A theoretical decision making continuum for projectized environments
Different definitions of what a decision is and involves abound in literature that spreads through the knowledge of many centuries of all disciplines (Sambharya, 1994; Shull, Delbecq & Cummings, 1970). Decision-making (DM) is of crucial importance to most companies, and modern organizational definitions can be traced back to von Neuman and Morgenstein (1947), who developed a normative decision theory from the mathematical elaboration of the utility theory applied to economic DM. Their approach was deeply rooted in sixteenth-century probability theory. This trend has persisted and can be found relatively intact in present decision analysis models such as those defined under the linear decision processes. This approach uses probability theory to structure and quantify the process of making choices among alternatives. Issues are structured and decomposed to small decisional levels, then these are re-aggregated with the underlying assumption that many less important good decisions will lead to a good large-scale decision. Analysis involves putting each fact in consequent order and deciding on its respective weight and importance.
This general approach to decision-making remains an important trend in both theoretical and practical models of DM, although most descriptive research in the area of DM concludes that humans tend to use both an automatic non-conscious thought process as well as a more controlled one when making decisions (Hastie & Dawes, 2001). This dual thought process is possible because of the human mind's capability to create patterns from facts and experiences that are stored in the registers of long-term memory, and accessed in the course of assessing and choosing options. Many authors refer to this mechanism as ‘intuitive decisionmaking’ a term that has not gained much credibility in the business milieu and is still looked down upon by many decision analysts.
In project management, most DM theories still focus mainly on the different steps involved in choosing among a given set of alternatives. These steps involve taking “smaller” or shorter-term decisions involving broken-up parts of the overall decision with the underlying assumption that the “sum” of many good decisions will amount to a better mega decision. Another strong underlying assumption is that, because these smaller decisions are based on probabilistic facts, they will subsequently be used in a “rational” way, thus presupposing that people think and decide in a ‘rational’ manner. In this model, rational behaviour is the basic assumption underlying all DM theories.
The word ‘Rational’ comes from the Latin word “ratio,” which refers to the concepts of “to calculate and reason” (Flyvbjerg, 2001). Brunsson (2000) elaborates on the use of the concept of rationality, linking it closely to the later consequence of action. This author debates that, from a rational perspective, in making decisions, one chooses future actions. “Decision” and “choice” are often understood as synonyms. Along with this legacy, the concept of logic is another old tradition in philosophy. From the time of Aristotle, logic provides a standard of reasoning from which people sometimes systematically depart, and it has been the underlying mission of scholars and science to help them think more logically.
Recent definitions of the word “rational” have been proposed by several authors in the area of DM; Baron (1998) gives it a special meaning:
“…not the kind of thinking that denies emotions and desires: the kind of thinking we would all want to do, if we were aware of our best interests, in order to achieve our goals” (Baron, 1998, p.5)
However, in the traditional context of project management, when project management DM has been discussed, it has often been in the context of project selection and, until more recent developments in portfolio management, it has not even been common to discuss project selection. The project manager has typically had little or no say in the project funding decision, nor has he been expected to have much input concerning the development of organizational strategy (Meredith, 2005). With many organizations now moving to fully “projectized” structures involving the management of strategic decisions through programmes (Moore, 2000; Richards, 2001; Thiry, 2004), project selection has become only one of many decisions associated with project management.
This paper first traces the history of traditional “Rational” project management decision-making processes as reflected in project selection methods.
The paper then presents an overview of more recent developments in DM theories and practice. These are reviewed in the context of rational and irrational approaches to DM and PM's resistance to embrace newer models are explored within the context of “naïve theories” and “individual belief systems.”
The paper concludes by presenting a comprehensive framework in which different approaches to decision-making are presented on a continuum that spans from rational traditional models to recent models incorporating intuitive and irrational aspects of the DM process. It is hoped that this can enhance interest and future developments of PM in this particular area.
Traditional Project Management Decisional Models
“Linear programming tools” are good examples of the traditional linear thought process model of DM from which they were derived. These tools are typically used in contexts of “high certainty.” They seek an optimal solution to the problem which has been modelled with the two essential requirements: a) that each of the variables involved in the decision-making process behaves in a linear fashion of the decisional range; and b) that the number of feasible solutions is limited by constraints on the solution. The underlying assumptions of such models are those of proportionality, Additivity, divisibility, certainty and constants (seldom met as they refer to a future period of time when the decision will be implemented and cannot be fully known in advance).
The use of linear programming tools, techniques, and procedures to make short-term decisions is common and has a number of advantages. For applications such as determining weekly production levels, it is usually reasonable to assume that the estimates included in the model are reliable. If the model is used repeatedly, those values that prove to be wrong can be refined by gaining further information (Jennings & Wattam in Jennings, 1998, Ch.7; Johnson, 1986; Neelamkavil, 1987). A simple example of such a decision process is found if you consider two products, Product A and Product B. The manufacture of both is constrained by three criteria: labour, machine time, and materials. Together, these give the operating space that the company can use to maximise profits. Let x be the production of Product A per time period and y the production of Product B per time period: each product A takes two units of labour; each Product A three units of machine time; each Product A five units of materials. The figures for a Product B are two, one, and two, respectively (Jennings & Wattam, 1998, Ch.7). This type of tool is readily used at project-level decisions.
