Scribd
Upload
23 views22 pages

Hailpern

Uploaded by

Iqbal Banyu Sunarya
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
23 views22 pages

Hailpern

Uploaded by

Iqbal Banyu Sunarya
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 22

Teaching Statistics

Using Stata Software

Susan Hailpern BSN MPH MS


Department of Epidemiology and
Population Health
Albert Einstein College of Medicine
Bronx, New York
Albert Einstein College of Medicine
Clinical Research Training Program
 Clinical Research Training program
(CRTP) began in 1998
 Program funded by the NIH under Clinical
Research Curriculum Award Program
Program Components
 CRTP is a 2 year program consisting of two
complementary components:
 Didactic curriculum with emphasis on

epidemiology, biostatistics, study design,


computer methods, and research ethics
 Mentored experience in clinical research

leading to masters thesis


Program Description:
 Program provides future investigators (physicians) with
basic and applied knowledge in core disciplines of clinical
investigation:
 Biostatistics

 Epidemiology

 Students learn how to


 design clinical research studies

 ensure that high ethical standards are met

 organize study team

 apply and obtain funding

 collect, manage and analyze data

 report results
Statistical Software
 SPSS
 SPSS has been used since the start of CRTP

 Pull-down menus were thought to be easy to

use and learn


 Students were easily successful in using

SPSS software and obtaining statistical


results
 As students progressed in statistical analyses

and clinical research, SPSS was found to be too


limited in scope
 Alternative software was sought
Statistical Software
 STATA (and Stat/Transfer v.7)
 Stata was taught in the CRTP for first time in

Summer ‘04 using Version 8


 Stata was chosen because

 It has an excellent reputation

 It is easy to use

 Addition of new pull-down menus (version

8) was attractive feature and was thought to


make learning software easier for students
Statistical Software
 STATA (continued)
 It has substantially wider range of procedures

and options than SPSS


 Multinomial and ordinal logistic regressions

 Frailty models for multivariate survival

analysis (semi-parametric and parametric)


 Immediate commands
Teaching Stata
 Interesting challenge: instructors teaching
statistics with Stata come from very different
backgrounds
 Senior instructor has had extensive experience
using SPSS and is a relative newcomer to Stata.
 Other instructor has had extensive experience
using Stata, but with expertise in writing Stata
programs (unfamiliar with using “pull-down”
menus in version 8).
Teaching Stata
 Students were asked to install software prior to
first class
 All students used Windows 98, 2000, or XP

 All had some computer experience

 Brief introduction to the software was distributed


prior to first class
 Description of the windows

 Variable storage types

 Most importantly: how to get Help!


Teaching Stata: First Class
 Stata Structure
 Windows: Data Editor, Results, Review, Variables,

Command, Viewer
 Menus and Syntax

 Course was taught primarily using pull-down menus

 Data File

 Data structure: Files, Variables, Subjects

 Variable Elements: Name, Type, Format, Label,

Values (labels), Missing


 Descriptive Statistics
Teaching Stata: Second Class
 Modifying Variables and Subsets and Conditions
 Creating new variables using: gen and replace

 Functions

 Recoding recode, gen, encode, decode

 if , in, sort, drop, keep, save

 Comparison of means and medians: one and two sample


 Comparing means and medians for >2 groups: oneway
anova and Kruskal-Wallis
 Comparing categorical data 2x2, 2x k tables, ordered
categories, RR and OR
Teaching Stata
 Additional weekly lab session was added
 Review

 Learn additional nice features of Stata

 Immediate commands

 Date functions

 Do file

 Graphing
Teaching Stata: Third Class
 File handling:
 use, clear, importing files

 merge

 update

 append

 collapse, reshape

 Correlation & Linear Regression


Teaching Stata: Fourth Class
 Multivariate Analyses:
 Multiple Linear Regression

 Two-way Anova

 Interpreting the outputs


Teaching Stata: Fifth Class
 Statistical power and sample size using
NCSS/PASS software
 Review
Success
 Stata statistical software was easy to teach
 At the end of 6 week summer session
students were successful in learning and
using Stata
Failures
 None!
Recommendations for Next Year
 Use Stata for power and sample size
calculations
 Continue to teach Stata software
 This was a terrific success!
Interesting Problems
 Largest problem encountered was
communicating with mentors and
statisticians unfamiliar with Stata
Student Wishes and Grumbles
 THE BIG GRUMBLE
 Windows do not “hang together”
 Windows were all over the screen

 Windows sometimes got “lost”


Student Wishes and Grumbles
 Wishes - Students would like
 “Un-do” button

 Ability to order variables as desired, and then

return to original ordering


 Description of statistical tests

 As in SPSS where placing the cursor over

the test will give a brief description of the


statistical test
 Data window to remain “open” while running

commands (as in SPSS)


Any Questions?

You might also like