How to analyze political attention with minimal assumptions and costs

Kevin M. Quinn, Burt L. Monroe, Michael Colaresi, Michael H. Crespin, Dragomir R. Radev

Research output: Contribution to journalArticle

213 Citations (Scopus)

Abstract

Previous methods of analyzing the substance of political attention have had to make several restrictive assumptions or been prohibitively costly when applied to large-scale political texts. Here, we describe a topic model for legislative speech, a statistical learning model that uses word choices to infer topical categories covered in a set of speeches and to identify the topic of specific speeches. Our method estimates, rather than assumes, the substance of topics, the keywords that identify topics, and the hierarchical nesting of topics. We use the topic model to examine the agenda in the U.S. Senate from 1997 to 2004. Using a new database of over 118,000 speeches (70,000,000 words) from the Congressional Record, our model reveals speech topic categories that are both distinctive and meaningfully interrelated and a richer view of democratic agenda dynamics than had previously been possible.

Original languageEnglish (US)
Pages (from-to)209-228
Number of pages20
JournalAmerican Journal of Political Science
Volume54
Issue number1
DOIs
StatePublished - Jan 1 2010

Fingerprint

costs
senate
learning

All Science Journal Classification (ASJC) codes

  • Sociology and Political Science
  • Political Science and International Relations

Cite this

Quinn, Kevin M. ; Monroe, Burt L. ; Colaresi, Michael ; Crespin, Michael H. ; Radev, Dragomir R. / How to analyze political attention with minimal assumptions and costs. In: American Journal of Political Science. 2010 ; Vol. 54, No. 1. pp. 209-228.
@article{1f976df783954cb9bafb040219fd5a6a,
title = "How to analyze political attention with minimal assumptions and costs",
abstract = "Previous methods of analyzing the substance of political attention have had to make several restrictive assumptions or been prohibitively costly when applied to large-scale political texts. Here, we describe a topic model for legislative speech, a statistical learning model that uses word choices to infer topical categories covered in a set of speeches and to identify the topic of specific speeches. Our method estimates, rather than assumes, the substance of topics, the keywords that identify topics, and the hierarchical nesting of topics. We use the topic model to examine the agenda in the U.S. Senate from 1997 to 2004. Using a new database of over 118,000 speeches (70,000,000 words) from the Congressional Record, our model reveals speech topic categories that are both distinctive and meaningfully interrelated and a richer view of democratic agenda dynamics than had previously been possible.",
author = "Quinn, {Kevin M.} and Monroe, {Burt L.} and Michael Colaresi and Crespin, {Michael H.} and Radev, {Dragomir R.}",
year = "2010",
month = "1",
day = "1",
doi = "10.1111/j.1540-5907.2009.00427.x",
language = "English (US)",
volume = "54",
pages = "209--228",
journal = "American Journal of Political Science",
issn = "0092-5853",
publisher = "Wiley-Blackwell",
number = "1",

}

How to analyze political attention with minimal assumptions and costs. / Quinn, Kevin M.; Monroe, Burt L.; Colaresi, Michael; Crespin, Michael H.; Radev, Dragomir R.

In: American Journal of Political Science, Vol. 54, No. 1, 01.01.2010, p. 209-228.

Research output: Contribution to journalArticle

TY - JOUR

T1 - How to analyze political attention with minimal assumptions and costs

AU - Quinn, Kevin M.

AU - Monroe, Burt L.

AU - Colaresi, Michael

AU - Crespin, Michael H.

AU - Radev, Dragomir R.

PY - 2010/1/1

Y1 - 2010/1/1

N2 - Previous methods of analyzing the substance of political attention have had to make several restrictive assumptions or been prohibitively costly when applied to large-scale political texts. Here, we describe a topic model for legislative speech, a statistical learning model that uses word choices to infer topical categories covered in a set of speeches and to identify the topic of specific speeches. Our method estimates, rather than assumes, the substance of topics, the keywords that identify topics, and the hierarchical nesting of topics. We use the topic model to examine the agenda in the U.S. Senate from 1997 to 2004. Using a new database of over 118,000 speeches (70,000,000 words) from the Congressional Record, our model reveals speech topic categories that are both distinctive and meaningfully interrelated and a richer view of democratic agenda dynamics than had previously been possible.

AB - Previous methods of analyzing the substance of political attention have had to make several restrictive assumptions or been prohibitively costly when applied to large-scale political texts. Here, we describe a topic model for legislative speech, a statistical learning model that uses word choices to infer topical categories covered in a set of speeches and to identify the topic of specific speeches. Our method estimates, rather than assumes, the substance of topics, the keywords that identify topics, and the hierarchical nesting of topics. We use the topic model to examine the agenda in the U.S. Senate from 1997 to 2004. Using a new database of over 118,000 speeches (70,000,000 words) from the Congressional Record, our model reveals speech topic categories that are both distinctive and meaningfully interrelated and a richer view of democratic agenda dynamics than had previously been possible.

UR - http://www.scopus.com/inward/record.url?scp=73649142099&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=73649142099&partnerID=8YFLogxK

U2 - 10.1111/j.1540-5907.2009.00427.x

DO - 10.1111/j.1540-5907.2009.00427.x

M3 - Article

AN - SCOPUS:73649142099

VL - 54

SP - 209

EP - 228

JO - American Journal of Political Science

JF - American Journal of Political Science

SN - 0092-5853

IS - 1

ER -