The Impact of Measurement Model Misspecification on Coefficient Omega Estimates of Composite Reliability

被引:10
作者
Bell, Stephanie M. [1 ]
Chalmers, R. Philip [1 ]
Flora, David B. [1 ,2 ]
机构
[1] York Univ, Toronto, ON, Canada
[2] York Univ, Dept Psychol, 4700 Keele St, Toronto, ON M3J 1P3, Canada
关键词
coefficient omega; reliability; measurement model; omega-hierarchical; item factor analysis; factor reliability; BIFACTOR MODELS; ALPHA; SCORES;
D O I
10.1177/00131644231155804
中图分类号
G44 [教育心理学];
学科分类号
0402 ; 040202 ;
摘要
Coefficient omega indices are model-based composite reliability estimates that have become increasingly popular. A coefficient omega index estimates how reliably an observed composite score measures a target construct as represented by a factor in a factor-analysis model; as such, the accuracy of omega estimates is likely to depend on correct model specification. The current paper presents a simulation study to investigate the performance of omega-unidimensional (based on the parameters of a one-factor model) and omega-hierarchical (based on a bifactor model) under correct and incorrect model misspecification for high and low reliability composites and different scale lengths. Our results show that coefficient omega estimates are unbiased when calculated from the parameter estimates of a properly specified model. However, omega-unidimensional produced positively biased estimates when the population model was characterized by unmodeled error correlations or multidimensionality, whereas omega-hierarchical was only slightly biased when the population model was either a one-factor model with correlated errors or a higher-order model. These biases were higher when population reliability was lower and increased with scale length. Researchers should carefully evaluate the feasibility of a one-factor model before estimating and reporting omega-unidimensional.
引用
收藏
页码:5 / 39
页数:35
相关论文
共 47 条
[1]   Alpha, Dimension-Free, and Model-Based Internal Consistency Reliability [J].
Bentler, Peter M. .
PSYCHOMETRIKA, 2009, 74 (01) :137-143
[2]  
Bollen K. A., 1989, Structural equations with latent variables
[3]   On the Complexity of Item Response Theory Models [J].
Bonifay, Wes ;
Cai, Li .
MULTIVARIATE BEHAVIORAL RESEARCH, 2017, 52 (04) :465-484
[4]   Three Concerns With Applying a Bifactor Model as a Structure of Psychopathology [J].
Bonifay, Wes ;
Lane, Sean P. ;
Reise, Steven P. .
CLINICAL PSYCHOLOGICAL SCIENCE, 2017, 5 (01) :184-186
[5]   True scores, latent variables, and constructs: A comment on Schmidt and Hunter [J].
Borsboom, D ;
Mellenbergh, GJ .
INTELLIGENCE, 2002, 30 (06) :505-514
[6]  
Borsboom D., 2005, Measuring the mind: Conceptual issues in contemporary psychometrics
[7]   Writing effective and reliable Monte Carlo simulations with the SimDesign package [J].
Chalmers, R. Philip ;
Adkins, Mark C. .
QUANTITATIVE METHODS FOR PSYCHOLOGY, 2020, 16 (04) :248-280
[8]   Reliability and Omega Hierarchical in Multidimensional Data: A Comparison of Various Estimators [J].
Cho, Eunseong .
PSYCHOLOGICAL METHODS, 2022,
[9]   From alpha to omega: A practical solution to the pervasive problem of internal consistency estimation [J].
Dunn, Thomas J. ;
Baguley, Thom ;
Brunsden, Vivienne .
BRITISH JOURNAL OF PSYCHOLOGY, 2014, 105 (03) :399-412
[10]   A Test Can Have Multiple Reliabilities [J].
Ellis, Jules L. .
PSYCHOMETRIKA, 2021, 86 (04) :869-876