Did COVID-19 pandemic period increased the italian right-wing online interest? Did right-wing related keywords spread on Italian Google searches during the pandemic?
Inspired by a Rovetta A. study published on JMIR, the aim of this project is testing the hypothesis claiming that the volume of queries on Google Italia about right-wing topics, actors, websites and news has raised during the pandemic.
To test the hypothesis, I create a dataset that contains as many right-wing related keywords as possible. In order to do it, I start from scraped list of organizations and people linked to right-wing parties and then look for related queries trying to identify the trending keywords among users.
The starting list is obtained by scraping an already existing map made by patriaindipendente.it.
The best way to import a dataset with all trending queries on Google is by using Google Trends and, since I'm exclusively using RStudio to work on this project, I employ gtrendsR package (more info).
This package allows us to get a dataset with different variables: geo, time, keywords and onlyInterest (a boolean that allows to call ONLY the interest_over_time or not).
Once obtained the dataframe, and cleaned it with regex, I can analyze it and then try to answer to three questions:
- Is there a seasonality to the queries?
- Crossing
dateandhitsvariables, will we observe a positive linear trend? - Overall, how effective can be Google Trends on following socio-political movements?
My researches about the topic suggest me that lockdowns and, more generally, COVID-19 Pandemic, has raised internet traffic and extremism has gained popularity among world citizens1, therefore it's likely that my plot visualization will show a positive linear trend.
No prediction could be made for the first question, since that the answer could be affected by important socio-political events occurred during the pandemic or even changes in the lockdown severity.
Last question is a secondary quest I added for a more personal curiosity2, so the answer will be more subjective and won't be answered with a statistical approach but rather with a researcher point of view that is testing Rstudio's package gtrendsR and Google Trends for my socio-political studies! ✌️
| Folder | Contents |
|---|---|
| makingof | checkpoints and backups |
| data | datasets |
| src | scripts |
| figs | plots and images |
| output | reports (pdf/Rmd) |
- Rovetta A. "The Impact of COVID-19 on Conspiracy Hypotheses and Risk Perception in Italy: Infodemiological Survey Study Using Google Trends" JMIR Infodemiology 2021;1(1):e29929
- Davies G., Wu E. & Frank R. "A Witch’s Brew of Grievances: The Potential Effects of COVID-19 on Radicalization to Violent Extremism" Studies in Conflict & Terrorism; Taylor & Francis 1-24, 2021
- Burchill R. "Extremism in the Time of COVID-19"; Bussola Research Paper No. 6, July 15, 2020
- Singh K. "Effect of COVID-19 on Violent Extremism and Terrorism in Digital Spaces", Centre for Air Power Studies (CAPS), December 2020
- Patria Indipendente: Cronache dalla galassia nera
- Patria Indipendente: Progetto Facebook, lista