From the very beginning a criticism made against meta-analysis is that it is based on the indiscriminate mixing of good and bad studies. This garbage-in, garbage-out complaint originated with Eysenck (1978: 517), who was dismayed with the apparently low standards of inclusion used in meta-analysis. "A mass of reports - good, bad, indifferent - are fed into the computer in the hope that people will cease caring about the quality of the material on which the conclusions are based." According to Eysenck, there was little to be gained by trying to distill knowledge from poorly designed studies. A similar view was espoused by Shapiro (1994: 771) in his article entitled “Meta-analysis/shmeta-analysis." Shapiro argued that the quality of a meta-analysis was contingent upon the quality of the individual studies being combined. As the highest standard of research is the experimental design, he proposed that meta-analyses based on the accumulation of nonexperimental data should be abandoned. Feinstein (1995) was also concerned with the mixing of good and bad studies which he considered statistical alchemy. He argued that insufficient attention to issues of quality control had given rise to the situation where reviewers could dredge up data to support almost any hypothesis. The solution, according to Feinstein, was to exclude biased studies and combine only "excellent individual studies" or "studies that seem unequivocally good" (1995: 77).