2017/18 Taught Postgraduate Module Catalogue

EPIB5025M Multilevel and Latent variable Modelling

15 creditsClass Size: 21

Module manager: Dr Darren Greenwood
Email: d.c.greenwood@leeds.ac.uk

Taught: Semester 2 (Jan to Jun) View Timetable

Year running 2017/18

Pre-requisite qualifications

Normally a first degree in a science allied with medicine, including biology, ecology, biochemistry, statistics, mathematics, computing, psychology, economics or biomedical science (at least 2:2). We will also consider working experience (two years or more) of research in a quantitative subject area.

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.

Pre-requisites

EPIB5022M	Core Epidemiology
EPIB5023M	Introduction to Modelling
EPIB5024M	Statistical Inference

This 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 epidemiological 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 epidemiological 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 (via signposted links to the professional 'spine') towards statistical epidemiology.

Learning outcomes

Multilevel and latent variable modelling

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 the MLwiN and Mplus 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)
- MLwiN and Mplus 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:
- attendance
- 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 1,500 word	50.00
Report	Summative 1,500 word	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 website

Last updated: 28/02/2014

Disclaimer

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