Figure 1: A conflict-theory model of decision making applicable to all consequential decisions (Janis and Mann, 1977, p. 70)

Summarised, due to many cognitive limitations individual decision makers tend to reflect a number of characteristics. Harrison states the following overview (Harrison, 1987, p. 10):

·        Human subjects tend to overestimate low probabilities of events and underestimate high probabilities.

·        Individuals appear to be insensitive to the sample size of their observations.

·        Individuals adjust their first approximations to an estimate on the basis of additional evidence.

·        Individuals tend to be overconfident of their ability to estimate the probability of an uncertain event.

·        Individuals tend to overestimate the probability of events that actually occur, as well as the extent to which they or others would have been able to predict past events.

·        Human subjects reveal a tendency to compare pairs of alternatives rather than a whole list.

·        Individuals tend to minimise reliance on explicit trade-offs or other numerical computations.

·        Individuals exhibit choices that are sometimes inconsistent and intransitive.

·        On balance, human subjects often behave in ways that dispel the notion that they are proper decision theorists with well defined objective functions.

3 Groups and decision making

Group decision making is not simply an extension of individual decision making. When people interact in a group a new entity is created with its own complexity and dynamics, also influencing the behaviour and perception of the group members. Groups are defined as: “a number of persons who communicate with one another often over a span of time, and who are few enough so that each person is able to communicate with all the others, not at second-hand, through other people, but face to face” (Homans, 1950, p. 1).

Groups as decision makers have both strengths and weaknesses. The strengths of groups are a greater sum of knowledge and information, resulting in a greater sum of approaches to a problem. Besides that, participation in decision making increases general acceptance of the final choice, whereas direct participation in the communication leads to a better comprehension of the decision (Harrison, 1977, p. 273).

Examples of weaknesses are (Harrison, 1977, p. 274):

·        Social pressure. The desire to be accepted as a group member tends to silence disagreement and favour consensus.

·        Acceptance of solutions. The first solution that appears to receive strong support from the majority of the members or even from a vocal minority, tends to be accepted. Solutions of higher quality introduced after this first solution has been accepted have little chance of receiving serious consideration.

·        Individual domination. In most leaderless groups a dominant individual emerges and exerts more than his, or her, share of influence on the decision. Even a leader who is appointed by the group tends to exert a major influence on the selection of a preferred alternative.

·        Winning the decision. The appearance of several alternatives often causes the members to support a particular position. These preferences often take precedence over finding the best solution and the result is a compromise decision of lower quality.

·        Reduction of responsibility. In a group situation the sense of personal responsibility for the decision is reduced to the extent that the decision is shared (Harrison, 1977, p. 298).

There can be many reasons why group decision making goes wrong. They all relate to conflicting objectives of the members: twist for resources, different value systems, and interdependency.

Presence of conflicts can be recognised by signals as interference, or the lack of consideration shown, overstatement of issues, withholding of information, annoyed reactions and distrust (Corwin, 1969, p. 507-520).

Another problem with group decision making is groupthink: a collective pattern of defensive avoidance (Janis and Mann, 1977). The eight symptoms of groupthink are:

·        An illusion of invulnerability, shared by most or all of the members, which creates excessive optimism and encourages taking extreme risks.

·        Collective efforts to rationalise in order to discount warnings which might lead the members to reconsider their assumptions before they recommit themselves to their past policy decisions.

·        An unquestioned belief in the group’s inherent morality, inclining the members to ignore the ethical or moral consequences of their decisions.

·        Stereotyped views of rivals and enemies as too evil to warrant genuine attempts to negotiate, or as too weak or stupid to counter whatever risky attempts are made to defeat their purposes.

·        Direct pressure on any member who expresses strong arguments against any of the group’s stereotypes, illusions, or commitments, making clear that such dissent is contrary to what is expected of all loyal members.

·        Self-censorship of deviations from the apparent group consensus, reflecting each member’s inclination to minimise to himself, or herself, the importance of doubts and counter-arguments.

·        A shared illusion of unanimity. partly resulting from this self-censorship and augmented by the false assumption that silence implies consent.

·        The emergence of self-appointed mindguards: members who protect the group from adverse information that might shatter their complacency about the effectiveness and morality of their decisions.

Conditions that increase the chances of groupthink are: high cohesiveness, insulation of the group, lack of methodical procedures for search and appraisal, directive leadership and high stress with a low degree of hope for finding a better solution than the one favoured by the leader or other influential persons (Janis and Mann, 1977, p. 132).