Because of the underlying linear assumptions of these DM models, they are not applicable when uncertainty rises and managers must put assets “at risk” to achieve their objectives. The assessment of business risk, then, becomes the key to successful future operations. Weighing opportunities versus potential losses is part of the issue. Past and present data are useful in documenting the past, but such data cannot predict the future. In fact, reliance on data lulls management into a false sense of security, and it becomes difficult to imagine anything different. Hence, the key success factor in risk assessment is the ability to imagine events that are different. Jennings and Wattam (1998) suggest completing a risk balance analysis for each project within the overall environment; they call this the “generic risk balance.” Such a risk balance is identified for each of the main stakeholder groups affected by the control options being considered, where stakeholder analysis may be based on soft systems or other similar techniques (Jennings & Wattam in Jennings, 1998, Ch.7). Then, by evaluating the options, the most effective control mechanisms can be selected; these should allow correction, should things go wrong. This analysis is subsequently linked to the “satisficing” of possible solutions to a given problem, given a set of defined constraints (Simon, 1964). The risk can then be reduced in two ways: the probability that an event that gives cause for risk is reduced (preferably removed), or the impact on the project is reduced by looking for contingency or alternative methods of provision.
These decision-making techniques rely almost entirely on the logic of statistical analysis, regression analysis, past examples, and the linear expectations and predictions that they stimulate. However, even in mathematical probabilistic terms, the idea of “fuzzy thinking” can be illustrated by considering a set of questions that appear simple, but have a multivariate outcome, or an outcome that is not clear. Therefore, fuzzy ideas, risk, uncertainty, and the complexity of situations are closely related, and, together, they make the decision domain a difficult one. To take account of fuzzy behaviour, most people will use phrases such as “in general or normally...,” and, in this model, most solutions rely heavily on probability (Jennings & Wattam in Jennings, 1998, Ch.7; Johnson, 1986).
An interesting example of such models has been studied extensively to support choice among projects with an “optimum project portfolio” mathematical model (Cooper, Edgett, & Klienschmidt, 1997; Henriksen & Traynor, 1999). These authors came to the conclusion that the models are simply too complex to be used by practitioners. Further investigations by Graves, Rinquest, and Case (2000) showed comparable conclusions and argued that this was due to the lack of theoretical grounding of such models over and above the fact that they are simply too difficult to implement.
Information is seen as a key aspect of such models to reduce uncertainty and to complete the different stages involved in the decision process. To name but one example, financial calculations are most often integrated as pertinently linked to “uncertainty”; however, research has shown that managers tend to make “overly optimistic forecasts” (Cooper, Edgett, & Klienschmidt, 2000; Lovallo & Kahnemann, 2003), and some authors such as Cooper, Edgett, and Kleinschmidt (2000) find that it is so hazardous to base decision-making on such forecasts that “one might be better off tossing a coin”! Additionally, traditional project evaluation or prioritization techniques are almost solely based on accounting and financial measures (Dyson & Berry, 1998). According to Barwise, Marsh, and Wensley (1989) and Brewer, Chandra, and Hock (1999), financial-based tools are “myopic” instruments as they undervalue overall strategic benefits against pure financial measures.
As discussed by Janney and Dess (2004), a concern that cannot be avoided when using formal applications of these tools involves setting appropriate parameters for the model. For example, one of the key variables for modeling is the amount of statistical variance estimated for potential outcomes, and the way that this variance is estimated has enormous implications on the decision-making value of the tool and decisional process (Dixit, Pindyck, & Pindyck, 1995). Variance is a formal way of modeling uncertainty. This means that it describes (for any decision) how wide the range is between the worst-case and best-case outcome scenarios. It is generally measured in terms of standard deviations. For investment tools, the worst-case is generally limited to the cost of the initial decision. The best-case scenario, however, is rarely limited because it represents “potential unlimited results.” However, as this upside potential gets larger, so does the estimate of overall variance. While variance can be measured on past outcomes, it must be estimated for future ones.
Unfortunately, these subjective estimates may yield very different outputs from the decision-making tool when the estimate of variance changes just slightly, creating “small difference, big effect” repercussions (Dixit et al., 1995). The practical implications are that, depending on the amount of variance one assumes in the decision model, this may yield either an approval or a disapproval decision. With a low variance rate, more projects get rejected; with a high rate, too many might get approved. This is similar to the criticisms often raised against using Net Present Value for evaluating investment options. Several authors feel that this point needs to be raised for managers to understand just how sensitive outcomes are to selected inputs and the limit of such decision tools (Janney & Dess, 2004).
When people understand how to estimate variance, this can create yet a secondary problem as it also means that they have the know-how to “game the system” and most electronic spreadsheets permit users to simply back-solve any formula. This means that a manager can type in the desired answer and ask what values are needed in a formula to get that answer. If managers know that a certain value must be met in order for the proposal to get approved, they can back-solve the model to find the variance estimate needed to arrive at the answer that best suits their needs. The problem in this instance is best understood through the lens of the Agency Theory, where the interests of managers and owners are not necessarily co-aligned (Clarke, 2004; Daily, Dalton, & Cannella, 2003; Jensen & Meckling, 1976). Agency theory “is an extremely simple theory, in which large corporations are reduced to two participants – managers and shareholders – and the interests of each are assumed to be both clear and consistent with the notion of humans as self-interested and as generally unwilling to sacrifice personal interests for the interests of others” (Clarke, 2004; Daily et al., 2003). Agency theory suggests that, as managerial and owner interests diverge, managers will follow the path of their own self-interests. Agency problems arise because managers who propose projects may believe that if their projects are approved, they stand a better chance of getting promoted. So while managers may have an incentive to propose projects that are successful, they also have an incentive to propose projects that are simply chosen, thus to choose variance values that increase the likelihood of approval (Janney & Dess, 2004).
