SEM Analysis SMARTPLS

Download as pdf or txt
Download as pdf or txt
You are on page 1of 21

SEM

Analysis
SMART PLS
Converting Data
• Raw data to be converted into Excel / SPSS format requirement

• Smart PLS data analysis is via Excel csv file

• After data conversion – setup SPSS to input the data for respondent profile analysis
SPSS – Respondent Profile
• Analyze / Descriptive Statistics / Frequency
• Statistics – Mean / Median / Mode / Std Deviation
• Charts – Bar/ Pie / Histogram, Chart Values – Percentage
• Display Frequency Tables
• OK
Common Method Variance /
Common Method Bias - SPSS
Harman's Single Factor Test
▪ Analyze / Dimension Reduction / Factors
▪ We are conducting Exploratory Factor Analysis [EFA]
▪ Input all variables
▪ Extraction / method: principle axis factoring / Fixed number of factors: 1
▪ Rotation / No rotation / OK
Common Method Variance /
Common Method Bias - SPSS
Harman's Single Factor Test
▪ Results
▪ Total Variance Explained Table /% of Variance
▪ If the result is not more than 50%, we can conclude that there is no
common method variance
Confirmatory Factor Analysis
(SPSS & AMOS)
▪ Run Factor Analysis on SPSS on the data

▪ Analyze / Dimension Reduction/ Factor / Input the variables

▪ Descriptive: Initial Solution / Correlation Matrix: KMO and Barlett's test / Reproduced/

▪ Extraction/ Principal Components / Correlation matrix / Unrotated factor solution / Based on


Eigenvalues greater than 1

▪ Rotation: Promax

▪ Options: Exclude cases listwise / suppress small coefficients 0.4


Go to AMOS to calculate CFA
Run AMOS

▪ Plugins / Pattern Matrix Builder / Pattern Matrix Input

▪ Click Data File Icon / File Name/ Select File / Ok

▪ Copy Pattern Maxtrix from SPSS into the AMOS Pattern Matrix Input /Create Diagram/

▪ Click Standardized Estimates / Calculate Estimates

▪ View Text Output


AMOS - CFA
▪ CFA Results

▪ modelling all items as the


indicators of a single factor
indicated a poor fit
(CMIN/DF=5.846, p=0.000). In
conclusion, this study does not
suffer from the CMV problem.
▪ The suggested acceptable value for
relative chi-square, CMIN/DF should
be as high as 5 which are used to
reduce dependency on sample size.
SmartPLS
▪ Click New Project

▪ Name the Project

▪ Double click to import the data

▪ Double click Brightspace Motivation

▪ Drag indicators to build the path

▪ Drag the connect line to link the variables / variable will turn blue if correctly connected

▪ Rename all your variables


Smart PLS Calculation
▪ Calculate PLS Algorithm / Start calculations

▪ From results, check Construct Reliability and Validity

▪ Construct Reliability should be above 0.7

▪ AVE should be greater than 0.5


Path Coefficient Analysis
▪ Go back to first calculations from
PLS Algorithm

▪ Look for Collinearity Statistics (VIF)

▪ Results rule of thumb: VIF should


be 5 or lower for no collinearity
problem
Assessing Discriminant Validity:
HTMT test
▪ Calculate / Consistent PLS Algorithm

▪ Setup – checked connect LVs for intial calculations and click start

▪ Go to Quality Criteria at the bottom / Discriminant Validity / click Heterotrait-Monotrait Tab

▪ HTMT should be lower than 0.9 to indicate no discriminant validity


Checking Structural Path
Significance - Bootstrapping
▪ Calculate bootstrapping

▪ Subsamples 5000 / Start Calculations

▪ Results

▪ Path coefficient – T statistics should be larger than 1.96


SmartPLS Reporting
▪ Common Method Variance (CMV)

▪ Assessment of Measurement Model

▪ Assessment of Structural Model


▪ Assessment of Coefficient of Determination, R2
▪ Level of Path Coefficients Analysis

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy