Decision-making and problem solving skills
the rational versus the garbage can model of decision-making
Albert A. Einsiedel, Jr., Ph.D.
University of Alberta
To be a manager means to make decisions (Herbert A. Simon). Everyone has to make decisions. Managers are special in that they are paid good salaries to make decisions. Other ways in which managerial decision-making differs from nonmanagerial decision-making are:
• Managers have to deal with a large volume of decisions every day. For example, managers must make decisions about:
- organizational priorities, goals, and objectives
- organizational design
- hiring, promotions, transfers, training, leaves, and termination of personnel
- performance standards, office rules, and smoking
- budget, purchases, vendors, and vouchers
- public relations strategies
- linkages, coalitions, alliances, and networks
• The consequences of a manager’s decisions influence many people. Decisions, particularly policy decisions, affect:
- clients and their advocates
- constituents, public, public interest groups, and media
- bosses, peers, and subordinates
- current and future members of the above groups
• The stress involved in making managerial decisions is often high due to:
- time pressure
- few really good options to choose from
- a lot of poor options to choose from
- little or no useful information about most options
- existence of a dilemma between a correct but unpopular decision on one hand, and an incorrect but popular one on the other
- lack of expertise to evaluate properly each choice and arrive at a reasonably defensible decision
- no external, objective standard against which to evaluate a decision for its effectiveness, adequacy, and efficiency
The Typology Method of Dealing with Stressful Decisions
The manager who is continually and persistently faced with stressful decision-making tasks must develop effective coping strategies or eventually succumb to burnout and other negative effects of chronic stress. One method that is often used is the typology method of coping. Its effectiveness appears to be very, very limited. This is how it works. The manager sorts out the type of decision to be made. Typical categories (and their characteristics) used by many managers are displayed in Figure 1.
Types of Decisions
Categorizing decisions using such typologies gives some managers the confidence (derived, no doubt, from the illusion of having organized the problematic situation) they need to deal with the real task at hand, which is to make a choice of a solution and to implement it. This technique is obviously not a bad one if the manager can move beyond the typology. If all that occurs is the classification of the problem, then the manager is well advised to consider other possibilities, of which the following are highly recommended.
The Rational, Normative Approaches
Some managers are serious students of decisionmaking and problem-solving theory, especially those fresh out of school who have not had the experience of testing textbook approaches to managing organizations. They try their best to apply what experts say managers “ought” to do when solving problems. Let us examine some of the rational, normative approaches of some of these experts.
• Maier  summarizes in four sentences what numerous desision-making and group dynamics experts have said:
1. Determine the problem boundaries by focusing on what is or is not a part of the problem.
2. Examine the changes in the decision environment or decision maker which may have precipitated the problem.
3. Divide complex problems into subproblems. For example, distinguishing clearly the means from the ends is one technique for mapping out the components of a complex set of events which are casually linked.
4. Focus upon the controllable components of a decision situation.
• Easton  describes the steps involved in the Satisficing and the Optimizing Models of decision-making which are shown in Figure 2. According to Easton, the Satisficing Model is the simplest form of the decision process; it involves the search “for an acceptable, workable solution, and not for the best possible.” A skillful and experienced decision maker may be able to identify a workable solution on the first try. Easton describes two kinds of optimizing approaches: (1) finding the best of all possible alternative courses of action; and (2) finding which alternative among those available is best.
• A comparison of the Satisficing and Optimizing Models reveals the following differences (See Figure 2), in Easton’s words:
“The satisficing model devises only one action at a time in Step 4 of the process. If the action is acceptable, the process goes on to the end without need for the recursive loop of Step 6, but if it is not acceptable, there must be a path back to the ‘devise action’ activity of Step 4. The optimizing model, on the other hand, identifies all feasible alternatives in Step 4. One of these must be chosen in Step 6. Thus, the optimizing process has no need for a recursive loop back to Step 4.”
“If the decision involves an emergency for which the costs of inaction or delay are high, the satisficing model is useful. This is mainly because an experienced decision maker can usually devise an action on the first try that will cause a degree of improvement. After the emergency is over, he could take additional, optimizing action to bring the situation to its desired state.”
