Which to use, SAS or SUDAAN?
in [SAS]
|
From: PingHe9 on 10 Jan 2007 11:43 Hello, SAS's survey procedures (proc surveyfreq and proc surveylogistic) can pretty much accomplish the same tasks that were used to be done by SUDAAN (proc crosstab and proc rlogist). When comparing the results, you may find they provide the same point estimates, but slightly different standard errors. SUDAAN has been the software used in most published NHANES-related articles including those published in recent years. I would appreciate your opinion as which software to use as I am preparing a menuscript using NHANES data. Thank you! Ping
From: rjf2 on 10 Jan 2007 15:47 > From: PingHe9(a)gmail.com > SAS's survey procedures (proc surveyfreq and proc > surveylogistic) can pretty much accomplish the same tasks that were > used to be done by SUDAAN (proc crosstab and proc rlogist). When > comparing the results, you may find they provide the same point > estimates, but slightly different standard errors. SUDAAN has been the > software used in most published NHANES-related articles including > those published in recent years. I would appreciate your opinion as > which software to use as I am preparing a menuscript using NHANES > data. There are a number of SUGI papers written about Sudaan http://www.lexjansen.com/sugi/ .... hit send too soon: sugi25.238 Analyzing Complex Sample Survey Data: A New Beginning Arlene B. Siller, National Center for Health Statistics, Hyattsville, MD. ABSTRACT ... and an alternative program, SUDAAN. The paper will also include ... http://www2.sas.com/proceedings/sugi25/25/po/25p238.pdf Ronald J. Fehd, IT Spec team lead: Office of Statistical Software Support ... including Sudaan
From: rjf2 on 10 Jan 2007 15:44 > From: PingHe9(a)gmail.com > SAS's survey procedures (proc surveyfreq and proc > surveylogistic) can pretty much accomplish the same tasks > that were used to be done by SUDAAN (proc crosstab and proc > rlogist). When comparing the results, you may find they > provide the same point estimates, but slightly different > standard errors. SUDAAN has been the software used in most > published NHANES-related articles including those published > in recent years. I would appreciate your opinion as which > software to use as I am preparing a menuscript using NHANES data. There are a number of SUGI papers written about Sudaan http://www.lexjansen.com/sugi/ paper31.130: SAS/AF, SAS/IntrNet, and ODS: Automation of State Health Switched from using Sas(r)-callable Sudaan(r) procedures to Proc Surveymeans. Switching from SAS(r) version 8.2 to SAS(r) version 9.1 changed the output from Proc ... http://www2.sas.com/proceedings/sugi31/130-31.pdf - Similar pages paper31.193: Statistics and Data Analysis Wait Wait, Don't Tell Me! You're Using the Wrong Proc! David L. Cassell, Design Pathways, Corvallis, OR. http://www2.sas.com/proceedings/sugi31/193-31.pdf - Similar pages see also: 1997 paper by Donna Brogan of Emory University discussing BRFSS data comparing SAS v6.11 and Sudaan v7. http://www.rti.org/sudaan/pdf_files/brogan.pdf
From: David L Cassell on 11 Jan 2007 02:46 PingHe9(a)GMAIL.COM wrote: > >Hello, > >SAS's survey procedures (proc surveyfreq and proc surveylogistic) can >pretty much accomplish the same tasks that were used to be done by >SUDAAN (proc crosstab and proc rlogist). When comparing the results, >you may find they provide the same point estimates, but slightly >different standard errors. SUDAAN has been the software used in most >published NHANES-related articles including those published in recent >years. I would appreciate your opinion as which software to use as I >am preparing a menuscript using NHANES data. > >Thank you! > >Ping My experience has been that the point estimates are likely to have almost exactly the same standard errors between the two tools. There are people who have investigated these kinds of issues, and you can use references to their papers in your own work. Here are two which apply directly to your concerns: U.S. Health and Nutrition: SAS� Survey Procedures and NHANES http://www2.sas.com/proceedings/sugi31/140-31.pdf The Big Four: Analyzing Complex Sample Survey Data Using SAS, SPSS, STATA, and SUDAAN http://www2.sas.com/proceedings/sugi31/172-31.pdf So, if people kvetch about your using SAS, you can wave these papers at them. Or you can try it both ways and publish your own comparison! Once you have published such a comparison, you can cite it over and over as evidence for your position in future work... HTH, David -- David L. Cassell mathematical statistician Design Pathways 3115 NW Norwood Pl. Corvallis OR 97330 _________________________________________________________________ The MSN Entertainment Guide to Golden Globes is here. Get all the scoop. http://tv.msn.com/tv/globes2007/?icid=nctagline2
From: rjf2 on 11 Jan 2007 12:05 > From: PingHe > Subject: Which to use, SAS or SUDAAN? see also: http://en.wikipedia.org/wiki/Comparison_of_statistical_packages Ron Fehd the google maven CDC Atlanta GA USA RJF2 at cdc dot gov
|
Next
|
Last
Pages: 1 2 Prev: clear the libref after calling a stored macro Next: Safe way to test if a date is valid ? |