APPROACHES FOR ASSIGNING A VALUE TO INFORMATION
The absence of a reliable basis for assigning a value to information means that organizations' investments in information technology are unlikely to based on the formalistic methods (e.g., capital budgeting, cost-benefit analysis) used to discipline asset-acquisition decisions in other areas. All we have are more or less imperfect approaches:
· Denominated (Computed) Valuations: . The three most obvious and direct bases for assigning a dollar value to information are not really of much relevance in the context of management support systems:
─ Cost (acquisition-related expenses)
─ Exchange Value (what someone is willing to pay/trade to secure information)
─ Suppression Value (what would be paid to prevent the dissemination of information)
As in many other areas, cost may not be at all well-related to value (e.g., some people pay money to buy “lottery numbers”, an investment which has no value whatsoever in random drawing scheme. The other two approaches are suspect if only because value can generally only be determined entirely after-the-fact and with respect to particular items ─rather than classes of─ information
· Value-in-Use/Ordinal Assessments (re: relative leverage): In certain problem-solving situations, a categorical (qualitative or fuzzy) assessment of the value of a parcel/quantum of information may be possible. For example, in the case of the TWA flight that was destroyed near Long Island a few years ago, the relatively most important or highest-leverage informational items would be those that would point directly to a singular cause (sabotage, pilot error, mechanical failure, attack by aliens, etc.); all else would be of lesser value. Note also this correlative suggestion: Decisions should generally be made by those who need the least in the way of additional information!
· Academic: The value ascribed to an item of information is a reflection of the extent to which it comes as something of a surprise to those obtaining it, as per these thoughts from Richard Dawkins' recent book [UNWEAVING THE RAINBOW, 1998]:
Information, in the technical sense, is surprise value, measured as the inverse of expected probability. Redundancy is the opposite of information, a measure of unsurprisingness…Redundant messages…are not information because the receiver…already knows what is coming.…Nervous systems exploit the massive redundancy in all sensory information [as] most nerve cells are biased to signal changes in the world….exploitation of redundancy ¾screening out of [sic] the sameness of the world¾ goes on in cells that tell the brain about changes in light, changes in temperature, changes in pressure….Everything about the world is signaled as change, and this is a major economy.
A practical derivative of this approach for the design of management support systems (particularly pertinent to real-time operations) is the emphasis on recognizing and discarding redundancies…the IT counterpart to the concept of management by exceptions. It may sometimes be possible to get a rough estimate of the 'surprise-value' associated with an information item by looking at the magnitude of the changes it evokes in the structure and/or substance of the previously existing information base or predicate set. A practical corollary is that you continue collecting additional data until they cease to excite any changes in our appreciation of a situation. Harder to operationalize is the information economics rule that expenditure on information acquisition should cease at the point where its apparent marginal cost approaches its apparent marginal benefit (incremental value). But in any event, what in many real-world decision situations puts an end to the information acquisition process is a deadline…you run out of time to analyze and must act! Note also that the recognition/elimination of redundancies is also the technical key to well-paced (effectively real-time, streamed) digital video; in refreshing a screen, only those items that have changed since the previous version need to be communicated.
· Imputational: For any decision situation, the value of information has to be determined prospectively…before a decision commitment is made or a course-of-action undertaken. Seldom will we be able to compute an impact value for an item/set of information on an organization's aggregate performance, or bottom-line. Often, the best we can hope to is try to impute a value to information in reflection of any reduction in the Expected Value of Decision Error that results from its acquisition, as per the following: