Comparing GEE and robust standard errors for conditionally dependent data

Research output: Contribution to journalArticle

61 Citations (Scopus)

Abstract

In recent years political scientists have become increasingly sensitive to questions of conditional dependence in their data. I outline and compare two general, widely-used approaches for addressing such dependence - robust variance estimators and generalized estimating equations (GEEs) - using data on votes in Supreme Court search and seizure decisions between 1963 and 1981. The results make clear that choices about the unit on which data are grouped, i.e., clustered, are typically of far greater significance than are decisions about which type estimator is used.

Original languageEnglish (US)
Pages (from-to)329-341
Number of pages13
JournalPolitical Research Quarterly
Volume59
Issue number3
DOIs
StatePublished - Jan 1 2006

Fingerprint

seizure
political scientist
Supreme Court
voter

All Science Journal Classification (ASJC) codes

  • Sociology and Political Science

Cite this

@article{284a347e44bf4b7d93c75682e8521d65,
title = "Comparing GEE and robust standard errors for conditionally dependent data",
abstract = "In recent years political scientists have become increasingly sensitive to questions of conditional dependence in their data. I outline and compare two general, widely-used approaches for addressing such dependence - robust variance estimators and generalized estimating equations (GEEs) - using data on votes in Supreme Court search and seizure decisions between 1963 and 1981. The results make clear that choices about the unit on which data are grouped, i.e., clustered, are typically of far greater significance than are decisions about which type estimator is used.",
author = "Zorn, {Christopher Jon}",
year = "2006",
month = "1",
day = "1",
doi = "10.1177/106591290605900301",
language = "English (US)",
volume = "59",
pages = "329--341",
journal = "Political Research Quarterly",
issn = "1065-9129",
publisher = "SAGE Publications Inc.",
number = "3",

}

Comparing GEE and robust standard errors for conditionally dependent data. / Zorn, Christopher Jon.

In: Political Research Quarterly, Vol. 59, No. 3, 01.01.2006, p. 329-341.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Comparing GEE and robust standard errors for conditionally dependent data

AU - Zorn, Christopher Jon

PY - 2006/1/1

Y1 - 2006/1/1

N2 - In recent years political scientists have become increasingly sensitive to questions of conditional dependence in their data. I outline and compare two general, widely-used approaches for addressing such dependence - robust variance estimators and generalized estimating equations (GEEs) - using data on votes in Supreme Court search and seizure decisions between 1963 and 1981. The results make clear that choices about the unit on which data are grouped, i.e., clustered, are typically of far greater significance than are decisions about which type estimator is used.

AB - In recent years political scientists have become increasingly sensitive to questions of conditional dependence in their data. I outline and compare two general, widely-used approaches for addressing such dependence - robust variance estimators and generalized estimating equations (GEEs) - using data on votes in Supreme Court search and seizure decisions between 1963 and 1981. The results make clear that choices about the unit on which data are grouped, i.e., clustered, are typically of far greater significance than are decisions about which type estimator is used.

UR - http://www.scopus.com/inward/record.url?scp=33745719096&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33745719096&partnerID=8YFLogxK

U2 - 10.1177/106591290605900301

DO - 10.1177/106591290605900301

M3 - Article

AN - SCOPUS:33745719096

VL - 59

SP - 329

EP - 341

JO - Political Research Quarterly

JF - Political Research Quarterly

SN - 1065-9129

IS - 3

ER -