Pre-election polling

Identifying likely voters using iterative expert data mining

Gregory Roy Murray, Chris Riley, Anthony Scime

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

One often-noted difficulty in pre-election polling is the identification of likely voters. Our objective is to build a likely voter model for presidential elections that efficiently balances accuracy and number of questions used. We employ the Iterative Expert Data Mining technique and data from the American National Election Studies to identify a small number of survey questions that can be used to classify likely voters while maintaining or surpassing the accuracy rates of other models. Specifically, we propose two survey items that together correctly classify 78 percent of respondents as voters or nonvoters over a multielection, multidecade period. We argue that our proposed model compares favorably to competing models by capturing the successful elements of those models while ignoring other elements that constrain identification. We end by suggesting that our model offers a new approach to identifying and evaluating likely voters that may maintain or increase accuracy without also increasing cost.

Original languageEnglish (US)
Pages (from-to)159-171
Number of pages13
JournalPublic Opinion Quarterly
Volume73
Issue number1
DOIs
StatePublished - Aug 14 2009
Externally publishedYes

Fingerprint

Data mining
election
expert
election research
presidential election
Data Mining
Polling
Voters
Elections
Identification (control systems)
voter
costs
Costs

ASJC Scopus subject areas

  • Communication
  • History
  • Sociology and Political Science
  • Social Sciences(all)
  • History and Philosophy of Science

Cite this

Pre-election polling : Identifying likely voters using iterative expert data mining. / Murray, Gregory Roy; Riley, Chris; Scime, Anthony.

In: Public Opinion Quarterly, Vol. 73, No. 1, 14.08.2009, p. 159-171.

Research output: Contribution to journalArticle

@article{401ff28c6a0e49669515aebe0439f3f9,
title = "Pre-election polling: Identifying likely voters using iterative expert data mining",
abstract = "One often-noted difficulty in pre-election polling is the identification of likely voters. Our objective is to build a likely voter model for presidential elections that efficiently balances accuracy and number of questions used. We employ the Iterative Expert Data Mining technique and data from the American National Election Studies to identify a small number of survey questions that can be used to classify likely voters while maintaining or surpassing the accuracy rates of other models. Specifically, we propose two survey items that together correctly classify 78 percent of respondents as voters or nonvoters over a multielection, multidecade period. We argue that our proposed model compares favorably to competing models by capturing the successful elements of those models while ignoring other elements that constrain identification. We end by suggesting that our model offers a new approach to identifying and evaluating likely voters that may maintain or increase accuracy without also increasing cost.",
author = "Murray, {Gregory Roy} and Chris Riley and Anthony Scime",
year = "2009",
month = "8",
day = "14",
doi = "10.1093/poq/nfp004",
language = "English (US)",
volume = "73",
pages = "159--171",
journal = "Public Opinion Quarterly",
issn = "0033-362X",
publisher = "Oxford University Press",
number = "1",

}

TY - JOUR

T1 - Pre-election polling

T2 - Identifying likely voters using iterative expert data mining

AU - Murray, Gregory Roy

AU - Riley, Chris

AU - Scime, Anthony

PY - 2009/8/14

Y1 - 2009/8/14

N2 - One often-noted difficulty in pre-election polling is the identification of likely voters. Our objective is to build a likely voter model for presidential elections that efficiently balances accuracy and number of questions used. We employ the Iterative Expert Data Mining technique and data from the American National Election Studies to identify a small number of survey questions that can be used to classify likely voters while maintaining or surpassing the accuracy rates of other models. Specifically, we propose two survey items that together correctly classify 78 percent of respondents as voters or nonvoters over a multielection, multidecade period. We argue that our proposed model compares favorably to competing models by capturing the successful elements of those models while ignoring other elements that constrain identification. We end by suggesting that our model offers a new approach to identifying and evaluating likely voters that may maintain or increase accuracy without also increasing cost.

AB - One often-noted difficulty in pre-election polling is the identification of likely voters. Our objective is to build a likely voter model for presidential elections that efficiently balances accuracy and number of questions used. We employ the Iterative Expert Data Mining technique and data from the American National Election Studies to identify a small number of survey questions that can be used to classify likely voters while maintaining or surpassing the accuracy rates of other models. Specifically, we propose two survey items that together correctly classify 78 percent of respondents as voters or nonvoters over a multielection, multidecade period. We argue that our proposed model compares favorably to competing models by capturing the successful elements of those models while ignoring other elements that constrain identification. We end by suggesting that our model offers a new approach to identifying and evaluating likely voters that may maintain or increase accuracy without also increasing cost.

UR - http://www.scopus.com/inward/record.url?scp=68349134893&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=68349134893&partnerID=8YFLogxK

U2 - 10.1093/poq/nfp004

DO - 10.1093/poq/nfp004

M3 - Article

VL - 73

SP - 159

EP - 171

JO - Public Opinion Quarterly

JF - Public Opinion Quarterly

SN - 0033-362X

IS - 1

ER -