Professional evaluators are often called upon to analyze data produced by a catastrophically inadequate evaluation design. A remedial strategy involving diagnosis of error, application of a corrective procedure and sensitization of program personnel of the need for a more sophisticated stance, is proposed and illustrated with a case study.…

This article responds to the "Standards for Evaluation of Educational Programs, Projects and Materials" by addressing three issues in qualitative research: (1) data quality; (2) data verification; and (3) collusion between evaluators and their sponsors. To be of maximum international value, the standards report needs more emphasis on…

An introduction and overview to document analysis as a retrospective data collection technique are presented. (PN)

It is asserted that various factors, such as policy interests and available funding, influence the development of social indicators. However, it is stated that there is still enough uncertainty about significant social indicators to allow for experimentation. (Author/MH)

Three standard techniques for the analysis of geographic data are presented and their use in evaluation is discussed. These three techniques are: geocode analysis, trend surface analysis, and social area or ecological analysis. (CTM)

Three types of macro-level variables for program analysis--cultural, economic and demographic--provide a context for evaluation and offer alternative explanations for program success or failure. The problems of causality and availability of data, as well as the usefulness of macro variables in meta evaluation are discussed. (Author/CM)

Social area analysis (SAA) is defined, and its conceptual roots and principal applications identified. The utility of SAA is assessed by reviewing use of demographic data among several disciplines. Authors review seminal and recent contributions to the field. Ecological fallacies and other problematic facets of SAA are considered. (Author/DWH)

The various costs of Type I and Type II errors of inference from data are discussed. Six methods for minimizing each error type are presented, which may be employed even after data collection for Type I and which minimizes Type II errors by a study design and analytical means combination. (Author/CM)