Distinguishing state variability from trait change in longitudinal data: The role of measurement (non)invariance in latent state-trait analyses

Research output: Journal contributionsJournal articlesResearchpeer-review

Standard

Distinguishing state variability from trait change in longitudinal data: The role of measurement (non)invariance in latent state-trait analyses. / Geiser, Christian; Keller, Brian T.; Lockhart, Ginger et al.
In: Behavior Research Methods, Vol. 47, No. 1, 03.2015, p. 172-203.

Research output: Journal contributionsJournal articlesResearchpeer-review

Harvard

APA

Vancouver

Geiser C, Keller BT, Lockhart G, Eid M, Cole DA, Koch T. Distinguishing state variability from trait change in longitudinal data: The role of measurement (non)invariance in latent state-trait analyses. Behavior Research Methods. 2015 Mar;47(1):172-203. Epub 2014 Mar 21. doi: 10.3758/s13428-014-0457-z

Bibtex

@article{c86bc1f8df26452f9ed2fbf8b2a849fb,
title = "Distinguishing state variability from trait change in longitudinal data: The role of measurement (non)invariance in latent state-trait analyses",
abstract = "Researchers analyzing longitudinal data often want to find out whether the process they study is characterized by (1) short-term state variability, (2) long-term trait change, or (3) a combination of state variability and trait change. Classical latent state-trait (LST) models are designed to measure reversible state variability around a fixed set-point or trait, whereas latent growth curve (LGC) models focus on long-lasting and often irreversible trait changes. In the present article, we contrast LST and LGC models from the perspective of measurement invariance testing. We show that establishing a pure state-variability process requires (1) the inclusion of a mean structure and (2) establishing strong factorial invariance in LST analyses. Analytical derivations and simulations demonstrate that LST models with noninvariant parameters can mask the fact that a trait-change or hybrid process has generated the data. Furthermore, the inappropriate application of LST models to trait change or hybrid data can lead to bias in the estimates of consistency and occasion specificity, which are typically of key interest in LST analyses. Four tips for the proper application of LST models are provided.",
keywords = "Sociology, state variability versus trait change, Latent stat-trait analysis, measurement invariance, Latent growth curve models, Model misspecification",
author = "Christian Geiser and Keller, {Brian T.} and Ginger Lockhart and Michael Eid and Cole, {David A.} and Tobias Koch",
note = "Publisher Copyright: {\textcopyright} 2014, Psychonomic Society, Inc.",
year = "2015",
month = mar,
doi = "10.3758/s13428-014-0457-z",
language = "English",
volume = "47",
pages = "172--203",
journal = "Behavior Research Methods",
issn = "1554-351X",
publisher = "Springer New York LLC",
number = "1",

}

RIS

TY - JOUR

T1 - Distinguishing state variability from trait change in longitudinal data

T2 - The role of measurement (non)invariance in latent state-trait analyses

AU - Geiser, Christian

AU - Keller, Brian T.

AU - Lockhart, Ginger

AU - Eid, Michael

AU - Cole, David A.

AU - Koch, Tobias

N1 - Publisher Copyright: © 2014, Psychonomic Society, Inc.

PY - 2015/3

Y1 - 2015/3

N2 - Researchers analyzing longitudinal data often want to find out whether the process they study is characterized by (1) short-term state variability, (2) long-term trait change, or (3) a combination of state variability and trait change. Classical latent state-trait (LST) models are designed to measure reversible state variability around a fixed set-point or trait, whereas latent growth curve (LGC) models focus on long-lasting and often irreversible trait changes. In the present article, we contrast LST and LGC models from the perspective of measurement invariance testing. We show that establishing a pure state-variability process requires (1) the inclusion of a mean structure and (2) establishing strong factorial invariance in LST analyses. Analytical derivations and simulations demonstrate that LST models with noninvariant parameters can mask the fact that a trait-change or hybrid process has generated the data. Furthermore, the inappropriate application of LST models to trait change or hybrid data can lead to bias in the estimates of consistency and occasion specificity, which are typically of key interest in LST analyses. Four tips for the proper application of LST models are provided.

AB - Researchers analyzing longitudinal data often want to find out whether the process they study is characterized by (1) short-term state variability, (2) long-term trait change, or (3) a combination of state variability and trait change. Classical latent state-trait (LST) models are designed to measure reversible state variability around a fixed set-point or trait, whereas latent growth curve (LGC) models focus on long-lasting and often irreversible trait changes. In the present article, we contrast LST and LGC models from the perspective of measurement invariance testing. We show that establishing a pure state-variability process requires (1) the inclusion of a mean structure and (2) establishing strong factorial invariance in LST analyses. Analytical derivations and simulations demonstrate that LST models with noninvariant parameters can mask the fact that a trait-change or hybrid process has generated the data. Furthermore, the inappropriate application of LST models to trait change or hybrid data can lead to bias in the estimates of consistency and occasion specificity, which are typically of key interest in LST analyses. Four tips for the proper application of LST models are provided.

KW - Sociology

KW - state variability versus trait change

KW - Latent stat-trait analysis

KW - measurement invariance

KW - Latent growth curve models

KW - Model misspecification

UR - http://www.scopus.com/inward/record.url?scp=84896419003&partnerID=8YFLogxK

U2 - 10.3758/s13428-014-0457-z

DO - 10.3758/s13428-014-0457-z

M3 - Journal articles

C2 - 24652650

AN - SCOPUS:84896419003

VL - 47

SP - 172

EP - 203

JO - Behavior Research Methods

JF - Behavior Research Methods

SN - 1554-351X

IS - 1

ER -