Bayesian multilevel mimic modeling for studying measurement invariance in cross-group comparisons

Luk Bruyneel, Baoyue Li, Allison Squires, Sara Spotbeen, Bart Meuleman, Emmanuel Lesaffre, Walter Sermeus

Research output: Contribution to journalArticle

Abstract

Background: Recent methodological advancements should catalyze the evaluation of measurement invariance across groups, which is required for conducting meaningful cross-group comparisons. Objective: The aim of this study was to apply a state-of-the-art statistical method for comparing latent mean scores and evaluating measurement invariance across managers' and frontline workers' ratings of the organization of hospital care. Methods: On the 87 nursing units in a single institution, French-speaking and Dutch-speaking nursing unit managers' and staff nurses' ratings of their work environment were measured using the multidimensional 32-item practice environment scale of the nursing work index (PES-NWI). Measurement invariance and latent mean scores were evaluated in the form of a Bayesian 2-level multiple indicators multiple causes model with covariates at the individual nurse and nursing unit level. Role (manager, staff nurse) and language (French, Dutch) are of primary interest. Results: Language group membership accounted for 7 of 11 PES-NWI items showing measurement noninvariance. Cross-group comparisons also showed that covariates at both within-level and between-level had significant effects on PES-NWI latent mean scores. Most notably, nursing unit managers, when compared with staff nurses, hold more positive views of several PES-NWI dimensions. Conclusions: Using a widely used instrument for measuring nurses' work environment, this study shows that precautions for the potential threat of measurement noninvariance are necessary in all stages of a study that relies on survey data to compare groups, particularly in multilingual settings. A Bayesian multilevel multiple indicators multiple causes approach can accommodate for detecting all possible instances of noninvariance for multiple covariates of interest at the within-level and between-level jointly.

Original languageEnglish (US)
Pages (from-to)e25-e35
JournalMedical Care
Volume55
Issue number4
DOIs
StatePublished - 2017

Fingerprint

Nursing
Nurses
Language

Keywords

  • Bayesian structural equation modeling
  • cross-sectional studies
  • hospital work environment
  • nursing
  • psychometrics
  • statistics and numerical data

ASJC Scopus subject areas

  • Public Health, Environmental and Occupational Health

Cite this

Bayesian multilevel mimic modeling for studying measurement invariance in cross-group comparisons. / Bruyneel, Luk; Li, Baoyue; Squires, Allison; Spotbeen, Sara; Meuleman, Bart; Lesaffre, Emmanuel; Sermeus, Walter.

In: Medical Care, Vol. 55, No. 4, 2017, p. e25-e35.

Research output: Contribution to journalArticle

Bruyneel, L, Li, B, Squires, A, Spotbeen, S, Meuleman, B, Lesaffre, E & Sermeus, W 2017, 'Bayesian multilevel mimic modeling for studying measurement invariance in cross-group comparisons', Medical Care, vol. 55, no. 4, pp. e25-e35. https://doi.org/10.1097/MLR.0000000000000164
Bruyneel, Luk ; Li, Baoyue ; Squires, Allison ; Spotbeen, Sara ; Meuleman, Bart ; Lesaffre, Emmanuel ; Sermeus, Walter. / Bayesian multilevel mimic modeling for studying measurement invariance in cross-group comparisons. In: Medical Care. 2017 ; Vol. 55, No. 4. pp. e25-e35.
@article{0dfc47640d0244e39753c1121113d664,
title = "Bayesian multilevel mimic modeling for studying measurement invariance in cross-group comparisons",
abstract = "Background: Recent methodological advancements should catalyze the evaluation of measurement invariance across groups, which is required for conducting meaningful cross-group comparisons. Objective: The aim of this study was to apply a state-of-the-art statistical method for comparing latent mean scores and evaluating measurement invariance across managers' and frontline workers' ratings of the organization of hospital care. Methods: On the 87 nursing units in a single institution, French-speaking and Dutch-speaking nursing unit managers' and staff nurses' ratings of their work environment were measured using the multidimensional 32-item practice environment scale of the nursing work index (PES-NWI). Measurement invariance and latent mean scores were evaluated in the form of a Bayesian 2-level multiple indicators multiple causes model with covariates at the individual nurse and nursing unit level. Role (manager, staff nurse) and language (French, Dutch) are of primary interest. Results: Language group membership accounted for 7 of 11 PES-NWI items showing measurement noninvariance. Cross-group comparisons also showed that covariates at both within-level and between-level had significant effects on PES-NWI latent mean scores. Most notably, nursing unit managers, when compared with staff nurses, hold more positive views of several PES-NWI dimensions. Conclusions: Using a widely used instrument for measuring nurses' work environment, this study shows that precautions for the potential threat of measurement noninvariance are necessary in all stages of a study that relies on survey data to compare groups, particularly in multilingual settings. A Bayesian multilevel multiple indicators multiple causes approach can accommodate for detecting all possible instances of noninvariance for multiple covariates of interest at the within-level and between-level jointly.",
keywords = "Bayesian structural equation modeling, cross-sectional studies, hospital work environment, nursing, psychometrics, statistics and numerical data",
author = "Luk Bruyneel and Baoyue Li and Allison Squires and Sara Spotbeen and Bart Meuleman and Emmanuel Lesaffre and Walter Sermeus",
year = "2017",
doi = "10.1097/MLR.0000000000000164",
language = "English (US)",
volume = "55",
pages = "e25--e35",
journal = "Medical Care",
issn = "0025-7079",
publisher = "Lippincott Williams and Wilkins",
number = "4",

