University Of Belize Management Decision Making PDF

UNIVERSITY OF BELIZE FACULTY OF MANAGEMENT & SOCIAL SCIENCE COURSE: Decision Making in Management (MGMT 4023) LECTURER: Dr. Romaldo Isaac Lewis (DBA) Chapter No.6; The individual in decision making: Heuristics in Management Decision Making OBJECTIVES After studying this chapter, you should be able t...

UNIVERSITY OF BELIZE FACULTY OF MANAGEMENT & SOCIAL SCIENCE COURSE: Decision Making in Management (MGMT 4023) LECTURER: Dr. Romaldo Isaac Lewis (DBA) Chapter No.6; The individual in decision making: Heuristics in Management Decision Making OBJECTIVES After studying this chapter, you should be able to understand: 1. Understand the application of individual in decision making 2. Understand the application of heuristics in Management Decision Making: 6.1) Heuristics in Management Decision Making Moving on from the frameworks presented by Soft Systems Methods as guides to structure organizational intervention and research (Check land, 1985), if we are to consider the systems view of problems and decisions, then we need to also need to consider how individuals can interpret the decision context. This chapter therefore considers how individuals can both interpret a decision context and what problems emerge when we look at that interpretation. Smith and Sharma (2002) note, for example, that specifically considering the value and emotions associated with organizational staff has historically been perceived as a weakness in the development of a consistent strategic aim and its implementation (Maccoby, 1976 cited by Carr, 2002). Yet, as chapter 5 has illustrated, management methods need to embrace these organizational features fully as well. This becomes quite self-evident when the evidence is reviewed that organizational structure and processes can simply be the extension of the ‘self’ of the leader(s) (Kets de Vries and Miller, 1991). In addressing the first issue of interpreting the decision context, we can adopt an ideographic perspective of the individual or an inductive view, which views individuals as unique – but in doing this we then have the difficulty of developing sufficiently generalized (or nomothetic) views of individuals to allow managers to structure the decision context. The earlier three phased model of chapter 1 would be an example of a nomothetic model for example. Similarly, we can look for nomothetic themes in individuals to help understand what may be common amongst how individuals make decisions. Three distinctive methodologies have thus been developed to look at this question. These are: 1. descriptive methods 2. explanatory method 3. prescriptive methods These three methods have been used by four dominant but interdependent knowledge bases of understanding individual actions (or their epistemologies) (Gross, 1996): 1. Psychodynamic 2. Behavioural 3. Humanistic 4. Cognitive (including neurobiology) 1|Page All four approaches claim to offer the understanding of the individual’s decision making processes, but also they raise questions of being able to identify and observe these processes – for example, behavioural epistemologies offer stronger evidence of these process as they can be measured and observed in practice, unlike cognitive epistemologies. Hence we also need to consider the appropriateness and validity of each view and how evidence for that view is gathered. We will consider each of the epistemologies in turn. 6.1.1 Views of decision making The psychoanalytical view sees the individual, as opposed to the external environment or other factor, as dominant in determining human behavior. The premise of this view is that unconscious and irrational processes play an active role in individual decision making and organizational activities (Jarrett & Kellner,1996). There are therefore affective and behavioural factors shaping individual development, which are believed to become more evident as risk and anxiety associated with a given decision, increases (Jarrett & Kellner, 1996). Early insights presented to understand this anxiety in decision making were developed by Sigmund Freud, who argued that behavior of individuals is derived from non-conscious aspects of individuality, including biological development. This development generates three contributing non-conscious facets of individuality which have been labeled (Carr, 2002): The Id – also more famously known as the ‘pleasure principle’. For Freud this aspect of individuality represented the unfettered child, devoid of any constraints, who acts hedonistically at all possible opportunities. It also therefore encompasses self-interested actions and behaviours which could be negatively viewed by others. The Ego – was the facet which is inherited from our immediate family and their sets of value on right or wrong, good or bad. For Freud, it provided a moral framework for making informed decisions through logic, memory and judgement to appropriately satisfy human needs and wants (and when to do so). The Superego – was the facet which described the societal set of cultures and values within which individuals live and share community. This view argues that it is these facets in conflict, which drive individual behaviours’ and actions as individuals seek to minimize this conflict and internal mental discomfort (this is also called cognitive dissonance (acting because we feel uncomfortable holding two conflicting ideas at the same time)). It is argued that the extent of the individual’s awareness of these facets and of these facets in others, which contributes to effective communication and decision making. The Johari window (Luft & Ingham,1955 cited by Hase et al, 1999) is a practical outcome of considering the rigidities between effective communication in individuals (and is discussed in more detail in chapter 7). From this perspective of decision making, unfamiliar situations can provoke actions and behaviours that may lead the individual to implement childhood coping mechanisms that have been recalled from the sub conscious. Jung also proposed that there was a deeper layer of unconsciousness, from which humans derive primordial ideas, images and emotions (Carr, 2002), from which common global myths emerge amongst many apparently disparate branches of human society. 