Speakers
Description
Uncertainty has increasingly become an inherent feature of contemporary societies, especially after the Great Recession started in 2007 which undermined the stability of most national economies.
Such economic crisis not only directly affects many families and individuals by lowering their income but also indirectly, by engendering general perceptions of uncertainty about the future. This implies that narratives of the economy affect individual decision-making processes regardless of real structural economic constraints.
This paper aims to understand how media have framed economic issues over time, also examining the narratives of different newspapers in terms of topics and semantics.
To address these research questions, we use data from the LexisNexis database (a unique news archive containing almost two billion articles from the European press over the last 20 years). From the half of 2008 to the end of 2019, through a supervised machine learning approach, we select articles related to economic issues from the ten most important Italian newspapers.
We intend to delve qualitatively into media framing dynamics. For this purpose, we first apply LDA topic modelling techniques to detect different clusters of words and similar expressions that best characterize a specific set of articles, in order to grasp the themes and subjects most associated with the economy in the press. Then, through the pattern of topics co-occurrences, we build a dynamic network that allows us to understand how narrative structures evolve over time. Finally, we compare the narrative structures of different newspapers, based on visual clustering and graph analytical tools (e.g. modularity analysis, community detection), over time.
Topics | • Textual data analysis and network methods |
---|---|
Keywords | Media, Economy, Fertility, Italy |