Crain, Robert L.
Is Nineteen Really Better Than Ninety-Three?
John Hopkins University
Questions whether one should selectively review literature or review all of it .
Journal Name or Institutional Affiliation:
The National Institute of Education.
- They say that in this panel there is a rather neat correlation between the number of studies the author chooses to look at and the effect of desegregation the author finds.
- If the principle function of selecting a superior subgroup of studies is to find the consistency of results which is masked by error in an unselected sample of studies, we believe we did that, and that the panel did not.
- He plus his colleagues find consistent positive effects when including all 93 studies.
Method of Analysis:
Unit of Analysis:
- The author states that he believes that the prior selection of a small group of preferred studies from a pool using criteria chosen in advance in examining the studies is in principle a mistake. Because it assumes the methodological criteria which define a good study are known in advance.
- The author criticizes the other NIE panelists for eliminating all but 19 of the 93 extant studies of desegregation effects for their meta-analysis.