The
Transformations In The Five Tier Knowledge Management
Transformation Matrix
Richard C. Hicksı, Stuart D. Galup², Ronald Dattero³
ıTexas A&M
International University ,
² Florida Atlantic University, ³ Missouri
ABSTRACT:
The most common paradigm for discussing the relationships between data, information, and knowledge is the Knowledge Hierarchy. The Knowledge Hierarchy is based on the assumption that data will be transformed into information, and that information will be transformed into knowledge. Others add two transformations in the opposite direction -- knowledge into information and information into data.
We will show that this set of transformations is incomplete. In addition, the Knowledge Hierarchy is limited in its usefulness for Knowledge Management applications and research because it includes only codified systems and omits behavioral systems. In this paper, we describe a more complete and expressive five tier representation (matrix) and the twenty knowledge transformations within it.
Keywords: Knowledge Management, Data, Information, Knowledge
1. Introduction
Knowledge Management (KM) researchers and practitioners have adopted many
definitions and concepts over the years to explain the relationships between
data, information, and knowledge. The initial attempts to explain the
relationships are rooted in Management Information Systems (MIS). The
relationships are based on an evolving hierarchy that starts with transaction
processing data, evolves into management information as the data is aggregated,
and further aggregation of the data results in decision support and executive
information data. This common paradigm
is widely used in the KM literature as the Knowledge Hierarchy (Nissen and Espino, 2000;
Figure 1: Knowledge
Hierarchy
The Knowledge Hierarchy does not support direct transformations from data to knowledge, but only the transformations from data to information, and from information to knowledge. Extensions to the Knowledge Hierarchy support transformations from knowledge to information, and from information to data (Ackoff, 1996), but not directly from knowledge to data.
In this paper, we show that this set of four transformations is far from complete. Further, the Knowledge Hierarchy is limited in its usefulness for Knowledge Management applications and research because it includes only codified systems and omits behavioral systems. In the next section, an overview of some extensions to the Knowledge Hierarchy and the additional knowledge transformations will be presented. Then, we review an extension to the Knowledge Hierarchy developed by Hicks et al. (2005) called the Five Tier Knowledge Management Hierarchy. As representing these five tiers as a hierarchy implicitly limits the transformations to adjacent tiers, we represent their five tiers as a matrix to explicitly show that the transformations can be between any two tiers. Next, we discuss the complete set of transformations in the Five Tier Knowledge Management Transformation Matrix (5TKMTM). Finally, we present our conclusions and directions for future research.
2. Extensions Of The Knowledge Hierarchy
Variations on this central theme of the Knowledge Hierarchy include Tuomi (1999), who proposes an inverted hierarchy. His position is that knowledge is required to represent information, which must be done to store data. Nissen and Espino (2000) and Spiegler (2000) extend this concept with a model containing double hierarchies, as shown in Figure 2. One hierarchy models the view of the knowledge seeker, where the second hierarchy is inverted and represents the view of the knowledge creator. From the seeker’s perspective, data is placed in context to create information, and information that becomes actionable is knowledge. From the creator’s perspective, knowledge is necessary to create information, which in turn is necessary to create data.
Figure 2: Double
Hierarchy
The transformations supported by the Double Hierarchy are incomplete. The Double Hierarchy supports the transformations from Data to Information, and from Information to Knowledge, but does not support the direct transformation from Data to Knowledge. The Double Hierarchy supports the transformations from Knowledge to Information and Information to Data, but does not support the direct transformation from Knowledge to Data.
The Knowledge Hierarchy and Double Hierarchy do not fully express the relationships between the different forms of knowledge. The proliferation of Internet applications and the integration of information systems through the value and supply chains further the argument that a simple data-information-knowledge hierarchy does not support the complexity of knowledge and its interaction with data and information. More complex information systems (e.g. expert systems) clearly demonstrate the springboard effect that can occur when data transfers immediately to knowledge or when information or knowledge can become a source of data.
3. The Five Tier Knowledge
Management Hierarchy
The Knowledge Hierarchy was designed to support Management Information Systems (MIS) research, and focuses on codified knowledge. Knowledge Management, on the other hand, also includes the management of personal knowledge, making the Knowledge Hierarchy inappropriate for Knowledge Management research.
Hicks et al. (2005) extend the Knowledge Hierarchy to make it more appropriate for Knowledge Management research by adding a new Personal Knowledge Class consisting of two tiers -- the Individual Tier and the Innovation Tier. As Individual Knowledge creates, uses, and maintains all of the tiers of the Codified Knowledge Class, it is positioned as the foundation of the hierarchy. Innovation is the highest level because it integrates all of the other tiers, using strategy to exploit both Personal and Codified knowledge assets (Edvinsson et al. 2004). In an attempt to clearly distinguish the many definitions of data, information, and knowledge, Hicks et al. (2005) derived the following terms in their paper.
Tier 1: Individual (Knowledge) is defined as
“knowledge contained only in the mind of a person.”
