CSP 2016 Online Program

The course emphasizes principles of effective graphs, rather than how to create graphs with specific software. These principles apply to whatever software the attendees use. Participants will learn to:

• Present data more effectively in all media

• Display data so their structure is more apparent

• Understand principles of effective simple graphs to build upon when creating interactive or dynamic displays

• Become more critical and analytical when viewing graphs

Outline & Objectives

This course begins by reviewing human perception and our ability to decode graphical information. It continues by:

• Ranking elementary graphical perception tasks to identify those that we do the best.

• Showing the limitations of many common graphical constructions.

• Demonstrating newer, more effective graphical forms developed on the basis of the ranking.

• Providing general principles for creating effective graphs, as well as metrics on the quality of graphs.


• Commenting on software packages that produce graphs.


• Comparing the same data using different graph forms so the audience can see how understanding depends on the graphical construction used.


• Discussing Trellis Display (a framework for the visualization of multivariate data) and other innovative methods for presenting more than two variables.

• Presenting some graphical methods for categorical data.

• Discussing some examples of embedded graphs for complex data.

Since scales (the rulers along which we graph the data) have a profound effect on our interpretation of graphs, the section on general principles contains a detailed discussion of scales.

About the Instructor

• Chair of Statistical Graphics Section of ASA

• Author of Creating More Effective Graphs (Chart House, 2013; originally Wiley, 2005).

• Invited to blog for Forbes on Effective Graphs (see http://www.forbes.com/sites/naomirobbins/)

• Speaker on graphs to numerous universities and professional societies

• Presenter to government, corporate and non-profit organizations

• Delivered tutorials and/or presentations at most O’Reilly Strata Conferences

• Speaker at the 61st Deming Conference on December 5, 2005

• Short course instructor at the Joint Statistical Meetings in Seattle on August 7, 2006

• A.B. degree in mathematics from Bryn Mawr College

• M.A. in mathematics from Cornell University

• Ph.D. in mathematical statistics from Columbia University

• Founding member of the New Jersey Chapter of ASA and served as President, Vice-President, Secretary, Treasurer, and Chair of the Advisory Committee

• Member of ASA, the Society for Technical Communication, Chapters of ASTD, IEEE and other professional societies

• Associate Fellow of the Society of Technical Communication

• Formerly a Member of Technical Staff at Bell Laboratories

Relevance to Conference Goals

Attendees will be exposed to graphical techniques, some of which may be new to them. Ideas covered are immediately applicable.

The entire emphasis of the course is to use best graphical practices to communicate quantitative information better.

Effective charts and graphs and understanding data better lead to better decisions which have a positive impact on the company. Communicating data better saves time at meetings.

Better communication of data enhances one’s career and avoids the loss of credibility that comes with using confusing, misleading or deceptive figures.

The R Markdown package builds reports, documents, and presentations straight from your R code. You write your reports in markdown (an easy-to-write, plain text format) and embed R code chunks that create output to be included in the final document. R Markdown documents are fully reproducible (they can be automatically regenerated whenever underlying R code or data changes) and completely dynamic (you can export an R Markdown document as a slide show or an html, pdf, or MS Word file).

The Shiny package uses R to build interactive web apps, an ideal data product to share with consulting clients. Shiny apps reduce iteration and allow clients to explore their own data without relying on the consultant's resources. Clients can use a shiny app to explore data and run analyses---without needing to write or understand R code (you write the code for them when you create the app).

Outline & Objectives

You will learn to write reports with R Markdown and will create your own Shiny web apps. I will alternate brief lectures with hands-on practice: you’ll get plenty of experience actually using R Markdown and Shiny (not just hearing about them!) Outline:

1. Reports - Write reports in markdown. Embed R code chunks and customize how code output appears in the report. Render reports as html, pdf, MS Word docs and slideshows.

2. Interactivity - Add self-contained interactive data visualizations to your reports with htmlwidgets. Learn to use Shiny's reactive programming framework.

