A regression model for longitudinal change in the presence of measurement error

Norbert Yanez, Richard A. Kronmal, Lynn R. Shemanski, Bruce M. Psaty

Research output: Contribution to journalArticle

36 Citations (Scopus)

Abstract

PURPOSE: The analysis of change in measured variables has become quite popular in studies where data are collected repeatedly over time. The authors describe some of the potential pitfalls in the analysis of change when the variable for change is measured with error. They show that regression analysis is often biased, possibly leading to erroneous results. METHODS: A simple method to correct for measurement error bias in regression models that model change is presented. RESULTS AND CONCLUSIONS: The two examples illustrate how measurement error can adversely affect an analysis. The bias-corrected approach yields valid results.

Original languageEnglish (US)
Pages (from-to)34-38
Number of pages5
JournalAnnals of Epidemiology
Volume12
Issue number1
DOIs
StatePublished - 2002
Externally publishedYes

Fingerprint

Regression Analysis

Keywords

  • Bias-correction
  • Bootstrap standard errors

ASJC Scopus subject areas

  • Medicine(all)
  • Public Health, Environmental and Occupational Health
  • Epidemiology

Cite this

A regression model for longitudinal change in the presence of measurement error. / Yanez, Norbert; Kronmal, Richard A.; Shemanski, Lynn R.; Psaty, Bruce M.

In: Annals of Epidemiology, Vol. 12, No. 1, 2002, p. 34-38.

Research output: Contribution to journalArticle

Yanez, Norbert ; Kronmal, Richard A. ; Shemanski, Lynn R. ; Psaty, Bruce M. / A regression model for longitudinal change in the presence of measurement error. In: Annals of Epidemiology. 2002 ; Vol. 12, No. 1. pp. 34-38.
@article{46fd53c12cb8440cb7fbafce95fa362f,
title = "A regression model for longitudinal change in the presence of measurement error",
abstract = "PURPOSE: The analysis of change in measured variables has become quite popular in studies where data are collected repeatedly over time. The authors describe some of the potential pitfalls in the analysis of change when the variable for change is measured with error. They show that regression analysis is often biased, possibly leading to erroneous results. METHODS: A simple method to correct for measurement error bias in regression models that model change is presented. RESULTS AND CONCLUSIONS: The two examples illustrate how measurement error can adversely affect an analysis. The bias-corrected approach yields valid results.",
keywords = "Bias-correction, Bootstrap standard errors",
author = "Norbert Yanez and Kronmal, {Richard A.} and Shemanski, {Lynn R.} and Psaty, {Bruce M.}",
year = "2002",
doi = "10.1016/S1047-2797(01)00280-0",
language = "English (US)",
volume = "12",
pages = "34--38",
journal = "Annals of Epidemiology",
issn = "1047-2797",
publisher = "Elsevier Inc.",
number = "1",

}

TY - JOUR

T1 - A regression model for longitudinal change in the presence of measurement error

AU - Yanez, Norbert

AU - Kronmal, Richard A.

AU - Shemanski, Lynn R.

AU - Psaty, Bruce M.

PY - 2002

Y1 - 2002

N2 - PURPOSE: The analysis of change in measured variables has become quite popular in studies where data are collected repeatedly over time. The authors describe some of the potential pitfalls in the analysis of change when the variable for change is measured with error. They show that regression analysis is often biased, possibly leading to erroneous results. METHODS: A simple method to correct for measurement error bias in regression models that model change is presented. RESULTS AND CONCLUSIONS: The two examples illustrate how measurement error can adversely affect an analysis. The bias-corrected approach yields valid results.

AB - PURPOSE: The analysis of change in measured variables has become quite popular in studies where data are collected repeatedly over time. The authors describe some of the potential pitfalls in the analysis of change when the variable for change is measured with error. They show that regression analysis is often biased, possibly leading to erroneous results. METHODS: A simple method to correct for measurement error bias in regression models that model change is presented. RESULTS AND CONCLUSIONS: The two examples illustrate how measurement error can adversely affect an analysis. The bias-corrected approach yields valid results.

KW - Bias-correction

KW - Bootstrap standard errors

UR - http://www.scopus.com/inward/record.url?scp=0036135436&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036135436&partnerID=8YFLogxK

U2 - 10.1016/S1047-2797(01)00280-0

DO - 10.1016/S1047-2797(01)00280-0

M3 - Article

C2 - 11750238

AN - SCOPUS:0036135436

VL - 12

SP - 34

EP - 38

JO - Annals of Epidemiology

JF - Annals of Epidemiology

SN - 1047-2797

IS - 1

ER -