You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently, the documentation for standardizedsolution() does not explain how standard errors are calculated from standardizedsolution() (including whether the sampling probabilities are included in the likelihoods) nor does it note that different SEs may be estimated compared to the ones specified elsewhere as suggested in the mailing group because the SEs are for the standardized parameters and - as far as I can tell - there does not appear to be a way to change the type of SEs estimated from its default. It would be welcome to explicate this in the documentation.
More generally, one of the confusing parts of syntax in lavaan is how changing one thing (e.g. estimator) affects other things (e.g. type). This is fine and largely reflects the imposition of strongly-opinionated default relationships between arguments that reduce issues for many users. However, another example of this that would benefit from explication is how bootstrapping affects inferences. Per this post, SE and CIs are affected by se = "bootstrap" but not P-values.
The text was updated successfully, but these errors were encountered:
Currently, the documentation for standardizedsolution() does not explain how standard errors are calculated from standardizedsolution() (including whether the sampling probabilities are included in the likelihoods) nor does it note that different SEs may be estimated compared to the ones specified elsewhere as suggested in the mailing group because the SEs are for the standardized parameters and - as far as I can tell - there does not appear to be a way to change the type of SEs estimated from its default. It would be welcome to explicate this in the documentation.
More generally, one of the confusing parts of syntax in lavaan is how changing one thing (e.g. estimator) affects other things (e.g. type). This is fine and largely reflects the imposition of strongly-opinionated default relationships between arguments that reduce issues for many users. However, another example of this that would benefit from explication is how bootstrapping affects inferences. Per this post, SE and CIs are affected by se = "bootstrap" but not P-values.
The text was updated successfully, but these errors were encountered: