STATISTICS for BASIC MEDICAL SCIENCES
(BMS 1200)

Course Director : Carl I. Thompson, Ph.D.
Dept. of Physiology, Rm. 621
Phone # : (914) 594 - 4106

E-Mail: Carl_Thompson@NYMC.EDU

Statistics for Basic Medical Sciences

(BMS 1200)

Summer 2004 Schedule

4-6 Pm Rm 609 Physiology

 

Month

Date

No.

Day

Topic

Lecturer

 

 

 

 

 

 

June

08-Jun

1

Tuesday

Introduction / Why Statistics

Dr. Thompson

 

 

 

 

Basic Definitions

Dr. Thompson

 

10-Jun

2

Thursday

Normal Curve, Confidence Interval

Dr. Thompson

 

 

 

 

Poisson Distribution

Dr. Thompson

 

15-Jun

3

Tuesday

Binomial Dist./Descrete Probability

Dr. Moy

 

 

 

 

Bayes Theorem

Dr. Moy

 

17-Jun

4

Thursday

Test of a Hypothesis/T-test

Dr. Thompson

 

 

 

 

Experimental Design/Power Analysis

Dr. Thompson

 

22-Jun

5

Tuesday

Mid-Term Test

 

 

 

 

 

 

 

 

24-Jun

6

Thursday

Review of Test

Dr. Thompson

 

 

 

 

Multiple Groups -- ANOVA

Dr. Hamby

 

29-Jun

7

Tuesday

Review of ANOVA

Dr. Hamby

 

 

 

 

ANOVA Models

Dr. Thompson

 

01-Jul

8

Thursday

Review of ANOVA Models

Dr. Thompson

 

 

 

 

Post Hoc Analysis

Dr. Thompson

July

06-Jul

 

Tuesday

Review of Post Hoc Analysis

Dr. Thompson

 

 

 

 

Group Comparison--Non-Parametric

Dr. Thompson

 

08-Jul

9

Thursday

Review of Non-Parametric

Dr. Thompson

 

 

 

 

Survival Analysis

Dr. Hamby

 

13-Jul

10

Tuesday

Review of Survival Analysis

Dr. Hamby

 

 

 

 

Regression Analysis/Correlation

Dr. Hamby

 

15-Jul

11

Thursday

Review of Regression

Dr. Thompson

 

 

 

 

Multiple Regression

Dr. Thompson

 

20-Jul

12

Tuesday

Review of Regression

Dr. Thompson

 

 

 

 

Non-linear Curve Fitting Modeling

Dr. Thompson

 

22-Jul

13

Thursday

Review of Curve Fitting

Dr. Thompson

 

 

 

 

Models of ANCOVA, MANCOVA, etc.

Dr. Thompson

 

27-Jul

14

Tuesday

Epidemiology

Dr. Mamtani

 

 

 

 

Screening

Dr. Mamtani

 

29-Jul

15

Thursday

 

 

Last Revision: Jun 08, 2004

Course Description and Pre-requisites:

The course is directed toward students involved in scientific data analysis in the basic science disciplines. The techniques presented will be those currently in use throughout the basic medical sciences departments. While no prerequisite statistics course is required, some basic familiarity with statistics is expected and can be achieved from any elementary statistic text. Statistics for Basic Medical Sciences is open to all students in the Ph.D. program. All others wishing to register must do so with the permission of the course directors.

Class Meeting:

The course will meet twice a week for 2 hour periods starting at 4:00 and lasting until 6:00. The first 5 sessions will consist of mainly didactic lectures and assigned homework problems. This section will cover the fundamentals of basic statistical concepts as a foundation to the second portion of the course. Sessions 7 through 15 will be devoted to particular topics with the emphasis on problem solving. Each session will begin with an in-depth discussion of the problem assignment(s) from the preceding session. The second portion of each session will be a didactic presentation of the material for the next problem assignments and will include an introduction of the statistical problem to be worked on before the next class. Each of the problem sets will consist of representative data from research laboratories within the basic sciences illustrating a particular statistical approach to problem solving and data representation. The discussion of the problem set will include appropriate statistical tests, presentation and interpretation of analysis.

Grading policy:

The course will represent two (2) credit hours and meet for 16 two hour sessions. The mid-term exam will represent 20% of the course grade and will cover the basic material through the first 5 sessions. The emphasis of the exam will be on concepts and Astatistical vocabulary@. The final end-term project will consist of an appropriate analysis and statistical discussion of an experimental data set of the student=s choosing (with approval of instructor). This end term grade will be worth 20% of the overall grade. The remaining 60% of the course grade will be based on performance in class discussions and the problem sets which must be turned in each session.

Required Material :

NCSS 2001 and PASS 2000 Statistical Software Package
NCSS - Statistical Software that is easy to use. Includes Data Analysis, Statistical Graphics, and Statistical Analysis.
http://www.ncss.com/

Data file downloads for NCSS exercises :
Data File Downloads for NCSS

 

Recommended books:

Biostatistics. A Methodology for the Health Sciences. Lloyd D. fisher and Gerald vanBelle.
Biometry. Robert R. Sokal and F. James Rohlf
Biostats Basics. A student handbook. James L. Gould and Grant F. Gould.
   
http://www.whfreeman.com/gould/
Biostatistics: A Foundation for Analysis in the Health Sciences.
Wayne W. Daniel
Basic Statistics for the Health Sciences. Jan W. Kuzma
Epidemiology Biostatistics and Preventive Medicine. James F. Jekel, Joann G. Elmore and David L. Katz.
Biostatistics The bare essentials. Norman and Streiner.

