Rationality has been a key debate in decision making producing a number of definitions and models for rational decisions which will be discussed in this essay before concluding whether decisions can be rational exploring both group and individual dynamics. Whereas economic rationality - substantive rationality - focuses on optimization of choice other social sciences focus on the process of a decision - procedural rationality (Simon H. (1982)).Other general definitions of rationality refer to good reasoning and maximisation of information (Arrow K (1986)).
Perhaps the broadest concept of rational decisions is the three distinct definitions of rational belief (Bicchieri C. (1993)) Arguably by this model many decisions are compliant with weak subjective rational belief. Actions are often consistent with thoughts and desires of the decision maker and contain logical reasoning thus can be considered rational. Consistency expressed here is modelled by others as two types of rational choice;
Internal Consistency - Consistency of choice which directly links to consistency expressed in Bicchieri's model.
Self-Interest pursuit - Successfully pursing self-interests (Sen A. (1985)).These collaborate to the idea that a decision can be rational if following individual desires continually, which is probable. From these models, it can be inferred that what may seem irrational to others may in fact be rational due to value judgements, beliefs and interests of individuals (Rose A. (1957)). It is harder to assert a decision as a Strong subjective rational belief as it raises the issue of knowledge. Not only this but the assumptions that systematic mistakes cannot be made raises further issues if this is possible. It could be achievable in scientific arenas such as the Challenger launch in which Thiokol engineers had forecasted effects using previous data to conclude this kind of decision (Vaughan D. (1996)) but this assumes no errors were made in calculation. Finally, objectively rational belief assumes the correct decisions can be attached to a certain outcome. Even if the right outcome is achieved this does not follow that the process to get to that outcome was rational. Thus again this type of rationality is hard to assess as correct decisions are hard to attribute.
According to Decision theorists there is a series of stages which model rational decision making . If follows that if this series of stages is followed the decision can be seen as rational. However, there are a number of assumptions made with this model which must be assessed.
A number of counter arguments stem from these assumptions. Firstly, decision makers are often required to use personal judgements to assess the likelihood of events resulting in imprecision. Uncertainty is a clear example where precise probabilities and information are not fully known thus forcing personal judgements to be made (Knight F.(1921)), violating the assumptions in the model. This links the idea that confidence in personal knowledge and information in a situation causes decision makers to be more likely to bet on uncertain beliefs (Heath C. & Tversky A. (1991)) (Weber M. &Fox C.R(2002)), once more questioning the extent to which full knowledge is available and how far decisions are rational and founded on true and accurate knowledge. Further criticisms following issues of knowledge question whether all outcomes and alternative events are known and equally whether consequences of each event are fully known. Not only does this assume the prior knowledge of these outcomes but it leaves no room for unanticipated consequences and if complying with probabilistic rules these alternatives need to have a definite probability of outcomes which is another unlikely factor (Simon H. (1955)). A preference order also assumes the knowledge of which is best or worst raising the same issues as discussed above. Thus lack of knowledge is key to the discussion against the five stages model. Other criticisms question whether unlimited resources can ever be true, the concepts of economics bases itself on limited resources and many are unlikely to have a luxury of a variety of resources available to process and evaluate decisions. Actions are often cited to be constrained by "income, time, imperfect memory and calculating capacities"(Becker G. (1992)) resulting in "voluntary closure" of alternative solutions due to these prohibitive circumstances (Hansson S. (1994)). Finally some have argued that a number of these stages happen in parallel rather than sequentially so as information is obtained then simultaneously new alternatives are thought of (Witte E.1972)). This has lead to introductions of more cyclical descriptions of this processes incorporating the idea of re-evaluation and re-development (Hansson, S. (1994))
Another model for rational decisions emphasises just three assumptions (Ward E. (1954))
Often an individual has a clear preference regarding events and possibilities. Maximisation is also achievable, in fact individuals are often assumed to maximize welfare as they regard it, despite individual motives and behaviours (Becker G. (1992). However, transitivity has been countered by the idea of acyclicity which permits a preference of A to B and B to C but not necessarily linking preference of A to C due to perhaps incomparability of the two factors (Ward E. (1954)). But overall rationality appears more achievable and realistic in this model.
