Information Quality Principles and Practices

1
Information QualityPrinciples and Practices

• John P. Slone, Ph.D.
• Lockheed Martin Corporation

2
Disclaimer

• The opinions expressed herein are solely those
  of the author and do not represent the opinions
  of the Lockheed Martin Corporation.

3
In this presentation

• What is Information Quality (IQ)?
• Why should one care about it?
• What can be done about it?
• Taking the next steps

4
What is IQ?

• Data Quality vs. IQ
• Conventional view Data Quality Accuracy
• A more appropriate view Data Quality goes
  BEYOND Accuracy
  
• Additional definitions
• IQ in context

5
IQ Definition

• Data / IQ Beyond Accuracy

6
IQ Definition

• Data / IQ Beyond Accuracy

Data Quality
Intrinsic
Accessibility
Contextual
Representational
AccessSecurity
Relevance Value-add Timeliness Completeness Am
ount
InterpretabilityUnderstandability
Conciseness
Consistency
AccuracyObjectivityBelievabilityReputation
7
IQ Definition

• Sample definitions of dimensions

8
IQ Definition

• Other definitions in use
• Intuitive based on individuals experience or
  intuitive understanding of whats important
  
• Typically a small set of attributes is selection
  (e.g., accuracy, timeliness, availability,
  consistency)
  
• Systems perspective
• Focus is on deficiencies introduced during the
  process of manufacturing information
  
• Information consumer
• Focus is on fitness for use

9
IQ in Context
10
Relevance of IQ

• Why is IQ important?
• Information is a valuable asset
• Information of high quality is more valuable than
  information of low quality
  
• Information of high quality can
• Improve customer satisfaction
• Improve financial performance
• Improve strategic posture
• By contrast - information of low quality can
• Cost lives
• Cost customers
• Cost jobs

11
The Impact of Low IQ
The decision to launch 51-L was based on
incomplete and sometimes misleading information

-- Rogers Commission Report
12
The Impact of Low IQ
Relevant information from the National Security
Agency and the CIA often failed to make its way
to criminal investigators. -- The 9/11 Commiss
ion Report
13
The Impact of Low IQ
The information available about the foam impact
during the mission was adequate.
The structure of NASAs Shuttle Program blocked
the flow of critical information up the
hierarchy, If Program managers had understoo
d the threat . . . before Flight Day Seven, . . .
a rescue would have been conceivable.
-- Columbia Accident Investigation Report
14
The Impact of Low IQ

• Information quality problems have been identified
  as an underlying root cause in each of these
  incidents
  

15
Financial Perspectives on Low IQ

• Financial Results and Poor IQ
• The relationship between financial results and IQ
  has been widely researched, but is stubbornly
  difficult to quantify
  
• Research literature reveals numerous examples of
  hard evidence of a connection
  

16
What can be done about IQ?

• What can be done to improve IQ?
• Manage Information as a product
• Manage the provision of information as a service
• Adopt metrics for measuring IQ
• Apply tools and methodologies for improving IQ
• Recognize IQ improvement as a journey

17
Managing Information as a Product

• If companies believe that quality information is
  critical to their success, do they act on this
  belief?
  
• Is information treated as a product as an
  end-deliverable that satisfies consumer needs?
  
• Or is it treated as by-product?

18
Information Management Perspectives
19
Information as Product and Service
Adapted from Kahn, Strong, and Wang (2002)
20
IQ Measurement

• Metrics have been developed for
• Measurement of subjective IQ perceptions
• Measurement of objective qualities
• Context-independent qualities
• Context-dependent qualities

21
Tools and Methodologies

• Total Data Quality Management
• An adaptation of TQM
• Process-focused
• Basic steps are Define, Measure, Analyze,
  Improve
  
• IQ Assessment Survey
• A survey tool for measuring subjective and
  objective dimensions
  
• Data Production Maps
• Used to track relevant attributes of the
  information product
  
• Numerous Commercial Products
• Primarily focused on integrity checking and data
  cleansing, for example
  
• Entity integrity
• Referential integrity
• Column integrity
• User-defined integrity (e.g., business rules)

22
Tools and Methodologies

• The TDQM Cycle

Measure
Define
Improve
Analyze
23
Tools and Methodologies

• IQ Assessment Survey (sample)

Source Cedars-Sinai Information Needs and
Information Quality Survey, 2004
24
Tools and Methodologies

• Data Production Maps

Symbols used
Sample Data Production Map
Source Ballou, Wang, Pazer, Tayi (1998)
25
Organizational Benefits of IQ
26
Journey to IQ

• Because there is nothing more difficult to take
  in hand, more perilous to conduct, or more
  uncertain in its success, than to take the lead
  in the introduction of a new order of things.
  Because the innovator has for enemies all those
  who have done well under the old conditions, and
  lukewarm defenders in those who may do well under
  the new.
• Machiavelli, 1513

• Why is IQ so difficult to achieve?
• Because it requires change
• Why is change so difficult?

27
Journey to IQ

• Managing IQ takes time and effort, but it works!

Source Cedars-Sinai Information Needs and
Information Quality Survey, 2004
28
Taking the Next Step

• Learning more about IQ
• Numerous resources and links can be found at
  http//mitiq.mit.edu
  
• Getting involved in the IQ community
• For Researchers
• Numerous universities around the world have
  IQ-focused research activities
  
• Among the most active are
• Massachusetts Institute of Technology
• University of Arkansas at Little Rock
• Dublin City University
• Hasser-Plottner Institute
• University of South Australia
• Annual International Conference on Information
  Quality (each November)
  
• 2009 conference in Potsdam, Germany
• 2010 conference in Little Rock, Arkansas
• 2011 and beyond TBD
• For Practitioners
• International Association of Information and Data
  Quality (http//www.iaidq.org)
  
• Annual Information Quality Industry Symposium
• Held each July at MIT

29
IQ A Final Thought
All data are wrong. Some data are useful.
-- W. Edwards Demming
30
(No Transcript)
31
References

• The 9/11 Commission Report. (2004). Washington,
  DC National Commission on Terrorist Attacks upon
  the United States.
  
• Ballou, D., Wang, R. Y., Pazer, H., Tayi, G. K.
  (1998). Modeling information manufacturing
  systems to determine information product quality.
  Management Science, 44(4), 462-484.
• Columbia accident investigation board report.
  (2003). Washington, DC Columbia Accident
  Investigation Report.
  
• Kahn, B. K., Strong, D. M., Wang, R. Y. (2002).
  Information quality benchmarks Product and
  service performance. Communications of the ACM,
  45(4), 184-192.
• Report of the presidential commission on the
  space shuttle challenger accident. (1987).
  Washington, DC Presidential Commission on the
  Space Shuttle Challenger Accident.
• Wang, R. Y., Allen, T., Harris, W., Madnick, S.
  (2003). An information product approach for total
  information awareness. Paper presented at the
  IEEE Aerospace Conference, Big Sky, Montana.
• Wang, R. Y., Strong, D. M. (1996). Beyond
  accuracy What data quality means to data
  consumers. Journal of Management Information
  Systems, 12(4), 5-34.