Fundamentals of

AMOS
By
Imran Khan
 Contents
What is AMOS?
Why we use AMOS?
Structural Equation Modeling
Comparison of AMOS with other Statistical Software
AMOS Interface
Installation Procedure
Familiarity with AMOS Interface
Factor Analysis
Exploratory Factor Analysis (EFA)
Confirmatory Factor Analysis by using AMOS
Practical Exposure
 What is AMOS?

• Amos (Analysis of Moment Structures) is an IBM SPSS

Statistics module designed for the analysis of covariance
structure models, including structural equation modeling
(SEM), path analysis, and confirmatory factor analysis
(Barnidge & zúñiga, 2017).
 Structural Equation Modeling
• Structural Equation Modeling (SEM) is a statistical technique
designed to analyze relationships among variables,
combining aspects of factor analysis and multiple regression.
• Purpose
• To test theoretical models that specify complex causal
relationships between observed and latent (unmeasured)
variables.
• To evaluate how well their data fits these hypothesized
models.
 Uses of SEM
Understanding Complex Relationships:
• To explore and confirm the relationships between variables, including direct,
indirect, and mediating effects.
• It accounts for measurement errors, providing more accurate and reliable
results than simpler methods like multiple regression.
Testing Theoretical Models:
• test theories that hypothesize relationships between constructs.
Combining Measurement and Structural Components:
• SEM integrates measurement models (e.g., factor analysis) with structural
models (e.g., path analysis).
 Use of SEM

Comparing Alternative Models:

• SEM can compare competing theoretical models to determine which best
fits the data.
• It supports hypothesis testing and model refinement.
Evaluating Mediation and Moderation:
• SEM can test mediation and moderation
Longitudinal and Multigroup Analysis:
• SEM can analyze relationships over time (longitudinal studies) or compare
models across groups (e.g., different demographic groups).
 Path Analysis

• Path analysis is a statistical technique used to examine

direct and indirect relationships among a set of variables
within a hypothesized causal framework.
 Path Diagram

• A visual representation of the hypothesized relationships

among variables.
• Variables are depicted as nodes (circles or rectangles), and
causal relationships are shown using arrows.
• Single-headed arrows represent causal relationships (e.g., A → B).
• Double-headed arrows represent correlations (e.g., A B).
 Why we Use AMOS

•User-Friendly Interface:
•Structural Equation Modeling (SEM):
•AMOS specializes in SEM, allowing users to test relationships between
observed and latent variables.
Path Analysis:
explore direct, indirect, and total effects among variables in a
hypothesized model.
 Why We Use AMOS

•Latent Variable Modeling:

•AMOS handles latent variables
•Model Fit Assessment:
•AMOS provides a variety of goodness-of-fit indices (e.g., Chi-square,
RMSEA, CFI, TLI)
•Visualization:
•AMOS generates publication-ready path diagrams, making it easier to
communicate results in academic and professional settings..
•Compatibility:
•Since AMOS integrates with SPSS, it can directly import SPSS data files,
simplifying the workflow for researchers already using SPSS.
Comparison of AMOS with other Statistical Software

Feature AMOS Mplus

Graphical, user-friendly, Syntax-based, with some
User Interface
drag-and-drop GUI tools
Easy to learn for More complex, but
Complexity
beginners extremely powerful
Latent Variables Well-supported Well-supported
Excellent support (e.g.,
Categorical Data Limited support
binary, ordinal)
Strong multilevel
Multilevel Modeling Limited
modeling features
Monte Carlo
Basic Advanced
Simulations
Group Comparisons Yes Yes
Expensive standalone
Cost Included in SPSS suite
software
 C o m p a ris o n o fA M O S w ith o th e rS ta tis tic a lS o ftw a re

Feature AMOS R (lavaan)

Graphical, drag-and-
User Interface Syntax-based
drop
Highly flexible and
Flexibility Limited to SEM
customizable
Cost Paid (part of SPSS) Free and open-source

Latent Variables Yes Yes

Categorical Data Limited Good support

Model Fit Indices Extensive Extensive

Steeper, requires
Learning Curve Easy for beginners
coding skills
 Comparison of AMOS with other Statistical
Software
Feature AMOS LISREL

Command-line and
User Interface Graphical
syntax-based

Highly flexible and

Flexibility Limited to SEM
supports SEM, CFA

Ease of Use Beginner-friendly Advanced, less intuitive

Categorical Data Limited Good support

Advanced Modeling Limited Stronger capabilities

One of the earliest SEM

History Newer
tools
 Comparison of AMOS with other Statistical Software

Feature AMOS Stata

Syntax-based, with GUI

User Interface Graphical
options

Supports SEM plus other

SEM Capabilities Focused on SEM
statistical analyses

Flexibility Limited Extremely flexible

Excellent support for all

Data Types Primarily continuous
types

Multilevel Models Limited Strong

Cost Paid (SPSS module) Paid (but standalone)

