Message/Author 

Marisa posted on Wednesday, June 29, 2005  8:20 am



Dear Linda, two questions about the use of weighting variables. Why isn´t it possible to use weights in ML estimation? Why do weights under frequency option need to be integer? What does the settig "sampling" realy do to my data, if it does not touch the number of cases. Regards, Marisa. 


It is possible to use weights in ML estimation. I'm not sure why you think it isn't. Ususally frequency weights are used to represent more than one observation and usually these are not fractions. It tells Mplus that the weight is a sampling weight not a frequency weight. A sampling weight is used when data have been collected with unequal selection probabilities. 


I wonder if I may have a similar issue to the one implied above; I'm using weights for a nationally representative sample which are both frequency and sampling weights combined in one number (i.e. they are all greater than 1 and most have fractions). I thought that perhaps I could get this to work in MPlus by using the weight as a sampling weight and specifying the number of observations (NOBSERVATIONS) as equal to the weighted population (my sample has about 2000 cases that represent about 200,000 individuals), but that does not seem to work. Are there any other solutions within Mplus? Thank you, Inna 


I can't think of a way this can be done in Mplus. 

Nara Jang posted on Sunday, March 30, 2014  3:16 pm



Dear Dr. Linda, My surveyed data include high proportion of female and low proportion of male. Also the range of age is skewed. Would you tell me if I need to weight the variable, based on gender and age? Thank you very much for your expert advice in advance! 


A data set that is not a random sample should have a weight variable that can be used in the analysis. 

Nara Jang posted on Monday, March 31, 2014  3:16 pm



Dear Dr. Muthen, Thank you very much!! Best regards, Nara Jang 

Back to top 