Introduction to Models & Micro-Targeting

Module Overview

Models and microtargeting are dramatically changing the landscape in Canadian and American elections. What are the applications for smaller political and advocacy campaigns? Could deeper insights into your data provide the competitive advantage you need? This module will provide a compelling, accessible introduction to topics such as segmentation, microtargeting and predictive analytics using well-known examples and simple, back-of-the-envelope exercises. Talia Borodin's clear, engaging teaching style will help you master basic concepts, improve your organization's data collection efforts and use data analysis to maximize your limited resources.

Learning Objectives

  1. Understand the connections between modelling, microtargeting, machine learning and data mining and their applications to modern campaigns.
  2. Interpret a basic turnout model to identify get-out-the-vote (GOTV) and persuasion targets in a campaign.
  3. Understand the risks of bad data and overfitting models.
  4. Identify opportunities to expand your campaign or organization's data collection and analysis efforts.

Presenters

Talia Borodin | Founder, Amaro Science

Talia Borodin is the founder of Amaro Science, a new kind of business dedicated to helping organizations implement data driven decision-making.

After receiving her MSc. in Mathematics, Ms. Borodin moved from Canada to Washington D.C. to work for IBM. Ms. Borodin left IBM two years later to build models to detect tax fraud before being swept up by the D.C. political machine. She is credited with building some of the first national micro-targeting models used in progressive politics. Her work helped pave the road for a new era in data-driven political campaigns.

Ms. Borodin then returned to her native Toronto to build and run the first Data Science team at Kobo, a global e-Reading company. After completely revising the way Kobo measures their business she decided to do the same for a broader audience and founded Amaro Science.