| Date: | Tue, 17 Aug 2004 10:15:33 -0400 |
|---|
| Reply-To: | "Chen, Jian" <OZZ6@CDC.GOV> |
|---|
| Sender: | "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU> |
|---|
| From: | "Chen, Jian" <OZZ6@CDC.GOV> |
|---|
| Subject: | Proc tabluate |
|---|
| Content-Type: | text/plain; charset="windows-1255" |
|---|
Is there any option in proc tabulate that would give 95% confidence interval?
For example:
proc tabulate data=indsmk format=8.2 missing;
format eversmk utc YESNOf.;
class eversmk agecat utc/preloadfmt;
table eversmk='Smoking Category:' all='Total',
(agecat= "Age Category" all='Total Ages')
*(n='Number'*f=9.
pctn<utc all> = 'Percent of row total'
pctn<agecat all>='Pcnt of column total per Smoke Cat'
pctn ='Percent of total'),
utc="Cervix Uteri Cancer" all='Total'/
box=_page_
rts=25 printmiss;
format agecat agecatf.;
QUIT;
Output:(sorry for that means ---------)
Smoking Category: Stomach Cancer
yes
Missing No Yes Total
Age
Category
5 - 14 yrs Percent of
row total 1.00 45.00 54.00 100.00
Pcnt of
column
total per
Smoke Cat . 40.00 . 40.00
Percent of
total 0.00 25.00 0.00 25.00
15 - 24 yrsNumber . . . .
Percent of
row total . . . .
Pcnt of
column
total per
Smoke Cat . 0.00 . 0.00
Percent of
total 0.00 0.00 0.00 0.00
25 - 34 yrsNumber . . . .
Percent of
row total . . . .
Pcnt of
column
total per
Smoke Cat . 0.00 . 0.00
Percent of
total 0.00 0.00 0.00 0.00
I hope to get 95%CI, for example:
The first row in the first table 1.00 45.00 54.00 100.00
Would become something like:
1.00 ± 2 45.00 ± 2 54.00 ± 5 100.00 ± 9
| 