3. Web Apps - Layout interactive components into a web app with a customizable user interface. Learn how to share apps on the internet (with a password), or a private intranet.

This class will be a good fit for you if you already use R, but are new to R Markdown and Shiny.

Come ready to learn. You'll need your laptop and the latest versions of R and RStudio. Install the shiny and reportsWS packages ahead of time with install.packages(c("shiny", "devtools")) and devtools::install_github("rstudio/reportsWS"). R Markdown comes with RStudio. Each of these is free and open source, as is R and RStudio.

About the Instructor

The course will be led by RStudio Master Instructor and author Garrett Grolemund, Ph.D. Garrett is a teacher and statistician who has used R to analyze data for over 10 years. He works closely with the developers of Shiny and R Markdown, and is Editor in Chief of the Shiny Development Center web page. Garrett created and maintains the popular lubridate R package and is the author of Hands-On Programming with R as well as Data Science with R, an upcoming book by O'Reilly Media.

Garrett gave a very well-received talk at the 2015 Conference on Statistical Practice titled "Reporting results with R, R Markdown, and Shiny."

Relevance to Conference Goals

This course teaches students two ways to better communicate with customers. Embedded in the methods are benefits that will have a positive impact on the student's organization. The two methods that improve communication are:

1. An easy and quick way to create reports from the code that statisticians use to generate results. This method has the benefits that it

* Creates reproducible reports, which can quickly be regenerated on new data to update an analysis or to prove the scientific validity of the results.

* Lets students export the same report in multiple formats, which saves time and makes content easier to share through different channels.

2. A way to build interactive data products that clients can use to explore their own data and explore the results of the student's analysis. This method has the benefits that it

* Conveys large amounts of information efficiently, and provides a navigation system for accessing the information.

* Shortens cycles of iteration. Clients can use the app to fine tune analyses without the aid of the consultant.

* Lets clients recreate analysis and explore data without needing to know how to program in R.

* Looks impressive.

Relevant R and SAS software packages for implementing data analyses will be discussed in detail. Specific topics to be covered include guidelines for constructing a propensity score model, creating matched pairs for binary group comparisons, assessing baseline covariate balance after matching, and using inverse propensity score weighting techniques. Illustrative examples will accompany each topic and a brief review of recent relevant developments and their implementation will be discussed.

Outline & Objectives

Outline:
- Observational Studies: definition, examples, causal effects, confounding control.

- Propensity Scores: definition, properties, modeling techniques.

- Propensity Score Approaches in Observational Studies: weighting, matching, sub-classification; graphical methods to assess covariate balance after matching; R and SAS software implementation of these techniques.

- Guidelines on how to best describe the methodology utilized and the results obtained when presenting to a non-technical audience.

- Brief review of the most recent methods developments and discussion of their potential for immediate use in practice.

Objectives and Scope:
The first objective is to provide an overview of some of the most commonly used propensity score-based methods in observational studies, while focusing on the practical aspects. The second objective is to present the practical implementation of these methods. The third objective is to discuss the advantages and disadvantages associated with these methods.

About the Instructor

Dr. Andrei has obtained a Ph.D. degree in Biostatistics from the University of Michigan in 2005. He is currently a Research Associate Professor in the Feinberg School of Medicine at Northwestern University, where he enjoys successful collaborations in cardiac surgery and cardiovascular outcomes research, of which observational studies are a major component. Dr. Andrei has developed expertise in propensity score techniques, as reflected by national conference presentations and his participation as primary biostatistician in a series of publications that utilize propensity score methods. He has developed practice-inspired and -oriented statistical methods in survival analysis, recurrent events, group sequential monitoring methods, hierarchical methods, and predictive modeling.

In the last 10 years, Dr. Andrei has collaborated with medical researchers in fields such as pulmonary/critical care, organ transplantation, nursing, prostate and breast cancer, anesthesiology and thoracic surgery.

