Statistics Core Workshop


How to Interpret Logistic Regression Coefficients

This class will briefly review the interpretation of regression coefficients in multiple linear regression and then show how regression coefficients in logistic regression are different. The class will alternate didactic material with demonstrations and exercises in the CSSCR computer lab. Students will be shown how to calculate odds ratios, probabilities, and relative risks from the raw logistic regression coefficients using Excel; and will learn how to do logistic regression in Stata including plotting predicted probabilities. Presentation of the results in units meaningful to the research question will be emphasized.

Instructors

Cori Mar
Raitt Hall 218C
cmmar@u.washington.edu
(206) 616-6183

Anita Rocha
Raitt Hall 218D
alrocha@u.washington.edu
(206) 616-6687

Time and Location

Thursday, June 22, 2006 - 9am - 12pm and 1pm - 4pm

Savery Hall Rooms 149 (classroom) and 135 (computer lab)

Pre-requisite

Course work or equivalent experience in multiple linear regression

Registration

https://csde.washington.edu/training/register/


Class Materials

Part 1: Introduction

Lecture Notes

Part 2: Equations

Lecture Notes
Exercise 1
Exercise 1 Answers
Exercise 1 Excel Spreadsheet

Part 3: Example 1 Analysis

Lecture Notes
Exercise 2
Exercise 2 Answers

Part 4: Doing It In Stata

Example 1 Data Set
Stata code

Part 5: Summary

Lecture Notes

Part 6: Post-class Data

Data Set
Questions
   CONTACT CSDE                                         UW LINKS: MyUW | Web Pine | Libraries | Computing | UWired | UW Directory | Campus Maps