A methodology to reveal hidden assumptions of stakeholders in a group or organization confronted with some difficult situation.
This methodology, proposed by I.I. MITROFF and H.A. LINSTONE, is described by them in their 1993 book. Their general view is that it is "far better to debate a question without necessarily settling it than to settle a question without debating it" (p.146) Indeed, this last way is the best recipe to obtain an incomplete or false view of a situation or problem.
However, questions are settled without debate at least partly because their do not come out clearly and as a result, are easily ignored or jumbled. To make them appear clearly is specially important when they include conflicting views as held by the different stakeholders, because this is the only way to obtain a critical evaluation of the whole matter under consideration.
According to MITROFF and LINSTONE, the first step is to find out who are the stakeholders, that can affect or be affected by the situation as it is or as is could evolve. This they do by establishing a graph of all stakeholders (p.141). For example, in the case of a drug company, these are: the customers and patients; the physicians; the pharmacists; the suppliers; the salesmen; the competitors; government agencies; the holding company; the company management and the stockholders (not to be confused with the much more general class of the stakeholders!). In this way, it will become possible to obtain a most complete view of the situation.
The second step is to make the different groups of stakeholders expose their assumptions, which are mapped one by one into grids whose coordinates are "certainty-uncertainty" and "least to most important" (p.144)
In this way most hidden aspects can be brought to light and seriously debated. J. WARFIELD 's Generic Design provides quite similar methods to avoid underconceptualization and incoherent debate (1994b).
Anyhow, assumptional analysis is by nature an open process. M. JACKSON considers that even after a synthetic list of assumptions has been established and a rating chart constructed ( see hereafter), this is still a mere compromise among different- and possibly divergent- opinions. Thus "assumptions (should) continue to be negotiated and modified�for as long as progress is being made "(p. 228, 2000)
It is moreover obvious that the issue or situation may evolve. Accordingly assumptions may never be definitive, and corrective iterations of the analysis may be needed.