Currently, he serves as Statistical Co-Editor of the Journal of the American College of Surgeons and deputy Statistical Editor of the Journal of Thoracic and Cardiovascular Surgery.

Relevance to Conference Goals

Upon attending this course, participants will become more familiar with propensity score-based methods for estimating causal effects in observational studies. Implementation in R and SAS software will be discussed in great detail, thus permitting participants to integrate these newly-acquired statistical techniques into their professional activities and projects. Learning how to produce simple yet powerful graphics to assess the propensity score model adequacy, check covariate balance and display the results, will undoubtedly benefit every participant. By leveraging their enhanced set of skills, individuals across industries will be adequately positioned to become more effective communicators in their interactions with customers and clients. Continued professional development is key to one’s career growth and can enhance the overall analytical capabilities within their respective organizations and institutions.

Outline & Objectives

A. Course Outline
1) Fundamental Principles of Adaptive Designs
2) Three Examples of Adaptive Designs
3) Adaptive Design versus Conventional Designs
a. Advantages of adaptive designs
b. Disadvantages and risks associated with adaptive designs
4) Recent Advances in Adaptive Designs
5) The Basic Adaptive Design Toolkit: What you need to get started
a. Essential design elements for your proposal or protocol designs
b. Software implementation
6) Hands-on Practice

B. Course Objectives
1) Gain familiarity with the basic principles of adaptive designs
2) See of examples of adaptive designs in practice
3) Understand strengths and weaknesses, and when to use and when not to use adaptive designs
4) See the latest application of adaptive designs
5) Obtain a starter toolkit to start using adaptive designs in your work

About the Instructor

Dr. Minh Huynh is a Principal Research Associate and Managing Director at IMPAQ International, LLC. Dr. Huynh was a former Institute Staff Scientist at the NIH Clinical Research Center. Dr. Aaron Heuser is a Senior Research Associate at IMPAQ International, LLC. Dr. Heuser was also a former Mathematical Statistician at the NIH Clinical Research Center in Bethesda, Maryland.

Relevance to Conference Goals

This course is closely related to Theme 2, Data Modeling and Analysis. The course will provide all attendees with theoretical and practical knowledge and techniques related adaptive designs through the application of state-of-the-art methods and findings. Course presentations will cover all aspects of adaptive designs and present information relevant to a broad range of applied statisticians, regardless of industry or field of expertise.

Outline & Objectives

This interactive workshop will cover:
1. The Coward in Us All – examine your own reasons for avoiding conflict and the top 10 reasons people avoid conflict
2. Motivate Yourself to Address Conflict – examine the pain and pleasure of conflict
3. Common Causes of Conflict – Proactively avoid conflict
4. Techniques for Handling Conflict

You will complete interactive exercises and learn step by step approaches that will help you work through conflict more easily, gracefully, and courageously.

About the Instructor

Colleen Mangeot's diverse career includes 10 years in the actuarial field, 10 years in coaching and leadership development, and 6 years in biostatistics.
Highlights of her 7 year coaching business include: Successfully working with clients to increase efficiency and sales by 30% or more; Graduate of the world’s largest coach training organization, Coach U; Attained the Professional Coach Certification from the International Coach Federation in 2003; Monthly columnist for the Dayton Business Journal; National speaker with over 200 hours of paid speaking engagements; Contractor with the Anthony Robbins Companies.
She received her MS in Statistics from Miami University, received the NSA National Research Council Fellowship at NIOSH, and worked in statistical quality improvement at the VA. Now, in addition to working in the Biostatistical Consulting Unit at Cincinnati Children’s Hospital Medical Center, she is also an internal coach working with executives to further their careers.
She was a panelist for the invited session at JSM 2013, Secrets to Effective Communication for Statistical Consultants, and presented two sessions at last years’ Conference on Statistical Practice.

Relevance to Conference Goals

This session will develop important communication skills for career advancement, leadership and management effectiveness. We all must face conflict from time to time. The most successful people transform conflict into positive results. The approaches and steps taught in this session will apply equally well in all environments, academia, government and industry. The presenter has personally utilized these in her career in government and industry.