• Finally, there is the Garbage Can Model of decision-making which was described by Cohen, March, and Olsen  as the process of making decisions in an “organized anarchy,” which they define as, “decision situations or organizations which are characterized by inconsistent and illdefined preferences, unclear technology, and a fluid definition in the decision-making process.” The Garbage Can Model is not a normative model, one which prescribes how one ought to make decisions. Rather, it is more of a metaphor that attempts to describe how decisions are actually made in many organizations.
Easton’s Models of Decision-Making
- There are four variables or “streams” that circulate in a kind of Brownian movement in a fixed decision space, that decision space being the garbage can; those four variables are: problems, decision participants, choice opportunities, and solutions.
Note: This is the author's graphic representation of the Garbage Can Model. Apologies to Cohen, et al.)
- Unlike the rational models that assume a sequential ordering of steps in the decision process, the garbage can model makes no such requirement. Solutions can be in search of problems; the expertise of participants can determine which problems to address, and; the choice opportunities that exist can determine which solutions are selected. For example, the organization may have some surplus funds to spend before the end of the fiscal year (or else lose them). The manager in charge of the decision would be looking for a legitimate project to fund. Some choices are readily available and a match is made between project and funds.
- The garbage can model does not assume a feedback loop. Indeed, in contrast to the rational models in which the content of the solution to a problem is not known precisely until the end, the garbage can model anticipates the content of the solution, and influences the availability of resources which, in turn, affects the choice of the intervention. For example, the project’s outcome may not be clearly predicted at the onset. The project evolves depending on which resources are available, who is working on the project, and the time spent on the project.
- Rational models view resources as mere means to an end, and tend to regard them as passive elements in the decision process that ought not to influence the choice of the problem and, at times, even the solution. The garbage can model, on the other hand, considers resources as practical and significant determinants of choices. For example, funding made available for demonstration projects in developing social services for the elderly who suffer speech difficulties is likely to stimulate novel interests; gerontologists will develop an interest in speech pathology, and speech therapists will develop a deeper concern for the elderly. In another example, a manager who finds himself inside an Italian restaurant at noon is more likely to have a pizza, spaghetti, or calzone instead of egg drop soup.
- The garbage can model probably describes how many managers “muddle through” decisions, charting their way by the “seat of their pants,” relying on all the managerial “street smarts” that they have acquired from “winging it” through life, not to mention their “gift of gab” that sometimes conceals the lack of substantive reason and logic with eloquence and a few gimmicks.
An Integrative Approach
The Garbage Can Model, although very realistic, can produce reactive rather than proactive solutions. Reactive solutions are usually short-term and crisisoriented. Proactive solutions are usually part of a long-term plan of action or series of actions based on a more systematic analysis of the problem situation. Relying only on reactive solutions may, in the long run, cause some important problems to be overlooked and some solutions to be overworked. Combining the rational models and the Garbage Can Model may be a better approach since most decisions, to be effective, have to be objectively correct and acceptable.
• Correctness refers to the objective quality of the choice. The correctness of the choice can be measured against a conventional or scientific standard. Examples of this would be compliance with quality control standards, job specifications, rules, and regulations.
• Acceptance refers to the subjective attractiveness or popularity and desirability of the choice, particularly to the individuals who are expected to carry out the decision.
• According to Maier , for a decision to be effective, it has to have the right combination of both variables. Thus,
Effectiveness = Correctness x Acceptance
• If you want to get mathematical, you can assign values from 0 to 10 to each variable. Since the model assumes that an effective decision is the product of the multiplicative2 combination of correctness and acceptance, you can figure out what happens when both are extremely low (0x0 = 0); when one is low, the other high (0 x 10 = 0 or 10 x 0 = 0); and when both are extremly high (10 x 10 = 100). A difficult decision is one where you need to be correct, but the most correct option is very unpopular and is apt to be rejected. Another is where the most popular choice is also known to be objectively wrong. Since not all decisions require that both variables be high, the first task of the decision maker is to...you’re right... consult a typology of problem situations and decision strategies (See the “recommended choice of decision-making strategies” which follow later).
