__EPSE 596: Correlational Designs & Analysis__

2021/22, Winter Term 1

**Course overview:**

This course will cover all the basics and essentials of linear regression analysis for application to correlational designs. These methods have a long history and enjoy broad application in all of the natural, social, and health sciences. We will discuss theory and applications. Mathematical foundations will form a component of the material, though our main focus will be on practical applications for researchers in a variety of disciplines.

**Recommended textbooks:**

*- Applied Linear Regression Models,*M.H. Kutner, C.J. Nachtscheim, J. Neter. The quintessential reference for linear regression.

*- Introduction to Linear Regression Analysis*, D.C. Montgomery, E.A. Peck, G.G. Vining. A personal favourite.

**Class Notes:**-Week 1: Introduction and Stats Review 1 (random variables, statistics, standard errors, confidence intervals, Central Limit Theorem)

- Extra notes on confounding variables from "Handbook of Biological Statistics" (HBS) by John H. McDonald
- Extra notes on probability from HBS

-Week 2: Stats Review 2 (conditional probability, p-values, Bayes' Theorem, t-tests, F-tests)

- Extra notes on standard error of the mean from HBS
- Extra notes on confidence intervals from HBS
- Extra notes on hypothesis testing from HBS

-Week 3: Data structures and relationships

-Week 4: Simple linear regression basics

-Week 5: Prediction; residual diagnostics; curvature

-Week 6: More curvature; interactions; model building

-Week 7: Categorical predictors

-Week 8: Data transformations

-Week 9: Mediation and moderation

-Week 10: Likelihood theory and maximum likelihood estimation

-Week 11: Tools for assessing model fit/quality

-Week 12: Introduction to logistic regression for binary response data