 |
 |
Office of Surveillance, Epidemiology, and Laboratory Services
Behavioral Risk Factor Surveillance System
BRFSS Home | Contact Us
|
|
||||||||
|
|
|
|
|
|
|
||
|
Office of Surveillance, Epidemiology, and Laboratory Services Behavioral Risk Factor Surveillance System BRFSS Home | Contact Us |
|
|
||||||
|
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|
|
|
|
||
Note: You must register for the conference to enroll in the training sessions. Two, 4-hour classes may be taken at a discounted rate of $250.
| Saturday, March 18, 2006 | |
|---|---|
| 8 AM Noon | Comparison of SAS and SUDAAN for BRFSS Descriptive Analyses. |
| 8 AM Noon | Part I: Bayesian Perspectives for Epidemiologic Research: Rationale and Basic Methods |
| 8 AM 5 PM | Logistic Regression Modeling in Epidemiologic Research |
| 1 PM 5 PM | Comparison of SAS and SUDAAN for BRFSS Standardized Rates and Modeling Analyses |
| 1 PM 5 PM | Part II: Bayesian Perspectives for Epidemiologic Research: Regression and Bias Modeling |
| Sunday, March 19, 2006 | |
| 8 AM Noon | Using SAS for Descriptive Analyses of BRFSS Data |
| 8 AM Noon | Multi-level Modeling (MLM) Techniques for Use with Complex Survey Data |
| 8 AM Noon | BRFSS Coordinator Training in Physical Activity & Fruits and Vegetables Survey Questions |
| 1 PM 5 PM | Combining and Reweighting BRFSS Data |
| 1 PM 5 PM | Time Series Analysis - A Practical Overview |
| 1 PM 5 PM | CATI Users Group Meeting |
Saturday, March 18, 2006
8 AM Noon
Comparison of SAS and SUDAAN for BRFSS Descriptive Analyses.
Instructor: Donna Brogan, Ph.D., Professor of Biostatistics, Rollins School of Public Health, Emory University, Atlanta, GA
Cost: $150
Course Description:
SAS Release 9 offers several PROCS for analysis of sample survey data,
making it a possible alternative to SUDAAN for BRFSS data analyses. In this
course the descriptive analytical capabilities of each software package are
compared using the following criteria:
Prerequisites:
Experience with descriptive analyses of BRFSS survey data using SUDAAN or
SAS. Basic statistical or epidemiological methods.
Saturday, March 18, 2006
8 AM Noon
Part I: Bayesian Perspectives for Epidemiologic Research: Rationale and Basic Methods
Instructor: Sander Greenland, Dr.P.H., M.A., M.S., Professor of Epidemiology, University of California, Los Angeles
Cost: $150
Course Description:
Bayesian methods continue to grow more popular in advanced statistical
modeling, but have as yet had little impact on basic teaching and analysis.
This lag may be due to the common misconception that Bayesian methods are
computationally difficult and require special software. Nonetheless,
perfectly adequate Bayesian analyses can be carried out with ordinary
software for standard (frequentist) analysis, with no special programming
required. Under a wide range of priors, the accuracy of these approximations
is just as good as the frequentist accuracy of the software, and more than
adequate for observational studies in health and social sciences. An easy
way to do Bayesian analyses is via inverse-variance (information) weighted
averaging of the prior with the frequentist estimate. A more general method
expresses the prior distributions in the form of prior data or data
equivalents, which are then entered in the analysis as a new data stratum.
That form reveals the strength of the prior judgments being introduced, and
leads to methods for modeling biases, which will be discussed in Part II.
Prerequisites:
Coursework in epidemiology and working knowledge of basic epidemiologic
statistics (summary estimates and tests and heterogeneity tests, such as
Mantel-Haenszel, "effect modification" tests, etc.).
Saturday, March 18, 2006
8 AM 5 PM, with a break from Noon 1
PM
Logistic Regression Modeling in Epidemiologic Research
Instructor: David Kleinbaum, Ph.D., Professor of Biostatistics, Emory University
Cost: $250
Course Description:
This course considers the basic assumptions
and methods of logistic regression modeling techniques as relevant for
epidemiologic research. Logistic modeling is appropriate for binary outcomes
that are typically considered in epidemiologic studies. The mathematical
form of the logistic regression model will be described, as well as methods
of estimating model parameters, particularly the odds ratio measure of
effect obtained from a logistic model. Maximum likelihood techniques for
testing hypotheses and obtaining confidence interval estimates will also be
described. Also, modeling strategy guidelines for obtaining a "best" model
will be presented. The concepts and methods described above will be
illustrated with several numerical examples. Although hands-on practice
using the computer with standard software packages will not be available,
several practice-exercises that use SAS computer output will be provided.
