Hello AMOS connaisseur. I have a problem with my model. I'm doing a CFA, using the bifactor model to validate a new scale with 7 factors and a general factor. I tried to run the model but got this error message : The model is probably unidentified. In order to achieve identifiability, it will probably be necessary to impose 7 additional constraint. Here's my model. Do anyone have a suggestion ? Thanks again.
Raphaëlle, a bifactor model includes a general factor on which all items load, and several specific or residual factors with only a subset of the items loading on each specific factor. Being residuals, the specific factors cannot be correlated. Therefore, it is necessary to fix all correlations between the specific factors to zero and it is also necessary to fix all correlations between the general factor and the specific factors to zero.
You may also wish to think about a bifactor(S-1) model, in which one of the specific factors is fixed to zero. This is explained in the following article:
Eid, M., Geiser, C., Koch, T., & Heene, M. (2017). Anomalous results in G-factor models: Explanations and alternatives. Psychological Methods, 22(3), 541–562. https://doi.org/10.1037/met0000083
Abstract:
G-factor models such as the bifactor model and the hierarchical G-factor model are increasingly applied in psychology. Many applications of these models have produced anomalous and unexpected results that are often not in line with the theoretical assumptions on which these applications are based. Examples of such anomalous results are vanishing specific factors and irregular loading patterns. In this article, the authors show that from the perspective of stochastic measurement theory anomalous results have to be expected when G-factor models are applied to a single-level (rather than a 2-level) sampling process. The authors argue that the application of the bifactor model and related models require a 2-level sampling process that is usually not present in empirical studies. We demonstrate how alternative models with a G-factor and specific factors can be derived that are more well-defined for the actual single-level sampling design that underlies most empirical studies. It is shown in detail how 2 alternative models, the bifactor-(S − 1) model and the bifactor-(S·I − 1) model, can be defined. The properties of these models are described and illustrated with an empirical example. Finally, further alternatives for analyzing multidimensional models are discussed.
HTH, Karin
Exactly my thought when I saw this model. It appears that this is a bifactor model (in contrast to a higher-order model) and as such, there would be direct paths from the group factors to the observed indicator variables but the covariances specified among the group factors should not be present. Remove all the covariance paths among your group factors and your model may work. This isn’t a problem of AMOS but any SEM program would likely flag this model as problematic.
I agree. The main problem is that you are estimating all of those paths from the "Total" latent variable to all the manifest variables. Instead, if you are looking to estimate a second order CFA, you should have paths from the "Total" latent variable go only to the first order latent variables. I am attaching a document that may be useful.
- Badges
- Science method