SPSS AMOS Training β Structural Equation Modelling (SEM) in Kenya
SPSS AMOS is the leading software for Structural Equation Modelling β a powerful statistical technique that allows researchers to test complex theoretical models involving both observed and unobserved (latent) variables simultaneously. Our expert-led, hands-on training takes you from the fundamentals of SEM and path diagrams through to Confirmatory Factor Analysis, model fit assessment, mediation, and moderation analysis.
Course Details
Certificate awarded on completion
What is Structural Equation Modelling & Why Does It Matter?
Structural Equation Modelling (SEM) is a powerful statistical technique used to test and estimate causal relationships using a combination of statistical data and theoretical assumptions. Unlike standard regression, SEM allows researchers to simultaneously examine multiple dependent and independent variables β including variables that cannot be directly observed.
The use of SEM is commonly justified in the social sciences because of its ability to impute relationships between unobserved constructs (latent variables) from observable variables. To provide a simple example: the concept of human intelligence cannot be measured directly as one could measure height or weight. Instead, researchers develop a hypothesis of intelligence and write measurement instruments designed to measure it β then use SEM to test that hypothesis using data gathered from respondents. In this case, “intelligence” is the latent variable and the test items are the observed variables.
SPSS AMOS (Analysis of Moment Structures) is the leading software tool for SEM. It offers a user-friendly graphical interface allowing users to draw path diagrams and specify models visually β making it far more accessible than other SEM tools, while remaining highly rigorous in its analytical output.
Latent Variables
Unobservable constructs inferred from observed data β such as customer satisfaction, motivation, or organizational commitment. Drawn as ovals in AMOS path diagrams.
Observed Variables
Directly measured variables from your questionnaire or dataset β individual survey items or scale scores. Drawn as rectangles in AMOS path diagrams.
Path Coefficients
Represent the strength and direction of relationships between variables. Used to test whether hypothesized relationships are statistically significant.
Measurement vs Structural Model
The measurement model links latent variables to their observed indicators (CFA). The structural model tests hypothesized relationships between latent constructs.
| Feature | Standard Regression | SEM with AMOS β |
|---|---|---|
| Latent variables | Not supported | Fully supported |
| Multiple outcomes | One at a time | Simultaneous |
| Measurement error | Not accounted for | Explicitly modelled |
| Mediation testing | Approximate | Full bootstrap |
| Model fit assessment | RΒ² only | Multiple indices |
What You Will Achieve
By the end of this training programme, participants will have the practical skills to apply SPSS AMOS confidently in their own research projects and academic work.
Apply the SPSS AMOS software for data analysis in research problems at school, university, and in the workplace β using Structural Equation Modelling to test complex theoretical frameworks.
Apply SPSS AMOS and complementary software in academic projects, dissertations, and thesis work β from building path diagrams and running CFA to reporting results in APA format for publication.
Test and validate measurement models β including confirmatory factor analysis, construct reliability (Cronbach Alpha, AVE, CR), convergent validity, and discriminant validity using AMOS.
Run and interpret full structural models β testing path coefficients, hypothesis significance, and producing publication-ready outputs including mediation and moderation analyses.
Training Mode & Approach
Class-Based Training
Structured, expert-led sessions β either online via video call or in person at our Nairobi offices β with clear step-by-step instruction.
Practical AMOS Use
Every session is hands-on. You will build real path diagrams, run live analyses, and interpret actual AMOS output β not just watch demonstrations.
Take-Away Assignments
Each module includes practical take-away assignments so you can consolidate your learning and apply SEM techniques to your own research data.
One-on-One Sessions
Individual training sessions are available on request β ideal if your research has unique data structures or you need personalised guidance on your specific thesis model.
The Three-Module Learning Path
Each module builds progressively β from SEM fundamentals through measurement model validation to advanced structural analysis and testing.
Introduction to SEM & AMOS Basics
Understanding SEM concepts, latent vs observed variables, the AMOS graphical interface, drawing path diagrams, data preparation, and EFA vs CFA.
Measurement Model & CFA
Building and running a CFA, assessing model fit (CFI, RMSEA, SRMR, GFI), reliability and validity, modification indices, and improving model fit.
Full Structural Model, Mediation & Moderation
Running the full structural model, hypothesis testing, bootstrapping for mediation, moderation analysis, multi-group analysis, and reporting results.
Detailed Module Content
Every topic is covered hands-on in a live AMOS session. Each module is Ksh 5,000 per individual and runs for 5 hours. Modules can be taken individually or as a complete programme.
