Decision Making / 4 / 4.9 / 4.9.7

4.9.7 Tame Vs. Wicked Problems—Level Of Analysis

Every tame problem has an identifiable, certain natural form; there is no need to argue about the level of the problem. The proper level of generality can be found for bounding the problem and identifying its root cause. Every wicked problem can be considered as a symptom of another problem. It has no [single] identifiable root cause; since curing symptoms does not cure problems, one is never sure the problem is being attacked at the proper level.

In each of the accident cases studied, relevant concepts were found to be strongly-coupled. That is, there were many instances of one problem being a symptom of another. Concept mapping was used to reveal the true nature of causality. For example, Coroner Madden’s 657-page report on the death of Katie Bender was dissected. A series of complex maps were built. Some contained over 300 concepts linked in highly interrelated ways. The structure of the problems mapped suggest that there were many issues to be addressed through multi-pronged strategies.

In his summation regarding one aspect of ‘systemic failure in the project’, for example, Coroner Madden focused on nine contributory factors. Seven (± 2) key concepts (or factors) were found in many such summaries. This accords with the ‘magic number’ identified by Miller (1956). However, the act of reducing the detail in a document to a level expected by readers can be problematic. In this case Coroner Madden carefully considered many tightly-coupled factors to produce his summary list. Not all analysts have the skills needed to produce such incisive summaries.

As aids to selecting the correct level of aggregation for problem consideration, that is the level of analysis:

• Coyle (1996) offers a ‘cone of influence diagrams’ as a conceptualisation tool, encouraging us to view and analyse problems at various levels, moving between those levels as needed to enhance understanding and facilitate analysis.
• Eden and Ackermann (1998a) suggest we might consider a hierarchy of ‘tear drops’ within which goals, aspirations, and strategies are contemplated. They also suggest, at times, we take a ‘helicopter view’ of problems to enable thinking at various levels of aggregation.
• Checkland and Scholes (1999) suggests we take a world-view, weltanschauung, made up of many different perspectives.

Checkland and Scholes (1999) also suggests we look for systemicity, which is the systemic behaviour (produced by multiple feedbacks and delays) exhibited by systems or systems-of-systems. They and Flood (1999) caution against attempting to model complete systems or systems-of-systems. Rather they suggest we seek to understand the causes of systemic behaviour within a defined problem space and learn to manage within the complexity.