We consider a variety of statistics (mean, trimmed mean, regression, etc.) and a number of sampling situations (one-sample, two-sample, stratified, finite-population), stressing the common techniques that apply in these situations. We'll look at applications from a variety of fields, including telecommunications, finance, and biopharm.

These methods let us do confidence intervals and hypothesis tests when formulas are not available. This lets us do better statistics (e.g., use robust methods (we can use a median or trimmed mean instead of a mean, for example)). They can help clients understand statistical variability. And some of the methods are more accurate than standard methods.

Outline & Objectives

Introduction to Bootstrapping
General procedure
Why does bootstrapping work?
Sampling distribution and bootstrap distribution

Bootstrap Distributions and Standard Errors
Distribution of the sample mean
Bootstrap distributions of other statistics
Simple confidence intervals
Two-sample applications

How Accurate Is a Bootstrap Distribution?

Bootstrap Confidence Intervals
Bootstrap percentiles as a check for standard intervals
More accurate bootstrap confidence intervals

Significance Testing Using Permutation Tests
Two-sample applications
Other settings

Wider variety of statistics
Variety of applications
Examples where things go wrong, and what to look for

Wider variety of sampling methods
Stratified sampling, hierarchical sampling
Finite population
Regression
Time series

Participants will learn how to use resampling methods:
* to compute standard errors,
* to check the accuracy of the usual Gaussian-based methods,
* to compute both quick and more accurate confidence intervals,
* for a variety of statistics and
* for a variety of sampling methods, and
* to perform significance tests in some settings.

About the Instructor

Dr. Tim Hesterberg is a Senior Statistician at Google. He previously worked at Insightful (S-PLUS), Franklin & Marshall College, and Pacific Gas & Electric Co. He received his Ph.D. in Statistics from Stanford University, under Brad Efron.

Hesterberg is author of the "Resample" package for R and primary author of the "S+Resample" package for bootstrapping, permutation tests, jackknife, and other resampling procedures, is co-author of Chihara and Hesterberg "Mathematical Statistics with Resampling and R" (2011), and is lead author of "Bootstrap Methods and Permutation Tests" (2010), W. H. Freeman, ISBN 0-7167-5726-5, and technical articles on resampling. See http://www.timhesterberg.net/bootstrap.

Hesterberg is on the executive boards of the National Institute of Statistical Sciences and the Interface Foundation of North America (Interface between Computing Science and Statistics).

He teaches kids to make water bottle rockets, leads groups of high school students to set up computer labs abroad, and actively fights climate chaos.

Relevance to Conference Goals

Resampling methods are important in statistical practice, but have been omitted or poorly covered in may old-style statistics courses. These methods are an important part of the toolbox of any practicing statistician.

It is important when using these methods to have some understanding of the ideas behind these methods, to understand when they should or should not be used.

They are not a panacea. People tend to think of bootstrapping in small samples, when they don't trust the central limit theorem. They are not a panacea. People tend to think of bootstrapping in small samples, when they don't trust the central limit theorem. However, the common combinations of nonparametric bootstrap and percentile intervals is actually accurate than t procedures. We discuss why, remedies, and better procedures that are only slightly more complicated.

These tools also show how poor common rules of thumb are -- in particular, n >= 30 is woefully inadequate for judging whether t procedures should be OK.

Outline & Objectives

Session 1:
• Introduction to R
• Overview to meta-analysis for both fixed-effects and random-effects models in meta-analysis. Real datasets in public health are introduced along with two commonly used R packages of "meta" and "rmeta"
Session 2:
• Meta-analysis models for binary data, such as for risk-ratio, risk difference and odds-ratio
• Methods to quantify heterogeneity and test the significance of heterogeneity among studies in a meta-analysis and then introduce meta-regression with R package of "metafor".

