American Statistical Association
New York City
Metropolitan Area Chapter

Refresher Course in Statistics
March - May 2011



The New York Metro Area Chapter of the American Statistical Association

In Collaboration With

The Graduate School of Psychology and School of Health Sciences at Touro College

Are Pleased to Invite You to A

REFRESHER COURSE IN STATISTICS

Consisting of a Series of Six Sessions, With One Session Per Week

Schedule and Topics
Session Date/Time Topic/Speaker
Monday, March 21, 2011
4:00 - 6:00 P.M.
DESCRIPTIVE STATISTICS
Dr. Guy Cohen
Monday, March 28, 2011
Check-in: 3:30 P.M.
4:00 - 6:00 P.M.
CATEGORICAL DATA ANALYSIS
Dr. Bernard S. Gorman
Monday, April 4, 2011
Check-in: 3:30 P.M.
4:00 - 6:00 P.M.
SIMPLE LINEAR REGRESSION
Dr. Louis H. Primavera
Monday, April 11, 2011
Check-in: 3:30 P.M.
4:00 - 6:00 P.M.
INFERENTIAL STATISTICS
Dr. Kelly Zou
Monday, May 9, 2011
Check-in: 3:30 P.M.
4:00 - 6:00 P.M.
MULTIPLE REGRESSION
Prof. Dominick Fortugno
Monday, May 16, 2011
Check-in: 3:30 P.M.
4:00 - 6:00 P.M.
BAYESIAN STATISTICS
Dr. David Rindskopf

Location
Touro College
33 West 23rd Street
(between 5th and 6th Avenues)
Room Number was Provided in Your Registration E-mail Confirmation
New York, New York

REGISTRATION IS CLOSED

Fees
NYC Metro Area Chapter Members: $20 per session or $100 for series
Non-Chapter Members: $30 per session or $150 for series
Students: $10 per session or $50 for series (include copy of current student ID with payment)

Additional Information
For questions, send an e-mail to
nycasa@mindspring.com.



DESCRIPTIVE STATISTICS
by
Guy Cohen

Summary Agenda

  1. Definition of population and sample
  2. Graphical summaries: stem and leaf, histograms, etc.
  3. Measures of central tendency, including geometric means and ROBUST estimates
  4. Measures of variability
  5. SAS output review: PROC UNIVARIATE, PROC MEANS
  6. Brief mention of SAMPLE Outliers
  7. Reference distributions
Guy Cohen is a professionally trained statistician who has been working in the health care industry for the past 43 years. He has supported the development of new medicines in the therapeutic areas of neuroscience, cardiovascular, metabolic, and anti-infectives and has been an adjunct lecture at St Johnís University, Mercy College, and Long Island University. In addition, he has also counted whales, Canadian Geese, estimated NYC populations, predicted elections, and assisted in survey methodologies to assess sensitive issues. He received his professional training at Cornell University, University of North Carolina, and Yale University.

Descriptive Statistics Presentation Slides
Descriptive Statistics Presentation Handout

Schedule and Topics | Location | Additional Information



CATEGORICAL DATA ANALYSIS
by
Bernard S. Gorman, Ph.D.
SUNY Distinguished Professor, Nassau Community College
and
Adjunct Professor, Hofstra University

Many years ago, S.S. Stevens and others drew a distinction between categorical (nominal) scale data and ordinal, interval, and ratio scale data. While categorical data imposes fewer restrictions than other data forms, it is not an "inferior" form of data. In fact, many phenomena of interest to psychologists are naturally categorical. Unfortunately, the statistical training of most social scientists emphasizes data models and methods geared toward interval and ratio scale data. While some statistic courses give cursory coverage of contingency tables and a few nonparametric techniques, many people are unaware of the rich possibilities that other research fields have enjoyed through the use of categorical data methods. Advances in both statistical theory and software development can deliver the promises of strong categorical analyses.

The workshop will show participants how to gain the most information from two-way categorical contingency table analyses and will then provide glimpses and overviews of advanced categorical procedures such as loglinear modeling, categorical logistic regression, configural frequency analysis, correspondence analysis, categorical regression trees, and latent class analyses.

