Oh dear, oh dear! I'm really at the deep end now!!!
You are correct in imputing non-random selection of
areas:
The provinces were purposively selected (they had
willing governors).
Cities were picked based on socio-demographic factors
(one upper class, one middle class and one lower
class).
The villages were picked based on geography (one
coastal, one inland, one mountainous).
The sampling of the control area was the same and
loosely patterned after the intervention area
sampling, i.e. one each of upper, middle and lower
class cities then one each of coastal, inland and
mountainous villages per city.
The women within the villages were the only ones truly
randomly selected from a list of all the women in the
village.
The intervention was a health care delivery system: a
more streamlined and organized system for delivering
iron supplements to the women using spot mapping,
clustering of households, more frequent monitoring
visits, etc. which heavily involved the village health
care worker. A representative health care worker from
each intervention village was trained on the new
health care delivery system and they in turn trained
the other health care workers in their village
(between 15 to 30 village health care workers per
village). No such training was given to the control
areas.
The outcome variables include hemoglobin levels
(continuous), anemia (yes/no), number of iron tablets
taken per month (continuous).
If I understand correctly, the city and village have
random effects.
Thus:
xi:xtmixed y i.group*i.survey || city: || village:
Is this correct or am I totally misunderstanding
xtmixed?
Pura
**********
Pura:
You will probably need stata's multi-level commands,
like xtmixed and
gllamm, not the survey commands.
On rereading your description, I am not sure that you
took a
probability ("random") sample of each province, or
even of the
province's cities.
To advise you better, we would need a detailed
description of the
study and intervention.
Steve
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/