Construct Validity in Health Behavior Research: Interpreting Latent Variable Models Involving Self-Report and Objective Measures

Raymond F. Palmer, John W. Graham, Bonnie Taylor, James Tatterson

Research output: Contribution to journalArticle

25 Citations (Scopus)

Abstract

Latent variable models assess the common variance across multiple indicators of a specific construct and are often used when measurement error may bias parameter estimates. However, care must be taken when interpreting the meaning of the latent construct when using item indicators that come from different measurement domains (e.g., self-report and biochemical indicators of smoking). Utilizing simulated data, we demonstrate that even though a model may be considered to have a "good fit" based on conventional criteria, data interpretation may be misleading or erroneous if precautions are not taken when specifying residual covariances. These findings have important implications for health-related research. Whenever different kinds of data are used to define latent variables in a health domain, exactly what items are used, and what biases may be present can affect, sometimes dramatically, (a) the definition of the latent variables and (b) the effects of the latent variables on other variables of interest.

Original languageEnglish (US)
Pages (from-to)525-550
Number of pages26
JournalJournal of Behavioral Medicine
Volume25
Issue number6
DOIs
StatePublished - Dec 1 2002

Fingerprint

Health Behavior
Self Report
Health
Research
Smoking

All Science Journal Classification (ASJC) codes

  • Psychology(all)
  • Psychiatry and Mental health

Cite this

Palmer, Raymond F. ; Graham, John W. ; Taylor, Bonnie ; Tatterson, James. / Construct Validity in Health Behavior Research : Interpreting Latent Variable Models Involving Self-Report and Objective Measures. In: Journal of Behavioral Medicine. 2002 ; Vol. 25, No. 6. pp. 525-550.
@article{b3cb1b3fc9994937bf39df1f0d37b908,
title = "Construct Validity in Health Behavior Research: Interpreting Latent Variable Models Involving Self-Report and Objective Measures",
abstract = "Latent variable models assess the common variance across multiple indicators of a specific construct and are often used when measurement error may bias parameter estimates. However, care must be taken when interpreting the meaning of the latent construct when using item indicators that come from different measurement domains (e.g., self-report and biochemical indicators of smoking). Utilizing simulated data, we demonstrate that even though a model may be considered to have a {"}good fit{"} based on conventional criteria, data interpretation may be misleading or erroneous if precautions are not taken when specifying residual covariances. These findings have important implications for health-related research. Whenever different kinds of data are used to define latent variables in a health domain, exactly what items are used, and what biases may be present can affect, sometimes dramatically, (a) the definition of the latent variables and (b) the effects of the latent variables on other variables of interest.",
author = "Palmer, {Raymond F.} and Graham, {John W.} and Bonnie Taylor and James Tatterson",
year = "2002",
month = "12",
day = "1",
doi = "10.1023/A:1020689316518",
language = "English (US)",
volume = "25",
pages = "525--550",
journal = "Journal of Behavioral Medicine",
issn = "0160-7715",
publisher = "Springer New York",
number = "6",

}

Construct Validity in Health Behavior Research : Interpreting Latent Variable Models Involving Self-Report and Objective Measures. / Palmer, Raymond F.; Graham, John W.; Taylor, Bonnie; Tatterson, James.

In: Journal of Behavioral Medicine, Vol. 25, No. 6, 01.12.2002, p. 525-550.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Construct Validity in Health Behavior Research

T2 - Interpreting Latent Variable Models Involving Self-Report and Objective Measures

AU - Palmer, Raymond F.

AU - Graham, John W.

AU - Taylor, Bonnie

AU - Tatterson, James

PY - 2002/12/1

Y1 - 2002/12/1

N2 - Latent variable models assess the common variance across multiple indicators of a specific construct and are often used when measurement error may bias parameter estimates. However, care must be taken when interpreting the meaning of the latent construct when using item indicators that come from different measurement domains (e.g., self-report and biochemical indicators of smoking). Utilizing simulated data, we demonstrate that even though a model may be considered to have a "good fit" based on conventional criteria, data interpretation may be misleading or erroneous if precautions are not taken when specifying residual covariances. These findings have important implications for health-related research. Whenever different kinds of data are used to define latent variables in a health domain, exactly what items are used, and what biases may be present can affect, sometimes dramatically, (a) the definition of the latent variables and (b) the effects of the latent variables on other variables of interest.

AB - Latent variable models assess the common variance across multiple indicators of a specific construct and are often used when measurement error may bias parameter estimates. However, care must be taken when interpreting the meaning of the latent construct when using item indicators that come from different measurement domains (e.g., self-report and biochemical indicators of smoking). Utilizing simulated data, we demonstrate that even though a model may be considered to have a "good fit" based on conventional criteria, data interpretation may be misleading or erroneous if precautions are not taken when specifying residual covariances. These findings have important implications for health-related research. Whenever different kinds of data are used to define latent variables in a health domain, exactly what items are used, and what biases may be present can affect, sometimes dramatically, (a) the definition of the latent variables and (b) the effects of the latent variables on other variables of interest.

UR - http://www.scopus.com/inward/record.url?scp=0036886538&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036886538&partnerID=8YFLogxK

U2 - 10.1023/A:1020689316518

DO - 10.1023/A:1020689316518

M3 - Article

C2 - 12462957

AN - SCOPUS:0036886538

VL - 25

SP - 525

EP - 550

JO - Journal of Behavioral Medicine

JF - Journal of Behavioral Medicine

SN - 0160-7715

IS - 6

ER -