Articles and reports: 12-001-X198000254949Description: This paper deals with the desirability of designing surveys in such a way that results can be compared to previous existing data. The writer explains why there are practical difficulties in assessing the significance of data collected in a one-time survey where these data stand alone and are not readily comparable to other existing data, i.e., where control group data or other benchmarks do not exist. Issue Number: 1980002Author(s): Watson, KenMain Product:Survey Methodology
