to shaping the national agenda, and the framing of political actors and their respective political parties. The Lexicoder Sentiment Dictionary and the Latent Semantic Scaling evidenced that commercial media publishers exhibited positive tonality bias for the Democratic Alliance during the 2021 Municipal Elections. South African media publishers were found to exhibit consistent negative tonality bias when reporting on protest action. The state-owned media publisher was found to drive a pro ruling party sentiment whereas commercial media publishers’ sentiment was anti- populist and agenda-setting. The congruency in political news reporting gave grounds to the call for diversity in publishing