Principles and Practices for Making Statistics Relevant for Economic
Decision Making, by J. Steven Landefeld
Presented at the Goskomstat Workshop: Developing the State Statistical
System of the Russian Federation, Moscow, Russia, May 28-29, 2000
One of the most important goals of agencies producing economic statistics
is that those statistics be relevant and regularly used by public and
private decision makers. Achieving this goal requires several elements
including accuracy, timeliness, and sound concepts and methods. However,
perhaps most important, in a world of constrained resources, it requires
focusing on those statistics that are most important to decision makers
in the public and private sector. And this may be the most difficult challenge.
Statisticians are by nature perfectionists and most are data producers
rather than data users. The result is a tendency to focus on a level of
accuracy, conceptual consistency, or statistical integration that is too
high relative the quality of the underlying source data or the needs of
the users. A jewel-like set of integrated accounts that comes out so late
as to be irrelevant to decision makers, or that is not benchmarked to
comprehensive information, is of limited value. This short paper will
discuss these and other issues important to developing and maintaining
a set of economic statistics that are relevant to economic decision making.
The focus will be on the U.S. experience and lessons that can be learned
-- positive and negative -- from that experience.
Last changed: August 24, 2001