--- Nikolaos Pandis wrote:
> If I only enter child2 and child3 (independent variables)
> in the model how would child1 be used as a reference for
> child2 and child3?
Lets consider the simple case of a linear model:
y = b0 + b1 child2 + b2 child3
Now look at what happens at the three possible
scenarios:
1 kid means child2 = 0 and child3 = 0:
y = b0 + b1*0 + b2*0 = b0
2 kids means child2 = 1 and child3 = 0:
y = b0 + b1*1 + b2*0 = b0 + b1
3+ kisd means child2 = 0 and child3 = 1:
y = b0 + b1*0 + b2*1 = b0 + b2
So, b0 measure the expected value of y when
there is 1 kid, b1 and b2 measures the
difference in y relative to the situation
with only 1 kid.
So leaving child1 out makes it the reference
category.
In the example below you can see that the
same logic also works in a logistic regression.
*------------------- begin example ------------------
sysuse nlsw88, clear
// create a categorical variable from dummy variables
gen marst = cond(never_married, 1, ///
cond(married, 2, 3)) ///
if !missing(married, never_married)
label define marst 1 "never married" ///
2 "married" ///
3 "divorced/widowed"
//set the reference category to be divorced/widowed
char marst[omit] 3
// both -xi- and using the dummy variables lead to
// exactly the same results
xi : logit union i.marst
logit union never_married married
*----------------- end example ---------------------
( For more on how to use examples I sent to statalist see:
http://www.maartenbuis.nl/stata/exampleFAQ.html )
Hope this helps,
Maarten
--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany
http://www.maartenbuis.nl
--------------------------
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