As IT evaluation requires a lot of knowledge of the entire organization, it is usually performed in a group process. A composite approach that encourages the strengths and weaknesses of both individual and group decision making seems appropriate. Harrison states the following suggestions (Harrison, 1977, p. 300):

·        In setting objectives, groups may be selectively superior to individuals simply because more knowledge is available from the combined membership.

·        In the search for alternatives, many individual efforts, either by members of the group or their supporting staff, are necessary to ensure a broad search in the areas that apply.

·        In evaluating the alternatives revealed mainly by individual efforts, the collective judgement of the group, with its broader range of critical viewpoints, has a definite potential to be superior.

·        The moment of choice in group decision making joins individuals who desire risk and those who wish to avoid it. The actual choice, therefore, may involve less a risk than one made by an individual, depending upon the individual’s willingness to accept uncertainty. In any event the group decision is more likely to be generally accepted as a result of the participation of those who will be affected by its consequences.

·        A decision made by an individual or a group must be implemented by individual managers who are personally responsible for the outcome of their actions. It is good to recall that it is not practical to hold a group responsible for its choice. Therefore, responsibility falls on the individual manager.

·        Follow-up and control procedures must also be carried out by the individual managers at the point of implementation. Of course, a decision making group can, and perhaps should, keep an eye on such procedures. Still, the direct responsibility to ensure an outcome that will meet the original objective lies with the individual manager.

The scope of decision making does not stop at the level of a group. After all, groups are part of an organization, where, numerous groups tend to accomplish their objectives. Consequently, decision making does not stop at the level of the organization, because organizations are related to each other in society, etcetera
 

4 Quality characteristics of decision making

Janis and Mann extracted seven important criteria to determine whether a decision making process is of high quality (Janis and Mann, 1977, p. 11). These criteria are stated in table 2.

Table 2: Quality characteristics of a decision making process (Janis and Mann, 1977, p. 11)

The concept of quality in  the above enumeration is closely related to the concept of rationality. Simon defines the concept of bounded rationality as follows (Simon, 1979, p. 501):

·        Look at satisfactory choices in stead of optimal ones.

·        Replace abstract, global goals with tangible subgoals, whose achievement can be observed and measured.

·        Divide up the decision making task among many specialists, co-ordinating their work by means of a structure of communications and authority relations.

The concept of rationality can refer to many characteristics of the decision making process. It can refer to the search for alternatives, the search criteria, the weighing of alternatives, etcetera. Harrison is one of the very few who defines the concept of rationality in more detail (Harrison, 1977, p. 105). He points at the difference between rational decisions and the attitude of decision makers towards the decision making process. Harrison defines rational decisions: “all that is necessary to make a choice a rational one is that an objective exists and that the decision maker perceives and selects some alternative that promises to meet the objective” (Harrison, 1977, p. 107). This implies that it can indeed be rational to perceive an emotional objective. In contradiction to the situation where an emotion obstructs a rational weighing of the consequences. “In managerial decision making, rationality and emotionality are not the opposite ends of the spectrum” (Harrison, 1987, p. 108). Taylor emphasises to train decision makers in order to better cope with cognitive strain and expand the decision making performance (Taylor, 1975, p. 419). 

5 Decision supporting methods

Many methods and techniques have been proposed to assist in the evaluation of information systems. A number of overviews have been published (Swinkels and Irsel, 1992; Farbey et al., 1992; Willcocks, 1992; Blackler and Brown, 1988; Powell, 1992; Berghout and Renkema, 1994). Berghout and Renkema (1994) refer to over sixty methods. Research showed that more mechanical methods are easier to apply and therefore might lead to better results (Klompé and Berghout, 1995). However, the disadvantage of these methods is that decision makers in spite of slightly better results, are less confident of their decision and criticise the applied method more. Applying more mechanical methods means that (dis-)advantages of alternatives have to be expressed in figures after which the ranking is produced applying some algorithm (some method of weighing and scoring). Not only it is hard to express all (dis-)advantages in figures, many decision makers experience this unauthorised addition of distinct units of measure as unpleasant. Decision makers also feel that they have too little influence on the actual assignment of priorities to alternatives and many of them believe that the outcomes of the method are not correct, because they are not corresponding with their expectations (Klompé and Berghout, 1995).

Most methods in literature support just a little part of the decision making process, being the evaluation of the various alternatives. Others phases are hardly supported by these methods, although a number of software development methods offer some support during problem analyses and searching for possible alternatives. More structured evaluations limit the chance of overlooking important (dis-)advantages of an investment proposal, but do not guarantee a decision of high quality, as the actual evaluation of the alternatives, as well as the final decision making, is still done by the decision makers involved.

