comp.soft-sys.sas - The SAS statistics package.
Hey, guys, I have a program as below: data try; input a b; cards; 1 2 3 4 3 8 ; run; data try2; set try; where a=5; run; data _null_; set try2 end=eof; if end=0 then put "Warning: data set doesn't exist"; run; I hope to have warning in log, but the from the log it says Variable end is uninitialized. Even I change condition to a=3, the warning variable end is unintialized is still there. Can anyone explain? thanks. Mindy
Hi users I used proc print but could not wrap this long record. After that I used proc report and used flow and stuff but still I am having problem wrapping this specific date record. Help me define PI*** /order left width=10 'PatientID^(pids)' flow; Is there any quick way to wrap the text. This all the dates stays on one row. ~7/14/2005~7/14/2005~7/14/2005~7/14/2005~7/14/2005~7/14/2005~7/14/2005~7/14/2005~7/14/2005~7/14/2005~7/14/2005~7/14/2005~7/14/2005~5/10/2001~5/10/2001~5/10/2001~5/10/2001~8/16/2001~8/17/2001~8/16/2001~8/16/2001~7/14/2005
It appears to me that you might have inadvertently changed the scale from percent to proportion. Paul R. Swank, Ph.D. Professor, Developmental Pediatrics Medical School UT Health Science Center at Houston -----Original Message----- From: SAS(r) Discussion [mailto: XXXX@XXXXX.COM ] On Behalf Of Kerri Rivers Sent: Thursday, October 23, 2003 3:58 PM To: XXXX@XXXXX.COM Subject: OT: statistical question regarding standard error hi all, sorry for posting this here - i thought i'd tap the wealth of knowledge on this list to see if anyone can explain this to me. i'm working with 2000 and 2001 supplementary survey data (ACS data from the census bureau). i've calculated estimates and percents for educational attainment by race for 2000 and 2001 for the states. b/c i'm breaking this down by different race groups (and the unweighted observations in some cases are small), some standard errors are quite large. i decided to do a 2-year average to decrease the standard error. the standard error did, in fact, decrease. the SE just decreased so substantially, that i'm wondering why/how. i've checked and rechecked all of my calculations, and everything is correct. can anyone hazard a guess as to why the standard error dropped so dramatically using a 2 year average? i need to be able to explain this to my funder. here's an example of the standard error for percents that i'm talking about. all data are for non-hispanic, black alone ages 18-64 at different levels of educational attainment. the sample size for 2001 data is roughly 3xs a large as the sample size for 2000. yr 2000 SEperc , yr 2001 SEperc, 2yravg SE perc 1.91 , 1.07 , .01 2.02 , 1.26 , .02 2.10 , 1.32 , .02 6.17 , 5.59 , .04 12.45, 8.80 , .10 13.32, 9.44 , .20 4.14 , 2.69 , .03 i just can't fathom how a standard error of 12.45 one year and 8.8 the next year drops to .1 with a 2 year average. that seems HUGE to me. any thoughts would be greatly appreciated. if anyone can help and you need more info, i'm happy to provide more. :) thanks much, ~kerri Kerri Rivers Research Associate Population Reference Bureau 1875 Connecticut Avenue, NW Suite 520 Washington, DC 20009