Predicting election results is a crucial component of political science and data analytics, influencing not just academic discussion but also public policy and democratic participation. This research looks into a prediction analysis of the overall popular vote for the approaching 2025 Canadian federal election.
This repository contains the RProject used to analyze the GSS and CES datasets through regression models with post-stratification. It includes samples from various demographic groups and their voting patterns, along with logs of their interactions with the political landscape. A detailed report of our methodology, analysis, and findings is also included.
The data required for this analysis can be downloaded from the following links:
The central research question guiding this study is: “Can the overall popular vote in the 2025 Canadian federal election be accurately predicted using a regression model with post-stratification based on current socio-demographic data?”
Based on the GSS and CES datasets, we hypothesize that a combination of demographic, socioeconomic, and geographic factors will be able to forecast the winning party in the Canadian federal election of 2025 in terms of the popular vote. We predict that a number of factors, including age, education, region, and past voting behavior, will be important predictors of the majority party choice among Canadian voters.