How is a social scientist to cope with the cornucopia of already existing studies in his or her area? How to draw useable conclusions from a body of work that might run to 5000 items? Traditional narrative integration fails to usefully portray such accumulated knowledge. Meta-analysis is an approach that systematically analyzes and synthesizes research. This book is its first full explanation. Meta-analysis treats a field of research as a complex set of data to be accumulated and integrated. As such it has much in common with survey research -- though, as causal relationships may have already been established by the studies being surveyed meta-analysis need not suffer from the limitations of survey research as a tool for establishing causes. Besides showing how to derive generalizations from very large and divergent bodies of research, the authors also provide ways for enhancing the findings of few or small research studies, and techniques for evaluating the findings of individual experiments by contrasting them with the combined weight of findings from other studies. Their approach does not enforce uniformity on different research. Instead, it is a way to enhance clarity, explicitness and openness in research reviews. Its use will speed the first step of most research projects -- to see what has been done before -- and will help researchers to avoid costly research duplication.