Cardiff University  


Booking form Evaluation form Contact RCBN




Introduction | RCBN Activities | References | Examples | Links | Software

Introduction (return to top)

There are a significant number of researchers who want to be able to ‘consume’ research that uses multi-level modelling and/or time-series analysis. This must be considered in light of the limited knowledge most educational researchers have in this area. Rather than concentrate on one particular method of analysing numerical datasets this programme theme will explore the analysis of longitudinal data (drawing from the work of the large-scale secondary datasets programme theme) using a range of methods – perhaps concentrating on the relationships and relative merits of different ways of analysing such data. Due to the limited use of such data analysis in educational research this theme should identify ways of how such data analysis could be used in educational research. However, this theme must also address the needs of a few educational researchers who have particular, more sophisticated needs (since there are so few with these skills already). It should also be noted that there was a high need to undertake longitudinal studies – this may be incorporated in this programme theme. See also the theme 'Role of Numbers'.

RCBN Activities (return to top)

December 13-14 2004, Multilevel Modelling Workshop 2, Cardiff University
This second workshop is designed to be more ‘hands on’ with a mixture of demonstrations by the tutors and working through data analysis examples by the participants using MLwiN. Issues to be covered include an introduction to multilevel models, continuous and binary responses, multivariate data, repeated measures data, cross classified and multiple membership models and the use of MCMC methods. The emphasis throughout will be on conceptual understanding and an appreciation of the weaknesses as well as the strengths of the technique. It is anticipated that participants will be able to discuss their own data sets, and there will be provision for some analysis of participants’ own data.The presenters are Harvey Goldstein and Jon Rasbash who work in the Centre for Multilevel Modelling at the Institute of Education, University of London and the Centre for Multilevel Modelling at the University of Bristol.

Participants should have a good working knowledge and experience with analysis of statistical data, including the use of multiple regression.Please click here for more information.

March 19 2004, Introduction to Multilevel Modelling, University of Leicester
This is a one-day workshop being led by Professor Harvey Goldstein and John Rasbash of the Centre for Multilevel Modelling at the Institute of Education, University of London. Issues to be covered include the calculation of confidence intervals for value added scores, the importance of allowing for 'differential value added effects’ e.g. that depend on initial achievement and the use of aggregate data such as school means. The emphasis throughout will be on conceptual understanding and an appreciation of the weaknesses as well as the strengths of the technique. Please click here for more information.

This introductory session may be followed by a further hands-on practical workshop - please revisit the RCBN website for updates information.

Assistance in complex statistical analysis
The RCBN has the resources to help provide immediate and direct consultation from experts in the area of complex statistics. The RCBN can help identify the type of expertise you may require and who could provide this. Alternatively, you may already have a good idea of the expertise you require. In either case the RCBN can provide some of the resources necessary to access and share such expertise. Please contact the RCBN to discuss your particular requirements.

Similarly, if you are able and willing to offer your expertise in complex statistical analysis to other researchers then please contact the RCBN. We can cover many of the costs incurred from assisting other researchers in analysing their data.

May 2004
Outcome evaluations: introduction to the use and application of outcome evaluations in teaching and learning research; 1-day training workshop (Southampton); 10 places

References (return to top

Building Research Capacity journal

Coe, R (2002) What is an effect size?, Building Research Capacity, 4, pp.6-8

Gorard, S (2003) Anyone can calculate conditional probabilities, Building Research Capacity, 5, pp.9-11

Gorard, S (2002) What do statistical tests signify?, Building Research Capacity, 2, pp.4-5

Prandy, K (2002) Measuring quantities: the qualitative foundation of quantity, Building Research Capacity, 2, pp.3-4

Roberts, K (2002) Belief and subjectivity in research: an introduction to Bayesian theory, Building Research Capacity, 3, pp.5-7

Examples (return to top)

Links (return to top)

Use of Numeric Data in Learning and Teaching
This website provides resources, case studies, recommendations and research on the use of numeric data in learning and teaching, primarily for undergraduate and postgraduate students. It focuses upon the use of existing national data resources that are readily available to all learners.

The Centre for Applied Statistics, Lancaster University
The Centre's activities include
statistical research, postgraduate training, the provision of statistical consultancy services for University staff and research students and a short course programme open to external participants.

The Royal Statistical Society
The RSS is one of the premier statistical societies in the world, with a high international reputation.

Statistical Services Centre, University of Reading
Specialists in statistical consultancy, short courses and training UK and overseas.

Timberlake Consultants
Timberlake Consultants provide a total solution to clients working in the statistics, econometrics, operational research or mathematical fields.

TRAMSS aims to develop a web-based learning and teaching resource for quantitative social science researchers, students and trainers.


Software (return to top)


Home > Activities > Complex statistical analysis
Return to top of page
This page was last updated 28th October 2004