This paper focuses on the sample design issues related to the collection of wealth data in
the U.S. Survey of Consumer Finances (SCF). Sample surveys are complex “measurement
engines” that may be viewed as falling into three important parts: selection and pursuit of
participants, collection of data, and statistical processing of the resulting information. The
sample design provides the most fundamental measurable statistical basis for it all. It is obvious
that a good design should provide the most efficient and unbiased representation of the
population relevant for the measurement task at hand that is feasible with the available resources.
At the same time, however, a design that is well integrated with the objectives of a survey may
also have implications for the management of data collection and post-survey processing.
Indeed, a survey that fails to capture at least some such benefits must either be inefficient or
highly constrained.