This book combines the authors experiences on the topic and brings out a wealth of new information relevant to the study of meta-analysis. Applications ranging from business to education to environment to health sciences in both univariate and multivariate cases are presented alongside and subservient to theory. The treatment of the common mean of univariate normal populations, tests ofhomogeneity, one-way random effects model, categorical data, recovery of inter-block information, and combination of polls is entirely new. A special feature includes the incorporation of detailed discussions on the computational aspects and related software to carry out statistical meta-analysis in practice. Most of the methods described can be understood and applied with a solid masters level background in statistics. Practitioners will benefit from the book owing to a host of worked-out examples from numerous and various contexts. INDICE: Preface. 1. Introduction. 2. Various Measures of Effect Size. 2.1 Effect Size based on Means. 2.2 Effect Size based on Proportions. 2.3 Effect Size based on - Coefficient and Odds Ratio. 2.4 Effect Size based on Correlation. 3. Combining Independent Tests. 3.1 Introduction. 3.2 Description of Combined Tests. 4. Methods of Combining Effect Sizes. 5. Inference about a Common Mean of Several Univariate Normal Populations. 5.1 Results on Common Mean Estimation. 5.2 Asymptotic Comparison of Some Estimates of Common Mean for k = 2 Populations. 5.3 Confidence Intervals for the Common Mean. 5.4 Applications. 5.5 Appendix: Theory of Fishers Method. 6. Tests of Homogeneity in Meta-Analysis. 6.1 Model and Test Statistics. 6.2 An Exact Test of Homogeneity. 6.3 Applications. 7. One-Way Random Effects Model. 7.1 Introduction. 7.2 Homogeneous Error Variances. 7.3 Heterogeneous Error Variances. 8. Combining Controlled Trials with Normal Outcomes. 8.1 Difference of Means. 8.2 Standardized Difference of Means. 8.3 Ratio of Means. 9. Combining Controlled Trials with Discrete Outcomes. 9.1 Binary Data. 9.2 Ordinal Data. 10. Meta-Regression. 10.1 Model with OneCovariate. 10.2 Model with More Than One Covariate. 10.3 Further Extensions and Applications. 11. Multivariate Meta-Analysis. 11.1 Combining Multiple Dependent Variables from a Single Study. 11.2 Modeling Multivariate Effect Sizes. 12. Bayesian Meta-Analysis. 12.1 A General Bayesian Model for Meta-Analysis under Normality. 12.2 Further Examples of Bayesian Analyses. 12.3 A Unified Bayesian Approach to Meta-Analysis. 12.4 Further Results on Bayesian Meta-Analysis.13. Publication Bias. 14. Recovery of Inter-Block Information. 14.1 Notationsand Test Statistics. 14.2 BIBD with Fixed Treatment Effects. 15. Combination of Polls. 15.1 Formulation of the Problem. 15.2 Meta-Analysis of Polls. 16. Vote Counting Procedures. 17. Computational Aspects. 17.1 Extracting Summary Statistics. 17.2 Combining Tests. 17.3 Generalized P-values. 17.4 Combining Effect Sizes. 18. Data Sets. 18.1 Validity Studies. 18.2 Effects of Teacher Expectance on Pupil IQ. 18.3 Dentifrice Data. 18.4 Effectiveness of Amlodipine on Work Capacity. 18.5 Effectiveness of Cisapride on the Treatment of Nonulcer Dyspepsia. 18.6 Secondhand Smoking. 18.7 Effectiveness of Misoprostol in PreventingGastrointestinal Damage. 18.8 Prevention of Tuberculosis. References. Index.
- ISBN: 978-0-470-29089-7
- Editorial: John Wiley & Sons
- Encuadernacion: Cartoné
- Páginas: 280
- Fecha Publicación: 01/08/2008
- Nº Volúmenes: 1
- Idioma: Inglés