Educational Statistics: Home

Statistics are everywhere. Their power and their undoubted efficacy in many areas have given rise to faith in measurement and metrics. More of them will tell us all that we need to know. Their use carries with it a number of presuppositions: that reality can be satisfactorily represented and that it can be controlled or the risks managed. The papers in this book interpret the ethics and aesthetics of statistics in terms of representation, visualisation and accessibility, focus on the appeal of 'simplicity', of technical languages, numbers, diagrams and pictures, and pay attention to their connection with action plans.

Practical Statistics for Educators, 4th edition focuses on the application of research and statistics as applied specifically to education. Since the first edition came out in 1994, thousand of students in educational statistics courses and their professors have found it to be an excellent textbook. Educational practitioners have also appreciated keeping this book on their reference shelf. Now in its fourth edition, this well-regarded text is a clear and easy-to-follow manual for use in introductory statistics or action research courses. Ruth Ravid concentrates on the essential concepts in educational statistics including when to use various statistical tests and how to interpret the results.

Throughout the social, medical and other sciences the importance of understanding complex hierarchical data structures is well understood. Multilevel modelling is now the accepted statistical technique for handling such data and is widely available in computer software packages. A thorough understanding of these techniques is therefore important for all those working in these areas. This new edition of Multilevel Statistical Models brings these techniques together, starting from basic ideas and illustrating how more complex models are derived. Bayesian methodology using MCMC has been extended along with new material on smoothing models, multivariate responses, missing data, latent normal transformations for discrete responses, structural equation modeling and survival models.