Book contents
- Frontmatter
- Contents
- Preface
- Acknowledgements
- 1 Introduction
- 2 Basic principles of multilevel analysis
- 3 What do we gain by applying multilevel analysis?
- 4 Multilevel analysis with different outcome variables
- 5 Multilevel modelling
- 6 Multilevel analysis in longitudinal studies
- 7 Multivariate multilevel analysis
- 8 Sample-size calculations in multilevel studies
- 9 Software for multilevel analysis
- References
- Index
9 - Software for multilevel analysis
Published online by Cambridge University Press: 26 March 2010
- Frontmatter
- Contents
- Preface
- Acknowledgements
- 1 Introduction
- 2 Basic principles of multilevel analysis
- 3 What do we gain by applying multilevel analysis?
- 4 Multilevel analysis with different outcome variables
- 5 Multilevel modelling
- 6 Multilevel analysis in longitudinal studies
- 7 Multivariate multilevel analysis
- 8 Sample-size calculations in multilevel studies
- 9 Software for multilevel analysis
- References
- Index
Summary
Introduction
In the foregoing chapters, all examples of multilevel analysis were analysed in MLwiN. Although this software package is specially developed for performing multilevel analysis, there are also other software packages that can be used for multilevel analysis. In this chapter the example dataset(s) will be reanalysed with other software packages, and any differences in the results will be compared and discussed. For continuous outcome variables the research question concerned the relationship between total cholesterol and age (see Sections 2.2, 2.5 and 2.6.1), for dichotomous outcome variables it was the relationship between hypercholesterolemia and age (see Section 4.2), and for ‘count’ outcome variables the relationship between ‘the number of risk factors’ and age (see Section 4.4). For multinomial logistic multilevel analysis the population was divided into three groups, i.e. a group with relatively ‘low’ cholesterol values, a group with relatively ‘moderate’ cholesterol values, and a group with relatively ‘high’ cholesterol values (see Section 4.3). For linear multilevel analysis (i.e. multilevel analysis with a continuous outcome variable) both a two-level structure (i.e. patients clustered within medical doctors) and a three-level structure (patients clustered within medical doctors and medical doctors are clustered within institutions) will be used in the comparison. Only a two-level structure will be used for logistic multilevel analysis (i.e. multilevel analysis with a dichotomous outcome variable), for Poisson multilevel analysis (i.e. multilevel analysis with a ‘count’ outcome variable), and for multinomial logistic multilevel analysis (i.e. multilevel analysis with a categorical outcome variable).
- Type
- Chapter
- Information
- Applied Multilevel AnalysisA Practical Guide for Medical Researchers, pp. 130 - 166Publisher: Cambridge University PressPrint publication year: 2006