Cross Classified Models
Presenter(s): Professor William Browne, Professor George Leckie
In this online resource, Professor William Browne and Professor George Leckie, from the Centre for Multilevel Modelling based at the University of Bristol talk about a set of statistical models called cross-classified models that extend multilevel models that are used widely in the social sciences to account for dependency structures in data. This resource comprises a series of three videos and a practical walkthrough.
Cross Classified Models Part 1: Introduction
In this first lecture on the topic, Professor Bill Browne recaps the idea of multilevel models showing how not all data structures have a hierarchical or nested structure. He then provides historical background to modelling non-nested data and introduces an example from education research that has a cross-classified structure.
Cross Classified Models Part 2: Model fitting
In this second lecture, Professor George Leckie fits several cross-classified models to the Fife dataset introducing notation and diagrams to represent cross-classified structures. He also explains how to extend the concepts of the ICC and VPC to cross-classified models and how to fit predictor variables, showing how they explain different sources of variation.
Cross Classified Models Part 3: Extensions and Further Applications
In this third lecture, Professor Bill Browne shows how we can extend cross-classified models to more than 2 higher-level classifications and different response types and how the notation and classification diagrams can be extended. Two examples from epidemiology and ecology are used to look at how cross-classified models can be used to estimate the importance of different factors using 3 and 4 higher-level classifications.
Cross Classified Models Part 4: Practical
Having worked through the previous three lectures you now have a good grounding in cross-classified models and are ready to try the cross-classified practical or watch the walkthrough video. The practical document to go with this walkthrough is available under the Download slides link
About the author
William Browne is a Professor of Statistics and School Education Director, for the School of Education at the University of Bristol. His research spans the area of statistical modelling, from the development of statistical methods to fit realistically complex statistical models to describe real-life problems, through the implementation of those models in statistical software to the application of the methods in several application areas.
George Leckie is a Professor of Social Statistics and the Co-Director of the Centre for Multilevel Modelling (CMM) at the School of Education, University of Bristol, UK.
His methodological interests are in the development, application and dissemination of multilevel and related models to analyse educational and other data. Substantive interests focus on design, analysis, and communication issues surrounding school performance measures and league tables, especially the use of value-added models for estimating school effects on student achievement for accountability and choice purposes.
- Published on: 10 June 2021
- Event hosted by: University of Bristol
- Keywords: Multilevel models | Multilevel Modelling | Markov Chain Monte Carlo (MCMC) | Random effects | Mlwin |
- To cite this resource:
Professor William Browne, Professor George Leckie. (2021). Cross Classified Models. National Centre for Research Methods online learning resource. Available at https://www.ncrm.ac.uk/resources/online/all/?id=20773 [accessed: 24 April 2024]
⌃BACK TO TOP