2|Page These were called the Jungian archetypes from the collective unconscious, although common usage of the term ‘archetype’ has lessened its correct interpretation. For Jung, the importance lies with between the shape of the archetype and the content of the archetype (i.e. what story is used to identify the message of the archetype). For Jung, the individual’s personality development was a continual process. Wilfren Bion and George Homans are well known researchers who have explored this belief. Whilst their primary area of interest has been in the understanding of group decision making, their insights also help our understanding of individual actions and decision making. For George Homans (1950), the actions of individuals are shaped by the perception of the task to be resolved, the individual satisfaction to be derived in resolving that task and the personal development that results from that. Individuals contributing towards a decision in a group are influenced by how cohesive that group is and the interdependency between the three areas of task, satisfaction and personal relevance. Satisfaction for example increases by an individual conforming more to group norms and gaining that social approval. For example, identity has both a distinctive individual quality and a social quality – people wish to belong to a social norm with an accepted status but also be recognised as unique within that environment. The influence upon a motivation to perform a behaviour is derived from both a self-identity and a social identity construct (Terry et al. 1999; Anderson & Warren 2005). In short, people work in groups for the group and for themselves. Wilfred Bion (1961) developed the Freudian perspective (and latterly the Berne view (discussed shortly)), where individual conflicts and/or agreements between tasks and belief in those tasks causes members to not focus on the task at hand in group decision making. Bion recognized that groups of individuals tasked with resolving a decision have two important aspects to manage – the task at hand and assumptions about the required actions and behaviours to achieve that task. Under conditions of stress, anxiety and uncertainty, these assumptions of behavior can block the effectiveness of decision making by the individual in the group (McKenna & Martin-Smith, 2005), as individuals in the group swing between these assumptions in guiding their behavior. The identified assumptions in practice are: 1. A dependency assumption – that the group seeks a charismatic leader to resolve the task 2. Fight-Flight assumption – that the group acts to fight or flee an enemy (or construction of an enemy) and the perception of a win or loss by the group 3. Pairing assumption – that the group members hope an expert amongst their membership will step forward to provide the answers needed 4. Oneness assumption – that the group will seek and join an external force to resolve the task and allowing the group to continue passively (this is a fourth Bion ‘assumption’ identified by Torquet (and cited by Stacey (2003) cited by McKenna & Martin-Smith, 2005). Freud’s early view of the origins of individual behavior through unconscious cognitive conflict was then developed by from Wilfred Penfield (1951) and later Eric Berne (post 1968), in the area of Transaction Analysis (TA). It has also been explored in other management subject areas such as entrepreneurialism by Manfred Kets de Vries and his studies on entrepreneurial social deviants and misfits and Elizabeth Chell in her studies of critical incident theory (whereby early events in individual’s lives have impact upon their future development). 3|Page Transaction Analysis, has become an important area of both psychoanalysis study and management consultancy. Berne developed early transaction analysis by proposing that individuals are comprised of three mental modes: the child ego state – where behaviours, thoughts and feelings are recalled and replayed from childhood, the parent ego state – where behaviours, thoughts and feelings are copied and learned from parental figures and the adult ego state – where behaviours, thoughts and feelings are direct responses to the individual’s situational context (the here and now). This was developed in to the so called ‘descriptive model’. This development views the Adult state now as the ‘Accounting Mode’- through which individuals are able to interpret communications and choose how to respond to those communications. These are called ‘strokes’. In both aspects of the Parent and Child state, strokes can be positive or negative and invite appropriate responses to those strokes, through the Adult state. For example, a negative child state stroke which is driven by the desire to express their freedom and ‘play’, should draw a stroked response through the Adult accounting mode of the negative punitive parent, to control the child. Berne argued this is the basis of effective communication, the selection of appropriate responses. Equally, ineffective communication can then occur by a poorly or inappropriately chosen response to a given stroke. It has been noted in the discussion, that extending beyond the psychoanalytical view of decision making, we need to consider the task, the situational context and the relevance to the individual of those factors. For example, as an organization develops and grows, there is a need to ensure an appropriate fit between the decision maker’s style and source of judgement and that growth. a) In the emergent and entrepreneurial stage, an organization may be more accepting of a behaviour – even tantrums (the Child state) - in senior managers due to the unformed basic assumption set, unformed organizational power structure and relatively wild free-flowing environment (McKenna & Martin-Smith,2005). b) In a mature stage of growth with strong, functional basic assumptions and culture- an organization through its staff, would not accept petty or pedantic behaviour, except in individuals with very high levels of personal power (the Parent) – in other words, the organizational members could not reconcile the behaviour with the accepted organizational culture (McKenna & Martin-Smith,2005). c) In a transformational stage where boundary breaking decisions which oppose the basic assumptions and culture are often required to “drag” the organization into a new environment, these decisions would also depend greatly on the power position of the decision maker (Adult) (McKenna & Martin-Smith, 2005). 