NCJ Number
38442
Editor(s)
I N BERNSTEIN
Date Published
1976
Length
134 pages
Annotation
SERIES OF FIVE PAPERS WHICH EXPLICATE METHODOLOGICAL PROBLEMS PARTICULARLY RELEVANT TO EVALUATION RESEARCH FOR THE PURPOSE OF ALERTING RESEARCHERS TO PROSPECTIVE PROBLEMS.
Abstract
DISCUSSIONS ARE PRESENTED THAT CRITICALLY EXAMINE A VARIETY OF TECHNIQUES THAT CAN BE USED TO INCREASE THE VALIDITY OF RESEARCH RESULTS. IN SPITE OF DEFECTS IN RESEARCH DESIGN. INDIVIDUAL PAPERS CONSIDER SOME OF THE MAJOR PROBLEMS THAT OCCUR IN THE PROCESS OF CONDUCTING LARGE-SCALE EXPERIMENTS, THE CONSEQUENCES OF USING AN OPTIMUM ALLOCATION MODEL FOR SAMPLING IN LARGE-SCALE EXPERIMENTS, THE USE OF ANALYSIS OF COVARIANCE PROCEDURES WHEN RANDOM ASSIGNMENT HAS NOT OCCURRED, AND A CODIFICATION OF PROBLEMS OF EXTERNAL VALIDITY IN EVALUATION RESEARCH. A FIFTH PAPER ADVOCATES THE USE OF TRUE EXPERIMENTS AND TRUE EXPERIMENTS COUPLED WITH APPROXIMATIONS RATHER THAN APPROXIMATIONS ALONE, ASSUMING THAT A SPECIFIED VARIETY OF CONDITIONS ARE PRESENT.