This special issue of the Zeitschrift fur Psychologie / Journal of Psychology is on meta-analysis, a methodological topic. For a journal that does not generally focus on methods and statistics, this may be considered odd, but it is not. Meta-analysis is a special case. No matter what specific area of psychology the reader may be most interested in, he or she will certainly know some meta-analyses in his or her field. This is indicative of the fact that this method is not specifically relevant only for a certain area of psychology or predominantly used just to address a limited number of research questions. It can be applied in almost any empirical research domain in psychology and beyond. Meta-analysis started out in psychology, but it diffused into other scientific disciplines long ago and has become part of the methods toolbox in fields like medicine or sport and exercise research (Hagger, 2006; Schulze, Holling, & Bohning, 2003; Sutton, Abrams, Jones, Sheldon, & Song, 2000). Remarkably, methods of meta-analysis have even been discussed and used in such diverse and remote fields (from the perspective of psychology) as industry, software engineering, and phytopathology (see Bohning & Dammann, 2003; Pickard, Kitchenham, & Jones, 1998; Rosenberg, Garrett, Su, & Bowden, 2004). More than 30 years after the term was coined (Glass, 1976)1, meta-analysis has earned its place in the pantheon of scientific methods. It became a standard method of research synthesis in many empirical research fields, especially in the social sciences. There are probably no major areas left in psychology where there are no highly cited meta-analyses that are referred to in basic psychology seminars. As a consequence, every young and senior researcher has to be acquainted with at least the fundamentals of metaanalysis to understand and critically appraise the results of many applications of the method. This does not mean, however, that meta-analysis has not been heavily criticized ever since the term was coined. Au contraire, a number of critical commentaries (e.g., Sohn, 1997) and almost derogatory remarks (e.g., Eysenck, 1978) in psychology journals have been published. Very skeptical views on meta-analysis can also be found from the perspective of statisticians and in the field of mathematical statistics (see, e.g., Finney, 1995; Nelder, 1986). In fact, Nelder wrote in his presidential address delivered to the Royal Statistical Society in London that “The use of this, to me, rather pretentious term [meta-analysis] for a basic activity of science is a clear indication of how far some statisticians’ view of statistics has diverged from the basic procedures of science” (Nelder, 1986, p. 113). Nevertheless, meta-analysis is not a fad, it is here to stay. An almost unmistakable indication is given by the fact that the number of publications on methods of meta-analysis and applications of the methods has been steadily growing since 1980. Figure 1 supports this assertion with data from the probably most comprehensive literature database available for psychology, PsycINFO. The absolute number of publications that use the term meta-analysis2 in this database is shown for a 31-year period from 1975 to 2006 as squares connected by a solid line (a lowess curve resulting from locally weighted polynomial regression). The number of publications per year can be seen by looking at the right ordinate. It is well known that the total number of publications per year is increasing over the years. For example, at the time of writing this guest editorial there are 58,314 publications in the PsycINFO database with a publication date of 1990 and 110,766 for the year 2006. Hence, the absolute number of publications is increasing overall and maybe not for meta-analysis in particular. The second curve, with a dotted line connecting the circles, shows the percentage for the number of meta-analysis publications. It can be seen from inspection of the graph that meta-analytic publications are growing in absolute and relative terms. When reading off the value for 2005 from the left ordinate shown in Figure 1, it becomes evident that the percentage is cur-