Anonymous posted on Monday, September 30, 2002 - 1:00 pm
Hello. I am writing up the results on some latent growth modeling analyses with missing data. I did "Type=meanstructure missing". I would like to report the means, SDs, and the correlation of all the variables in the paper. Is it more informative to report the estimated sample statistics in the print out? Or should I report the sample statistics calculated from the data before running the estimation. Thank you very much.
It seems to me the relevant sample statistics would be those estimated under TYPE=MISSING because these are the sample statistics involved in estimating the model. But others may have differing opinions.