SEM convergence not achieved

[Adam Mitchell]

I am running a SEM with four indicator variables and one latent variable. (For context in this same data set I have ran other successful SEM using different indicator variables including up to six indicators but that have the same structure meaning all indicator variables are dichotomous).

However when I run the specific model with these four specific variables I receive this error message.

Note: The LR test of model vs. saturated is not reported because the fitted
model is not full rank.
convergence not achieved

Could anybody suggest a possible reason for this and a solution?

Thanks in advance.

[Erik Ruzek]

Adam,

It is hard to tell what might be going on. Is this really a CFA model with just 4 indicators and a latent variable explaining them? No structural parameters in which you are using the latent variable to predict other variables? Can you please post the model code and results?

Tell us more about the indicators. Are the Likert scaled? What proportion of responses are in each category?