Forecasting Tomorrow's Election Today

About the lecture

Why does political polling sometimes fail to predict election results? Polling is not a central concern of the academic field of political science, but insights from political science are potentially beneficial to the project of predicting elections.

This lecture discusses standard polling methods as well as newer modelling frameworks like "multilevel regression and post-stratification" that are increasingly used for projecting election polling data onto "first-past-the-post" outcomes at the constituency/district/state level in countries like the UK and US. These methods are only likely to perform well where their assumptions are well-grounded in the enduring findings of the political science research literature regarding patterns of voting behaviour across elections and the properties of self-reported political attitudes and intentions.

About the speaker

Benjamin Lauderdale is a Professor of Political Science at UCL. He obtained his PhD from Princeton in 2010 and taught at the London School of Economics from 2011 to 2018. He is currently an Associate Editor of the American Political Science Review (2016-2020) as well as a Senior Data Science Advisor to YouGov. His research is focused on the measurement of political preferences from survey, voting, network and text data. Applications of these methods have included citizens, legislators and judges in the US, UK and EU.

Image by MetsikGarden from Pixabay

Missed the lecture? Catch up below:

• Read the blog post by Sanjana Balakrishnan, BSc Politics and International Relations student

Inaugural Lecture Series 2019/20

This lecture is part of the 2019/20 series for UCL's Faculty of Arts & Humanities and Faculty of Social & Historical Sciences. The series provides an opportunity to recognise and celebrate the achievements of our professors who are undertaking research and scholarship of international significance, and offers an insight into the strength and vitality of the arts, humanities and social sciences at UCL.

All our lectures are free to attend and open to all. You don't have to be a UCL staff member or student to come along.

Lectures begin at 18:30 and are typically one hour long. A drinks reception will follow, to which everyone is welcome to join.

We look forward to meeting you at one of our events.

For information on other upcoming lectures please visit: https://www.ucl.ac.uk/social-historical-sciences/news-events/inaugural-lectures