|
Abstract:
|
Seven decades ago, statisticians at the U.S. Census Bureau perceived the need of quantifying census quality, and so conceived of the idea a follow-on survey to measure the level of census error; thus, the post-enumeration survey was born. Since then, the Census Bureau has refined and changed the design, sometimes to reflect new methods, sometime to reflect changed goals. Along the way, this research environment made important contributions to the practice of statistics. The statistical agencies of other countries made their own important contributions. This talk reviews at how and when decisions were made in the U.S. and why other nations adopted other designs to fit their purposes. It looks at the problems that needed to be solved, those which were sufficiently addressed, and those which remain. Since the original goal of the PES was census quality measurement, it is fair to ask what we have learned about census-taking methods due to this effort. The talk will offer some thoughts future research areas on census coverage evaluation.
|