• There are trade-offs that can be made between correctness and acceptance, depending on the situation. This is the important lesson to be learned from this approach. Depending on the relative need for correctness and acceptance, any of these four strategies can be effective:
- The leader makes the choice with some assistance from experts. The decison is announced to the group.
- The group makes the choice, with the leader acting as discussion leader only.
- The leader makes the choice and uses persuasive skills to gain acceptance for it.
- The group makes the choice with the leader using conference leadership skills to gain quality.
• It may seem complex at first, but go over the typology below slowly and you will see that it is rather intuitive. Examples of problems are given to clarify the points.
The recommended choice of decision-making strategies is as follows:
|Type of Problem||Type of Decision||Strategy|
|1a.||Decisions regarding expansion, new services, decentralization||1. High need for, correct decision but low need for acceptance||1. Leader makes choice with aid from experts|
|1b.||Decisions regarding determining costs, setting prices, establishing eligibility|
|1c.||Decisions regarding purchases of services, supplies|
|1d.||Decisions requiring specialized skills and technical knowledge|
The acceptance of the decision can be enhanced by the effective use of group participation. However, the decision maker must weigh the cost of getting a group solution that is discrepant from the experts' solution against the cost of discontent brought about by nonparticipation of the group before soliciting group participation.
|Type of Problem||Type of Decision||Strategy|
|2a.||Schedule of overtime, vacation, coffee breaks||2. High need for acceptance and low need for correctness||2. Group makes choice with leader acting as discussion leader|
|2b.||Distributing scarce but desirable resources such as office furniture, money|
|2c.||Getting something undesirable accomplished such as unpleasant work, undesirable office space|
|2d.||Settling problems involving violations of regulations or lack of cooperation|
The leader presents the problem to the group and leads a discussion during which differences in opinions surface. The leader maintains a neutral position and confines his or her role to clarifying the issues, encouraging discussion, supplying information, and making appropriate summaries. The leader’s objective is to facilitate the group’s arrival at a consensus of a unanimous agreement on a decision that resulted from effective group interaction.
|Type of Problem||Type of Decision||Strategy|
|3.||Problems of efficiency and cost-reduction requiring technical solutions and widespread support||3. High need for acceptance and high need Tor correctness||3. Choice is made by leader who then uses persuasive skills to gain acceptance|
|Choice is made by group with leader using effective conference skills to gain quality|
This type of decision can fail if the decision is incorrect and/or if it is not supported by the group.
Several assumptions have to made in using this integrative approach:
• First of all, that it is possible to determine the desired levels of correctness and acceptance of the solution—whatever they may be—before the decision strategy is selected. This means that the leader has to set the initial criteria with which to evaluate the type of problem; this, in turn, enables him/her to select the type of decision and the type of strategy.
• Secondly, that it is possible to actually elicit participation of the group in a decision-making activity. In some organizations, where the practice is to have all the decisions made by the leader, breaking tradition may only confuse the group.
The ultimate effectiveness of a decision can only be assessed after it has been made and carried out. Fortunately, for most managers, the majority of their decisions are such that even if their initial choice were not the best, they can try again. The trial-and-erroruntil-yousucceed experiences, an innovative organization, a forgiving boss, supportive coworkers, and a creative mind are the ingredients that produce the savvy in experienced decision makers that justifies their robust salaries and reputation for knowing how to handle stressful problem situations. Knowing when to apply the rational and garbage can models of decision making is another factor than can enhance their managerial effectiveness.
1. Cohen, D., March, J.G., & Olsen, J.P., A Garbage Can Model of Organizational Choice, Administrative Science Quarterly, 1972, 17, 1-25.
2. Easton, A., Decision Making: A Short Course for Professionals.,Volume I, New York: John Wiley, 1976.
3. Maier, N.R.F., Problem-Solving Discussions and Conferences: Leadership, Methods and Skills., New York: McGraw-Hill, 1963.
A.A. Einsiedel is Director of the Local Government Studies Program and Associate Professor of Extension at the University of Alberta. He has been a management consultant and trainer, and is the author of the forthcoming book, Improving Project Management.
2This is analogous to the wind chill factor, which takes into account ambient temperature and wind velocity. A slight increase in the value of one variable produces a dramatic change in the resulting stimulus.