Prerequisites:
Epidemiologists and statisticians who are familiar with hypothesis testing,
confidence intervals, basic epidemiologic study designs, measures of
association, stratified analysis, interaction, and confounding.
Saturday, March 18, 2006
1 PM 5
PM
Comparison of SAS and SUDAAN for BRFSS Standardized Rates and Modeling Analyses
Instructor: Donna Brogan, Ph.D., Professor of Biostatistics, Rollins School of Public Health, Emory University, Atlanta, GA
Cost: $150
Course Description:
SAS Release 9 offers several PROCS for analysis of sample survey data,
making it a possible alternative to SUDAAN for BRFSS data analyses. In this
course the advanced descriptive analytical capabilities and the modeling
capabilities of each software package are compared on (1) programming
syntax, (2) available analytical procedures, and (3) numerical answers
obtained. The SUDAAN PROCS LOGISTIC and REGRESS are compared to the SAS
PROCS SURVEYLOGISTIC and SURVEYREG for conducting logistic regression and
linear regression. SUDAAN DESCRIPT has several analytical options not
available directly in SAS: standardized rates (STDVAR & STDWGT), comparison
of domains on means or proportions (CONTRAST, PAIRWISE, DIFFVAR), and trend
over the levels of some variable (e.g. time; POLYNOMIAL). However, it will
be illustrated that these analyses can be conducted in SAS SURVEYREG using
appropriate syntax. All examples use BRFSS data. After the course,
participants should be able to decide whether they can use SAS, rather than
SUDAAN, for the types of advanced descriptive and modeling analyses they
usually conduct with BRFSS data.
Prerequisites:
Experience with analysis of BRFSS survey data using either SUDAAN or SAS.
Experience with logistic regression and linear regression. Basic
understanding of linear contrasts would be useful.
Saturday, March 18, 2006
1 PM 5
PM
Part II: Bayesian Perspectives for Epidemiologic Research: Regression and
Bias Modeling
Instructor: Sander Greenland, Dr.P.H., M.A., M.S., Professor of Epidemiology, University of California, Los Angeles
Cost: $150
Course Description:
This session describes extensions of the basic Bayesian methods in part I to
analyses involving non-normal priors and regression analysis, including
hierarchical (multilevel) modeling. These methods provide an alternative to
the parsimony-oriented approaches of standard regression analysis. In
particular, they replace arbitrary variable-selection criteria by prior
distributions, and by doing so facilitate realistic use of vague but
important prior information. They also allow Bayesian analyses to be
conducted with standard regression packages without any special programming;
one need only be able to add variables and records to the data set. The
methods thus facilitate the use of Bayesian solutions to problems of sparse
data, multiple comparisons, and study bias.
Prerequisites:
You must have taken Bayesian Perspectives Part I (offered in the morning) to
register for this training class. You should also have working knowledge of
epidemiologic regression (i.e., logistic, conditional logistic, Poisson
regression).
Sunday, March 19, 2006
8 AM Noon
Using SAS for Descriptive Analyses of BRFSS Data
Instructor: Donna Brogan, Ph.D., Professor of Biostatistics, Rollins School of Public Health, Emory University, Atlanta, GA
Cost: $150
Course Description:
This course will accomplish the following:
Specific analytical methods include estimating population totals, percentages (rates) or means, chi-square tests for survey data, and estimating prevalence ratios or odds ratios. Examples use the typical descriptive analyses conducted with BRFSS data. After the course, participants should be able to conduct basic and descriptive analyses of BRFSS data, using SAS.
Prerequisites:
Basic statistical or epidemiological methods. Experience using SAS for data
management and statistical analysis. No experience with survey data analysis
is necessary.
