Howard Dean's presidential bid was notable for many things,
including the mixed reaction it drew from political scientists. Many
scholars found Dean's ultimate failure predictable. Longstanding
political science wisdom suggests several explanations for Dean's
defeat: the central issue of electability, which seemed to weigh heavily
against his campaign; the fact that primary voters are more moderate than
party activists; the well-documented difficulty of regaining lost
momentum. Less systematic factors—such as numerous verbal gaffes and
one infamous scream—surely contributed as well.Matthew Hindman is an assistant professor of political science
at Arizona State University (matthew.hindman@asu.edu). This research was
supported by the National Center for Digital Government, with funding from
the National Science Foundation under grant no. 0131923. The author thanks
Jennifer Hochschild, Larry Bartels, Chris Karpowitz, Gabriel Lenz, David
Lazer, Alan Abramowitz, James McCann, and the three anonymous reviewers
for their contributions.
Analytic techniques developed for big data have much broader applications in the social sciences, outperforming standard regression models even—or rather especially—in smaller datasets. This article offers an overview of machine learning methods well-suited to social science problems, including decision trees, dimension reduction methods, nearest neighbor algorithms, support vector models, and penalized regression. In addition to novel algorithms, machine learning places great emphasis on model checking (through holdout samples and cross-validation) and model shrinkage (adjusting predictions toward the mean to reduce overfitting). This article advocates replacing typical regression analyses with two different sorts of models used in concert. A multi-algorithm ensemble approach should be used to determine the noise floor of a given dataset, while simpler methods such as penalized regression or decision trees should be used for theory building and hypothesis testing.
This chapter studies the implications of the various technical developments for citizenship, relying mostly on the term ‘Internet’ as a synecdoche. It then identifies the four dimensions of citizenship, looking at how they relate to media, and also tries to determine how the Internet interacted – and possibly revolutionized – these dimensions of citizenship.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.