Module MATH43515: Multilevel Modelling

Department: Mathematical Sciences

MATH43515: Multilevel Modelling

Prerequisites

• For students on an MA Research Methods programme: SGIA49915 Quantitative Research Methods, or completion of the DRMC R School, or equivalent

Corequisites

• None

Excluded Combination of Modules

• None

Aims

• To provide alternative advanced training in quantitative methods which will enable students to advance from more traditional concepts of independent data.
• To introduce the notion that social processes rarely exist in isolation, similar to underlying health factors.
• To demonstrate that there are natural clustering of events, which affect how society functions (from households and moving through to governance).
• To introduce students to advanced techniques for analysing correlated data resulting from clustering (such as in household, schools, hospital administrative authority) and repeated data on the same entity, over time.

Content

• Indicative content will include:
• Introduction to hierarchical data structures
• Revisiting general linear models
• Revisiting generalised linear models
• Multivariate general linear model
• Multivariate generalised linear model
• Two-stage techniques
• Linear mixed effects models
• Generalised linear mixed effects models
• Incomplete data techniques
• Nonlinear models

Learning Outcomes

• On completion of this module, students should be able to:
• Critically explore the concepts of repeated measures
• Demonstrate advanced understanding of structure and methods for longitudinal data analysis
• Demonstrate advanced knowledge and understanding of incomplete data techniques

• By the end of the module students should be able to:
• Apply multivariate generalised linear models to repeated measures, repeated cross-sectional data and longitudinal data
• Apply two-stage analytical methods to repeated measures, repeated cross-sectional data and longitudinal data
• Apply linear and generalised linear mixed effects models to repeated measures, repeated cross-sectional data and longitudinal data
• Perform sensitivity analysis for incomplete data

• Students will also develop some important key skills, suitable for underpinning study at this and subsequent levels, such as:
• An ability to critically evaluate hierarchical data structures and communicate conclusions to specialist and non-specialist audiences
• An ability to demonstrate a high degree of self-direction in analysing and interpreting correlated data
• An ability to work autonomously in planning and implementing multilevel models, including knowledge of how to use software

Modes of Teaching, Learning and Assessment and how these contribute to the learning outcomes of the module

• This module will be delivered by the Department of Mathematical Sciences.
• Teaching will be delivered primarily by lectures, practicals and group tutorials. These modes of teaching will ensure that statistical methods are not taught in abstractions and instead the learning and teaching approach for this module will consider students as apprentices in quantitative research methods.
• Lectures demonstrate what is required to be learned and the application of the theory to practical examples.
• Tutorials and practicals enable students to test and develop their understanding of the material by applying it to practical problems, and provide feedback and encourage active engagement.
• Tutorials will be led by tutors with experience in the student’s primary discipline, wherever possible.
• Lectures, tutorials and practicals will be supported by the distribution of materials such as video content, directed reading, e-assessments, reflective activities, opportunities for self-assessment, and peer-to-peer learning within a tutor-facilitated discussion board.
• Students will be able to obtain further help in their studies via scheduled office hours or surgeries as well as by approaching their lecturers by email.
• Students will be expected to work in between sessions, and to discuss their own work during the practicals and tutorials. This work will be guided by the module leader, but will be organised by the students themselves, thereby enabling them to demonstrate their time management skills.
• Students will undertake independent research to further their knowledge of the topic and self-directed learning to further their technical and transferable skills.
• The practicals and tutorials also provide opportunities for module leaders to monitor progress and to provide feedback and guidance on the development of ideas for the project, and for students to gauge their progress throughout the duration of the module.
• Student performance will be assessed through one written project based on data provided to students and analysed and interpreted by themselves.

Teaching Methods and Learning Hours

Summative Assessment

Formative Assessment:

There will be a short formative assignment in which students will carry out a series of operations and interpret the results of the operations. Although not all procedures will be covered in this work, and different data will be used, the formative assignment follows a similar format to the summative assignment. It is therefore aimed to support students to become familiar with requirements and expectations of their summative work. Students will receive peer-feedback, as well as generic feedback on the formative.

■ Attendance at all activities marked with this symbol will be monitored. Students who fail to attend these activities, or to complete the summative or formative assessment specified above, will be subject to the procedures defined in the University's General Regulation V, and may be required to leave the University