[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: NHIS data, SVYLOGIT, and Subpopulations
I have a perplexing dilema that someone on this list may have
encountered before. I am using National Health Interview Survey (NHIS)
data for a secondary data analysis for my dissertation. I need to run
several logisitic regressions. The data that I need is actually a
subsample of the total data set (those 65 years of age and over and with
at least on functional limitation).
I have the entire data set in Stata (all 97,000 or so respondents - all
ages and functional limitations). I tried to run the first logistic
regression (SVYLOGIT) after doing SVYSET. I put in an IF statement so
that the analysis would be done on only those 65 yrs of age and over and
with at least 1 functional limitation. I get "stratum with only one PSU
r(460)". I loaded the entire data set from the go get because I was
getting this error with the subpopulation.
Any suggestions as to how I can run SVYLOGIT on only those people 65 +
and with atleast on functional limitation and not get thos error?
* For searches and help try: