home contact team research learning MLwiN MLwiN FAQs publications links

• review home
• aML
• EGRET
• GENSTAT
• GLLAMM
• HLM
• LIMDEP
• LISREL
• MIXREG
• MLwiN
• Mplus
• R
• SAS
• S-Plus
• SPSS
• STATA
• SYSTAT
• WINBUGS
• Comparative Timings
• Datasets
• Tables

Multilevel Modelling Software

Reviews

The Centre for Multilevel Modelling aims to provide a comprehensive set of reviews of packages for multilevel modelling. The first set of reviews is now nearing completion and they may be accessed by clicking on the appropriate package name at the left. This will take you to a brief introduction from where you can download the full review. The remaining reviews will be provided as they become available.

We have attempted to cover all the major packages that allow the fitting of multilevel models as well as smaller and stand-alone software. We intend to update the reviews as new software or major enhancements to existing software become available. We are very grateful to the providers of software for their help in supplying materials and commenting on draft reviews. If software providers or authors remain unhappy with certain aspects of these reviews it is our policy to publish their comments alongside the review itself, together with a response from the review author where appropriate.

In carrying out these reviews we are well aware that there are potential conflicts of interest, since the Centre for Multilevel Modelling has its own software package, MLwiN, which is also reviewed (see MLwiN review). We realise that the choice of reviewers and also of data sets for analysis is not a neutral activity. We hope that reviewers have been as objective as they can and they were each given a set of clear guidelines. The position with data sets is more difficult since all the software packages have their strengths and weaknesses. The reviews themselves discuss these, but inevitably, in our summary tables, some packages may seem relatively advantaged simply because of choice of data. Please therefore bear in mind these issues when looking at the results. What we have produced is a first step in providing information to the research community.

The reviews carry details of how to use each package together with timings and estimates. While common datasets are used for each review, not all packages are able to fit every dataset. For details go to Datasets. In addition a single summary comparative table of estimates and timings has been compiled - go to tables summary and the tables ( 0.2 mb) .

From time to time we will add new reviews and update existing ones. We welcome feedback. In particular please tell us if you think there are any mistakes or omissions, please email hilary.browne@bristol.ac.uk