Recent Trends in DM
Some of the more recent contributions to the descriptive theory of DM have not come from observations of people deciding, but rather from attempts to make computers think (Baron, 1998). The idea is to try to program the computer to do the task. As such, the program becomes an embodiment or model of a particular theory. With computers, it became possible to explore complex models; one could now tell the computer “what to do” without telling it “how to do it” (Holland, 1998). Traditional computer-based DM models have become increasingly popular and have common roots in the study of board games.
Although computers have opened new horizons, they still have great limitations when compared to humans. For example, humans can parse an unfamiliar scene into familiar objects. Computers cannot do this task because it is too complex. We have no plausible computer-based models of human parsing procedures. Though the number of building blocks that humans use for parsing may be relatively small, the process of discovering them is never ending, and to implement a computer model, one must first determine the model's principal components. Beyond its principal components, the elaboration of computer-based modelling involves establishing functions for which mathematicians use the term “mapping,” and it turns out that there is a close relationship between these maps and games (Holland, 1998).
It is with the publication of their Games Theory in 1947 that von Neumann and Morgenstern strongly influenced both computer modeling and DM theories through statistics, information theory, and economics. Among important concepts derived from the Games Theory, the state of the game (position of pieces), state space (collection of all arrangements), the Tree of moves (initial state and branches) and, last but not least, Strategies are useful to understand the process and nature of DM.
A strategy specifies a sequence of decisions and the game tree provides a way of making this rough idea more precise. A sequence of decisions traces a path in the game tree, so we can define a strategy in terms of the branches it chooses. In game theory, a complete strategy prescribes a branch (move) for each state (board arrangement). A complete strategy tells us what to do in any possible situation. In a multi-person game, one can attribute a strategy to each player. The combined strategies select a path through the move trees. Much as in business, this process does not remove interest or surprise, as each player does not know its opponent's strategy and therefore cannot predict the outcome of the game. However, when a game is played repeatedly, the unknown aspects of the other players’ strategies may become clearer and can be used to develop a model that will lack detail because there are just too many possibilities; nevertheless, if the model is correct, we can hope to do better with it than without it (Holland, 1998).
The idea that a small number of rules or laws can generate systems of great complexity, which has been demonstrated many times, over is central to the understanding of DM in complex environments. Even more important is the fact that this complexity does not just evolve in random patterns, but that recognizable features exist, and that the systems that form are animated and change over time. Many authors have labeled this recurring and recognizable phenomenon emergence. Recurrent and recognizable does not mean that it is easy to understand or explain. On the contrary, this task is difficult even when the rules and laws of a given game are known. In chess, it took many centuries to understand certain patterns, such as control of pawn formation; however, once recognized, these patterns greatly enhance the possibility of winning the game.
Many organizations have felt a need to further develop towards a fully “projectized” structure, which goes beyond a simple portfolio approach and involves the management of strategic decisions through programs (Moore, 2000; Richards, 2001; Thiry, 2004) This move has somewhat shifted the responsibilities and decision-making roles of project and program managers in the same direction as the general evolution of DM modeling theories. The implications of such a move are readily felt at the decisional level; when several variables are added to a system or when the environment is changed, the relationships quickly lose any resemblance to linearity (Begun, 1994).
Strategic decisions serve the purpose of answering fundamental questions, such as: What activities should the organization be involved in and how will it compete in its various business areas? They state the fundamental means by which the organization seeks to achieve its goals; they are concerned with the future and they also have a purpose in relation to the internal world of the organization and its people (Jennings, 1998). In an ever-changing environment, the outcomes are no longer predictable. Even the most elaborate linear techniques can be misleading when applied to non-linear phenomena (Begun, 1994).
In his work on DM during the implementation of strategic projects, Grundy (2000) further found that cognitive, emotional, and territorial aspects were so intrinsically interwoven to the decision-making process that he developed the concept of “muddling through” originally introduced by Lindblom in 1959. Similarly unsatisfied with the rational model of decision-making at top management levels, Isenberg (1984) had come to the conclusion that managers “rely heavily on a mix of intuition and disciplined analysis” and “might improve their thinking by combining rational analysis with intuition, imagination and rules of thumb.”
Humans’ attachment to linear frameworks and thought processes can be better understood through some ideas about knowledge imported from the field of cognitive psychology, such as “Naïve Theories.” Naïve theories are systems of beliefs that result from incomplete thinking. They are analogous to scientific theories, but what makes them naïve is that they have been superseded by better theories. Many modern theories will become naïve in the light of future theories. The same holds true for individual beliefs (Baron, 1998). The development of thought process from childhood to maturity is well documented in psychology, and we know that it usually evolves in three stages: first, the old concepts are replaced with new concepts; second, the relationship among the concepts change; and, finally, the new system explains different phenomena (Vosiadou & Brewer, 1987). Many experiments have demonstrated how change in thought process can linger on long after the necessary information needed to change has been learnt. Students having taken and passed a physics course covering the topic of “a body in motion” still held a theory (even after the course) similar to the one Galileo held in his earlier work. These students thought that a body in motion always required a force to stay in motion rather than holding the more actual view of it keeping going unless some force stopped it or slowed it down (Clement, 1983).
Even to the casual observer, it is clear that business and, indeed, life in general are not entirely predictable. Chaos theory suggests that a whole range of phenomena are inherently unpredictable. A chaotic situation is characterized by the absence of regularities which prevent the accurate prediction of what will happen next, and an important concept underpinning chaos theory is that of non-linearity (Gleick, 1998; Gribbin, 2004). As a consequence, to try and foresee the future may be a futile and wasteful activity; we may need to make decisions knowing that we can never be sure of the outcomes.
