A Short Course In Inferential Statistics and Multiple Regression

All 4 days only R11,500.00 (Excl. VAT) per delegate

Mon 12th – Thu 15th Mar | Mon 27th – Thu 30th Aug (4 days)  

Johannesburg, South Africa

About this course

Every researcher wants to ensure that the associations he observes are valid i.e. not due to confounders or bias. Multiple regression models are used to determine risk factors and predictors, adjusting for confounders. Therefore researchers and programme managers should be able to understand and interpret multiple regression methods and outputs.

This course teaches intermediate to advanced statistics using Stata. Participants will be able to conduct inferential statistics and estimate measures of effect. They will be able to build causal or predictive statistical models, and interpret outputs.

They will be able to adjust for confounders and estimate adjusted odds ratios (or other measures of effect).

Course Content

Day 1
Overview of study design and descriptive statistics
Bi-variate analysis
Measures of effect
Is there an association/hypothesis testing
Which statistical test to use?
Is there a difference?
Isthedifferencestatisticalsignificant?P-values and 95%C

Day 2
Is the association confounded?
Bias, confounding and interaction
How to control for confounders: design and analysis
From stratified analysis to regression models
Unadjusted and adjusted odds ratios
Introducing regression models; causal and predictive
Generalised linear models
Uses of regression models

Day 3
Logistic regression models for binary outcomes
Odds, log odds, and logistic regression
Application, variable selection techniques, fitting the model
Interpretation of output and regression diagnostics

Day 4
Linear regression models for numerical outcomes
T-test, ANOVA, linear regression
Application, variable selection, fitting the model,
Centering, transformation
Interpretation of output and regression diagnostics

Objectives of this course

Bivariate hypothesis testing.
Understanding validity: bias and confounding
Use of regression models to control for confounding.
Logistic and liAnear regression models.
Univariate (unadjusted) and multivariate (adjusted) models

Building regression models using Stata
Variable selection techniques
Model assumptions
Understanding and interpreting outputs

Who Should attend
  • Researchers
  • Biostatisticians
  • Research Analysts
  • Data analysts
  • Economists
  • HODs
  • Clinicians
  • Epidemiologists
  • Programme managers
  • Postgraduate students
  • Market Researchers
  • Clinical and Medical researchers
  • Scientists
  • Government Practitioners

For more details about our services contact:

Rose More
eMail: rose.more@cesar-africa.com
Tel: +27 11 403 1411 / +27 72 509 1861

Price Includes
  • Course attendance
  • Full refreshments: lunch
  • Welcome tea
  • Two breaks for tea including pastries
  • Course lecture notes and training manual
  • Complimentary parking
  • Certificate of attendance