|
Activity Number:
|
275
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, August 8, 2006 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Government Statistics
|
| Abstract - #306447 |
|
Title:
|
How Do We Know If We Aren't Looking? An Investigation of Data Quality in the SCF
|
|
Author(s):
|
Arthur Kennickell*+
|
|
Companies:
|
Federal Reserve Board
|
|
Address:
|
20th C Street, NW, Mail Stop 153, Washington, DC, 20551,
|
|
Keywords:
|
interviewers ; data quality
|
|
Abstract:
|
"Data quality" is a term often used vaguely to indicate some cluster of desirable traits. This paper argues that the most useful notion of data quality is one that turns on the utility of data for the analytical purposes for which it was intended. Although the standard attributes, such as missing data rates, are important, more subtle matters can be critical. Unfortunately, many such factors are difficult---even impossible---in the absence of advanced AI technology, to identify mechanically. This paper focuses on an exercise undertaken with the 2004 Survey of Consumer Finances (SCF) involving review of the data by subject matter experts and the use of feedback to interviewers based on that review.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2006 program |