Many of the formal techniques of DM rest on assumptions of linearity. Budgeting, capital investment appraisal, regression analysis, and linear programming are all examples of projecting past trends to provide a profile of future events. Past and current information is used to predict, on the assumption that the future is like an unexplored part that will directly follow from the explored part. On the other hand, the mathematics of Chaos and Complexity document that events can be discrete and that, despite having full information about past events, the next occurrence may follow a pattern different from previous occurrences. Fundamental to the concept of chaos is the perspective that many events, both in the physical and social world, are complex and hence intrinsically difficult to predict with any certainty (Gleick, 1998; Gribbin, 2004).
Chaos theory has led to developments which provide new insights and techniques progressively refining the quality of short-term forecasts. An example of this can be found in weather forecasting. While the weather system has been identified as the archetypal chaotic system, this has led to improvements in short-term weather forecasts. Such lessons are easily transferred to business: “if long-term forecasting is pointless, competitive edge might be gained from having the best short-term forecasts” (Lye, 1998). This author summarizes the main influences of chaos theory on DM as follows:
- The need to focus on short-term rather than long-term DM and the importance of contingency planning as part of any organization's DM process
- The need to value intuitive and imaginative approaches to DM
- The significance of developing temporary structures and systems
- The modification of corporate cultures to incorporate new and relevant values and norms
- The need to look for order within chaos.
Logical procedures are built into the traditional rational approaches of DM, and its steps follow a logical sequence from inception to final decision. Chaos theory questions the value of always adhering to this logical process. If logic has limitations, then non-logical methods cannot be dismissed as inappropriate and ineffective. Intuitive and creative thinking that does not follow the canons of formal logic has an inherent role in formulating good decisions (Stewart 1998).
The Role of Intuition in DM
Baron (1998) defines the three basic kinds of thinking that we need to go through to achieve our goals and calls this the “search-inference model”: Thinking about decisions, about beliefs, and about goals themselves. In this framework, a decision is a choice of action; decisions are made to achieve goals based on beliefs about what set of actions will achieve these goals. This process calls upon judgment to evaluate one or more possibilities with respect to a specific set of evidence and goals. This author also argues that one of our main problems with DM is our lack of open-mindedness, because of which we ignore possibilities, evidence, and goals that we ought to consider; and that we make inferences in ways that protect our favored ideas.
Like Baron, most authors now agree that decision-making is not necessarily a rational process (Brunsson, 2000; Langley, Mintzberg, Pitcher, Posada, & Saint-Macary, 1995; March, 1995; Nutt, 2002). Langley et al. (1995) further feels that the label “insightful man” must be added to those of “rational or administrative” man in order to better understand the processes at work in decision-making. This author provides a definition of insightful as “one who listens to his subconscious inner voice,” and the concepts he refers to are similar to those of intuition. Etzioni (1989) goes so far as to say that rational models of DM simply do not meet the needs of the organization in a complex environment because we are not capable of meeting the expectation of predictability stimulated by rationality.
Beyond these concerns, the impressionistic theory of DM (Brunsson, 2000) has been gaining credibility over the last few years. Advocates of this theory feel that it might be wiser to act irrationally, if needed, to reflect “organization-specific beliefs and norms,” a criterion which seems to take precedence over future consequences in the evaluation of decisions. In this context, one should not expect a causal relationship between decisions and actions. Brunsson (2000) outlines the fact that, in any case, we do not have empirical evidence verifying a close relationship between decision-making and choice of action, a necessary condition for coherence in a rationalistic approach to decision-making.
A recent research project gathering 126 project manager interviews in 32 companies to understand DM at project portfolio level concludes that the rationalistic model is better suited to situations in which uncertainty about consequences is low, and in which it is possible to estimate more accurately bottom-line effects of successful completion (Eskerod, Blichfeldt, & Toft, 2004). These findings are coherent with Brunsson's suggestion that the impressionistic perspective is of value mainly in complex decisions involving high levels of coordination of action, and for which consequences are long-term and far- reaching; thus, involving high levels of uncertainty.
Already in 1986, supporting the positive role of intuition in decision-making processes, Agor provides one definition of intuition as being neither the opposite of rationality nor a random process of guessing; for this author, intuition corresponds to thoughts, conclusions, or choices produced largely or in part through subconscious mental processes (Agor, 1986).
Another interesting working definition of intuition is provided by Miller and Ireland (2005): “Intuition is knowledge gained without rational thought. And since it comes from some stratum of awareness just below the conscious level, it is slippery and elusive, to say the least…. New ideas spring from a mind that organizes experiences, facts, and relationships to discern a [mental] path that bas not been taken before.” (p. 21)
As early as 1994, Parikh had found that most managers supported intuition as playing a positive role in the DM process. More recent studies of intuition in relation to DM, although recognizing that it is a potential source of competitive advantage for companies in the increasingly complex global economy, suggest that, at best, intuition needs to be examined because it can be a troublesome decision tool (Miller & Ireland, 2005; Sadler-Smith & Shefy, 2004).
Several authors now agree on intuition's necessity in times of change and that intuitively dominated decisions are likely to increase in the fast-paced 21st century. Indeed, recent publications in the business press and in academic literature support this, as do many surveys of executives and managers. A recent survey found that almost half of corporate executives use intuition more than formal analysis to run their companies (Christian & Timbers, 2003), and that as many as two- thirds of managers feel that it leads to better decisions (Harris, 2001; Burke, & Miller, 1999). Intuition seems to come as a particularly useful concept when executives cannot articulate how they make decisions that defy logical analysis.
