.
Hello,
I am trying to create a table of descriptive statistics using imputed
data, however I want to calculate the mean math score (my dependent
variable) by poverty level (an ordinal independent variable which is
also imputed) across all of my imputed data sets collectively. I've
tried using the following code:
mi estimate: mean mathf, by pov_stat
Stata doesn't like this and gives me an error message. Since the
poverty status variable is imputed, an individual coded 1 in the first
imputed data set might be coded 2 in the second. So then I tried
this:
mi xeq: by pov_stat, sort: sum mathf
This gives me the mean math scores by poverty status independently for
each imputed data set (I have one original data set and five imputed
copies and thus 6 sets of means).
I have two questions:
First, Is there a stata command that can give me means and standard
deviations by poverty status across all of my imputed data sets
simultatneously?
Second, is it incorrect to calculate the average of the mean math
scores across all my imputed data sets (exluding the incomplete data
set _mi_m=0)? how about the standard deviations?
Thank you,
Caitlin
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/