6.1.2 The judgement context – schemata Friedman (2004) asks the question about why manager’s often have misperceived understanding of their organizational problems or issues. We will discuss the issue of ineffective communication between individual decision makers in chapter 7 with the Johari window, but we can note for this discussion that one outcome of this weakness is then that managers tend to make decisions which have unintended consequences for their organization. This misperception can arise due to (Friedman, 2004): 4|Page a) Cognitive bias - can arise due to an individual’s experience, learning environment and organizational culture and context. There may also be neurological imperatives shaping these biases too. Later in this chapter we consider one well known area of cognitive bias, that of heuristics b) Mental models and schemata – these are the ways in which cause and effect are interpreted by the manager and therefore shape how that individual manager believes a problem should be resolved. This then leads c) to the selection of appropriate judgements to use – called schemata. These are cognitive processes (further discussed below) which for example can direct how and individual searches for a solution, how information is organized, how information is sorted and used etc. d) Emotions – have tended to be viewed negatively in an organizational context, although ample evidence of their importance as factors shaping cognitive structures (chapters 5,6 and 7). e) Perseverance of ineffective beliefs- occurs when managers resist changing their beliefs to correspond to actual observed evidence. This resistance can have many sources, from questioning the validity of evidence presented, to how we attribute success or failure to ourselves or our external environment. McKenna & Martin-Smith (2005) further note that individual schemata (discussed shortly), from which cognitive actions are chosen and applied (heuristics) can be affected by perceptions of those individuals (such as the locus of power in an organization) which may not be externally perceived as rational. Consider the situations in Box. 6.1 for example: Box 6.1 Irrational rationality? Lewin et al (1990) make numerous references to instances where trade union action to strike can appear irrational to observers, but is in fact rational to the union (despite counter claims actions are based upon incorrect forecasts of benefits and costs) as the action still seeks to maximize self-interested behavior. Tiwana (2006) discuss the observed phenomenon that failing software projects are often in receipt of escalated resources, yet from the project manager’s point of view, this escalation can be in pursuit of a variety of other real options which can capture much of the original project premise value. Moon et al (2003) similarly cite that project escalation of a failing activity can also originate (at least in part) from a group bias dynamic (see chapter 7 for a further discussion). Vander Shee (2008:244) notes “…although irrationality is commonplace, it does not necessarily mean that we are helpless. Once we understand when and where we make erroneous decisions, we can try to be more vigilant, force ourselves to think differently about these decisions, or use technology to overcome our inherent shortcomings.” One of the key aims of this text is to present in an accessible format, insight on the causes of these erroneous decisions so that a manager is better able to anticipate and mitigate the symptoms which generate false judgements. 5|Page The behavioural view of decision making rejects the ‘internal mind’ of psychoanalysis (the nomothetic view) and instead argues that observation is the important focus as this can be measured and observed. Decision makers are rational and analytical and use dispassionate processes which may include ethical and moral standards (Mckenna & Martin-Smith, 2005). It also argues that individuals are shaped by their experiences and by environmental forces, whereby they are defined through a series of collected learned responses acquired throughout all ages of development (Gross, 1996). Thus this view seeks to make a ‘better claim to ‘truth’. Hence, experience is now perceived as the key factor that motivates actions. Classically, this has been also explored through reflexive & stimulus / response and operant conditioning (voluntary & reinforcement). One observed difficulty of this view is the reliance upon past experience as a guide to future decision making actions can shape and limit those future decisions, when there is limited prior experience of a given situation. This has been previously identified with the incrementalism approach to decision making discussed in chapter 5 and moreover, the preceding chapter and this one, have sought to emphasize that decision making is not a linear nor rational process in general (McKenna & Martin-Smith, 2005). Smith & Sharma (2002) note that fundamentally, human beings are social animals and hence the reason why meetings between individuals in organizations play such a prominent role in most decision making activities (although this does not negate the need in many people to understand how formal meetings work within an organizational context and the importance of self-disclosure (see chapter 7 for a further discussion)). The cognitive view of decision making proposes a synthesis of the preceding views to understand individual decision making behaviours, where in other words, decision actions are shaped by our experiences, behaviour, environment and an individual’s processing abilities (Shackleton et al, 1990: Gross, 1996: McKenna & Martin-Smith. 