The rational theory of choice defends a decisions as rational on the premise that decision makers are usually effective in pursuing goals and objectives thus can be seen to act rationally as long as they follow a process of weighing up costs and benefits (Quattrone, G. A. and Tversky, A. (1988)). Thus an appropriate decision given personal beliefs can be achieved and regarded as rational (Little D (1991)). The assumptions in this model include the idea that an agent is resourceful, scarcity results in choices being made, subjective probabilities of future events are held by the decision maker, evaluation is key in the process and maximising utility is the aim (Swatos W. (1998)). The idea that maximisation stems from a cognitive process rather than a motivational one has been questioned as it has been suggested that emotions interfere with the calculation of rationality. The idea of maximisation also raises two alternatives; Satisficing whereby a decision is chosen on it's adequacy to meet criterion whereas Optimizing refers to the choice of the best alternative according to these sets of criterion. (Simon H (1956)) This raises a discussion about what true maximisation refers to and thus which represents true rationality. Satisficing is often used as it is far easier, cheaper and less complex. it allows for omission of some options,and the selection of immediate solutions which meet minimum requirements. It should then be considered whether this reflects true rationality.
Pfeffer's Rational Model is another alternative for the basis of a rational decision. This model however extends the idea of rational decision making into a group complex. Again a number of assumptions are made to achieve a rational decision:
Once more, a number of arguments stem from these assumptions. It is unlikely goals of all decision makers are identical. A group of individuals is likely to have own values unless organizational cultureis remarkably effective. Although seen as dysfunctional, power often plays a key role in organisations and it is perhaps unrealistic to not account for this factor. Power advantages are often exploited and strategies such as cliques and coalitions are often present as interest groups seek to exploit their own aims (Knights D., Willmott H. (2007)).Furthermore, information is again a key issue and accuracy debatable.
The idea of group rationality is also discussed in the groupthink concept (Janis I. (1972)) diagnosed by a series of 8 symptoms. Deterioration of rationally occurs from group pressures and ignorance of alternatives seen in a number of real life events such as the Bay of Pigs invasion. Another example frequently cited is the Asch experiment where around 30% of the subjects conformed to the wrong decision due to group pressure (this countered to nearly no errors in the control group) (Asch S. (1951)). Groupthink is re-enforced by the Risky Shift Phenomenon (Stoner J. (1961)) which emphasises a tendency for groups to be less risk averse than when individually are making low or high risk decisions. Inherently high risk-takers feel more confident in a group as they cannot be wholly responsible for the decision thus their ability to persuade others creates higher risk decisions.(Collins B. & Guetzkow H.(1964) ). There is also a tendency for decisions to become more turbulent than those of an individual who would be more likely to take medium risk decisions. Although evaluation of the size of this effect is necessary. It should however be noted that groupthink has a higher probability under certain conditions such as homogeneous backgrounds of decision makers, lack of external opinion and lack of rules for decision making. Thus it has been argued that by altering group dynamics and combating these issues one can avoid the groupthink phenomenon.
Irrelevant of model, general irrationality has also been cited to occur due to a number of factors such as conflicting/changing goals, limited/misled information, logic errors, personalities and personal ambition. Examples include ineffective use of logic resulting in incorrect decisions (Wason & Shapiro D.(1971))and illogical decisions despite familiarity with situations (Anand P (1993). Other sources of irrationality originate from decision makers' errors such as lack of thought, poor reasoning or weakness of will (Sen A. (1985)). Two clear irrationalities are thus outlined, Correspondence Irrationality resulting in rejection of the decision maker's own choice on reflection and Reflection Irrationality referring to failure to be rational due to poor reasoning capabilities or inability to identify reasoning which would achieve objectives more successfully (Sen A. (1985)). These seem common occurrences showing human inability to create a rational decision. It could thus be argued that these are often inherent to human behaviour so questioning whether decisions can be seen as rational. Bounded rationality (Simon H. (1955)) also questions whether rationality is achievable. Bounded rationality expresses the "limitations of human capacity to process information" (March J. & Simon H. (1958)). These include issues such as: incomplete knowledge, personal and political influences, inability to process information and limits on ability to forecast (Simon, H. (1982)). Thus is should be debated whether bounded rationality results in irrational decisions or should simply be regarded as a limitation on rationality.
In conclusion, it is clear that there are a number of different definitions, criteria and models for rational decisions. Based on this is it difficult to say whether decisions can be unequivocally rational but certainly can be rational within given criteria of models. In particular Bicchieri's model highlights broad conditions for which decisions can be seen as rational, whereas other models such as the five decision stages have a number of assumptions which are questionable although the cognitive process has rational merit. Group decisions threatened by the groupthink concept seem more likely to result in an irrational decision. However, if this concept is avoided groups are capable of similarly rational decisions to those of individuals, perhaps more so on knowledge decisions and less so on motivational ones. On using these terms it must be considered whether failure to meet conditions outlined in any of these models leads to irrationality or act as just limitation on rationality itself.
Reference List
