Answers are grounded in this paper and cite their sources.
Federal statistical agencies are funded and supervised by elected and appointed politicians. What counts as politial interference is not self-evident. This article offers a working definition of interference, emphasizing the importance of an agency offering its best judgment regarding accurate measurement of a given phenomenon, its ability to apply state-of-the-art science in that measurement, and its protection from preclearance of the resulting statistical product. Interference is indicated by efforts to shape statistical products to achieve political advantage. Statistical adjustment to correct the differential undercount in the decennial census is used to illustrate political interference.
No takes yet. Share an insight, caveat, or question.
Kenneth Prewitt (2010) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: