Test whether a path in the latent variable model correspond to a null effect.
Similar to lava::effects but with small sample correction.
So far it only work for paths composed of two edges.
effects2(object, link, ...)# S3 method for lvmfit
effects2(object, link, df = TRUE, bias.correct = TRUE, ...)
# S3 method for lvmfit2
effects2(object, link, ...)
an object that inherits from lvmfit.
[character vector] The path for which the effect should be assessed (e.g. "A~B"),
i.e. the effect of the right variable (B) on the left variable (A).
[internal] only used by the generic method.
[logical] should the degree of freedoms of the Wald statistic be computed using the Satterthwaite correction?
Otherwise the degree of freedoms are set to Inf, i.e. a normal distribution is used instead of a Student's t distribution when computing the p-values.
[logical] should the standard errors of the coefficients be corrected for small sample bias? Argument passed to sCorrect.