Isaac Newton is known to have humbly explained his success: “If I have seen further it is by standing upon the shoulders of giants” (1675; from Columbia World of Quotations, 1996). Although the history of science suggests that Newton may have been as likely to kick his fellow scientists down as he was to collaboratively stand on their shoulders (e.g., Boorstin, 1983, Chs. 52-53; Gribbin, 2002, Ch. 5), this statement does eloquently portray a central principle in science: That the advancement of scientific knowledge is based on systematic building of one study on top of a foundation of prior studies, the accumulation of which takes our understanding to ever increasing heights. A closely related tenet is replication—that findings of studies are confirmed (or not) through repetition by other scientists.
Together, the principles of orderly accumulation and replication of empirical research suggest that scientific knowledge should steadily progress. However, it is reasonable to ask if this is really the case. One obstacle to this progression is that scientists are humans with finite abilities to retain, organize, and synthesize empirical findings. In most areas of research, stud-ies are being conducted at an increasing rate, making it difficult for scholars to stay informed of research in all but the narrowest areas of specialization. I argue that many areas of social science research are in less need of further research than they are in need of organization of the existing research. A second obstacle is that studies are rarely exact replications of one another, but instead commonly use slightly different methods, measures, and/or samples.1 This imperfect replication makes it difficult (1) to separate meaningful differences in results from expectable sampling fluctuations, and (2) if there are meaningful differences in results across studies, to determine which of the several differences in studies account for the differences in results.
An apparent solution to these obstacles is that scientists systematically review results from the numerous studies, synthesizing results to draw conclusions regarding typical findings and sources of variability across studies. One method of conducting such systematic syntheses of the empirical literature is through meta-analysis, which is a methodological and statistical approach to drawing conclusions from empirical literature. As I hope to demonstrate in this book, meta-analysis is a particularly powerful tool in drawing these sorts of conclusions from the existing empirical literature. Before describing this tool in the remainder of the book, in this chapter I introduce some terminology of this approach, provide a brief history of meta-analysis, further describe the process of research synthesis as a scientific endeavor, and then provide a more detailed preview of the remainder of this book.
Source: Card Noel A. (2015), Applied Meta-Analysis for Social Science Research, The Guilford Press; Annotated edition.