Causal Indicator Models: Identification, Estimation, and Testing

By Kenneth A. Bollen, W. R. Davis. 2009

184 citations recorded by the source corpus.

Subfields: Sociology and Political Science, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/b3c26aea0001547d4d0306e62adbabf711afe973

How researchers see this paper

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

In Sociology and Political Science

In Statistics and Probability

Citations in the corpus

Incorporating Formative Measures into Covariance-Based Structural Equation Models (2011, background): “Unfortunately, this is not the case because the error variances of η 1 (i.e., ψ 11 ) and η 2 (i.e., ψ 22 ) cannot be separately identified when only a single path emanates from the formatively measured latent variable to an outcome variable (Bollen and Davis 2009; MacCallum and Browne 1993).”
Conspiracy theories as barriers to controlling the spread of COVID-19 in the U.S. (2020, methodology): “Standard goodness of fit measures were used to assess model fit (Bollen and Davis, 2009).”
12 Structural Equation Modeling in Management Research: A Guide for Improved Analysis (2009, methodology): “…problems will occur with formative measurement models such as the one shown in Figure 12.2b, unless the latent variable directly or indirectly causes at least two indicators in addition to the ones used as antecedents of the latent variable (Bollen & Davis, 1994; MacCallum & Browne, 1993).”
Advancing formative measurement models (2008, methodology): “…for identifying the disturbance term ζ at the construct level, the formative latent variable needs to emit at least two paths to other (reflective) constructs or indicators (MacCallum and Browne, 1993); literature also refers to this condition as 2+ emitted paths rule (Bollen and Davis, 1994).”
Partial least squares path modeling: Time for some serious second thoughts (2016, methodology): “The use of PLS is often justified by stating that formative models have certain requirements to be identified (Bollen and Davis, 2009), and by claiming that PLS would either not require identification or could somehow solve the issue of identification (e.”
Emergent Ghosts of the Emotion Machine (2010, methodology): “Their use of SES as an example refers primarily to simple linear combinations of causes, whereas my use of it holds out at least the possibility of irreducibility.”
Using Partial Least Squares in Operations Management Research: A Practical Guideline and Summary of Past Research (2012, background): “The additional eflective indicators and the set of formative items together allow esearchers to estimate a multiple indicators and multiple causes MIMIC) model (Bollen and Davis, 2009; Diamantopoulos and inklhofer, 2001) to evaluate the external validity of formative onstructs.”
An assessment of the use of partial least squares structural equation modeling in marketing research (2011, methodology): “While CB-SEM is applicable for formative outer model specifications only under certain conditions (e.g., Bollen and Davies 2009; Diamantopoulos and Riefler 2011), PLS-SEM can almost unrestrictedly handle both reflective and formative measures (e.g., Chin 1998).”
Current Methodological Considerations in Exploratory and Confirmatory Factor Analysis (2011, methodology): “…statistic estimation, robust LS as ULSMV, and robust ML as MLR.4 Two additional FA approaches that are important to mention because they are becoming more common in FA are formative measurement (e.g., Bollen & Davis, 2009) and Bayesian estimation (e.g., Muthén, 2010; Muthén & Asparouhov, 2010).”
Collaborative Leadership Effects on School Improvement: Integrating Unidirectional- and Reciprocal-Effects Models (2010, background): “…of nonredundant elements in the observed covariance matrix, every latent variable with unrestricted variance must emit at least two directed paths to observed indicators or other latent variables when the latter variables have unrestricted error variances), as summarized by Bollen and Davis (2009).”
The Fallacy of Formative Measurement (2011, methodology): “To achieve identification, the model must be supplemented by at least two reflective measures that are caused directly or indirectly by the latent variable (Bollen & Davis, 2009; MacCallum & Browne, 1993).”
Two Rules of Identification for Structural Equation Models (2009, methodology): “Bollen and Davis (2009) discussed methods for estimating and testing some parameters in causal indicator models (i.e., those with a model structure as in Equation 4) even when the model is underidentified.”