2019/20 Taught Postgraduate Module Catalogue
EPIB5046M Latent Variable Methods
15 creditsClass Size: 21
Module manager: Dr Richard Feltbower
Email: r.g.feltbower@leeds.ac.uk
Taught: Semester 2 (Jan to Jun) View Timetable
Year running 2019/20
Pre-requisite qualifications
Academic entry requirementsA 1st degree in a quantitative or scientific subject area with substantial mathematical, statistical or numeracy components (at least 2:1). We also consider working experience (two years or more) of research in a quantitative subject area. Non-graduates who: have successfully completed three years of a UK medical degree; are normally ranked in the top 50% of the year 3 cohort; and wish to take the Health Data Analytics MSc as an intercalated programme, will also be accepted.
English language requirements
An overall score of 7.0 on IELTS (International English Language Testing System) with at least 6.0 in writing and no other skill below 6.5; from a TOEFL paper-based test the requirement is a minimum score of 600, with 4.5 in the Test of Written English (TWE); from a TOEFL computer-based test the requirement is a minimum score of 250, with 4.5 TWE; from a TOEFL Internet-based test the requirement is a minimum score of 100, with 25 in the "Writing Skills" score.
This module is mutually exclusive with
EPIB5025M | Multilevel and Latent variable Modelling |
Module replaces
EPIB5025M Multilevel and Latent Variable ModellingThis module is not approved as an Elective
Objectives
The objectives of this module are to:- introduce a range of advanced modelling techniques (multilevel, structural equation and latent variable), to analyse observational data with a complex structure;
- consider diagnostics of model fit;
- develop an understanding of latent variables underlining the observed variables in the data collected;
- develop causal thinking in observational research, via path diagrams, to understand the roles of latent variables in the causal relationships between observed variables;
- introduce appropriate software for multilevel and latent variable modelling;
- interpret results and draw appropriate inferences;
- develop qualities that are appropriate to future responsibilities to colleagues and society in general. As such, aim to develop a professional attitude towards health data analytics.
Learning outcomes
By the end of this module the student should be able to:
- define appropriate multilevel models
- assess model diagnostics to consider 'acceptability' of a model
- explain the statistical concepts of principal component analysis, exploratory factor analysis, confirmatory factor analysis, path analysis, and structural equation modelling
- illustrate the methods of multilevel and latent variable modelling using appropriate examples
- implement multilevel and latent variable modelling using statistical software to correctly specify and fit the statistical model
- interpret output in order to draw appropriate inferences.
Transferable skills
By the end of this module the student should be able to:
- access and use web based information to complete formative assessments on-line
- undertake on-line tasks, posting work.
Syllabus
The module will be delivered over 10 weeks, as a blend of face-to-face small group work and lectures, practical exercises, online written material, online formative assessments, vodcasts (audio-visual presentations), answers and feedback, presentation and an online discussion forum.
The course will cover the following subjects:
- Introduction to multilevel modelling (MLM), simple hierarchies, assumptions and consequence of ignoring hierarchy
- Model specification, variance components (VC) model, random intercept, complex level 1, random slope, complex random slope; Model fit,
- Residuals, diagnostics, model comparison, predictions
- MCMC
- Modelling other outcome distributions, binomial
- Complex hierarchies, cross-classified, multiple-membership
- Introduction to latent variables, principle components analysis (PCA)
- Exploratory (EFA) and confirmatory factor analysis (CFA)
- Causation and path analysis
- Structural equation models (SEM)
- Appropriate statistical software.
Teaching methods
Delivery type | Number | Length hours | Student hours |
Lecture | 5 | 1.00 | 5.00 |
Lecture | 5 | 2.00 | 10.00 |
Practical | 5 | 1.00 | 5.00 |
Practical | 5 | 2.00 | 10.00 |
Independent online learning hours | 30.00 | ||
Private study hours | 90.00 | ||
Total Contact hours | 30.00 | ||
Total hours (100hr per 10 credits) | 150.00 |
Private study
The module will exploit (staged) online independent learning and private study as follows:- a series of reusable learning objects, covering the fundamentals of multilevel and latent variable modelling, the software and examples will be available in the VLE
- comprehensive material covering multilevel and latent variable modelling, supplementing and expanding on the most pertinent aspects introduced during face-to-face sessions, will heavily direct private study
- required and recommended reading from selected textbooks and significant papers will be set via the VLE. Private study tasks based on these readings will be set
- each student will be given a published paper with data and be required to produce a report of the reanalysis and interpretation of the results
- 3 formative assessments will be set.
Opportunities for Formative Feedback
This will be done in a number of ways:- contribution made during learning and teaching exercises (both face-to-face and online)
- 3 sets of formative assessment exercises.
Methods of assessment
Coursework
Assessment type | Notes | % of formal assessment |
Report | Summative 1500 words | 50.00 |
Report | Summative 1500 words | 50.00 |
Total percentage (Assessment Coursework) | 100.00 |
Normally resits will be assessed by the same methodology as the first attempt, unless otherwise stated
Reading list
The reading list is available from the Library websiteLast updated: 22/01/2020
Browse Other Catalogues
- Undergraduate module catalogue
- Taught Postgraduate module catalogue
- Undergraduate programme catalogue
- Taught Postgraduate programme catalogue
Errors, omissions, failed links etc should be notified to the Catalogue Team.PROD