The October issue of JASIST has an article about measuring information quality. (Cite: Besiki Stvilia, Les Gasser, Michael B. Twidale, Linda C. Smith (2007). “A framework for information quality assessment” JASIST, 58, 12 (1720-1733). Here is a copy of the paper in different format, although I think the text is exactly the same.
The authors start off with:
“Information is increasingly becoming a critical resource in contemporary societies and organizations. For institutional and individual processes that depend on information, the quality of information (IQ) is one of the key determinants of the quality of their decisions and actions. The familiar “garbage in, garbage out” mantra of computing expresses the problem succinctly. The amount and diversity of information available, and the number and variety of information publishers have grown at an unmanageable rate. Unfortunately, as more information becomes available for use, it becomes increasingly difficult to identify “garbage.” Historically, there have been culturally sanctioned mechanisms of IQ assurance, such as the peer review process for research, human screening and cleaning for database entries, and careful editing processes for books and magazines. However, these are breaking down for reasons of scale and cost (McCook, 2006).”
They go on with some academic bla-bla-bla-ing before getting to a framework for measuring IQ. This is like a list of heuristics broken into these three categories:
- Intrinsic IQ: This category includes dimensions of IQ that can be assessed by measuring internal attributes or characteristics of information in relation to some reference standard in a given culture. Examples include spelling mistakes (dictionary), conformance to formatting or representational standards (HTML validation), and information currency (age with respect to a standard index date, e.g., “today”).
- Relational or contextual IQ: This category of IQ dimensions measures relationships between information and some aspects of its usage context. One common subclass in this category includes the representational quality dimensions. Those dimensions measure how well an information entity reflects (maps) some external condition (e.g., actual accuracy of addresses in an address database) in a given context.
- Reputational IQ: This category of IQ dimensions measures the position of an information entity in a cultural or activity structure, often determined by its origin and record of mediation.
Here’s the full list of metrics:
Intrinsic
1. Accuracy/Validity
2. Cohesiveness
3. Complexity
4. Semantic Consistency
5. Structural Consistency
6. Currency
7. Informativeness/Redundancy
8. Naturalness
9. Precision/Completeness
Relational/Contextual
10. Accuracy
11. Accessibility
12. Complexity
13. Naturalness
14. Informativeness/Redundancy
15. Relevance
16. Precision/Completeness
17. Security
18. Semantic Consistency
19. Structural Consistency
20. Verifiability
21. Volatility
Reputational
22. Authority
Complete, ain’t it? Not really practical for us regular guys on the street. Someone needs to come along and slim this done before it has any real use outside of academic ivory towers.
I’m most interested in Authority and Credibility, but that seems to stand on its own in this framework, whereas other areas get a lot of detail and attention.
EXPERIENCING INFORMATION 