Plot X, Y
Delete Outliers
Yes
A
Fit Linear Model
Is Plot Linear?
G
F
Yes
No
No
Any Outliers?
B
C
Piecewise Linear?
Constant Variance?
Fit Piecewise Model
No
H
WLS or GLS Regression
No
D
Transform X and/or Y
E
Yes
Achieved Linearity?
Normality?
No
I
Try Sp
Practice Questions
1. Which of J.S. Mills canons gives the philosophical basis for experimental design?
2. Which canon calls for equating groups prior to treatment? It was the first approach
used by statisticians to achieve this goal.
3. Systematic design
Some Geometrical Considerations
James H. Steiger
Department of Psychology and Human Development
Vanderbilt University
James H. Steiger (Vanderbilt University)
Some Geometrical Considerations
1 / 20
Some Geometrical Considerations
1
Introduction
2
Projecti
Multiple Regression
James H. Steiger
Department of Psychology and Human Development
Vanderbilt University
James H. Steiger (Vanderbilt University)
Multiple Regression
1 / 19
Multiple Regression
1
The Multiple Regression Model
2
An Example: The U.N. Fertil
Matrix Algebra A Minimal Introduction
James H. Steiger
Department of Psychology and Human Development
Vanderbilt University
James H. Steiger (Vanderbilt University)
Matrix Algebra A Minimal Introduction
1 / 51
Matrix Algebra A Minimal Introduction
1
Why M
The Matrix Algebra of Sample Statistics
James H. Steiger
Department of Psychology and Human Development
Vanderbilt University
James H. Steiger (Vanderbilt University)
The Matrix Algebra of Sample Statistics
1 / 29
The Matrix Algebra of Sample Statistics
1
Random Vectors, Random Matrices, and Their Expected Values
James H. Steiger
Department of Psychology and Human Development
Vanderbilt University
James H. Steiger (Vanderbilt University)
Random Vectors, Random Matrices, and Their Expected Values
1 / 14
Ran
Matrix Operations in R A Minimal Introduction
James H. Steiger
Department of Psychology and Human Development
Vanderbilt University
James H. Steiger (Vanderbilt University)
Matrix Operations in R A Minimal Introduction
1 / 24
Matrix Operations in R A Mini
Regression and the 2-Sample t
James H. Steiger
Department of Psychology and Human Development
Vanderbilt University
James H. Steiger (Vanderbilt University)
Regression and the 2-Sample t
1 / 44
Regression and the 2-Sample t
1
Introduction
2
The 2-Sample S
Investigator-Designed Randomized Experiments
James H. Steiger
Department of Psychology and Human Development
Vanderbilt University
James H. Steiger (Vanderbilt University)
Investigator-Designed Randomized Experiments
1 / 23
Investigator-Designed Randomize
Introduction to Linear Regresson
James H. Steiger
Department of Psychology and Human Development
Vanderbilt University
James H. Steiger (Vanderbilt University)
Introduction to Linear Regresson
1 / 40
Introduction to Linear Regression
1
Fitting a Straight
Constructing and Using Scatterplots
James H. Steiger
Department of Psychology and Human Development
Vanderbilt University
James H. Steiger (Vanderbilt University)
Constructing and Using Scatterplots
1 / 56
Constructing and Using Scatterplots
1
Introductio
Review of Basic Statistical Concepts
James H. Steiger
Department of Psychology and Human Development
Vanderbilt University
James H. Steiger (Vanderbilt University)
Review of Basic Statistical Concepts
1 / 87
Review of Basic Statistical Concepts
1
Introduc
Piecewise Regression
James H. Steiger
Department of Psychology and Human Development
Vanderbilt University
James H. Steiger (Vanderbilt University)
Piecewise Regression
1 / 14
Piecewise Regression
1
Introduction
2
Infant Mortality vs. GNP
3
Electricity Co
The Simple Linear Regression Model
James H. Steiger
Department of Psychology and Human Development
Vanderbilt University
James H. Steiger (Vanderbilt University)
The Simple Linear Regression Model
1 / 49
The Simple Linear Regression Model
1
Introduction
2
Addressing Causality
James H. Steiger
Department of Psychology and Human Development
Vanderbilt University
James H. Steiger (Vanderbilt University)
Addressing Causality
1 / 43
Addressing Causality
1
Introduction
2
Conditions to Strive For
3
The Circle of
True/False Questions
1. The limiting form of the F and chi-square is the normal.
2. The F sampling distribution uses chi-square deviates in its construction.
3. When you estimate the population variance, you should use the normal to test
hypotheses about