Get Higher Performance with Structural Equation Modeling

1
Get Higher Performance with Structural Equation
Modeling

• Tim Daciuk
• David Matheson
• Arik Pelkey
• August 2008

2
Agenda

• What is SEM?
• What is AMOS
• Who uses AMOS?
• How does AMOS fit in?
• What are the benefits?
• Demonstration
• How do I get started?
• QA
• Summary

3
Structural Equation Modeling A Primer and
Simple Example
Structural Equation Modelling
Attend.Bi Temp. B0
Latent
4
What is Structural Equation Modeling (SEM)

• A general, powerful multivariate analysis
  technique
• Includes specialized versions of a number of
  other analysis methods as special cases
• Factor analysis, path analysis and regression all
  represent special cases of SEM.
• SEM largely confirmatory, rather than exploratory
• Usually focuses on latent constructs
• A relatively young field

5
What is SEM?

• Builds on traditional techniques
• Suitable for complex systems
• Models can include latent variables
• Diagnostics

6
Basic Terminology

• Standard Structural Equation
• Path Analytic Equation
• ev effect variable
• pc path coefficient
• sc structural coefficient
• cv causal variable
• p path
• d disturbance

7
Basic Terminology - II

• Standardized Variable
• Variable whose mean is zero and variance is one.
• Latent Variable
• A variable in the model that is not measured.
• Exogenous Variable
• A variable that is not caused by another variable
  in the model.
• Endogenous Variable
• A variable that is caused by one or more variable
  in the model.

8
Basic Terminology - III

• Structural Coefficient
• A measure of the amount of change in the effect
  variable expected given a one unit change in the
  causal variable and no change in any other
  variable.
• Disturbance
• The set of unspecified causes of the effect
  variable.
• Structural Model
• A set of structural equations.
• Path Diagram
• A diagram that pictorially represents a
  structural model.

9
The Traditional SEM Lifecycle

• Model Specification
• Identification
• Estimation
• Testing Fit
• Model Modification or Respecification

10
Process Diagram for Structured Equation Modeling
Equation System
Covariance Structure
Modify or Reject
No
Data Supports Covariance Structure
Gather Data
Yes
Tentative support For equation system
11
What is AMOS

• Our Structural Equation Modeling application
• Latest is 17.0
• Model complex systems
• Model effects that cant be directly measured
• Graphic user interface is easy to use
• Stand alone or used in conjunction with SPSS

12
And Whats New in AMOS 17.0

• Copy and paste a path diagram from one Amos
  Graphics window to another
• Also part of a path diagram
• Convert path diagram to equivalent VB
• Enhanced growth curve
• Automatically constrains parameters
• Appropriate for many growth curve models
• Program changes
• Default value for all groups checkbox
• New PathDiagrammer method, EditPaste
• New PathDiagrammer method, ToolsWriteAProgram
• ImputeNo now called Imputation_

13
Who uses AMOS?

• Social scientists
• Market researchers
• Political analysts
• Psychologists
• Educators
• Physicians

14
Major Applications of SEM

• Causal modeling, or path analysis
• Hypothesizes causal relationships among variables
  and tests the causal models with a linear
  equation system.
• Confirmatory factor analysis
• Extension of factor analysis in which specific
  hypotheses about the structure of the factor
  loadings and intercorrelations are tested
• Second order factor analysis
• Variation of factor analysis in which the
  correlation matrix of the common factors is
  itself factor analyzed to provide second order
  factors

15
Major Applications of SEM

• Regression models
• Extension of linear regression analysis in which
  regression weights may be constrained to be equal
  to each other, or to specified numerical values
• Covariance structure models
• Hypothesize that a covariance matrix has a
  particular form. For example, you can test the
  hypothesis that a set of variables all have equal
  variances
• Correlation structure models
• Hypothesize that a correlation matrix has a
  particular form. A classic example is the
  hypothesis that the correlation matrix has the
  structure of a circumplex

16
Structured Equation Modeling and the Path Diagram

• Y aX E
• All independent variables have arrows pointing to
  the dependent variable.
• The weighting coefficient is placed above the
  arrow.
• Variables are represented in ovals or rectangles.
  
• Manifest variables are placed in boxes in the
  path diagram.
• Latent variables are placed in an oval or circle.

E
a
X
Y
17
Structured Equation Modeling and the Path Diagram

• Y aX E
• For example, the variable E in the diagram can be
  thought of as a linear regression residual when Y
  is predicted from X. Such a residual is not
  observed directly, but calculated from Y and X,
  so we treat it as a latent variable and place it
  in an oval.

E
a
Y
X
18
Demonstration of AMOS
19
What can I do with AMOS?

• Multiple equation models
• Include factors that cant be directly measured
• More options for dealing with missing data

20
How Does AMOS Fit?

• Structural equation modeling is now a well
  accepted technique
• Required in many fields
• Natural progression from regression and factor
  analysis
• Diagnostics similar to traditional methods

21
Questions
?
22
What are the Benefits?

• Realism
• Accuracy
• Graphic representation
• Create models easily
• Options for power users

23
How do I Get Started?

• Demo on same disk as SPSS 17 and modules
• Easy to use graphic interface
• User guide
• Online help
• Training
• Bibliography

24
Introduction to AMOS 17.0

• Technical http//www.spss.com/amos/
• Product literature and white papers
• Recorded seminars
• Citations
• www.amosdevelopment.com
• Amos development Corp.
• Examples, videos, bibliography, links, etc.
• Pricing your rep or sales_at_spss.com
• Tim Daciuk,
• Director, Worldwide Demo Resources
• tdaciuk_at_spss.com
• [phone]