Introduction to SEM & AMOS Basics
Concepts Β· Graphical interface Β· Data preparation Β· Path diagrams Β· EFA vs CFA
Introduction to Structural Equation Modelling
The AMOS Graphical Interface
Data Preparation for SEM
EFA vs CFA & Model Identification
Measurement Model β CFA, Fit & Validity
Confirmatory Factor Analysis Β· Model fit indices Β· Reliability Β· Convergent & discriminant validity
Confirmatory Factor Analysis (CFA)
Model Fit Assessment
Reliability & Validity Testing
Second-Order CFA
Full Structural Model, Mediation & Moderation
Structural model Β· Hypothesis testing Β· Mediation Β· Moderation Β· Multi-group analysis
Full Structural Model Analysis
Mediation Analysis
Moderation Analysis
Reporting & Publication
SEM Model Fit Indices β What You Will Learn to Interpret
One of the most critical skills in SEM is correctly assessing and reporting model fit. Our training covers all major fit indices and their acceptable thresholds.
| Fit Index | What It Measures | Acceptable Threshold | Interpretation |
|---|---|---|---|
| ΟΒ²/df | Chi-square to degrees of freedom ratio β overall model-data discrepancy | β€ 3.0 | Values close to 1 indicate excellent fit; values > 5 suggest poor fit |
| CFI | Comparative Fit Index β compares model to a null/independence model | β₯ 0.90 | Values closer to 1.0 indicate better fit; β₯ 0.95 is excellent |
| TLI | Tucker-Lewis Index β penalises model complexity | β₯ 0.90 | More conservative than CFI; values above 1.0 suggest over-fitting |
| RMSEA | Root Mean Square Error of Approximation β parsimony-adjusted fit | β€ 0.08 | β€ 0.05 = close fit; 0.05β0.08 = reasonable; > 0.10 = poor fit |
| SRMR | Standardised Root Mean Square Residual β average residual correlation | β€ 0.05 | Values close to zero indicate better fit between model and data |
| GFI | Goodness of Fit Index β proportion of variance explained by the model | β₯ 0.90 | Values range from 0 to 1; values close to 1 indicate good fit |
| AVE | Average Variance Extracted β convergent validity of each construct | β₯ 0.50 | Indicates that the construct explains more than half the variance of its indicators |
| CR | Composite Reliability β internal consistency of latent constructs | β₯ 0.70 | Preferred reliability measure over Cronbach Alpha in SEM |
Key AMOS Capabilities You Will Master
IBM SPSS AMOS is the most accessible and widely used SEM software in social science and business research. Here are the core capabilities covered across all three modules.
Graphical Path Diagram Builder
Draw and specify complex SEM models visually using AMOS’s intuitive drag-and-drop interface β no programming required. Build full models with latent and observed variables in minutes.
Confirmatory Factor Analysis
Validate your measurement model by testing whether your survey items reliably measure the underlying constructs you theorised β with full reliability and validity statistics.
Multiple Model Fit Indices
AMOS outputs all major fit indices simultaneously β CFI, TLI, RMSEA, SRMR, GFI β giving you a comprehensive picture of how well your theoretical model matches the data.
Mediation & Bootstrapping
Test complex mediation pathways with confidence using AMOS’s built-in bootstrapping β producing bias-corrected confidence intervals for indirect effects required by journals.
Multi-Group Analysis
Test whether your structural model holds equally across different groups (gender, age, organisation type) using multi-group analysis and the Critical Ratio for group differences.
Publication-Ready Output
AMOS integrates directly with Word, Excel, and PowerPoint (OLE) β allowing you to copy tables and path diagrams directly into your thesis or research paper.
Who Should Attend SPSS AMOS Training?
SEM with AMOS is particularly suited to researchers in business, psychology, education, health, and social sciences who work with survey-based questionnaire data and theoretical frameworks involving latent constructs.
Masters & PhD Students in Business
If your thesis involves constructs like employee motivation, customer satisfaction, organisational commitment, or brand loyalty β SEM with AMOS is the required analytical tool. Our training covers exactly what you need for Chapters 4 and 5.
Psychology & Behavioural Researchers
SEM was developed specifically for testing psychological constructs that cannot be directly observed. AMOS is the standard tool for psychometric validation in behavioural and social science research.
Public Health & Nursing Researchers
Health outcomes, quality of life, patient satisfaction, and behavioural intention are all latent constructs that require SEM for rigorous analysis. AMOS is increasingly required in health research dissertations.
Education Researchers
Student engagement, learning motivation, academic self-efficacy β education research commonly involves latent constructs measured through questionnaire items, making SEM the ideal analytical approach.
Researchers Using Likert-Scale Questionnaires
If your data collection instrument uses Likert scales and your study involves theoretical constructs, SEM with CFA is the appropriate analytical technique β rather than standard regression which ignores measurement error.
SPSS Users Ready to Advance
Already comfortable with SPSS for basic analysis? AMOS is a natural progression β it uses the same data files and works within the SPSS ecosystem, making the transition straightforward with our step-by-step guidance.
Frequently Asked Questions
Everything you need to know before enrolling in SPSS AMOS training.
Book Your SPSS AMOS Training Session
Fill in our booking form or contact us directly by phone or WhatsApp. Our team is available Monday to Saturday to confirm your slot and answer any questions about the course content.
Have Questions? Contact Us Directly
We are happy to help you decide whether SEM is the right approach for your research and which module best fits your needs.