Sadler-Smith and Shefy (2004) explain this tendency as a reaction to the increase in the volume of data that executives now deal with that has the potential to be overwhelming. These authors relate the comment of one CEO: “…at the point when you've gathered enough data to be 99.99 per cent certain that the decision you're about to make is the correct one, that decision has become obsolete” (Hayashi, 2001, in Sadler-Smith & Shefy, 2004, p.77). The requirement for fast decisions and the limits of human beings’ rational information-processing capacities may combine to impose severe demands upon executives’ cognitive capabilities to handle masses of information at the necessary speed. This may result in a combination of volume-induced and/or complexity-induced information overload that may limit or even preclude in-depth and deliberate consideration and the balancing of alternative courses of action (Sadler-Smith & Shefy, 2004).
One of the more recent and interesting approaches to intuition remains that of Miller and Ireland (2005), who base their arguments on Crossan, Lane, and White's (1999) definitions and conceptualize intuition in two distinct ways:
- Holistic hunch: corresponds to judgment or choice made through a subconscious synthesis of information drawn from diverse experiences. Information stored in memory is subconsciously combined in complex ways to produce judgment or choice that feels right. “Gut feeling” is often used to describe the final choice.
- Automated expertise: knowledge gained without rational thought. And because it comes from some stratum of awareness just below the conscious level, it is slippery and elusive. New ideas spring from a mind that organizes experiences, facts, and relationships to discern a path that has not been taken before. Simon (1987) was the first to elaborate on this second concept of intuition, labeling it “expert intuition.” For Simon, accumulated expertise leads to steps in the analysis being dropped while others are completed in a rapid, subconscious fashion. Later, Crosson, Lane, and White (1999) used the term and defined it as “unconscious recollections,” and Burke and Miller (1999) seem to refer to the same process when describing what they call “subconscious mental processing.”
These authors all describe similar pattern recognition abilities in people with expertise. Overall, the key to automated expertise lies in a person's quick identification of an unfamiliar situation, and subsequent automatic access and application of (subconscious) stored knowledge related to the situation.
“I believe in intuition only if you discipline it. The ‘hunch’ artists, the ones who make a diagnosis but don't check it out with facts, with what they observe, are the ones…who kill businesses.” (Drucker in Nelson, 1985).
Most authors stress that individuals who play hunches should not be ignorant of available data and other facts, but that, simply put, data and facts can be incomplete, sometimes misleading or otherwise fail to provide clear guidance. A useful step may be to use “probes” to test a market and to assess reactions to hunch-based decisions in order to guide future decision possibilities.
Reliability, which means the consistency with which a decision-maker uses past learning over time, is one problem for intuition as automated expertise. Memory failures, fatigue, information overload, and distractions can create inconsistencies, and these inconsistencies can be difficult to trace because the process is not at a conscious level (Burke & Miller, 1999; Crosson et al., 1999; Klein, 1998; Miller & Ireland, 2005; Schoemaker & Russo, 1993; Simon, 1987). This thought is best expressed by Khatri (2000), who says that intuition is neither the opposite of rationality nor a random process of guessing; intuition corresponds to thoughts, conclusions, or choices produced largely or in part through subconscious mental processes.
In the whole, current research seems to suggest that the proportion of executives with an intuitive preference is likely to increase with seniority. Sadler-Smith and Shefy (2004), explain this tendency in two ways. Intuitive skills (holistic and visionary thinking) may help executives attain senior positions; hence, these types may become disproportionately represented at higher levels. Alternatively, top executives’ job responsibilities may demand that they develop their intuitive capabilities more fully than managers lower down the hierarchy.
Irrational Aspects of DM
Traditional models of DM vary in the extent to which they attribute rationality and irrationality to the processes engaged in by the individual decision-maker. It is interesting to note that prescriptive models tend to imply a greater degree of rationality than many of the explanatory theories would suggest (Stewart 1998). In the psychology of DM, Stewart (1998) summarizes the implications of irrational behaviors in DM as he highlights the different factors that limit one's rationality in decision-making. Steward also adds that the quality of DM in work organizations can only be improved if psychological factors are understood and taken into account, an area still in great need of development in both project and programme management literature. Beyond Steward's concerns, attempts at understanding the irrational component of DM have multiplied and the research literature has grown considerably in the last decade.
For example, although superstition has traditionally been perceived as inconsistent with modern business management principles, it has become the subject of much interest because of its importance in Chinese business practices (March, 1994; Tsang, 2004; Vyse, 1997). This is compounded by the fact that privately owned Chinese firms outside mainland China itself make up the world's fourth economic power after North America, Japan, and Europe (Kao, 1993).