Bernard Gorman received his Ph.D. (1971) in Personality and Social Psychology from the City University of New York, and completed postdoctoral studies in psychotherapy at the Institute for Rational Emotive Therapy. He has written numerous articles and presented many convention papers in the areas of personality assessment, multivariate analysis, and relationships between cognition and affect. He co-authored the textbook, Developmental Psychology (Van Nostrand, 1980) with Theron Alexander and Paul Roodin, and co-edited the research monograph, The Personal Experience of Time (Plenum, 1977) with Alden Wessman. He is the author of several instructional computer packages published by Random House and McGraw-Hill. His book, Design and Analysis of Single Case Research, with Ronald Franklin and David Allison, focuses on the intensive study of individuals over time.

Gorman is a SUNY Distinguished Professor of Psychology and State University of New York Faculty Exchange Scholar at Nassau Community College/SUNY, where he teaches courses in general psychology, abnormal psychology, and child and adult development. He holds an adjunct professorship in Hofstra University's Graduate Psychology and Gerontology Programs, where he teaches courses in gerontology, multivariate statistical analysis, computer applications in psychology, and psychometrics. He received the State University of New York Chancellorís Award for Excellence in College Teaching. For more than 15 years, he combined his interests in measurement research, clinical issues, and teaching as a psychologist in the New York State Office of Mental Health. He served as vice-president of the Metropolitan New York Chapter of the American Statistical Association from 1993-1998. He is a Senior Research Scientist in the Department of Psychiatry at Beth Israel Medical Center, New York, where he is part of a research team investigating the efficacy of psychotherapy. He served as a member of the National Science Foundation Research Coordination Network on DNA microarray technology, where he developed multivariate statistical analysis methods for studying gene expression.

Categorical Data Analysis Presentation Slides
Categorical Examples in R

Schedule and Topics | Location | Additional Information



SIMPLE LINEAR REGRESSION
by
Louis H. Primavera, Ph.D.

The linear model underlies most of inferential statistics. This session will cover the basic concepts of simple linear regression and correlation including model assumptions and applications.

Dr. Primavera received his B.A. in Psychology from St. Johnís University and his M.A. and Ph.D. in Experimental Psychology from the City University of New York. He is currently the Dean of the Graduate School of Psychology and the Dean of the School of Health Sciences at Touro College. Dr. Primavera has published extensively and has interests in quantitative methods, drug use, and stigma and discrimination. He was a consultant to the Department of Psychiatry and Behavioral at Memorial Sloan Kettering Cancer Center for ten years and has held a number of other consulting positions in medicine, business, and education. He has been a member of a number of professional organizations and has presented at a number of annual conventions and conferences. Dr. Primavera was President of the Academic Division of the New York State Psychological Association, President of the New York City Metro Chapter of the American Statistical Association, and long time Board Member of the New York State Metro Chapter of ASA. Dr. Primavera is a Fellow of the Division of General Psychology of the American Psychological Association and a Fellow of the American Educational Research Association and a Fellow of the Eastern Psychological Association.

Simple Linear Regression Presentation Slides

Schedule and Topics | Location | Additional Information



STATISTICAL INFERENCE: HISTORY, PRINCIPLES, AND DEVELOPMENTS
by
Kelly H. Zou, Ph.D.
Director of Statistics, Pfizer Inc., Specialty Care Business Unit
New York, New York

This presentation reviews several fundamental and important concepts concerning statistical inference, which is an extremely broad, if not all encompassing, topic. First, a historical perspective of statistical inference is provided by summarizing the developments over the last 200 years (e.g., the chi-square test constructed by Pearson and the likelihood-based statistical foundation established by Fisher). Both frequentist and Bayesian approaches are presented. Specifically, classical concepts reviewed include likelihood functions, exponential family, confidence intervals and regions, hypothesis tests, exact inference, large sample theory, and Bayesian methods. Advanced methods developed by the speaker include classification accuracy using a receiver operating characteristic analysis, monotone transformation theory, and statistical validation of predictive modeling.

This refresher course is suitable for researchers in statistics and other closely related fields. Practicing statisticians in academia, industry, and government may benefit from a review of such important and yet frequently overlooked topics. In his monograph entitled "Principles of Statistical Inference," which may serve as an important reference when reviewing these topics, Professor D.R. Cox wrote that "the object is to provide ideas and methods for the critical analysis and, as far as feasible, the interpretation of empirical data arising from a single experiment or observational study or from a collection of broadly similar studies bearing on a common target."

Key Words: Statistical Inference; Confidence Interval and Region; Maximum Likelihood Estimation; Hypothesis Tests; Transformation; Bayesian methods.

