Hi everyone!
Welcome back to the Spring semester of the Applied Statistics Workshop. This
week we will be welcoming *David Lazer*, Distinguished Professor of
Political Science, Computer Science, and Information Science at
Northeastern University. He is also a Visiting Scholar at the Harvard
Kennedy School. He will be presenting work titled "*Tools for 21st Century
Social Science*" (abstract below).
As usual, we will meet in CGIS Knafel Room 354 from noon to 1:30pm, and
lunch will be provided. See you all there! To view previous Applied
Statistics presentations, please visit the website
<http://projects.iq.harvard.edu/applied.stats.workshop-gov3009/videos>.
Best,
Aaron Kaufman
--
Title: *Tools for 21st Century Social Science*
Abstract: Developments at the intersection of the social sciences, computer
science, and the Internet have opened up new vistas for studying social
systems. These opportunities often come with substantial start up costs.
For example, the Internet enables experiments at larger scale/lower costs
than was previously conceivable. However, the start up cost for
managing/coding online experiments can still be substantial. I will discuss
two data infrastructures that my lab has been working on. The first is
Volunteer Science (www.VolunteerScience.com
<https://urldefense.proofpoint.com/v2/url?u=http-3A__www.VolunteerScience.co…>).
Volunteer Science is a platform which enables easy management of online
experiments, integrates with existing infrastructures, such as Qualitrix
and Mechanical Turk, and reduces coding overhead for experiments that
require synchronous communication amongst groups. It also offers a large,
international, volunteer sample of subjects. The second infrastructure is
an open, disambiguated version of political contribution (FEC) data.
Political contribution data are incredibly rich and complex, spanning 36
years, with geographic, occupational, temporal, and employment
information. However, their scale and their messiness—most notably,
absence of unique identifiers for individuals—have impeded scientific
progress. We will be offering an open methodology (as well as a published
data set with unique identifiers—with early access to seminar attendees)
for disambiguating political contribution data, which I will outline in
this talk.