Installation Process
AMOS Interface
AMOS Interface
 Factor Analysis

Factor Analysis includes

I. Exploratory Factor Analysis
II. Confirmatory Factor Analysis
 Exploratory Factor Analysis (EFA)

• In EFA, researcher is attempting to

ε V1
explore the relationships among items Factor
to determine if the 1
ε V2
• items can be grouped into a smaller
number of underlying factors.
ε V3
Factor
2
• In this analysis, all items are assumed to ε V4
be related to all factors.
 Exploratory Factor Analysis (EFA)

• Measured Variables or Indicators:

ε V1
Factor
1
• These variables are those that the ε V2
researcher has observed or measured.
ε V3
Factor
• In this example, they are the four items 2
on the scale. ε V4

• Note, they are drawn as rectangles or

squares.
 Exploratory Factor Analysis (EFA)

• Unmeasured or Latent Variables:

• These variables are not directly measurable, ε V1

rather the researcher only has indicators of Factor
these measures. 1
ε V2
• These variables are more often the more
interesting, but more difficult variables
ε V3
• to measure (e.g., self-efficacy). Factor
2
ε V4
• In this example, the latent variables are the
two factors.
• Note, they are drawn as eclipses
 Exploratory Factor Analysis (EFA)
• Factor Loadings:

• Measure the relationship between the items and the factors.

• Factor loadings can be interpreted like correlation coefficients;

• ranging between -1.0 and +1.0.

• The closer the value is to 1.0

• positive or negative, the stronger the relationship between the
factor and the item.

• Loadings can be both positive or negative.

 Exploratory Factor Analysis (EFA)

• Errors in Measurement:
• Each of the indicator variables ε V1
Factor
has some error in 1
V2
measurement. ε

• The small circles with the ε ε V3

Factor
indicate the error. 2
V4
• The error is composed of 'we ε

know not what' or are not

measured directly.
 Basic Assumption of EFA

Kaiser-Meyer-Olkin (KMO)
KMO value measure the sample Adequacy. KMO varies from 0
to 1 and should be .60 or higher to proceed (can use .50 more
lenient cut-off).
• High values (close to 1.0) generally indicate that a factor
analysis may be useful with your data.
• If the value is less than .50, the results of the factor analysis
probably won't be very useful.
 KMO Cut-off Values
• Marvelous - - - - - - .90s
• Meritorious - - - - - .80s
• Middling - - - - - - - .70s
• Mediocre - - - - - - - .60s
• Miserable - - - - - - .50s
• Unacceptable - - - below .50
 Bartlett Test of Sphericity
• Bartlett's sphericity test provides information about whether
the correlations in the data are strong enough to use a
dimension-reduction technique such as principal components
or common factor analysis
 Eigenvalues
• Eigenvalues are a measure of the amount of variance
accounted for by a factor, and so they can be useful in
determining the number of factors that we need to extract.
• In a scree plot, we simply plot the eigenvalues for all of our
factors, and then look to see where they drop off sharply.
• If Eigenvalues is greater than one, we should consider that a
factor and if Eigenvalues is less than one, then we should not
consider that a factor.
• According to the variance extraction rule, it should be more
than 0.7. If variance is less than 0.7, then we should not
consider that a factor.
 Communality

• Ideal: Communalities above 0.6 are considered ideal

• Acceptable: Values between 0.25 and 0.4 are often
considered acceptable
• Low: Communalities below 0.2 are typically considered for
removal
 Scree Plot
• In multivariate statistics, a scree plot is a line plot of the
eigenvalues of factors or principal components in an analysis.
• The scree plot is used to determine the number of factors to
retain in an exploratory factor analysis (EFA) or principal
components to keep in a principal component analysis (PCA).

Scree Plot
4

2
Eigenvalue

0
1 2 3 4 5 6 7 8 9 10 11 12

Factor Number
 Validity

• Face Validity (do they make sense?)

• Pattern Matrix
• Convergent (high loadings >.70)
• Discriminant (no cross-loadings)

• Factor Correlations
• ≤.7 is better
Steps for EFA
Steps for EFA
Confirmatory Factor Analysis
(CFA)
Model Fit Values
Measurement of Reliability and
Validity
 Methods

• Through Excel Sheet

• Through Amos Plugin
• Through Manual Method
 Composite Reliability(CR) & Average
Variance Extracted (AVE)
• A composite Reliability (CR) of 0.70 is recommended (Hair,1997)
• Fornell and Larker(1981) recommended a CR value of 0.60 or more
• Fornell and Larker(1981) recommended an AVE greater than 0.50
• Note
• AVE measure convergent Validity (Unifactor)
• CR measure internal consistency
• MSV Maximum Shared Variance discriminant validity (Subscales)
Manual Method of AVE
Manual Method of AVE