Quality of Data Standards: Empirical Findings from XBRL

Abstract:     
Certain data standards can help improve the quality of the data created according to the standards. But data standards do not always improve data quality. We introduce the notion of "quality of data standards" and argue that quality of data is affected by the quality of the standards used. We develop metrics for assessing quality of data standards. The metrics are evaluated empirically using company financial reports created using the eXtensible Business Reporting Language (XBRL) data standards. Our findings show the use frequency of standard data elements roughly follows a power law distribution. Tradeoffs exist between relevancy and completeness dimensions and between a single user perspective and user community perspective.

Number of Pages in PDF File: 8

Keywords: Data Quality, Interoperability, Data Standards, XBRL, Power Law Distribution, Long Tail

JEL Classification: Z00

working papers series