2005). The biological development of this view argues behaviour is determined through genetic, neurobiological and physiological factors and which takes time to develop in individuals, as they grow and mature (Gross, 1996). From a cognitive view, decision actions can therefore occur through cognitive dissonance for example holding conflicting moral views ‘I am a good person’, yet using the internet to pay for an assessed submission, or knowing that smoking is unhealthy yet being a smoker and still seeking to enjoy a healthy lifestyle. They can also occur through interaction with others and effective communications or as noted, through consideration of the relative access to power and influence in a given situation. People, in this school of thought are viewed as being information processors and therefore have limitations upon those abilities. It is a subjective view which further proposes that any external reality is subject to interpretation by the individual through mental constructs, schemas and narratives, all of which take time to develop. This viewpoint has emerged in the preceding discussions and which is explored in greater detail shortly. Shackleton et al (1990) and Hardin III (1999) give succinct summaries of the development of this approach to decision making with in particular, Hardin III (1999) stressing Herbert Simon’s original work on the limitations of human cognitive processing, but latterly also identifying developments that integrate this with the problems of short term memory and the decision making process (the judgement aspect as discussed in chapter 1). Shackleton et al (1990) present the Driver & Rowe (1987) cognitive decision style model, which is mentioned here as it is a helpful summary of the decision making issues raised so far and which identifies: 6|Page 1) the directive style of decision making – reflects individuals who act promptly, based upon organizational/ ethical/moral rules and covet power and authority. 2) the analytical side of decision making – reflects individuals focused upon creativity and innovation on resolving problems or exploiting opportunities. 3) the conceptual side of decision making – reflects individuals with a high orientation for achievement, a desire for independency and artistic preferences. 4) the behavioural side of decision making – reflects individuals seeking affiliation, easy and effective communications and using coalitions of interest to achieve preferred outcomes. We can also note that in Chapter 5, the systems view and SSM approach to problem analysis, solution development and selection for decision making similarly reflected left and right brain thinking, which is also notably at the core of the four aspects of the Driver & Rowe model of decision making styles. As a result of these insights, management interest in the development of cognitive schemas and heuristics has increased. Heuristics are discussed in more depth shortly, but in simple terms, they refer to organized elementary information processes (Simon, 1978 cited by Hardin III (1999)) and schemas are then describing the problem space within which those heuristics derive their meaning (i.e. the associated knowledge for a given management context). In short, this view of decision making sees the manager as interpreting the task at hand (the problem domain), developing a problem space for its resolution and deploying appropriate information processes (heuristics). In addition to this cognitive subjective focus, recent neurobiological research has identified relationships between biology and conceptual decision making of individuals. For example, interesting research has highlighted that economic decisions extend beyond cost/benefit analysis (see Naqvi, Shiv and Bechara, 2006). Further evidence of the importance of neurobiology to individual decision making has been found by considering the impact of ventromedial prefrontal cortex damage research (vmPFC) on individual decision making. Individuals with vmPFC were observed to make more frequent decisions that resulted in personal losses/injury compared with uninjured individuals, but did not seem to learn from those experiences, so as to reduce their losses in the future. The injury of vmPFC did not affect an individual’s intellect, problem solving abilities or memory but did seem to affect that individual’s ability to use emotions in their decision making. Evidence suggests that reasoning about moral dilemmas activates structures including the vmPFC and which is amplified when there are negative consequences for others. One final view of decision making to reflect on, is the humanist view of individual decision making, which progresses our understanding beyond cognition, to consider broader issues of the ‘World View idea’ and ‘the whole individual’ discussed in chapter 5. It is another branch of human psychology, but one that argues motivations and actions come from the need to self-actualize and freely pursue and realize personal ambition and aims (McKenna & Martin-Smith, 2005). Purposeful actions therefore are a mix of experiences, meanings and choices. Individuals are driven by the need to meet selfdevelopment aims and difficulties emerge in their actions, when those aims cannot be achieved (Gross, 1996). 