Knowing Data Quality

In more mainstream applications, there is a trend that might be seen as contrary to the big data movement. Executives are being required to vouch for the quality of the data they are providing.

This presentation models a data quality effort along the lines of the Food and Drug Administration. A cost of implementation model is proposed as well as a number of cost containment features.

Specific issues of data quality are described including methods of statistical measurement. The concept of skew is introduced.

Important non-objective measures of confidence and confidence building strategies are discussed.

Most importantly, a distribution of the responsibilities for data quality is proposed.

Bob Schmidt is a Data Steward, Business Systems Consultant at a HUGE bank, the author of "Data Modeling for Information Professionals" (Prentice Hall, 1997), as well as numerous other articles, the inventor of “a method for valuing data” (US patent:20130151423), an occasional speaker at various national and international conferences including DAMA since 1997, and the National winner of Oracle Developers and Users Group best speaker and separately best paper.

Bob has labored for decades in an industry that has spawned thousands of millionaires, and yet he remains not too big to fail. His experience spans everything from Life Sciences to the mortuary industry, the Extractive sector to debt collection, and Michelin man to the Minute Maid can. He has been everywhere, man.