Hi Alan -
I doubt things are so simple. Assume, I had the table of rural counties,
so consider my information, too.
Alan Acock schrieb:
>
> URBAN COUNTIES
1990 2000 diff
> % no kids working 80% 82% 2%
> % with kids work 42% 62% 20%
-----------------------
18
Now, my information for rural counties looks like:
rural counties
% no kids working 10% 20% 10%
% with kids work 5% 25% 20%
-----------------------
10
Would you still assume:
> Thus, the difference of difference of difference is 8% meaning that in urban
> counties, after welfare reform mothers increased laborforce participation by
> this much more than happened in rural counties.
?
In my view these data indicate with strong evidence, that the welfare program
led to much more change in the rural counties.... especially being supportive
for rural women with kids being able to go out and work, and as well allowing
rural women in general to approximate to the level of being-in-work of the
urban women.
Using percentages as coefficients, and then again derivatives (and derivatives...)
of these is always an adventure, I'd say...
Gottfried Helms
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
. http://jse.stat.ncsu.edu/ .
=================================================================