7|Page 6.2) More on the cognitive view – schemata’s and heuristics After this broad overview of key factors shaping effective individual decision making, the discussion now develops further the cognitive view of decision making. We noted that in chapter 2, methods of making decisions under uncertainty were presented, where probabilities attached to actual event outcomes were unknown (and in which case each event was identified as equally likely) – so that decisions could be made through the optimistic criteria (Laplace), evaluated so as to generate the best worst outcome – the conservative criteria (Wald) or evaluated so as to generate the least worst outcome through a regret analysis (the minimax regret (Savage) criteria). In all cases therefore, inevitably we do ascribe some probability likelihood to the range of outcomes. We now consider how to manage those situations where we have no available data at all, on the likelihood of outcomes. As this chapter was being written, there was significant publicity for an otherwise little known human psychologist Daniel Kahneman (or rather as much as there ever can be for a human and cognitive psychologist in the public press) (Burkeman, 2011: Lewis, 2011). In the learned academic environment however, Daniel Kahneman and Amos Tverskey are highly regarded for their pioneering work on cognitive psychology, which was briefly introduced in chapter 5. Indeed, it earned them the Nobel Prize in economics. The recent publicity was driven by the forthcoming publication (2011) of Kahneman’s first ‘mass market’ text on decision making (Thinking fast and slow). Their work was undertaken largely between 1971 and 1984 and which focused upon the reasons for why individuals reach decisions, and in particular how information is interpreted by them (i.e. their problem space, schema and heuristics). Schmidt (2011) identifies four types of schema employed by individuals in making decisions: 1 Person Schema – mental constructs concerning the attributes of a particular individual. 2 Event schema - are the mental constructs of the ways in which tasks and problems are approached. The task at hand shapes the selection of an appropriate problem methodology. 3 Role schema - are the mental constructs that are used to present normative understanding of individual role expectations (i.e. how we expect an individual occupying a certain role to behave). 4 Self-schema – are mental constructs that are maintained by the individual of the present situation and past experiences. This contributes to an individual’s self-image and where for example, self-efficacy is a type of self-schema that applies to a particular task. 8|Page Schmidt (2011) further clarifies the role of schemas as evaluative (i.e. where there is a mental comparison between individual actions for a given function or job), role playing (where they generate understanding about how individuals should act in a given situation), identity (where they categorize individuals by the roles they do) and prediction (where by mentally placing individuals in certain roles, they are then expected to behave in those roles). Clearly, this view of the source of decision making judgements is also manifest in behaviours of individuals as was also identified in the discussion of transactional analysis earlier and expected scripts (role plays) of individuals involved in a communication transaction (Berne, 1996). The origin of schemas deployed for the problem space comes from individual experiences. Carr (2002) has also proposed that Jungian archetypes offer an origin for individual schemata too. It is reasonable to argue that there may be a required sequence of experiences required to develop a particular view of a given decision making context from these origins (i.e. constructs require maturation) and they may be acquired directly (as discussed) or through the use of a communication vehicle (such as metaphors or storytelling (see chapter 5 and the use of storytelling within organizations for Mode 2 SSM)). As the SSM and holism discussion further stressed in chapter 5, the understanding of the problem domain (and here problem space) needs to be an iterative and ongoing process of learning. Without this as Schmidt (2011) states, the accurate diagnosis of a problem (or exploitation of an opportunity) can be blocked. Furthermore, Hardin III (1999) argues that individual bias in decision making may occur through the misapplication of schema and/or under developed knowledge of the problem domain. From a management perspective then, how that process of determining an action is made, is subject to bias and prejudice based upon information uncertainty, scarcity and individual interpretation. We noted in the first chapter from the Garbage Can model of decision making from Cohen et al (1972) that it is highly unlikely that all the necessary information, resources and contexts will be convergent in an organization to make an optimal decision. Hardin III (1999) presents the work of Shanteau (1992) and his ‘Theory of Expert Competence’ to help clarify the necessary decision making inputs and which also identifies the sources of bias in individual (expert) decision making that require management attention. These are: a) The decision maker needs to have problem domain knowledge (codified and tacit) b) A robust leadership style including self-confidence and belief c) A cognitive ability to identify key information from within the problem domain d) To be able to use heuristics and data simplification procedures to resolve complex scenarios e) Be able to define clearly the issue / problem to be addressed and judged. As we have noted from chapter 5 however, it is unlikely that all management problems will be clearly delineated and packaged so that the decision maker is able to determine a full and detailed understanding of the task. In these situations, it is likely that bias and incorrect use of heuristics (through incomplete understanding of the problem space) will be the result (with sub-optimal decision outcomes). 