Scribd
Upload
971 views54 pages

Smart PLS

This document provides an overview of topics to be covered in a training on Structure Equation Modeling (SEM) using SmartPLS. The following will be discussed: fundamentals of PLS-SEM, assumptions when using SmartPLS, how to use the SmartPLS interface, basic modeling in PLS, assessing the measurement model through tests of convergent validity, discriminant validity, and factor loadings, assessing the structural model, and how to conduct and interpret mediation and moderation analyses and report findings. Examples will also be provided of how to rename constructs, rotate construct items, run a confirmatory factor analysis, and calculate Q2 for predictive relevance assessment in PLS-SEM using SmartPLS.

Uploaded by

shahidlatifhcc
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
971 views54 pages

Smart PLS

This document provides an overview of topics to be covered in a training on Structure Equation Modeling (SEM) using SmartPLS. The following will be discussed: fundamentals of PLS-SEM, assumptions when using SmartPLS, how to use the SmartPLS interface, basic modeling in PLS, assessing the measurement model through tests of convergent validity, discriminant validity, and factor loadings, assessing the structural model, and how to conduct and interpret mediation and moderation analyses and report findings. Examples will also be provided of how to rename constructs, rotate construct items, run a confirmatory factor analysis, and calculate Q2 for predictive relevance assessment in PLS-SEM using SmartPLS.

Uploaded by

shahidlatifhcc
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 54

Instructor:

Dr. Shazia Kousar


Assistant Professor, Superior University
Lahore
Contents:

•Introduction to Structure Equation Modeling (SEM)


•Fundamentals of PLS-SEM
•Assumptions to use Smart PLS
•Introduction to Smart PLS Interface
•Basic Modeling in PLS
•Measurement Model Assessment (CFA)
Convergent validity
Discriminant Validity (HTMT)
Factor Loadings & Cross Loadings
•Structure Model Assessment (SEM)
•Mediation Analysis
•Moderation Analysis
•Interpretation & Reporting of Findings
1st vs 2nd Generation Technique
Primarily Primarily
exploratory confirmatory
• multiple • correspondence
.
regression analysis
• logistic
.
1st regression
• analysis of
Generation
variance
Techniques • cluster
analysis
• exploratory
factor analvsis
2nd CB-SEM,
including
Generation PLS-SEM
CFA
Techniaues
•
Structure Equation Modeling (SEM):

 SEM is a second-generation statistical modeling technique


that allows the simultaneous modeling of relationships among
multiple independent and dependent constructs (Gefen,
Straub, & Boudreau, 2000).

SEM is used to explore or confirm the theory


Structure
Equation
Modelin
g

PLS-SEM
CB-
(Variance
Based
Based SEM)
SEM
PLS-SEM is used to
develop theories and
CB-Based SEM used to usually for exploratory
confirm or reject studies
theories
Software's for CB-Based Software's for PLS-
SEM SEM

AMOS SMART PLS


EQS PLS-GUI
SEPATH PLS Graph
LISREL WarpPLS
MPLUS Visual PLS
lavaan SPAD-PLS
Ωnyx ADANCO
Rules of Thumb for choosing PLS-SEM

We prefer PLS-SEM technique when:


The sample size is small.
The data are not normally distributed
The structural model is complex (many constructs and many
indicators)
We have more than one underpinning theories to support model
The goal is to predicting key target constructs.
The purpose is to build or predict theory, not to test the theory
The formative constructs are the part of structural model
Smart PLS

 A structural equation modeling (SEM) process


requires
two steps:

1. Building and testing a measurement model.


2. Building and testing a structural model.
Modeling in PLS
Inner Model

Outer Model Outer Model


Exogenous Endogenous
Measurement Model
We use CFA technique in measurement model

Measurement model is used to measure the construct validity

Construct
Validity

Co
nve Discr
rgent imi
nant
Validity Validity
•Factor loadings. •Fornell-Larcker Criterion
•Average Variance extracted •HTMT (Heterotrait-Monotrait
(AVE).
Ratio)
•Reliability.
Validation of the measurement models is a
requirement for assessing the structural model
Exploratory Factor Analysis (EFA)

Vs

Confirmatory Factor Analysis (CFA)


Note:
Independent variable is also known as Exogenous Variables in
PLS
 Dependent variable is also known as Endogenous Variables in
PLS
Before going to start work in PLS, we need to save our SPSS
file in
CSV format.
After saving your file in CSV format, next step is to attach
CSV file with your Smart PLS.
Abbreviation used in SPSS:

Adv Advertisement
BI Brand Image
SP Sales Promotion
PI Purchase Intentions
How to
Rename Construct:

How to Rotate Construct Items:


How to Run CFA
For Measurement Model
 HTMT value near to 1 indicates the lack of discriminant validity.

Threshold values for HTMT:

< 0.85 (kline, 2011)


<0.90 (Gold et al., 2001)
HTMT Inference (Use Bootstrapping )
Structure Equation Modeling (SEM)
Structure Equation Modeling (SEM)
According to Barron and Kenny (1986)

According to Hayes (2008), Hayes (2013)


Mediation
Effect Size:
0.02……...Small
0.15……. Medium
0.35……..Large
How to Calculate Q2 (Blinfolding)

Note:
If Q2 >0, no issue with blind folding
Blind folding use for Predictive relevance (Stone, 1974;
Geisser,
1974).
Blindfolding is only applicable for latent variable and reflective
scale
measurement.
Reporting of
Results
SEM Results

You might also like