Short answers, pulled from the story.
The term meta-analysis entered the statistical lexicon in 1976 when Gene Glass defined it as the analysis of analyses. This coinage marked a shift toward aggregating measures of relationships and effects across independent studies.
Karl Pearson published a paper in 1904 that collated data from several typhoid inoculation studies within the British Medical Journal. That work aggregated outcomes from multiple clinical trials decades before the formal method existed.
By 1991 researchers had published 334 meta-analyses globally while that number surged to 9,135 by 2014 indicating massive acceptance across disciplines like psychology medicine and ecology.
The file drawer problem characterizes negative or non-significant results tucked away in cabinets creating serious base rate fallacies. Estimates suggest 25% of meta-analyses in psychological sciences suffered from publication bias though actual figures likely higher.
In 1998 a US federal judge found the Environmental Protection Agency abused the process claiming cancer risks from environmental tobacco smoke to influence policy makers. The methodology remains highly malleable allowing agenda-driven manipulation to distort scientific consensus effectively.