Vyhledávat v databázi titulů je možné dle ISBN, ISSN, EAN, č. ČNB, OCLC či vlastního identifikátoru. Vyhledávat lze i v databázi autorů dle id autority či jména.
Projekt ObalkyKnih.cz sdružuje různé zdroje informací o knížkách do jedné, snadno použitelné webové služby. Naše databáze v tuto chvíli obsahuje 3151204 obálek a 950590 obsahů českých a zahraničních publikací. Naše API využívá většina knihoven v ČR.
Meta-Analysis: A Comparision of Approaches
Rok: 2004
ISBN: 9780889372801
OKCZID: 110053478
Citace (dle ČSN ISO 690):
SCHULZE, Ralf. Meta-analysis: a comparison of approaches. Toronto: Hogrefe & Huber, c2004. xi, 242 p.
Anotace
Meta-analysis has become the standard method for summarizing research findings in many scientific fields. The number of published applications using the method has been steadily growing in the last 25 years and the statistical procedures of meta-analysis continue to become more and more advanced. Many different statistical models and procedures are available for the same purpose – to quantitatively integrate a research database consisting of effect sizes. This book provides a comprehensive treatment and comparison of the statistical procedures available for meta-analysis with correlations as effect sizes. In the first part of the book, the basic ideas, steps, and aspects of the methodological background of meta-analysis are presented. The second part focuses on the statistical methods of extant meta-analytical approaches. The statistical properties of effect sizes – with focus on the correlation coefficient – are explored in detail. As general frameworks for meta-analysis, fixed effects, random effects, mixture, and hierarchical linear models are described. Both the detailed description of the effect sizes' properties and the general frameworks provide important classification aspects for the specific approaches of meta-analysis. The major approaches used in practice, as well as refinements of these, are presented in a consistent manner to facilitate their comparison. In the third part, the results of a comprehensive Monte-Carlo study are presented to evaluate the performance of the approaches in a large set of possible situations. The last part summarizes and evaluates the theoretical and Monte Carlo study results. It is highlighted when the procedures of commonly applied approaches work and when they fail to provide reliable results. This book is relevant not only for researchers for whom correlations represent an effect size of vital interest, but also for those desiring an enhanced understanding of the methods of meta-analysis. The outcome of reading this work will be a much more informed choice for researchers of the procedures best suited to different situations of interest, previously unavailable in such a comprehensive manner.