Problems With Item Parceling for Confirmatory Factor Analytic Tests of Measurement Invariance

By Adam W. Meade, Christina M. Kroustalis. 2006

188 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Management Science and Operations Research, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/bc8b7e2800cdaeb9af3ab4390547a19d7dab49ad

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

In Management Science and Operations Research

In Statistics and Probability

Citations in the corpus

When Are Multidimensional Data Unidimensional Enough for Structural Equation Modeling? An Evaluation of the DETECT Multidimensionality Index (2015, background): “…the common practice of “hiding” multidimensionality through the formation of parcels has been subject to much criticism of late (Bandalos, 2002; Meade & Kroustalis, 2006; Sterba & MacCallum, 2010). of the departure from unidimensionality and ultimately the bias in parameter estimates.”
When Family Supportive Supervisors Meet Employees’ Need for Caring: Implications for Work–Family Enrichment and Thriving (2018, methodology): “To confirm the measurement invariance of the need for caring, we followed Cheung and Rensvold (2000) by calculating the delta RMSEA, given that this measure is not affected by model complexity (Meade & Kroustalis, 2006).”
Why the Items versus Parcels Controversy Needn’t Be One (2013, background): “…can lead to biased estimates of other model parameters (e.g., structural parameters; Bandalos & Finney, 2001; Hall et al., 1999; Kim & Hagtvet, 2003; Landis, Beal, & Tesluk, 2000; Little et al., 2002; Matsunaga, 2008; Meade & Kroustalis, 2006; Rogers & Schmitt, 2004; Sass & Smith, 2006).”
Advancing the Field of Family Business Research: Further Testing the Measurement Properties of the F-PEC (2010, background): “Although several scholars have demonstrated that parcels reduce lengthy scales and increase model fit (e.g., Bandalos & Finney, 2001; Klein et al., 2005), Meade and Kroustalis (2006) argued that parceling masked variance that existed at the item level.”
Spirituality in the Workplace: Theory and Measurement (2011, methodology): “Cronbach’s alphas for Sample 3 are in parentheses. a measurement model, whereas the latter is more appropriate for investigating structural relations among latent variables (Meade & Kroustalis, 2006).”
Testing for Factorial Invariance in the Context of Construct Validation (2010, methodology): “In addition, Meade and Kroustalis (2006) suggested that itemlevel testing for measurement invariance be conducted, even when the scale-level tests indicate measurement invariance, because a lack of measurement invariance for a small number of items can be obscured by grouping these items with items…”
How extended work availability affects well-being: The mediating roles of psychological detachment and work-family-conflict (2017, methodology): “Following the recommendation of Meade and Kroustalis (2006), we conducted the invariance test with the item-based measurement models.”
Multidimensionality and Structural Coefficient Bias in Structural Equation Modeling (2013, background): “The troubling practice of burying multidimensionality in the item responses in parcels, however, finally is attracting attention (e.g., Bandalos, 2002; Meade & Kroustalis, 2006; Sterba & MacCallum, 2012).”
Ethical Leadership and Team Ethical Voice and Citizenship Behavior in the Military: The Roles of Team Moral Efficacy and Ethical Climate (2020, methodology): “Thus, following recommendations (Bandalos & Finney, 2001; Little et al., 2002; Meade & Kroustalis, 2006; Sass & Smith, 2006), we used parceling (with random assignment of items to parcels) only for those constructs assessed through unidimensional scales.”
Power and Precision in Confirmatory Factor Analytic Tests of Measurement Invariance (2007, background): “Subsequent to tests of factor loadings' equality, several additional model parameters can also be tested for MI; however, there is firm consensus that tests of factor loadings are the most important for establishing that MI conditions exist ( Meade & Kroustalis, 2006 ), thus our focus on these tests”
A Direct Comparison Approach for Testing Measurement Invariance (2012, methodology): “However, a major problem associated with scale-level tests is that if only a few items in a multi-item construct have noninvariance, the scale-level tests may not have enough power to identify the existence of the noninvariance (Meade & Kroustalis, 2006).”