This module provides a postgraduate-level foundational course in multivariate modelling, with a particular focus on applications in psychology, public health and education. It is intended primarily for students of psychology and any other field of social science where relationships between multiple observations on humans and other subjects are of interest, but also for students with a mathematics/statistics background interested in such applications. With foundations taught in the pre-requisite module, PSYC8013 Psychometrics, 'measurement by modelling' is formally introduced using a Structural Equation Modelling (SEM) framework. Within this framework, specific techniques such as path analysis, confirmatory factor analysis, basic longitudinal analysis and multiple-group analysis are taught. Data analysis applications from various fields of psychological studies are considered, and it is demonstrated how to model/test statistically complex phenomena such as spurious and indirect effects, growth and change, measurement invariance, and others.
Total contact hours: 30
Total private study hours: 70
Total module study hours: 100
Compulsory to:
MSc Organisational and Business Psychology
MSc Cognitive Psychology/Neuropsychology
MSc Developmental Psychology
MSc Forensic Psychology
MSc Political Psychology
MSc Social Psychology
Also compulsory on Psychology Postgraduate Research Courses.
Also available as an elective module.
Main assessment methods
20% Online Test (10 x 15 minutes in workshop)
80% Exam (2 hours)*
*This element is pass compulsory and must be passed to achieve the learning outcomes of the module
Reassessment methods:
100% Exam
The University is committed to ensuring that core reading materials are in accessible electronic format in line with the Kent Inclusive Practices.
The most up to date reading list for each module can be found on the university's reading list pages.
1. Demonstrate a comprehensive understanding of statistical modelling by the means of structural equations, and fitting structural equation models to observed data;
2. Demonstrate a comprehensive understanding of the principles of inferring measurement on unobserved psychological constructs from observed data; the notions of the latent variable and the error of measurement;
3 Demonstrate a comprehensive understanding of the principles and techniques for establishing measurement invariance when comparing measurements across time and/or across populations;
4. Critically evaluate the appropriateness of particular multivariate models to research design and data;
5. Use appropriate statistical software to manage data, specify and test a variety of path, measurement and structural models, applying principles learnt;
6. Interpret and critically evaluate results of model fitting and outputs of statistical software; make inferences from the results in applied settings;
7 Understand and critically evaluate methods used and results of multivariate modelling reported across a variety of psychological literatures (e.g., social, forensic, clinical).
The intended generic learning outcomes.
1. Appreciate and critically discuss theoretical positions and controversies related to Structural Equation Modelling (SEM), for example regarding causality;
2. Demonstrate an appreciation of the diverse range of SEM applications and their relevance to student's field of study or interest, and behavioural/social sciences more broadly;
3. Acquire or improve competence in the use of statistical software to manage and code data, and to conduct multivariate analyses for a range of applications.
University of Kent makes every effort to ensure that module information is accurate for the relevant academic session and to provide educational services as described. However, courses, services and other matters may be subject to change. Please read our full disclaimer.