9|Page Now we begin to identify what consequences this lack of an understanding of the problem domain and space have upon effective decision making. Lewis (2011) makes the very valid point, that the more you begin to consider why a human endeavor to achieve a given goal fails or does not succeed as hoped, the more you begin to discover the work of Kahneman and Tversky. For example, Gigerenzer (1999) and his colleagues asked four randomly chosen groups of decision makers (in this case graduate students of finance and economics from both countries and similar sized groups of Munich and Chicago shoppers) to select from a listing of 798 American and German companies, which ones they recognised. From this listing, two portfolios of companies were selected (one for the US and for German companies). Where the volunteers were choosing from their own country firm listing, a 90% recognition across the sample people was enough to warrant that firm’s inclusion in the resulting stock portfolio. However, for the shoppers from both countries looking at the company names from the other country (i.e. not their own), this common citational requirement had to dip as low as 10% to ensure that sufficient companies were in fact selected for the stock portfolio. When the value of the company portfolios created were then compared six months later, the selection chosen by the shoppers in Munich had grown an impressive 47%, 10% higher than the German DAX overall, over the same time period. More importantly, all four portfolios where the firms chosen had been from the other country, had outperformed those country’s stock market measures. To ensure that chance had no significant impact upon the selection of companies in the portfolio lists, a control listing of random firms performed more poorly than the selected portfolio’s in 7 out of 8 instances. Clearly, the selection of stocks by individuals from their non-home country was a marked success in terms of the value of those stocks – why? The accepted viewpoint on why this has happened, is found in the works of Daniel Kahneman and Amos Tversky and their studies of the ‘availability heuristic’ - which has been used by the individuals concerned. Heuristics are mental (cognitive) functions that allow individuals to make decisions under a context of uncertainty or scarcity. When presented with a new situation, an individual will attempt to recall similar examples and draw relevant insights from that recollection. The more recent or vivid a relevant memory is the more of an impact it can have on the current situation being considered or decided upon. As with the example of the selection of stock market companies, this can be a marked improvement on random selection as an answer to a problem. However, it can also result in bias (when for example an event is recalled precisely because it is rare). Being able to easily recall an event does not mean it is more or less likely to actually occur. As Goodwin & Wright (2004) state, the fear of crime is disproportionately more observed within some parts of society than actual incidents of crime in those parts of society do make clear. Recent press and the language and tone of the press, is an important variable here that shapes our ability to recollect this information. Furthermore, a prejudice towards viewing the activities of the organization may result in illusory correlations between the contribution of those activities. For example, let’s assume an organization has established a manufacturing plant in Vietnam. Through having early quality problems with semi-finished goods shipped to the UK, that plant may then suffer an illusory correlation that all their semi-finished goods are of poor quality within the organization. Yee et al (2008) for example cited Iaffaldano and Muchinsky (1985), who had described that employee satisfaction and job performance was an illusory correlation. In other words, we expect there to be a correlation between the two variables, but there is no empirical evidence of that. 10 | P a g e Equally Grant (2010) in discussing the contribution of innovations to the profitability of organizations, identifies that where a correlation might be expected between the two activities, there is in fact very limited data for this. It is important therefore for managers to consider the persistent influence that preconceived views and the immediacy of recall for an event, can have on how effective the availability heuristic is. This heuristic can manifest in a variety of forms – for example in the earlier discussion through individuals being able to identify an organization through a recollection of that firm (say through a TV report or other announcement). This can explain for example, why increased marketing in a recessionary period (which might seem counter intuitive) may generate positive results for an organization as shoppers are more easily able to recollect that marketed organization. Srinivasan et al (2005) for example, in a survey of 154 senior marketing executives, identified a positive correlation between a proactive marketing approach in a recessionary period and brand developed products and services. Indeed, Awake (2009) suggests that aggressive marketing during a recession also tends to generate improved sales which continues after the end of the recession as the gains tend to be ‘sticky and persist’. More importantly, Asker (2009) citing work by Kamber (1992) identified that recessionary and post recessionary success was not dependent upon the existing competitive strength of the organization. Overall, there is less competition for the attention of consumers and organizations that engage in proactive marketing, raise their profile and consumers find it easier to identify with and recall that organization’s name and/or product/brand at the point of the buying decision. This can therefore offer a significant opportunity for the prepared organization, willing to exploit the opportunity of the recessionary period. Another simple example of the availability heuristic, is to ask an individual whether there are more words in the English language which have k as the third letter or as the first? The ease with which an individual can recall words beginning with k, does not mean that in reality there are more words beginning with k (as the opposite is in fact true). Nonetheless, this does not stop most individuals choosing the wrong answer here. Nor does it stop governments and the media exploiting the human preference towards viewing more recent events as more important than distant events and manipulating that to preferred outcomes. Clearly, this heuristic (and others discussed), allow individuals to make rapid decisions and can be effective in doing so. Yet accurate statistical data will be superior in every event, if it is available. However, much more important than this, is the consequence of the implications of identifying that all individuals possess these heuristic attributes and therefore the very foundation of classical economics – that it is comprised of rational, selfish agents, who’s tastes do not change – is flawed (Lewis, 2011). Behavioural economics was the resultant new discipline that emerged from the convergence of these views. In total, over 133 different forms and types of heuristics have been revealed by cognitive research (Goodwin & Wright, 2004). Several more will now be illustrated that are particularly relevant for managers. 