Date:         Tue, 15 Oct 2002 20:14:09 -0400
Reply-To:     Eduardo Faerstein <eduardof@umich.edu>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Eduardo Faerstein <eduardof@umich.edu>
Subject:      age-adjusted means and proportions?
Content-Type: text/plain; charset="iso-8859-1"

Dear list members,

I'm involved in analyses of cross-sectional data from a Brazilian
population.

My dependent variables are body mass index (BMI=kg/m2, a continuous
variable) and obesity (presence of BMI > 30 kg/m2, a dichotomous variable).

My independent variables of main interest are educational level (4 levels)
and household per capita income (quartiles).

To start, I would like...

1. to present  a) BMI means, and b) Prevalence (%) of obesity, adjusted for
AGE (poorer and less educated participants are much older, and BMI and
obesity increases with age)

2. to test for statistical significance of possible linear trends in these
age-adjusted means and proportions across levels of education and income.

Does the UNIVARIATE GLM procedure allows me to have these age-adjusted means
and proportions as outputs, in addition to the test for linear trends? How
do I do that?

Thanks a lot in advance for any hint.

Eduardo Faerstein
Institute of Social Medicine, Rio