About the Instructor

Dr. Din Chen is a professor at the University of Rochester. He was the Karl E. Peace endowed eminent scholar chair in biostatistics at Georgia Southern University. Professor Chen is also a senior statistics consultant for biopharmaceuticals and government agencies with extensive expertise in clinical trials and bioinformatics. He has more than 100 referred professional publications and co-authored six books on clinical trial methodology and public health applications.
Professor Chen was honored with the "Award of Recognition" in 2014 by the Deming Conference Committee for highly successful 4 advanced biostatistics workshop tutorials at 4 successive Deming conferences on 4 different books that he has written or co-edited. In 2013, he was invited to give a short course at the twentieth Annual Biopharmaceutical Applied Statistics Symposium (BASS XX, 2013) for his contribution in meta-analysis and received a "Plaque of Honor" for his short course.
The tutorial is based on his recently book "Applied Meta-Analysis using R" with Professor Karl E. Peace by Chapman and Hall/CRC in 2013.

Relevance to Conference Goals

1. To give a up-to-date methodology development in meta-analysis and to guide the participants to learn the meta-analysis models

2. To give an overview of R implementations for meta-analysis models with R packages and R programming

3. To emphasize the applied aspects of meta-analysis with real public health data compiled from systematic reviews to help applied statisticians to solve their real-life problems from research and consulting.

Outline & Objectives

Course outline:

This short course discusses both traditional and more recent statistical process control (SPC) charts and how to use them for analyzing big data. More specifically, the course topics include (i) traditional SPC charts, including the Shewhart, CUSUM and EWMA charts, (ii) some recent control charts based on change-point detection, (iii) fundamental multivariate SPC charts under the normality assumption, (iv) novel univariate and multivariate control charts for cases when the normality assumption is invalid, (v) control charts for profile monitoring, and (vi) hands-on examples to use conventional control charts or their modifications for monitoring different types of processes with big data.


Course Objectives:

After taking this short course, participants should be able to apply some basic and more advanced statistical process control (SPC) charts to applications with big data streams. More specifically, participants will learn (i) traditional SPC charts, (ii) some recent control charts in the literature, and (iii) how to use SPC charts to handle applications with big data streams.

About the Instructor

The instructor, Professor Peihua Qiu, is the current editor of Technometrics, which is the flagship journal in industrial statistics, co-sponsored by ASA and American Society in Quality. He has been working on various statistical process control (SPC) problems since 1998, and has made substantial contributions in several SPC areas, including nonparametric SPC, SPC by change-point detection, and profile monitoring. His recent book Qiu (2014, Chapman & Hall) gives a systematic description of both traditional and newer SPC methods. Professor Qiu is the elected fellow of ASA, IMS and ISI. After obtaining his Ph.D. in statistics from University of Wisconsin - Medison, he helped create the Biostatistics Center at the Ohio State University during 1996-1998. Then, he worked as an assistant (1998-2002), associate (2002-2007) and full professor (2007-2013) of the School of Statistics at the University of Minnesota, and taught 13 different courses on various topics. He moved to University of Florida as the founding chair of the Department of Biostatistics in 2013. During his career, Professor Qiu is constantly involved in statistical consulting and collaborative research.

Relevance to Conference Goals

This proposed short course fits well the following two themes of the conference: 1) Big Data Prediction and Analytics, and 2) Data Modeling and Analysis.

Statistical process control (SPC) charts are commonly used in industries, especially in manufacturing industries. However, most SPC charts used in practice are several decades old. For instance, the most commonly used chart is the Shewhart chart proposed by Walter Shewhart in 1931. One main purpose of this course is to let the participants know that 1) many traditional SPC charts could give misleading results if their assumptions are not carefully checked, and 2) some newer SPC charts have been developed in the literature and they often provide more reliable results. In various big data applications, big data often take the form of data streams with observations of certain processes collected
sequentially over time. Another main purpose of the course is to show that SPC charts provide an effective tool for handling such applications. By taking this course, participants will learn statistical techniques that can apply to their daily jobs and better communicate with their clients and customers.

