Dear all,
I am just wondering if I am right to point out that 'ztest' seems to
use rather odd labels in its output. My outcome variable is a binary
(0/1) variable representing low birth weight and the explanatory, ht2,
is also a binary (0/1) variable. The null hypothesis shown in the output
is as follows.
Ho: Pr(lbw = 1) - Pr(y = 1) = diff = 0
The labels used in the brackets seem odd. Should Pr( ..) represent the
probability of being a case (low birthweight) in this case, it would
have made more sense if the labels within the brackets were "ht2=1" and
"ht2=0", respectively. Can anyone shed any light on this oddity please?
Many thanks in advance.
Edmond
. ztest lbw, by(ht2)
Two-sample z test
------------------------------------------------------------------------------
Variable | Obs Prop Std. Err. Std. Dev. [95% Conf.
Interval]
---------+--------------------------------------------------------------------
lbw | 542 .097786 .0127583 .2970251 .0727801
.1227918
y | 99 .2727273 .0447605 .4453618 .1849982
.3604563
---------+--------------------------------------------------------------------
combined | 641 .124805 .0361227 .3304977 .0540058
.1956042
---------+--------------------------------------------------------------------
diff | 641 -.1749413 .0361227 .3304977 -.2457405
-.1041421
------------------------------------------------------------------------------
Ho: Pr(lbw = 1) - Pr(y = 1) = diff = 0
Ha: diff < 0 Ha: diff ~= 0 Ha: diff >
0
z = -4.8430 z = -4.8430 z =
-4.8430
P < z = 0.0000 P > |z| = 0.0000 P > z =
1.0000
*
* 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/