Sunday, March 19, 2006
8 AM Noon
Multilevel Modeling (MLM) Techniques for Use with Complex Survey Data
Instructor: Scott P. Novak, Ph.D., Research Triangle Institute International
Cost: $150
Course Description:
Multilevel data in survey research are typically characterized as those
that have a hierarchical, or clustered structure. These include data that
are collected within groups, such as respondents (e.g., students) within a
social context (e.g., school), or longitudinal data where multiple
observations are collected on the same group of persons over time. This
workshop will consider issues related to the following in data collected
through a complex survey design:
Some advanced extensions (e.g., Item response theory, latent variables, propensity scoring/inverse probability weighting) will also be introduced. Examples will be providing using the BRFSS.
Prerequisites:
The seminar presumes familiarity with the general linear model (e.g.,
continuous outcomes) and the generalized linear model (e.g., outcomes that
are distributed as dichotomous, ordinal, Poisson). Experience with
programming in one of the statistical packages typically used in the
epidemiological sciences is also required (e.g., SPSS, Splus, SAS, STATA, R).
Knowledge of specialty multilevel programs (e.g., HLM, Mplus, Mixor/Mixreg/Mixpreg,
MLM_win) is a plus. This course is intended for an applied audience (e.g.,
epidemiologists, sociologists, health psychologists) with experience in
using general statistical packages and an interest in testing research
hypotheses that are multilevel in nature.
Sunday, March 19, 2006
8 AM Noon
BRFSS Coordinator Training in Physical Activity & Fruits and Vegetables Survey Questions
Instructor: Sandra Ham, MS, and Susan Carlson, Division of Nutrition and Physical Activity, CDC
Cost: No fee
Course Description:
This course will accomplish the following:
There will be a few comments on analysis of BRFSS data when the unit of analysis is not an adult (e.g., the household or child).
Prerequisites:
Experience with analysis of BRFSS survey data using SUDAAN or SAS or STATA
or SPSS. Intermediate statistical or epidemiological methods. Basic
understanding of linear contrasts would be useful.
Sunday, March 19, 2006
1 PM 5
PM
Combining and Reweighting BRFSS Data
Instructor: Donna Brogan, Ph.D., Professor of Biostatistics, Rollins School of Public Health, Emory University, Atlanta, GA
Cost: $150
Course Description:
This course gives the latest information about the history, reliability and
validity, algorithms, and future plans for the physical activity and fruits
and vegetables survey questions. The CDC's Division of Nutrition and
Physical Activity will also present the recommended usage of these questions
for surveillance and program evaluation. Participants will also learn how to
use these questions to enhance the usefulness of BRFSS by analyzing data for
nutrition and physical activity along with related chronic diseases.
Prerequisites:
None
Sunday, March 19, 2006
1 PM 5
PM
Time Series AnalysisA Practical Overview
Instructor: Mansour Fahimi, Ph.D., Research Triangle Institute International
Cost: $150
Course Description:
Time Series is a collection of observations obtained sequentially in time.
Unlike most statistical analysis techniques that are applicable to
independent observations, Time Series Analysis deals with methods that
explicitly recognize the order in which the observations are made. These
methods provide opportunities for both explaining the past and predicting
the future behavior of the measurement of interest. While there are
different approaches to analysis of time series data, this workshop provides
a practical overview of the two primary methods of regression (structural
equation) and Box-Jenkins (transfer function). Examples will be provided
using SAS and Excel to illustrate the main objectives of analyzing a time
series, which include (a) description, (b) explanation, and (c) prediction.
Prerequisites:
This workshop assumes basic knowledge of statistical inference, regression
analysis , and familiarity with statistical software packages such as SAS.
Sunday, March 6, 2005
1 PM 5
PM
CATI Users Group Meeting
Presenter: Mr. Claude Comeau
Cost: No fee
Course Description:
This meeting will provide the BRFSS states and contractors an opportunity
to get together, learn, make suggestions, and share advice on CATI issues.
The topics will be general CATI issues and should be helpful to everyone.
This is a great session to find out how the BRFSS is working in different
locations and learn a few new tips. This group provides a lively,
informative discussion of important BRFSS CATI issues. Start saving your
CATI questions and bring them to this session. Interviewer, survey
supervisor, and coordinator questions are all encouraged.
Prerequisites:
None
|
|
|
* Links to non-Federal organizations are provided solely as a service to our users. Links do not constitute an endorsement of any organization by CDC or the Federal Government, and none should be inferred. The CDC is not responsible for the content of the individual organization Web pages found at this link. BRFSS Home | Contact Us | CDC Home | Search | Health Topics A-Z | Policies and Regulations Page last reviewed:
12/11/2008
United States Department of Health and Human Services
|
|