 

Suggested Links :

General Information

Guidelines for reporting statistics in journals published by the American Physiological Society.  Douglas Curran-Everett and Dale J. Benos. Am J Physiol Endocrinol Metab 287: E189-E191,  2004.
This Editorial provides a number of useful guidelines to consider in providing 
sufficient detail to properly describe and use statistical techniques in the preparation 
of a manuscript. 
http://ajpendo.physiology.org/cgi/content/full/287/2/E189?eto

Statistical tutorials and eBooks

Concepts and Applications of Inferential Statistics.  Has specific Java based calculators and a fairly complete introductory statistical text on line in PDF.  Also has PDF downloads of useful tables.
  Richard Lowry
  Professor of Psychology
  Vassar College
  Poughkeepsie, NY  USA
http://faculty.vassar.edu/lowry/webtext.html

The Statistics Homepage E-Book
http://www.statsoftinc.com/textbook/stathome.html

Statistics at Square One  This is an eBook on statistics from the British Medical Journal.  The portions covered are well presented and should provide a quick reference to facilitate the material presented in class.
http://bmj.com/collections/statsbk/index.shtml

Statnotes: An Online Textbook, by G. David Garson.  This is an excellent set of notes that concisely describes many useful statistical techniques.  The sections on ANOVA and related topics (ANCOVA for example) are very readable.
http://www2.chass.ncsu.edu/garson/pa765/statnote.htm
He also includes a link to Dave Garson's :
Guide to Writing Empirical Papers, Theses, and Dissertations

This is the Home page of  a course in Biometry, or Biostatistics, offered at Arizona State University by Dr. John Nagy.  His Handouts are particularly clear and concise and may provide useful examples and clarification.
http://lsvl.la.asu.edu/bio415/jnagy/

The Prism Guide to Interpreting Statistical Results which is a nearly complete guide to statistical analysis.  While tailored toward using the Prisim software, it is an excellent overall presentation of statistical techniques.  This site includes a library of technical articles that can provide useful insight into many important areas of specific data analysis :
http://www.graphpad.com/index.cfm?cmd=library.index
http://www.graphpad.com/articles/interpret/principles/stat_principles.htm

This site provides an extensive guide to Non-Linear Curve Fitting.  One can download an extensive guide to statistical techniques including curve fitting which is based on  Analyzing Data with GraphPad Prism by Dr. Harvey Motulsky.  The discussion is quite good.
http://www.curvefit.com/

This is written for researchers and students in the sport and exercise sciences.  However, it is a concise and well illustrated approach to understanding basic statistical analysis and data presentation.
http://www.sportsci.org/resource/stats/index.html

General WEB based links with Statistical Resources
http://www.e-stat.org/copss/teaching.htm

The is the site at the FDA that defines Bioequivalence and gives specific guidelines for the determination of bioequivalence and bioavailability.  In reference to the Federal Register notice on "Preliminary Draft Guidance for Industry on In Vivo Bioequivalence Studies Based on Population and Individual Bioequivalence Approaches: Availability", vol. 62, No. 249, Dec. 30, 1997, the Food and Drug Administration (FDA) is announcing the availability of data that were used by the Agency in support of the proposal and the detailed description of statistical methods for individual and population approaches. One motivation for individual bioequivalence is to identify a subject-by-formulation interaction.
http://www.fda.gov/cder/bioequivdata/

CDC in Atlanta makes available :  
Epi Info 2000 Version 1.1.2 Released November 2, 2001
EpiPatch Released March 26, 2002
http://www.cdc.gov/epiinfo/index.htm

STATISTICAL METHODS IN RESEARCH I.  Site contains some excellent PowerPoint presentations of ANOVA and related topics.
http://sta6166.ifas.ufl.edu/Welcome.htm

Summary of Survey Analysis Software. This page is a summary of available software for the analysis of surveys with complex sample designs. Specifically, it includes software that can do variance estimation with such survey data.
http://www.fas.harvard.edu/~stats/survey-soft/survey-soft.html

FTP site containing PowerPoint presentations from a Elementary Statistics II course taught at Rutgers University.  This course are principles and methods of statistics, including hypothesis testing, regression and correlation analysis, curve-fitting, and non-parametric methods.
http://www.stat.rutgers.edu/~mxie/stat212/

This site contains a fairly complete set of statistical lecture notes in PDF format covering many topics presented in this course.
http://www.dal.ca/~houlihan/psy3500/lectures.htm

A good primer to Kaplan-Meier curves and their use in interpreting and presenting Survival Analysis by the folks at the CDC.
http://www.cdc.gov/epiinfo/MANUAL/kapmeier.htm

Other Software:
GraphPad
SAS
SigmaStat