}

TY - JOUR

T1 - Bayesian multilevel mimic modeling for studying measurement invariance in cross-group comparisons

AU - Bruyneel, Luk

AU - Li, Baoyue

AU - Squires, Allison

AU - Spotbeen, Sara

AU - Meuleman, Bart

AU - Lesaffre, Emmanuel

AU - Sermeus, Walter

PY - 2017

Y1 - 2017

N2 - Background: Recent methodological advancements should catalyze the evaluation of measurement invariance across groups, which is required for conducting meaningful cross-group comparisons. Objective: The aim of this study was to apply a state-of-the-art statistical method for comparing latent mean scores and evaluating measurement invariance across managers' and frontline workers' ratings of the organization of hospital care. Methods: On the 87 nursing units in a single institution, French-speaking and Dutch-speaking nursing unit managers' and staff nurses' ratings of their work environment were measured using the multidimensional 32-item practice environment scale of the nursing work index (PES-NWI). Measurement invariance and latent mean scores were evaluated in the form of a Bayesian 2-level multiple indicators multiple causes model with covariates at the individual nurse and nursing unit level. Role (manager, staff nurse) and language (French, Dutch) are of primary interest. Results: Language group membership accounted for 7 of 11 PES-NWI items showing measurement noninvariance. Cross-group comparisons also showed that covariates at both within-level and between-level had significant effects on PES-NWI latent mean scores. Most notably, nursing unit managers, when compared with staff nurses, hold more positive views of several PES-NWI dimensions. Conclusions: Using a widely used instrument for measuring nurses' work environment, this study shows that precautions for the potential threat of measurement noninvariance are necessary in all stages of a study that relies on survey data to compare groups, particularly in multilingual settings. A Bayesian multilevel multiple indicators multiple causes approach can accommodate for detecting all possible instances of noninvariance for multiple covariates of interest at the within-level and between-level jointly.

AB - Background: Recent methodological advancements should catalyze the evaluation of measurement invariance across groups, which is required for conducting meaningful cross-group comparisons. Objective: The aim of this study was to apply a state-of-the-art statistical method for comparing latent mean scores and evaluating measurement invariance across managers' and frontline workers' ratings of the organization of hospital care. Methods: On the 87 nursing units in a single institution, French-speaking and Dutch-speaking nursing unit managers' and staff nurses' ratings of their work environment were measured using the multidimensional 32-item practice environment scale of the nursing work index (PES-NWI). Measurement invariance and latent mean scores were evaluated in the form of a Bayesian 2-level multiple indicators multiple causes model with covariates at the individual nurse and nursing unit level. Role (manager, staff nurse) and language (French, Dutch) are of primary interest. Results: Language group membership accounted for 7 of 11 PES-NWI items showing measurement noninvariance. Cross-group comparisons also showed that covariates at both within-level and between-level had significant effects on PES-NWI latent mean scores. Most notably, nursing unit managers, when compared with staff nurses, hold more positive views of several PES-NWI dimensions. Conclusions: Using a widely used instrument for measuring nurses' work environment, this study shows that precautions for the potential threat of measurement noninvariance are necessary in all stages of a study that relies on survey data to compare groups, particularly in multilingual settings. A Bayesian multilevel multiple indicators multiple causes approach can accommodate for detecting all possible instances of noninvariance for multiple covariates of interest at the within-level and between-level jointly.

KW - Bayesian structural equation modeling

KW - cross-sectional studies

KW - hospital work environment

KW - nursing

KW - psychometrics

KW - statistics and numerical data

UR - http://www.scopus.com/inward/record.url?scp=85015173954&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85015173954&partnerID=8YFLogxK

U2 - 10.1097/MLR.0000000000000164

DO - 10.1097/MLR.0000000000000164

M3 - Article

VL - 55

SP - e25-e35

JO - Medical Care

JF - Medical Care

SN - 0025-7079

IS - 4

ER -