3.4 Using Data

3.4 Using Data

The final area where data goes wrong is the process of putting it into business objects such as reports, query result screens, portal-accessible windows, and the subsequent use of this by business professionals.

Data may be accurate, but if users do not understand the meaning of the data or the context within which it is presented, their interpretation and use of the data may be inaccurate. This again points to the problem of not having a good metadata repository that is maintained 100% accurately all of the time. This repository should contain information on what each data element represents, how the data within it is encoded, how to interpret special values, the source of data, the time periods last updated, and the quality levels last known for this data.

Quality levels are important because they can let users judge whether the data satisfies their needs or not. They should be able to determine from the repository the probable inaccurate rate within the data. This can be derived from periodic assessment, data monitoring tools, or both.

The user should have easy access to this information. What is needed is either a comprehensive help system on the business objects or an easy-to-use corporate portal on corporate data stores. If users have to call someone or jump through hoops to find out information about their data, they will just not do it. They will make their own assumptions about what the data represents and use it accordingly. Unfortunately, most companies do not have a good metadata repository and certainly do not have a good mechanism for non-IT professionals to access the information they do have.



Data Quality(c) The Accuracy Dimension
Data Quality: The Accuracy Dimension (The Morgan Kaufmann Series in Data Management Systems)
ISBN: 1558608915
EAN: 2147483647
Year: 2003
Pages: 133
Authors: Jack E. Olson

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net