scaling
If TRUE
, the scores are scaled to reflect the specific
objective function used by lavaan. If FALSE
(the default), the
objective function is the loglikelihood function assuming multivariate
normality.
ignore.constraints
Logical. If TRUE
, the scores do not reflect
the (equality or inequality) constraints. If FALSE
, the scores are
computed by taking the unconstrained scores, and adding the term
t(R) lambda
, where lambda
are the (case-wise) Lagrange
Multipliers, and R
is
the Jacobian of the constraint function. Only in the latter case will
the sum of the columns be (almost) equal to zero.
remove.duplicated
If TRUE
, and all the equality constraints have
a simple form (eg. a == b), the unconstrained scores are post-multiplied with a
transformation matrix in order to remove the duplicated parameters.