Product Proposal - sewaneedata/PATHS-to-disinformation-2024 GitHub Wiki

Product Proposal

Final Product: Our product will display the positive or negative relationship between individuals' ideology and the ideology sites lean toward. We will look at how people reach fake news, whether it be directly or through referrals, and could their ideology influence the type of media presented/consumed.

Methods: We will use a linear regression model to quantify the relationship between political affiliation bias and the consumption of disinformation. Linear regression is chosen mainly due to its simplicity and interpretability, allowing us to understand the direct impact of political affiliation on disinformation consumption. The model will help us quantify how much more or less disinformation is consumed by individuals with different political biases, controlling for the amount of time spent online and the diversity of sites visited. This will provide insights into the role of political affiliation in disinformation consumption.

Reasoning: Having a safe news environment is of great importance. To have a clean and safe news environment means to provide the public with reliable information that will continue to prove its validity. Working with Professor Cardenal, the aim of this project is to provide insight into how the public and their decisions are affected by unsafe and unreliable news sources and thus have an idea of what makes a news environment unsafe and vice versa. Our models will have an x-axis and a y-axis, with variables interchanging between ideology, type of media, partisan media, and visits to unreliable/ reliable sources. Each model will represent the type of relationship (positive or negative) between each variable we look at to answer our question on whether people reach untrustworthy sites directly or through referrals and how many visits have people made to fake news sites and unreliable sources.