DWLS degrees of freedom

[katja.si...@gmail.com]

Hi,

Where can I find the formula that is used in lavaan to calculate degrees of freedom for DWLS estimator (ordinal data) and how is this calculations different form the df calculation for the ML estimator method for continuous data?

Kind regards,

Katja

[Terrence Jorgensen]

Where can I find the formula that is used in lavaan to calculate degrees of freedom for DWLS estimator (ordinal data) and how is this calculations different form the df calculation for the ML estimator method for continuous data?

It is the same.  DF = #knowns - #unknowns

• #knowns is the observed sample statistics, which is normally the number of unique elements in each group's covariance matrix, plus the means if meanstructure=TRUE.  For categorical indicators, this is the number of estimated polychoric correlations among the ordered outcomes (and polyserial correlations between ordered and continuous variables plus (co)variances among continuous variables, if you have a mixture), plus thresholds of ordered outcomes (plus means of continuous variables, if you have any).
• #unknowns is the number of estimated parameters

Terrence D. Jorgensen

Assistant Professor, Methods and Statistics

Research Institute for Child Development and Education, the University of Amsterdam

http://www.uva.nl/profile/t.d.jorgensen