Further on, working more or less carefully is just one aspect of a variety of aspects influenced by individual and social aspects, for the decision making process is still a human process in which both individual and social aspects take part and influence the final decision. Individual aspects concern mainly psychological characteristics like personality, motivation and expectations regarding the alternatives and the goal to reach. Social aspects on the other hand, consider decision making as a group process and include for example conflicts, powerstructures and the co-operation within the group. None of the mentioned decision supporting methods supports one of these aspects. All restrict themselves to what criteria to evaluate the alternatives on and how to define a priority list based on this evaluation. This priority list is defined as the result of the evaluation and is used to make a decision in the next phase of the decision making process.

Discussions are only successful when they have an open character, competitors don’t take up not-binding positions but feel committed to the goal of the discussion, motivate their points of view, remain consistent, give only relevant arguments and stay business-like. This leaves two questions. First, which circumstances increase the chance of good decision making processes and good discussions? Second, what measures are permitted when a discussion tends to go wrong and is there a solution for this from the social sciences? This solution exists and contains, without mentioning the profound philosophic differences in approach, two crucial factors: structural factors and behavioural factors.

The structural factors influencing the rationality of decision making, concern the composition of the group. Important are possible differences in power or contrasting interests, mutual dependence of the group’s members and the setting of the group.

Differences of power within a decision making group are often considered to be failing factors. They would be at the cost of the openness of the discussions, they would stimulate groupthink and have a negative influence on the quality of decision making. On the other hand, differences in power speed-up decision making and improve the co-ordination of the group process (Pröpper and Bleijenberg, 1995). Senior managers, whose position is not at stake, can allow themselves to be reasonable. The power-paradox says that it comes to power exercising when actual differences in power are smaller. Just when differences in power are small, a discussion often degenerates into an endless debate in which people try to steal a march upon somebody instead of discussing each other’s arguments. In the last case, a decision making process rather becomes a strategic game then a process in which the results and contents are central. Of course, it is not beneficial to the process when it is led by a dictator. The personal characteristics of the most influential participant are crucial. If this person already has an explicit opinion about the desired solution of a problem, this could cause an open discussion to fail due to differences in status.

Concerning the multifarious character of the decision making group two aspects can be discerned. First, there are the differences of interests between particular departments or sub-organizations. For example, in a situation where the costs of a decision are for the account of one division, whereas the benefits fall to the share of another. Second, a particular interest will often imply a particular way of thinking. For instance, in terms of economics, of sales, or product quality. In the first situation there is a significant chance of conformity and groupthink, by which new, creative solutions are easily rejected. In the second situation one has to watch out for polarisation, a lack of trust, communication and co-operation, by which, in advance, new creative ideas are immediately labelled ‘strategic’ instead of really stating their contents. Just when partial agreement of interests and heterogeneous views exist, a group tends to search for more information and to look at new ideas seriously. It is therefore important that the chairman takes care of different opinions right at the beginning.

A third factor forms the mutual dependence of the persons involved. As mutual dependence increases, the need for information will increase and so will the rationality of the decision making process. Consequently, one is forced to incorporate more information in the decision making process. Regarding the legitimacy of the decision making process it is important to emphasise which parties are concerned, what their conditions are, and what the consequences of the decision are for the various persons involved. The legitimacy of a decision is determined among others by the quality of the results, its comprehensibility, the extent to which the decision can be defended, and the accuracy with which the decision is made.

The setting of the group appears to be of considerable importance to the decision making process, as well. Does the process take place in a more private or a more open setting? Both forms have particular advantages. A private process has the advantage that discussions can take place more freely, that one can reach an agreement sooner and that competitors tend to stick less to their initial opinions: they do not want to fail in front of other although they might already be convinced by other arguments. An open process on the other hand, has the advantage that additional information can be included more easily and the legitimacy of the decision can be increased.

The quality of the evaluation of an investment is not just determined by the product itself, but also by the process producing the decision. The way the alternatives are discussed is as equally important as how the actual choice is made. Before making this choice, the selection of alternatives takes place, the criteria for evaluation are recorded and weights are assigned to these criteria. How can discussions be stimulated and how can negative influences of the group process be managed? First, one has to define what makes a discussion successful. The requirements that can be defined to make a discussion ‘reasonable’, are represented in table 3. These represent the behavioural factors. Binding is regarded essential. This implies that decision makers are bound to the goals of the decision and to their own statements. Decision makers have to found their statements on arguments when expected. They are also not allowed to contradict themselves, and they may only claim things they believe themselves. Also, they may only give relevant arguments and information, they must be business-driven and the debate has to be open. This implies that all decision makers must be able to participate in the discussion and that every point of view may be formulated, argued and rejected.