11 | P a g e 6.3) More on heuristics Hardin III (1999) and Goodwin & Wright (2004) identify that the 3 earliest generic types of heuristic identified from the work of Kahneman & Tversky were: 1) The availability heuristic 2) The representativeness heuristic 3) Anchoring and adjustment heuristic The availability heuristic has been discussed above and we now consider the other two generic types of heuristic. The representativeness heuristic is a cognitive process through which individuals consider how similar a given event/ person/process is to a family of events/persons/processes (in other words, is event X representative of other known events) with a tendency to view a sample as being more representative of a parent population that it actually is. So if you were to meet ‘Linda’ who works in the USA and described herself as an outgoing individual with interests in climate change and global warming– when faced with the question – do you think Linda is either a lawyer, or a lawyer with a support for environmentalism – what would you answer?’ Rationally you should identify that it is far more likely Linda is a lawyer. Selecting the second option is logically incorrect (as to be this, she must also be a lawyer). Also it is statistically far more likely that she is a lawyer in the USA as an occupation without any additional focus. This heuristic is a cognitive process through which an individual seeks to identify given characteristics which associated stereotypes. The human brain is particularly gifted in identifying and seeking patterns in data (Marques de Sa, 2001) which can then be erroneously interpreted as evidence of causal relationships (Roberts, 2004). Goodwin & Wright (2004) identify that the representativeness heuristic has a number of key perceptional biases associated with it. These include ignoring the base rate frequencies associated with an estimated probability. For example, consider this description in Box 6.2: Janet was a quiet, self-assuming individual but of high intelligence. She has a strong need for order, structure and ensuring details are recorded and identified. Whilst not the most creative person, her writing style is fluid and tidy with clear elements of control and discipline apparent. She has a strong drive for competency in her skills and in others she works with. This personality description has been chosen from a random group comprising 30 librarians and 70 TV presenters. What is the probability that Janet is a librarian or a TV presenter? If the view was taken that she was a librarian, then this answer ignores the stated fact of the base frequencies of the information presented (that this was in fact only 30% likely). It was much more likely that she was a TV presenter. The author has used variants of this question with both undergraduate and postgraduate classes with the usual result that most answers chosen are biased upon the perception of the stereotyped description. 12 | P a g e Goodwin & Wright (2004) further identify that bias in the representativeness heuristic can arise from expecting sequences of events to appear random, where rationally, none should be expected. In tossing a coin ten times, the two sequences of HHHHTTTTHH and HTTHTHTTHT are both equally likely, although the latter sequence seems more random. The primary risk for an organization is that managers respond too abruptly to perceived patterns in, for example, sales data. We also know from chapter 2 and 3, that events such as coin tossing are independent of their previous history, so having had a sequence of ten heads being tossed still means that the next toss is equally likely to be a head or a tail. However, it is not uncommon that individuals expect chance to be self-correcting (the so called Gambler’s fallacy argument, that is a gambler has had a losing streak, they are ‘due’ a winning streak). A further bias in the representativeness heuristic is that of regression to the mean. In this instance, individuals maintain a perception that when an event follows another unlikely event, that subsequent events will continue to be unlikely. Consider for example the question posed in Goodwin & Wright (2004) – think about the likelihood of boys versus girls being born at a small versus large hospital? In particular, would a small or large hospital record more days in which more than 60% of newborns are boys? The choices available are (1): The larger hospital would see more days on average where more than 60% of newborns are boys? (2): The smaller hospital would see more days on average where more than60% of newborns are boys? (3): Both hospitals would expect on average to see about the same proportion of male and female newborns on any given day. The correct answer would be (2) – as this is the smaller sample size where you would expect to see more skewed results in a statistical distribution. The issue of an expected gender balance does not reflect the more likely statistical outcome for this sample size. The third generic heuristic outlined by Good & Wright (2004) and Hardin III (1999) is that of the anchoring and adjustment heuristic. In chapters 2 and 3, we explored the quantification of event likelihood as probabilities. This developed from the rational (RAT) models with externally verified data, through to estimated probabilities in say data forecasting. It is at this stage in the management decision making process, that the risk of this heuristic being erroneously used is apparent. For example, estimating the duration of time needed to complete a task is often based upon previous experience (the anchor value) which is then changed to reflect perceived difference in this new task when compared with older completed tasks (the adjustment). However, it is common that there is insufficient adjustment of the anchor to reflect the likely (in this case time) outcome. In classes taught by the author, an interpretation of the classic Tversky & Kahneman presentation of this problem, can be used: Activity 6.1: The Anchoring and Adjustment heuristic Class individuals are presented with the following problem statement: Go to the tutor and draw a number at random from the blind bowl. Read the number and then replace it back into the bowl. Now answer the question ‘How many countries of Africa are members of the UN?’