It has since become evident that, despite their MBA's, and although motivated by Chinese business practices and the economic power of the business communities that adopt these practices, many Asian managers tend to integrate superstition into their belief systems (Vittachi, 1993). Japanese managers seek advice from Ebisu, the Shinto god of business and one of the seven Shinto gods of good luck. According to Lewis (Lewis, 1989 in Tsang 2004), about 96- million Japanese are Shintoists. Indian managers also seem to be superstitious judging by the popularity of vaastu shastra, a superstitious practice that tries to create harmony among the five critical environmental elements: earth, sky, fire, water, and air (Sarkar, 1996). These considerations cannot be ignored, given that the Chinese, Indians, and Japanese together constitute about 40 per cent of the world's population. As reported by Tsang (2004), it also seems a mistaken belief that only the less educated are superstitious. This author reports:
“that the above-mentioned company that moved to a new office due to bad feng shui was in the business of engineering consultancy. Most of its employees had master's degrees, and some had doctoral degrees.” (p. 103)
With the trend toward globalization, most companies have become “superstition-conscious,” if only to develop an understanding of the practices in countries where they operate. But with increasing research in all cultures, it is becoming more evident that superstition is probably universal, as the following American example demonstrates:
“Virtually every major move and decision the Reagans made during my time as White House Chief of Staff was cleared in advance with a woman in San Francisco who drew up horoscopes to make certain that the planets were in favorable alignment for the enterprise.” (Regan, R., 1988, in Tsang, 2004, p. 103)
The reasons for which research and practice have ignored the importance of this variable on DM are self-evident because of the irrationality normally associated with superstition and the supposed rationality of the DM process. In addition, the concept could create cognitive dissonance in respondents, become a source of psychological discomfort and anxiety, and simply falsify the outcomes of research in the area. During the course of Tsang's (2004) investigation, fifteen respondents felt that DM should follow a well-structured process and be based on relevant facts (i.e., things that they learned about decision-making in management courses); and that their superstitious practices violated the basic principles of DM. One respondent said he felt shameful to tell his former finance professor that he relied more on God's advice than discounted cash flow analysis when evaluating investment projects. But many also felt that the feng shui expert might suggest some ideas that they would never have thought of, thus encouraging them to think “outside the box.”
Although superstition is not the easiest concept to integrate to classical visions of DM, its positive role in reducing uncertainty-induced anxiety has been mentioned by numerous authors. Simon (1987) had noticed that uncertainty creates anxiety, which often lengthens a decision; that this could be counterproductive to the decision process; and that most human societies developed ways of dealing with the anxiety. In other words, in some cultures, superstition can be seen as a stress-management strategy (Chan & Chiang, 1994; Hofstede, 1997).
It was an anthropologist, Bronislaw Malinowski (1948), who created what has become the most well known of theories of superstition and, interestingly enough, the one that is most helpful in understanding its key role in decision-making. Malinowski lived among the Trobriand Islanders of Melanesia, off the coast of New Guinea, and observed that, when important events lay beyond the scope of the islanders’ stock of scientific knowledge, magic was used as a hedge against uncertainty. Islanders who fished in the calm waters of the lagoon employed highly standardized fishing methods and did not engage in any magical procedures. On the other hand, deep-sea fishermen performed elaborate magical rituals to ensure a safe trip and good results, because fishing in the open sea was dangerous and results were unpredictable. According to Malinowski's theory, superstition serves to fill the void of the unknown and to reduce anxiety. This theory has gained wide acceptance, and superstitious behavior is commonly perceived as a response to uncertainty (Eisenhardt, 1989). As emphasized by Kahneman, Slovic, and Tversky (1982), uncertainty would provide a natural link between superstition and DM because uncertainty and related concepts such as ambiguity and risk are central to its literature.
Malinowski's theory plunges one directly into the heart of the debate about the utility of rationality in DM by emphasizing the role of uncertainty and the availability of information upon which executives may base their decisions. As Sadler-Smith & Shefy (2004) note, the relationship between DM and uncertainty has been interpreted and explained in a number of different ways. Some have argued that uncertainty increases the degree of procedural rationality that is required in order that information gaps can be filled and uncertainty may be removed by further analyses (Leblebici & Salancik, 1981). Others argue that rational DM does have situational superiority under particular sets of circumstances, and that the relationship between procedural rationality and decision-making effectiveness was positive and statistically significant (Dean & Sharfman, 1996). Yet other research by the same authors has revealed a negative relationship between uncertainty and rationality with the correlation between rationality and uncertainty being negative and statistically significant (Dean & Sharfman, 1993). With contradictory quantitative results, these authors nevertheless observed that decision-makers engaged in less rational processes of strategic DM when faced with highly uncertain problems; whereas when there was little competitive threat, little perceived external control, and the issues being faced were well understood, executives used rational procedures.
Given that superstition is probably seen as one of the most irrational aspects of the DM literature and of the process itself, according to some authors, it is not the most destructive. Janney and Dess (2004) states that “…of the many pitfalls we have identified, we see irrational escalation of commitment as the most serious. Avoiding such escalation requires more than just being specific about costs and benefits, and formally stating the decision rules and process to be followed” (p. 67).
This escalation of commitment is often company-induced as organizations typically encourage managers to “own their decisions,” in order to motivate them. Once managers have invested themselves in their decisions, it proves harder to “lose face” by reversing course because it feels as if they made the wrong decision (even if it was initially a “good” decision). To counter the negative effects of this dynamic pattern, Janney and Dess (2004) suggest developing the concept and use of an “exit champion,” and suggests that external audits be done to help firms make more objective decisions about when to exit a decision.
It would seem that, although the rational method can undeniably lead to effective decisions in some circumstances, when outcomes are difficult to predict through rational means, executives need to acknowledge the uncertainties; be more tolerant of ambiguities; be able to respond to complexities in pragmatic, intelligent, and fast ways in the face of the unknown; and recognize the potential that their intuitive judgments may offer.
Moreover, when decisions need to be taken speedily and with cognitive economy in the face of an overwhelming mass of information or tight deadlines, executives may have no choice but to use anxiety-reducing methods, relieve stress, and rely upon intelligent intuitive judgments rather than on non-existent or not-yet-invented routines.