Table 3: Requirements of a reasonable discussion (Pröpper and Bleijenberg, 1995)

Many causes may underlie unreasonable debates and the following list does not intend to be exhaustive. First, the choice of an open attitude is often seen as a weakness. While a reasonable discussion requires competitors responding to each other, in practice people often hide behind external sources. Often there is no distinction made between the correctness and acceptability of a discussion. Differences in status can sometimes be responsible for not rejecting unreasonable arguments (groupthink). The way problems are presented may disturb the discussion of the solution too (framing). Also, the available time can, if it is insufficient, influence the decision making process negatively and might, if it is superfluous, lead to a shift of evaluation criteria (change of decision rule). Finally, the structure of the belief-systems of the competitors is of influence. As central values are closer tied to the decision makers, they are less willing to adjust their point of view on new information and they are more open for counter-arguments. In practice one can also observe that decision makers become unreasonable when they are faced with another unreasonable decision maker. A chairperson with insight in the decision making process and the personality characteristics of the decision makers involved, can avoid unreasonable debates.

6 Conclusions

This article argues that successful decision making requires more than professional skills. The structure in which the decision making process is executed, the groups composition and in particular the choices made by the leading person during the process are of crucial importance. Behavioural sciences illustrate that a variety of social structural aspects can have both positive and negative influences on the quality of the decision making process. These aspects are also easy to manipulate. Taking all aspects into account not only makes it possible to predict the rationality of the decision making process, but also makes it possible to adjust the composition of the groups and structure in which it operates in advance. In this perspective it is bitter to see that in practice these possibilities and restrictions are hardly perceived in decision making processes regarding investments in information systems. These evaluations, therefore, often result in a confirmation of existing budgets, instead of an optimal allocation of resources.

Literature

• Antonides G (1991), Psychology in economics and business. Kluwer Academic Publishers, Enschede, The Netherlands. 

• Beach L R (1990), Image theory: Decision making in Personal and Organizational Context. John Wiley & Sons, England. 

• Berghout E W and Renkema T J W (1994), Evaluating information system investment proposals: a comparative review of current methodologies. Delft University of Technology, Research report number 94-27. 

• Blackler F and Brown C (1988), Theory and practice in evaluation: the case of new information technologies. In: IS Assessment (Bjorn-Andersen and Davis, Eds.), Elsevier, The Netherlands. 

• Corwin R G (1969), Patterns of organizational conflict. Administrative Science Quarterly, December, p. 522-542. 

• Douma S and Schreuder H (1992), Economic approaches to organizations. Prentice Hall, London. 

• Farbey B, Land F and Target D A (1992), Evaluating investments in IT. Journal of Information Technology, no. 7. 

• Farbey B, Land F and Target D A (1994), A Taxonomy of Evaluation Methods. Proceedings of the First European Conference on IT Investment Evaluation, Henley, United Kingdom, July 1994. 

• Harisson F E (1987), The managerial Decision-making Process. Houghton Mifflin Company, Boston. 

• Homans G C (1950), The human group. Harcourt Brace, New York. 

• Huizenga F D (1995), Decision Theory. Reader Free University of Amsterdam, Department of Social Sciences and Policy Analysis, Fourth Edition. 

• Janis I L and Mann L (1977), Decision making, A Psychological Analyses of Conflict, Choice and Commitment. The Free Press, New York. 

• Kleindorfer P R, Kunreuther H C, and Schoemaker P J H (1993), Decision Sciences: An Integrative perspective. Cambridge UP. 

• Klompé R H and Berghout E W (1995), Method is critical success factor in IS evaluation, Proceedings of the Second European Conference on IT Investment Evaluation, Henley, United Kingdom, July 1995. 

• Koopman P and Pool J (1992), Management and decision making in organizations: a strategic perspective. Van Gorcum, Assen, The Netherlands, in Dutch. 

• Powell P (1992), Information technology evaluation, is it different? Journal of the Operations Research Society, no. 43, pp. 29-42. 

• Pröpper I and Bleijenberg I (1995), Discussions in Politics and Administration: Guide for practice, Samsom, Alphen a/d Rijn, The Netherlands, in Dutch. 

• Simon H A (1979), Rational decision making in business organizations. American Economic Review, May, p. 1-16. 

• Swinkels G J P and Irsel H G P (1992), Investments in information technology: take IT or leave IT. Compact, p. 3-13, Summer, 1992, in Dutch.  

• Taylor R N (1975), Psychological determinants of bounded rationality: implications for decision-making. Decision Sciences, p. 409-429, No. 6. 

• Willcocks L (1992), Evaluating information technology investments: research findings and reappraisal. Journal of Information Technology, No. 2, p. 243-268.