Tier 2: Facts are defined as “atomic
attribute values about the domain.”
Tier 3: Influences are defined as “data in context that has been processed and/or prepared for presentation.”
Tier 4: Solutions are defined as “clear
instructions and authority to perform a task.”
Knowledge
Tier KM Application Example INDIVIDUAL Human Mind Stored
in human mind, accessed by “yellow pages” Bain
and Company (Hansen et al, 1999), Shell (Earl, 2001) FACTS Documents Databases Data
Warehouses Databases,
data warehouses Frito-Lay
(Applegate, 1992), CIGNA (Turban et al, 1999) INFLUENCE Decision
Support Systems Learning
Systems Yellow Pages Reports Learning
systems, DSS, reports, yellow pages, knowledge pooling Visa (Turban et al, 1999),
Skandia (Earl, 94), IBM (Willigan &,
Mullen, 2000), Bain and Company (Hansen et al, 1999), Shell (Earl, 2001) SOLUTIONS Intelligent
Systems Best
Practices Best
Practices, expert systems Shell
(Earl, 2001) INNOVATION Reengineering Knowledge-based
goods and services Integrates
various KM systems with corporate strategy Frito-Lay
(Applegate, 1992),
Tier 5: Innovation is defined as “the exploitation of
knowledge-based resources.”
Table 1: The Five Tier Knowledge Management Hierarchy
Hicks et al. (2005) show that their hierarchy includes all of the KM systems
in Earl’s (2001) KM taxonomy.
Earl’s (2001) taxonomy consists of three major branches of
KM: the
The Individual Tier includes Earl’s
Hicks et al. (2005) represent their five tiers as a hierarchy. As representing these five tiers as a hierarchy implies that the transformations are only between adjacent tiers we represent their five tiers as a matrix because the transformations are not limited to adjacent tiers – the transformations can be between any two tiers. That is, there are 20 possible transformations: four transformations between each of the five tiers.
4. Knowledge Transformations
The traditional Knowledge Hierarchy supports the transformation of Data to
Information and the transformation of Information into Knowledge. The
transformations from Knowledge to Information and Information to Data are
supported in the double hierarchy and the reverse hierarchy as previously
noted. Table 2 shows the knowledge transformations supported by the 5TKMTM, and
identifies the transformations supported, not supported, and not possible in
the Knowledge Hierarchy.
INDIVIDUAL FACTS INFLUENCE SOLUTIONS INNOVATION INDIVIDUAL Human Mind --------------- Not possible in Knowledge Hierarchy Not possible in Knowledge Hierarchy Not possible in Knowledge Hierarchy Not possible in Knowledge Hierarchy FACTS Documents Databases Data Warehouses Not possible in Knowledge Hierarchy --------------- Supported by Double Hierarchy Not Supported by Double Hierarchy Not Possible in Knowledge Hierarchy INFLUENCE Decision Support Systems Learning Systems Yellow Pages Reports Not possible in Knowledge Hierarchy Supported by Double Hierarchy --------------- Supported by Double Hierarchy Not Possible in Knowledge Hierarchy SOLUTIONS Intelligent Systems Best Practices Not possible in Knowledge Hierarchy Not Supported by Double Hierarchy Supported by Double Hierarchy --------------- Not possible in Knowledge Hierarchy INNOVATION Reengineering Knowledge-based goods and services Not possible in Knowledge Hierarchy Not possible in Knowledge Hierarchy Not possible in Knowledge Hierarchy Not Possible in Knowledge Hierarchy ---------------
Table 2: Transformations In The Five Tier Knowledge Management Transformation Matrix
First, compare the six transformations possible in the Knowledge Hierarchy and the three corresponding components of the 5TKMTM, the Facts (corresponding to Data), Influences (corresponding to Information), and Solutions (corresponding to Knowledge) Tiers. As indicated previously, the Knowledge Hierarchy does not support direct transformations from Data to Knowledge, but only the transformations from Data to Information, and from Information to Knowledge. Similarly, the Double Hierarchy does not support the direct transformation from Knowledge to Data, but only the transformations from Knowledge to Information and Information to Data.
Despite their absence in the Knowledge Hierarchy, these transformations do occur. Neural Networks transpose Data into Knowledge without an intermediate transformation into Information, which provides evidence that direct transformations occur between Data and Knowledge. Higher levels of the 5TKMTM are transposed into any of the lower levels by the changing of context or by decomposition. The Best Practices used by Shell (Earl, 2001) can be used as a directory of Shell drilling sites, for example, demonstrating the transposition of Knowledge directly to Data by changing context.
Next, let us consider the remaining fourteen transformations that are possible in the 5TKMTM. These transformations all utilize Personal Knowledge, which is not included in the Knowledge Hierarchy.