. It is very unlikely that the class student would know the answer to this specific question and by selecting a number at random, the student is presented with an anchor upon which to adjust their recorded answer. The bowl contains a large selection of number lots to draw from, but in fact only features two different numbers (in this case 10 and 45). 13 | P a g e The use of this heuristic by students then tends to result in two groups of answers clustered around the low teens with another clustered around the 40-50s. A further bias in the anchoring and adjustment heuristic, is that of overestimating the likelihood of conjunctive events. For example, the likelihood that the reader secures a first class degree and wins the lottery can result in individuals overestimating the likelihood of the second event as they have anchored that probability to the first event (Goodwin & Wright, 2004). Let’s take another hypothetical example, if we assume we have some fortune telling apparatus and from that we can deduce the following: the chance of securing a first class degree is 90%, then the chance of securing a Master’s Degree is 90%, then the chance of securing an excellent job is 90% and then the chance of securing promotion in that job within the first year is 90% - we might assume that within a few short years , to have secured a highly successful career as we have anchored on the achievement of one outcome and adjusted the likelihood of the subsequent outcome. However, when the probabilities are actually determined (recall chapter 2), the actual probability of securing these outcomes are presented in this manner is 65% (or 0.94). Similarly, individuals tend to underestimate probabilities of disjunctive events occurring. For example, the likelihood that you will as a manager experience either poor sale or bad marketing press in say a week. As before, individuals will have a tendency to anchor on one of the associated events and as a result, under estimate the likelihood of the other event then occurring. The classic question posed in Good & Wright (2004: 249) is that of: “10 people work in an office and each has a 5% probability that at least one of these people will leave within the next year. Estimate the probability that at least one person will leave within the next twelve months (assume that the chances of any one person leaving is independent of whether the others stay or leave)” What would your answer be? If you had said 5% you would be wrong – the actual answer is just over 50%. An incorrect answer suggests that it is anchored to the 5% and was insufficiently adjusted to reflect the actual combinations of potential leavers: i.e. Probability of Person A OR B OR C OR D OR E OR F OR G OR H OR I OR J leaves = (10x0.05) =50% But of course you may have 2 or more people leave the office at the same time which also then needs to be determined. Hence Probability (2 or more people leave) =0.05x0.05= 0.0025, Probability (3 or more people leave) = (0.05)3 = 0.000125 …etc. – hence the sum of these other probabilities would be (through to at least 9 people leaving) = 0.00262. Hence the probability of at least one person leaving is 0.5+0.00262 = 50.26%. This chapter will close with one further reference to one final important heuristic identified by Tversky & Kahnemann called – Prospect Theory. This is the observed tendency of individuals to become more risk averse when situations involving gains, but risk seeking when situations are likely to lead to losses (Chapman, 2006). In other words, If a decision is framed in terms of gains, then individuals will tend to avoid risk and protect a small gain rather than risk it for a larger gain. Equally, if a decision is framed in terms of losses, individuals will tend to gamble more to avoid a large loss, rather than accept a smaller loss. It is an important heuristic, that also manifests at the collective level in group decision making dynamics (see chapter 7). 14 | P a g e 6.4) Summary This chapter has expanded on the issues presented in chapter 5, namely the contribution of the individual to the decision making process. It has sought to present a broad overview of the issues that shape individual perception, behaviour and action when making decisions and in doing so, give to the reader a broad framework to identify potential sources of erroneous decision making outcomes. Significant emphasis has been placed on the role and function of cognitive process in the decision making process and the selection of appropriate processes to make judgements. From a managerial perspective, to mitigate the incidence of bias in individual decision making activities, individuals can practice the following: 1. Being aware of multiple cognitive issues shaping decision outcomes – and trying to encourage and adopt multiple perspectives. 2. Acting deliberately negatively towards initial decision outcomes (by acting as Devil’s Advocate) 3. Always question assumptions made in reaching a decision outcome and validate any inferences made 4. Do not neglect the unpopular decision outcome 5. Try to make incremental decisions rather than seek (and fail) to achieve optimal outcomes 6. Rely on probabilistic relationships and statistics to guide the decision making process 15 | P a g e