Outline & Objectives

Outline:
1. Installing graphic packages into R
2. Importing data including csv and excel data
3. Basic commands to learn the data – types of data and variables
4. Graphs for different types of data:
a. Continuous variables including scatter plot and box plot
b. Categorical/ Ordinal variables – histograms
5. Key to display – Title, Colors, fit many graphs into a page
6. Saving Output graphs in different formats (jpeg, pdf, and etc)

Objectives: To introduce programming of R and to ignite interest beginning R users. This tutorial will able to use simple programming in R to learn the distribution of data and to learn programming to implement code for simple graphs.

About the Instructor

Minjung Yoon is a professional biostatistician in the pharmaceutical industry. She has over 7 years of experience in various phases of clinical trials conducting sample size and power calculations, proposing study design and randomization scheme, writing statistical analysis plans, conducting data analysis and strategic plans for publications, and authoring manuscripts. She has a Master in Public Health from Boston University concentrating in Biostatistics.

Relevance to Conference Goals

This tutorial is relevant to the conference goals since it will help novice users get a little flavor in R in graphics. These basic skills will professionally help applied statisticians effectively communicate data with their clients.

Outline & Objectives

Participants will:
1. Learn to distinguish between inside-out and outside-in approaches to performance improvement
2. Learn the G.R.O.W. coaching model
3. Practice the coaching model
4. Identify when the model fails and key distinctions that allow the model to work

About the Instructor

Colleen Mangeot's diverse career includes 10 years in the actuarial field, 10 years in coaching and leadership development, and 6 years in biostatistics. Highlights of her 7 year coaching business include: Successfully working with clients to increase efficiency and sales by 30% or more; Graduate of the world’s largest coach training organization, Coach U; Attained the Professional Coach Certification from the International Coach Federation in 2003; Monthly columnist for the Dayton Business Journal; National speaker with over 200 hours of paid speaking engagements; Contractor with the Anthony Robbins Companies. She received her MS in Statistics from Miami University, received the NSA National Research Council Fellowship at NIOSH, and worked in statistical quality improvement at the VA. Now, in addition to working in the Biostatistical Consulting Unit at Cincinnati Children’s Hospital Medical Center, she is also an internal coach working with executives to further their careers. She was a panelist for the invited session at JSM 2013, Secrets to Effective Communication for Statistical Consultants, and presented two sessions at last years’ Conference on Statistical Practice.

Relevance to Conference Goals

Leaders must have effective ways to communicate with others, whether they be peers, direct reports or managers. This workshop provides a step by step approach to effectively communicating with individuals in a way that allows the leader as well as those the leader is communicating with to grow and learn.

With the increased penetration of social media, we’ve seen growing demand for quantitative "proof" of return on investment (ROI) to justify resource allocation decisions. Applied statisticians can play a prominent role to fulfill these needs.

This tutorial will walk attendees through key steps of measuring the impact of social media marketing with three business cases in retail, telecom, and health care. Attendees will learn how to do the following within a social media measurement framework:

- Define a specific measurement objective

- Determine optimal test and control group sizes to ensure adequate power and statistical significance *

- Select appropriate attributes to balance test and control group

- Apply best practice measurement methodology

* Attendees will get a sample sizer tool to help them determine the optimal cell size.

Outline & Objectives

The instructors will share their best practices and lessons learned with three case studies of social media marketing measurement in the retail, telecommunication and healthcare industries. Applied statisticians can gain real world experience in grasping the big picture of social media measurement and following detailed steps with specific techniques to carry out the statistical analysis. Topics to be covered include:

1) Project Planning - building a solid foundation for your social media measurement:

- Identify and prioritize desired learning

- Select your Key Performance Indicators (KPIs) related to your business objective

- Determine your measurement and reporting windows and set up a rollout plan

- Collect and integrate data

2) Statistical Design - ensuring a robust and statistically sound measurement:

- Test and control group size determination

- Balancing attributes and cell assignment

3) Data Analysis - drawing meaningful business conclusions:

- Re-balance the test/control group, if necessary

- Apply the appropriate methodology

- Analysis reporting and recommendations

About the Instructor

Eleanor Tipa, Ph.D.