Half of these transformations are used to transform a higher tier into a lower tier. Decomposition of any higher tier or changing of context will produce instances of a lower level tier. For example, the Innovation of the XCON expert system (Bachant and Solloway, 1989), a Solution which was used to implement a corporate strategy, can be decomposed into a large number of Facts, such as the fact that a specific hard disk has a specific part number, or Influences, such as the information that a specific hard disk controller can be used with a certain number and type of hard disk drives, or a Solution in the form of the expert system without the corresponding corporate strategy. Any codified source can be transposed into Individual knowledge when the Individual uses the computer to acquire the knowledge. An Innovation, such as a selling philosophy, can be transformed into Individual knowledge without codified systems or the transformation into Solutions, Influences, or Facts.
The remaining seven transformations occur when knowledge at one level is transposed into a higher level. Individual knowledge can be transformed into the codified tiers of Facts, Influences, and Solutions through knowledge acquisition. Individual knowledge, especially when used strategically, can be transformed directly into Innovation. Facts can be transformed into Innovation by providing knowledge-based goods and services. Influences can be transformed into Innovations by strategy, such as process reengineering. Solutions, such as XCON, can also be transformed into Innovation by strategy.
5. Conclusions And Future Research
This paper discussed the paradox that exists in the KM literature regarding the existing descriptions of the relationships between data, information, and knowledge and the limitation it presents to KM researchers and practitioners. To address this issue, we presented the 5TKMTM, which is designed to facilitate KM research. Unlike the Knowledge Hierarchy, the 5TKMTM contains personal knowledge, which is one of the cornerstones of KM. Additionally, the 5TKMTM extends the understanding of knowledge transformations by identifying 20 transformations: four transformations between each of the five tiers
There is a common theme in all the variations of the Knowledge Hierarchy. The paradigm can be used to predict the action ability and volume of each tier in the hierarchy. By this we mean that Knowledge is the most actionable level but the rarest, where data is the least actionable level but has the greatest volume (Nissen and Espino, 2000). The next logical step in this research is to determine if the two tiers added by the 5TKMT maintain the predictive characteristics of the three tiers in the Knowledge Hierarchy. A starting set of hypothesis might include the following: KM assets at the lower end of the 5TKMT have larger volume, lower specialty, lower action ability, lower risk, lower cost, lower potential payback, and wider dissemination than those at the higher end of the 5TKMTM.
6. Acknowledgement
A preliminary version of this paper was presented at the 2005 Decision
Science Institute conference held in
7. References
Ackoff, Russell L. (1996) “On Learning and the Systems that Facilitate It”, Center for Quality of Management Journal, (5)2, pp. 27-35.
Alavi, M. & Leidner, D. (1999) “Knowledge Management Systems: Issues, Challenges,and Benefits”, Communications of the Association for Information Systems, 1, pp. 1-37.
Applegate, L. M. Frito-Lay, Inc.: a strategic transition (consolidated).
Case Study 9-193-040,
Barker, V., O'Connor, D., Bachant, J., and Solloway, E., (1989) “Expert systems for configuration at Digital: XCON and beyond”, Communications of the ACM, (32)3, pp. 298 – 318.
Earl, M. (2001) “Knowledge Management Strategies: Toward A Taxonomy”, Journal of Management Information Systems, (18)1, pp. 215-233.
Earl, M. J. Knowledge as a Strategy: reflections on Scandia International
and Shorko Films. In C. Ciborra
and T. Jelassi (eds)
Strategic Information Systems: A European Perspective. John
Wiley & Sons,
Edvinsson, L., Dvir, R., Roth, N., and Pasher, E. (2004) “Innovations: the new unit of analysis in the knowledge era”, Journal of Intellectual Capital, (6)1, pp. 40-58.
Hansen, M., Nohria, N., and Kierney, T. (1999), “What’s Your Strategy for Managing Knowledge?”, Harvard Business Review, (77) 2, pp. 106-117.
Hicks, R., Dattero, R., and Galup, S. (2006) “The Five Tier Knowledge Management Hierarchy” Journal of Knowledge Management, (10)1, pp. 19-31.
Speigler,
Tuomi,
Turban, E., Rainer, R. K. Jr., and Potter, R. E. Introduction to Information Technology, John Wiley and Sons, New York, 2001.
von Krough, G., Ichijo, K., and Nonaka, I. (2000) Enabling Knowledge Creation: How to Unlock the Mystery of Tacit Knowledge and Release the Power of Innovation, Oxford University Press, New York, NY.
Willigan, W. and Mullen, M. (2000), “How to make the most of
intellectual capital”, Financial Times,
Contact the Authors:
Richard C. Hicks, Department of MIS and Decision Science, Texas A&M International University, 5201 University Blvd, Laredo, TX 78041; Email: rhicks@tamiu.edu
Stuart D. Galup, Department of Information
Technology & Operations Management,
Ronald Dattero, Department of Computer Information Systems, Missouri State University, 901 South National Avenue, Springfield, MO 65897; Email: RonDattero@MissouriState.edu