Kelly H. Zou, Ph.D. is a Director of Statistics in the Specialty Care Business Unit at Pfizer Inc. She received both Masterís and Ph.D. degrees in Statistics from the University of Rochester, and completed Post-Doctoral training in Biostatistics and Radiology at Harvard Medical School and its affiliated Brigham and Womenís Hospital. She was an Associate Professor in Radiology at Harvard Medical School and a Lecturer in Health Care Policy at Harvard Medical School. She was the Principal Statistician in Radiology at Brigham and Womenís Hospital and Director of Biostatistics at Childrenís Hospital Boston.

Dr. Zouís research interests include health policy, outcomes research, clinical trials and biomarkers. She has served as the Principal Investigator on a number of research grants funded by the National Library of Medicine, National Institute of General Medical Sciences, and the Agency for Healthcare Research and Quality, respectively, of the National Institutes of Health in the US. Her work on ROC analysis and statistical classification has led her to be the recipient of the Stauffer Award for the best article published in "Academic Radiology," the First, Second and Third Place winner of the American Statistical Association and Biopharmaceutical Statistics Section poster competition during the 2009 and 2010 Joint Statistical Meetings and both Second and Third Place Winner during the 2010 meeting. She was the recipient of the Travel Stipend Award from the Society of Health Services Research in Radiology and Reviewer with Special Distinction Award for "Radiology."

Dr. Zou has published nearly 100 peer-reviewed articles listed via PubMed. She is serving as an Associate Editor of "Statistics in Medicine" and "Radiology," as well as a Referee for over 10 professional statistical and medical journals. She was the theme editor and lead author of two books entitled: "Mathematical and Statistical Methods for Diagnoses and Therapies" and "Statistical Evaluation of Diagnostic Performance: Topics in ROC Analysis."

She has served as the Vice Chair of the Committee on Applied Statisticians, American Statistical Association; Member of the Corporate Sponsorship Committee, Biopharmaceutical Section, American Statistical Association; Chair of Judiciary Committee, Radiology Alliance to Health Services Research, Association of University Radiologists; Member of the Faculty Taskforce, Joint Committee on the Status of Women, Harvard Medical School and Harvard School of Dental Medicine.

Inferential Statistics Presentation Slides

Schedule and Topics | Location | Additional Information



MULTIPLE REGRESSION
by
Dominick Fortugno

This presentation will introduce the mathematical processes of multiple regression, contrast this method to simple linear regression, highlight examples of multiple regression used in research today, and describe the limitations, assumptions, and best practices for using this method in psychological research. Attention will be given to specific topics such as significance testing using relative weights; incorporating power, effect size, and confidence intervals; and the use of multiple regression in test development.

Dominick A. Fortugno is a New York State certified school psychologist and Interim Director for the graduate program in school psychology at Touro College, where he teaches courses in introductory statistics, psychometric theory, and cognitive assessment. Mr. Fortugno is currently completing his doctoral dissertation in school psychology at Fordham University. His research interests include test theory and development, cognitive assessment, and the etiology and treatment of pervasive developmental disorders.

Schedule and Topics | Location | Additional Information



AN INTRODUCTION TO BAYESIAN STATISTICS
by
David Rindskopf
Distinguished Professor
Educational Psychology and Psychology
City University of New York Graduate Center
New York, New York

Topics will include Pearson and likelihood ratio tests of independence, partition of chi-square, loglinear and logit models, logistic regression, survival analysis (discrete case), and latent class analysis.

David Rindskopf is a Distinguished Professor of Educational Psychology and Psychology at the City University of New York Graduate Center, where he has taught since 1979. He is a Fellow of the American Statistical Association, and past President of the New York City Metro Area Chapter of the ASA. His teaching and research interests include categorical data analysis, missing data, structural equation models, and multilevel (hierarchical) models.

Bayesian Statistics Presentation Slides

Schedule and Topics | Location | Additional Information


Chapter News | Chapter Officers | Chapter Events
Other Metro Area Events | ASA National Home Page | Links To Other Websites
NYC ASA Chapter Constitution | NYC ASA Chapter By-Laws

Page last modified on May 17, 2011

Copyright © 1998-2011 by New York City Metropolitan Area Chapter of the ASA
Designed and maintained by Cynthia Scherer
Send questions or comments to nycasa@mindspring.com