Eleanor Tipa is a Vice President in the Analytic Consulting Group at Epsilon with over 15 years of experience in marketing analytics across various industries. Her responsibilities include the design, management, and execution of analytic projects that support the optimization of clients’ multi-channel marketing campaigns. Her areas of expertise include the use of advanced statistical techniques in predictive modeling, segmentation and profiling, experimental design, and multi-channel campaign performance measurement.

Eleanor holds a Ph.D. in Statistics from the Pennsylvania State University.

Danny Jin

Danny Jin is a Director in the Analytic Consulting Group at Epsilon with over 8 years of experience in predictive modeling, segmentation, measurement and profiling. Danny’s areas of expertise include modeling techniques, experimental design and multivariate analyses.

Danny holds a M.S. in Applied Statistics from Worcester Polytechnic Institute.

Relevance to Conference Goals

This course will show attendees the industry best practices in measuring social media marketing efforts. The steps that we will go through also demonstrate the vital role applied statisticians can play in quantifying social media marketing ROI. Knowledge of these practices will help broaden participants’ data analysis tool set and give them higher visibility in their organization.

Outline & Objectives

1) Overview of concepts
a) Why do power and sample size analysis?
b) Hypothesis testing and confidence intervals
c) Prospective vs. retrospective
2) How-to guide
a) Essential components of study planning
i) Study design
ii) Effects and variability
iii) Planned data analysis
iv) Goals
b) Postulating “effect size” and other parameters
c) Sensitivity analysis
d) Tabular, graphical, and narrative displays
3) Examples
a) Two-sample t test
b) WMW (rank-sum) test
c) Equivalence with lognormal data
d) Noninferiority for two proportions
e) Survival analysis (log-rank test, Cox regression)
f) Logistic regression
g) Repeated measures
h) Confidence interval precision
Attendees will gain an understanding of the essential methods, best practices, and common pitfalls of power and sample size determination.
This is an introductory-level tutorial intended for a broad audience of statisticians. A basic understanding of the theory and practice of statistical inference is assumed.

About the Instructor

John Castelloe is a Senior Research Statistician Developer at SAS and is the developer of power and sample size software in SAS. He has presented many workshops on the topic at statistical conferences and other venues and is the co-author of the chapter “Sample-Size Analysis for Traditional Hypothesis Testing” in the book Pharmaceutical Statistics Using SAS. John received his PhD in Statistics from the University of Iowa in 1998 and joined SAS in 1999.

Relevance to Conference Goals

Over the years, I've heard many people express interest in a tutorial for power and sample size that covers all the basics, highlights best practices and common pitfalls, and focuses on practical issues rather than specific use of software. I've designed this tutorial with exactly these goals in mind, and I think the CSP is an ideal venue for it.

In our presentation, we use the Latent GOLD program to illustrate a wide variety of applications in which the common thread is that latent classes (segments) are identified that are homogeneous in the sense of having similar response patterns (cluster analysis), having similar growth patterns (latent growth or transition models), or being identical with respect to certain regression coefficients (LC regression models). Such models address modern goals, such as to identify the particular therapy that works best for a particular patient (individualized medicine).

LC models can be applied with categorical or continuous variables or a combination of categorical, continuous, and count variables. Moreover, LC factor (also known as discrete factor) models can be used to group variables similar to factor analysis, but since the latent variables themselves are discrete, they also can be used to identify homogeneous segments of cases.

The committee highly values suggestions for improvements gathered during this time. You will have an opportunity to win door prizes and witness awarding of the best student posters. The closing session is also a great time to let members of the CSPSC know if you are interested in helping out with future conferences.