When deliberative rational thought is not achievable or desirable, one way of managing and coping with uncertainty and complexity and of “thinking outside of the box” is by relying upon superstition and intuition. Because of this situation, Sadler-Smith and Shefy (2004) argue that executives need to be able to recognize and understand irrational aspects of decision-making, accept them, establish ways in which they can be comfortable, and leverage their potential for success and well-being both for themselves and for those whom they lead.
This knowledge, understanding, and skill constitute an intuitive awareness, and the question of how this may be developed is important both for project managers themselves and for those educators and consultants whose aim it is to improve decision-making skills. This is especially important given that management education and training in general appear to be lacking in this regard (Sadler-Smith & Shefy, 2004).
Agor, W. H. (1986) Intuition in organizations: Leading and managing productively. Newbury Park, CA: Sage Publications.
Baron, J. (1998). Thinking and deciding (3rd ed.). Cambridge, UK: Cambridge University Press.
Barwise, P. Marsh, P. R., & Wensley, R. (1989, September-October) Must finance and strategy clash? Harvard Business Review, 67(5), 85-90.
Begun, J. W. (1994, December) Chaos and complexity frontiers of organization science. Journal of Management Inquiry, 3(4), 329-335.
Brewer, P. C., Chandra, G., & Hock, C. A. (1999, Spring). Economic value added (EVA): Its uses and limitations. S.A.M. Advanced Management Journal, 64(2), 4-11.
Brunsson, N. (2000). The irrational organization – Irrationality as a basis for organizational action and change (2nd ed.). Bergen Chichester, UK: John Wiley & Sons.
Burke, L. A., & Miller, M. K. (1999). Taking the mystery out of intuitive decision making. Academy of Management Executive, 13(4): 91-99.
Chan, K. B., & Chiang, C. S. N. (1994). Stepping out: The making of Chinese entrepreneurs. Singapore: Simon & Schuster.
Christian & Timbers Website. (2003). Retrieved June 2005, from http://www.ctnet.com/default.html
Clarke, T.. (2004). Introduction: Theories of governance – Reconceptualizing corporate governance theory after the Enron experience In T. Clarke (Ed.), Theories of corporate governance – The philosophical foundation of corporate governance (pp 1 – 30). London: Routledqe.
Clement, J. (1983). A conceptual model discussed by Galileo and used intuitively by physic students. In D. Getner & Al Stevens (Eds.), Mental models (pp. 340-352). Hillsdale, NJ: Erlbaum.
Cooper, R. G., Edgett, S. J., & Kleinschmidt, E. J. (1997). Portfolio management in new product development: Lessons from the leaders- Part 1. Research Technology Management, 40(Sept.-Oct.), 16-28.
Cooper, R. G., Edgett, S. J., & Kleinschmidt, E. J. (2000). New problems, new solutions: Making portfolio management more effective. Research Technology Management, 43(March-April), 18-33.
Crosson, M. M., Lane, H. W., & White, R. E. (1999). An organizational learning framework: From intuition to institution. Academy of Management Review, 24, 522-537.
Daily, C. M., Dalton, D. R., & Cannella, A. C. (2003). Corporate governance: Decades of dialogue and data. The Academy of Management Review, 28(3), 371-82.
Dean, J. W., & Sharfman, M. P. (1993). Procedural rationality in the strategic decision-making process. Journal of Management Studies, 30(4), 587-620.
Dean, J. W., & Sharfman, M. P. (1996). Decision-making effectiveness. Academy of Management Journal, 39(2), 368-396.
Dixit, A. K., Pindyck. R. I. A. K., & Pindyck, R. S. (1995). The options approach to capital investment. Harvard Business Review, 73, 105-113.
Dyson, R. G., & Berry, R. H. (1998). The Financial Evaluation of Strategic Investments. In: R. G. Dyson & F. A. O‘Brien (Ed.), Strategic development: Methods and models (pp. 269-297). Chichester, UK: John Wiley & Sons Ltd.
Eisenhardt, K. M. (1989). Making fast strategic decisions in high-velocity environments. Academy of Management Journal, 32(3), 543-576.
Eskerod, P., Blichfeldt, B. S., & Toft, A. S. (2004). Questioning the rational assumption underlying DM within PPM literature. Proceedings of the PMI Research Conference 2004, London, UK.
Etzioni, A. (1989). Humble Decision Making. Harvard Business Review, Jul-Aug, 122.
Flyvbjerg, B. (2001). Making social science matter: Why social inquiry fails and how it can succeed again. Cambridge, UK: Cambridge University Press.
Gleick, J. (1998). CHAOS. London: Random House.
Graves, S. B., Rinquest, J. L., & Case, R. H. (2000). Formulating optimal R & D Portfolios. Research Technology Management, 43(May/Jun), 47-51.
Gribbin, J. (2004). Deep simplicity: Chaos, complexity and the emergence of life. London, UK: Allen Lane, Penguin Group.
Grundy, T. (2000). Strategic project management and strategic behaviour. International Journal of Project Management, 18, 93-103.
Harris, T. W. (2001). Establishing effect- you better at what you do. New York: Doubleday.
Hastie, R., & Dawes, M. (2001). Rational Choice in an Uncertain World. Thousand Oaks: CA: Sage Publications, Inc.
Hayashi, A. M. (2001). When to trust your gut. Harvard Business Review, February, 59-65.
Henriksen, A. D., & Traynor, A. J. (1999). A practical R&D project-selection scoring tool. IEEE Transactions on Engineering, 46, 158-170.
Hofstede, G. (1997). Cultures and organizations; Software of the mind. New York: McGraw-Hill.
Holland, J. H. (1998). Emergence from chaos to order. _Oxford, UK. Oxford University Press.
