Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Graph higher than individual survey data

From   ann montgomery <[email protected]>
To   [email protected]
Subject   st: Graph higher than individual survey data
Date   Tue, 17 Jul 2012 21:22:53 -0400


I have individual level data, and I want to graph proportionate
estimates at a regional level.

I have four regions (region==1/4),
and four binary variables (var1, var2, var3, var3==0/1) describing
four types of health service used by each individual in the dataset,
and pweights.

I want to plot proportion of each health service used by individuals
in the region (prop of individuals from that region who are "yes",
with 95%CI bars), preferable on a single graph.

I've played with collapse but can't figure out how to tell Stata to
attach the upper and lower 95% CI to the new regional proportion.

Thanks in advance,

Ann Montgomery
Graduate student, Epidemiology
University of Toronto
*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index