July 2006 meeting
Meeting held 11 July (Americas/Europe/Africa) 12 July (Asia/Australia)
Articles or Books Selected:
We discussed the following chapters from Enterprise Knowledge Management: The Data Quality Approach by David Loshin, published by Morgan Kaufmann, © 2001, ISBN: 0-12-455840-2
Three chapters to be discussed:
- 5 — Dimensions of Data Quality
- 9 — Measurement and Current State Assessment
- 12 — Rule-based Data Quality
(We will discuss two more chapters at the August meeting)
See the July 2006 Reading Guide (requires IAIDQ members zone access) to help you prepare for the meeting.
July meeting report:
Members from 4 countries (and four continents) participated in the IAIDQ July, 2006 book club meeting discussing the following chapters from Enterprise Knowledge Management: The Data Quality Approach by David Loshin.
5 — Dimensions of Data Quality
9 — Measurement and Current State Assessment
12 — Rule-based Data Quality
Of the 5 dimensional categories David uses, two were discussed in detail:
- The data modeling category has the highest number of dimensions (~15). We discussed this in the context of the importance of good data models to the data quality program. This applies to new models and to model updates.
- Information Policies are a driving force for the data quality program. Participants discussed information policies in their organizations. A best practice mentioned is to have a policy for data modeling practices.
The dimension of data lineage shows up in other parts of David's work and its connection to regulatory requirements for financial services was highlighted. A key point was that keeping track of data lineage helped identify many other data quality issues and therefore enabled preventative data quality activities. The connection between data lineage and an information life cycle or information chain was also discussed.
The data quality dimensions have suggested measures in chapter 9 that offer readers wide variety of measures to pick from. It was suggested to use Loshin's sentinel concept to identify the “critical few” to start with.
The use of the business rules approach to data quality was discussed with the following points: Keep the rules updated and synchronized by the business, use the data stewards! There are good rule documentation methodologies available. Processes are needed to manage rule approvals and conflict resolution. Some of the participants described the business rules activities and successes they have had in their companies.