USP 654 Data Analysis II
This course takes an applied approach to statistical analysis and research methodology and is the second in a two-course sequence. The goal is to provide students with statistical background, conceptual understanding, technical writing skills, computer application, and the ability to apply these skills to realistic data analysis problems and research designs. Topics include simple regression and correlation, multiple regression, and logistic regression. The laboratory (USP 654L) must be taken concurrently. Recommended prerequisite: USP 634 Data Analysis or an equivalent course approved by the instructor and prior experience with statistical software.
All download files pdf unless otherwise noted.
Fall 2014 Syllabus
Click here for the class syllabus, reading list, and my contact information.
Click here to download readings in a zip file (9.1mb).
Homework Data Sets
Get the data you need here.
Final exam review sheet
Handouts and Overheads
(sorry, not available until it has been covered in class, not all overheads included)
Simultaneous Regression: SPSS Example
Hierarchical Regression: SPSS Example
Model Building Procedures
Overhead: Good Hierarchical Table Example
Overhead: Multiple Regression Venn Diagram
Overhead: Partial and Semi-partial Correlation
Partial and Semi-partial Correlation Example
Overhead: Suppression Illustration
Suppression: SPSS Example
Coding of Categorical Predictors and ANCOVA
Creating Indicator Variables for Four Categories
ANOVA and Regression Equivalence: Gender and Depression SPSS Example
Remedies for Assumption Violations
Overhead: Diagnostic Plots
Curvilinear SPSS Example
Odds Ratio Computation
Link Functions and the Generalized Linear Model
Regression Models for Ordinal Dependent Variables
Ordinal Logit and Probit Examples
Regression Models for Count Data
SPSS Computer Lab
Handouts and data, lab instructor: Ryan Eddings
Over 20 "Web lectures" on introductory graduate statistics
Automatic calculator for calculating
significance of indirect effects
(compliments of Preacher & Leonardelli at KU and U of Toronto)
Macros for calculating significance of indirect effects
(compliments of Hayes and Preacher at Ohio State & KU)