The course will teach students to apply basic analytical epidemiologic methods using SAS statistical software. Heavy emphasis will be placed on basic data manipulation and operationalization skills in preparation for applied epidemiologic analysis that builds on concepts introduced in epidemiology and biostatistics core courses. Each session is a combination of didactic lecture and hands-on practice. Students will conduct epidemiological analyses on actual data sets, and learn the importance of data preparation and cleaning, descriptive analyses, as well as how to conduct basic statistical analyses in SAS.

The primary objective of this course is to provide students with the tools to use the SAS statistical software package in the practical conduct of basic epidemiological research analyses. To reach that objective, students will gain facility in the use of SAS to (1) understand the SAS working environment; (2) generate SAS data sets from nonSAS data files; (3) read, clean, manipulate and operationalize data; and (4) conduct basic statistical analyses.

By the end of the course, students will be able to:

• Read raw data into SAS.
• Manipulate and clean data sets in SAS through printing, sorting, merging and the use of conditional expressions.
• Apply simple statistical and graphical procedures for the descriptive analysis of normally distributed data.
• Conduct TTESTS and ANOVA analyses in SAS and interpret SAS output.
• Conduct basic linear regression analysis in SAS.
• Interpret SAS output for these analyses.

If time allows, basic logistic regression analysis and multivariable regression techniques will be introduced.

PREREQUISITES

No experience with SAS is assumed, but students should be familiar with computing in the Windows environment. A basic understanding of statistical and epidemiological principles is strongly encouraged but brief introductions to key concepts will be offered as necessary although focus will naturally be on the application and interpretation of statistical and epidemiological concepts in the context of SAS.

TECHNICAL REQUIREMENTS

Students will need access to a computer with high-speed internet access and SAS software version 9.0 or higher installed. At minimum, Base SAS and SAS/STAT should be installed. Please read “Obtaining and installing SAS” below for helpful tips on acquiring and installing SAS software.

ACCESSING SAS

It is strongly recommended that registrants consult with the IT department of their respective institutions to inquire about free or discounted versions of SAS that might be available through an institutional affiliation.

SAS provides a free access to a produce called SAS OnDemand for Academics which provides a SAS interface via 'SAS Studio'. The look and feel is different than PC-SAS but the command line coding is identical. For those that can't access a licensed version of PC-SAS, SAS OnDemand is a good alternative. Please note that all lectures are given using PC-SAS. To sign up for SAS OnDemand for Academics go here: https://www.sas.com/en_us/software/on-demand-for-academics.html

For a brief overview on using the SAS studio interface (accessed via SAS OnDemand for Academics), the following video is very helpful: https://video.sas.com/detail/videos/sas-analytics-u/video/4573016757001/getting-started-with-sas-studio?autoStart=true

There are no required readings but the following textbook is highly recommended:

'The Little SAS Book: A Primer' by Lora D. Delwiche and Susan J. Slaughter, 3rd Edition, Copyright 2003, SAS Institute. (ISBN 1-59047-333-7). As necessary, readings from additional sources will be recommended.

“SAS for Epidemiologists. Applications and Methods”. By Charles DiMaggio. 1st Edition. Copyright 2012. Springer. ISBN: 978-1-4614-4853-2.

Additional material for the course is drawn from the following sources:

• Applied Statistics and the SAS Programming Language, 4th addition by Ronald P. Cody & Jeffrey K. Smith, Copyright 1997, Prentice-Hall Inc.
• Statistics I: Introduction to ANOVA, Regression and Logistic Regression (developed by Melinda Thielbar, Mike Patetta and Paul Marovich), Copyright 2005, SAS Institute, Cary NC
• Categorical Data Analysis Using Logistic Regression (developed by Mike Patetta), Copyright 2005, SAS Institute, Cary NC.
• Psychiatric Epidemiology, by Ezra Susser, Sharon Schwartz, Alfredo Morabia and Evelyn J. Bromet. Copyright 2006, Oxford University Press
• Applied Regression Analysis and Multivariable Methods by David Kleinbaum, Lawrence Kupper, Keith Muller and Azhar Nizam, Copyright 1998, Duxbury Press

Ryan Demmer, PhD, MPH

Dr. Demmer is an Associate Professor of Epidemiology with a general research interest in elucidating causes and correlates of cardiometabolic diseases, including diabetes, atherosclerotic vascular disease and congestive heart failure. He is particularly focused on understanding the emerging role of the human microbiome in the initiation and progression of cardiometabolic diseases and the potential role of inflammatory phenotype as a biological mediator connecting microbes to preclinical and clinical disease entities. He is involved in several collaborative projects on this general theme and is currently PI of an NIH funded cohort study: The Oral Infections Glucose Intolerance and Insulin Resistance Study (ORIGINS). The scientific aims of ORIGINS are to study the interplay between the subgingival microbiome and early risk for the development of type 2 diabetes. Dr. Demmer has taught applied analysis with SAS for over 10 years in episummer@columbia. For more information on Dr. Demmer, please visit the following site: https://directory.sph.umn.edu/bio/sph-a-z/ryan-demmer"