|
|
Applying Survey Weights (Millenium Co... |
|
Message/Author |
|
|
Hello, I am trying to apply the survey weights provided with the Millenium Cohort Study in Mplus. The millenium cohort study sample is complex stratified and clustered,with oversampling of ethnic minorities, 'disadvanted' children and children from the smaller countries of the UK. As the information is provided with the survey the weights are easy to apply in SPSS (complex samples add on) and in stata. In stata the weighting command is svyset sptn00 [pweight=covwt2], strata(pttype2) fpc(Nh2) In Mplus I have tried to use this weighting command: weight is covwt2; stratification is pttype2; cluster is sptn00; analysis: type = complex; but with this I am only able to see the unweighted number of observations. I am assuming it's because i haven't taken into account the finite population correction. How do i do this in Mplus? Any advice would be appreciated. Thankyou, |
|
|
The finite population correction is specified like this FINITE IS fpc; The analysis commands given above specify a weighted analysis and should produce the same parameter estimates as svyset. Mplus doesn't report population total estimates. It produces average values and model parameter estimates and standard errors. |
|
Back to top |
|
|