We investigate the statistical properties of PLS Path Modeling (PLS-PM) estimators in the case of misspecification of a measurement model. We also discuss possible criteria for choosing between reflective and formative measurement models. This is particularly useful for situations in which there is little theoretical basis to decide a priori whether a construct is formative or reflective. The study is based on a Monte Carlo simulation analysis and some real applications.
Minotti, S., Ciampi, A., Lamberti, G., Aluja Banet, T. (2014). Statistical properties of coefficient estimates in PLS-PM: reflective versus formative measurement model.. Intervento presentato a: PLS’14 (8th International Conference on PLS and Related Methods), Paris.
Statistical properties of coefficient estimates in PLS-PM: reflective versus formative measurement model.
MINOTTI, SIMONA CATERINAPrimo
;
2014
Abstract
We investigate the statistical properties of PLS Path Modeling (PLS-PM) estimators in the case of misspecification of a measurement model. We also discuss possible criteria for choosing between reflective and formative measurement models. This is particularly useful for situations in which there is little theoretical basis to decide a priori whether a construct is formative or reflective. The study is based on a Monte Carlo simulation analysis and some real applications.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