Isenberg, D. J. (1984). How senior managers think. Havard Business Review, Nov/Dec, 80.
Janney, J. J., & Dess, G. G. (2004). Can real-options analysis improve decision-making Promises and pitfalls. Academy of Management Executive, 18(4), 60-75.
Jennings, D. (1998). Strategic decision making. In D. Jennings & S. Wattam (Eds.), Decision making An integrated approach (2nd ed, pp. 251-282). Harlow, UK: Prentice Hall Pearson.
Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Reproduced from the Journal of Financial Economics (1974) in Theories of Corporate Governance, Thomas Clarke, Ed. (2004). Routledqe London and New York, 58-63.
Johnson, D. (1986) Quantitative Business Analysis. London, Butterworth
Kahneman, D., Slovic, P., & Tversky, A. (1982). Judgment under uncertainty: Heuristics and biases. New York: Cambridge University Press.
Kao, J. (1993). The worldwide web of Chinese business. Harvard Business Review, March-April, 24-36.
Khatri, N., & Ng, H. S. (2000). The role of intuition in strategic decision making.Human Relations, 53, 57-86.
Klein, G. (1998). Sources of power: How people make decisions. Boston: MIT Press.
Langley, A., Mintzberg, H., Pitcher, P., Posada, E., & Saint-Macary, J. (1995). Opening up decision making: The view from the black stool. Organization Science, 6, 260-279.
Leblebici, H., & Salancik, G. R. (1981). Effects of environmental uncertainty on information and decisions in banks. Administrative Science Quarterly, 26(4), 578-596.
Lewis, M. (1989) God for trade. Business, January, 68-69.
Lindblom, C. E. (1959). The science of muddling through. Public Administration Review, 19, 79-88.
Lovallo, D., & Kahnemann, D. (2003). Delusions of success: How optimism undermines executives’ decisions. Harvard Business Review, (July), 56.
Lye, R. (1998), Chaos and decision making. In D. Jennings & S. Wattam (Eds.) Decision making: An integrated approach (2nd ed, chapter 5). Harlow, UK: Prentice Hall Pearson.
Malinowski, B. (1948). Magic, science and religion. Garden City, NJ: Doubleday.
March, J. G. (1994). A primer on decision making: How decisions happen. New York: Free Press.
March, J. G. (1995). Questioning the rational assumption underlying DM within PPM literature. Proceedings, PMI Research Conference, London, 2004.
Meredith, J. R. (2005). Project management: A managerial approach (6th ed.). New York: Wiley VCH.
Miller, C. C., & Ireland, R. D. (2005). Intuition in strategic decision making: Friend or foe in the fast-paced 21st century? Academy of Management Executive, 19(1), 19-30.
Moore, T. J. (2000). An evolving program management maturity model: Integrating program and project management. Proceedings of the Project Management Institute's 31st Annual Seminars & Symposium Proceedings [CD-ROM].
Neelamkavil, F. (1987). Computer simulation and modelling. Chichester, UK: John Wiley & Sons.
Nelson, R. (1985). How to be a manager. Success, 22, 69.
Nutt, P. C. (2002). Why decisions fail-Avoiding the blunders and traps that lead to debacles. San Francisco: Berrett-Koehler Publishers.
Parikh, J. (1994). Intuition: The new frontier in management. Oxford, UK: Blackwell Business.
Regan, D. T. (1988). For the record. San Diego, CA: Harcourt Brace Jovanovich.
Richards, D. (2001). Implementing a corporate programme office. Proceedings of the 4th Project Management Institute-Europe Conference [CD-ROM].
Sadler-Smith, E., & Shefy, E. (2004). The intuitive executive: Understanding and applying “gut feel” in decisionmaking. Academy of Management Executive, 18(4), 78.
Sambharya, R. B. (1994). Organizational decisions in multinational corporations: An empirical study. International Journal of Management, 11, 827-838.
Sarkar, J. (1996). Positions of power. Far Eastern Economic Review, 159(3), 37.
Schoemaker, P. J. H., & Russo, J. E. (1993). A pyramid of decision approaches. California Management Review, 35(1), 9-31.
Shull, F. A., Delbecq, A. L., & Cummings, L. L. (1970). Organizational decision making. New York: McGraw-Hill.
Simon, H. A. (1964). On the concept of organisational goals. Administrative Science Quarterly, 9(1), 1-22.
Simon, H. (1987). Making management decisions: The role of intuition and emotion. Academy of Management Executive, 1(1), 57-64.
Stewart, J. (1998). The psychology of decision making. In D. Jennings & S. Wattam (Eds.), Decision making: An integrated approach (2nd ed., chapter 3). Harlow, UK: Prentice Hall Pearson.
Thiry, M. (2004). Program management: A strategic decision management process. In P. W. G. Morris & J. K. Pinto (Eds.), The Wiley guide for managing projects. New York: John Wiley & Sons.
Tsang, E. W. K. (2004). Superstition and decision-making: Contradiction or complement? Academy of Management Executive, 18(4), 92-104.
Vittachi, N. (1993). The old-fashioned way. Far Eastern Economic Review, 156(46), 58.
von Neuman, J., & Morgenstein, O. (1947). Theory of games and economic behavior. Princeton, NJ: Princeton University Press
Vosiadou, S., & Brewer, W. F. (1987). Theories of knowledge restructuring in development. Review of Educational Research, 57, 51-67.
Vyse, S. A. (1997). Believing in magic: The psychology of superstition. New York: Oxford University Press.
©2006 Project Management Institute