Risk taking and the decision making process


Wright-Patterson AFB

Lloyd A. Swanson

Syracuse University

All managers must deal with many uncertainties or “risks” as they make the decisions which constitute a major aspect of the managerial profession. Further, it is generally assumed that individuals differ in their ability to face and deal with risks, while specific managerial positions expose their incumbents to different levels of risk and uncertainy. A project manager in particular deals with high levels of both project and professional risks. He is held directly responsible for meeting the schedule, budget, and technical objectives of the project: his extreme visiblity in the event of failure accentuates these project risks. Professionally, long-term affiliation with a project can lead to obsolescence in the manager's own technical specialty, thus increasing the uncertainty with which he must face the scheduled termination of his project. Thus, project managers would make excellent subjects for investigating the possibility that a manager's propensity to take risks could affect the way he makes decisions and hence his overall job performance.

To explore these possibilities, a group of 79 potential project managers each “managed” a computer-based project simulation, designed to determine their relative propensity to take risks, for a period of nearly four months. Data concerning project performance was collected on a weekly basis. Information concerning the individual's method of arriving at decisions was collected near the end of the stimulation. The results show that those most prone to avoid risks spent more time in making decisions, placed significantly more emphasis on obtaining information, or worked in groups to a greater extent. Thus this research indicates that an individual's propensity to take risks does affect the method by which he makes decisions and could, therefore, affect the way in which a manager organizes and relates to his work team.


“Increasing technological complexity,” “increasing requirements for resources,” and “high organizational complexity” are terms often used to describe current project efforts of both governments and industry. These descriptions appear only too true. Because of its extensive use in the aerospace, military, and general research fields, project management aids in the practical application of much of our most advanced technical knowledge. The cost of developing these technical concepts is staggering: a $75-million project was recently described to one of the authors by its manager as a “rather modest” effort. And even such a “modest” project involves a complex array of power and authority relationships among many different levels of several independent government and business organizations.

A project is established to achieve some specified goal, and the focus of all project activity is the project manager. He is held responsible for meeting all project objectives in terms of time, cost, and quality, yet he must often depend on a number of individuals and organizations outside of his direct authority structure. A fixed goal and a well-defined time schedule make it easy for his supervisors to evaluate his progress and criticize his work. Delay is usually expensive in terms of project costs and, since the project manager is normally evaluated in terms of his project's success, extensive or continued delay can have a major impact on his career as well.

As a result, the project manager has long been recognized as one who must face and deal with many uncertainties or “risks” in performing his job. From the project manager's view, these uncertainties can be grouped into two categories.1 Project risk involves the ultimate responsibility for meeting and maintaining the project cost, quality, and time objectives. His success and recognition as a competent manager depends primarily on his achievements in these areas, and the risks involved are made even more pressing by his extreme visibility in the event of failure. Professional risk involves the overspecialization or obsolescence which the project manager may experience as a result of long-term affiliation with a single project, or with the project management field. This problem can severely limit his choice of future jobs, and it thus increases the uncertainty with which he must face the scheduled termination of the project. These two factors combine to indicate that the project manager operates in a relatively high risk environment. As a result it appears likely that, through a process of self-selection and elimination, only those individuals with a high propensity to take risks will expose themselves to such a perilous, high stress job situation.

The purpose of this research, then, is to measure the project manager's propensity to take risks and compare it to his decision methodology and his overall success in the project environment. It would appear plausible that the project manager's high propensity to take risks would affect the way he makes decisions and thus the overall organization and success of the project itself, particularly in light of recently reported research results.2 If this relationship is shown to exist in the current research, it might conceivably help explain why different project managers organize their projects differently, and could ultimately have a major impact on the selection of project management personnel.

Managerial Risk

Managerial risk is defined as the manager's perceived exposure to possible failure and penalty in accomplishing his job or task. No model is known to have been proposed relating a manager's propensity to take risks to his job performance.

Related work available from the literature of the psychological and managerial fields shows that individuals make decisions within a unique frame of reference or “psychological set.”3 Of particular interest here is the work of Scodel (1961), which demonstrates that the risk taking propensity of decision makers affects the way individuals solve problems.

At about the same time, Wallach and Kogan (1959) began a series of researches into the nature of risk by developing a “Dilemmas of Choice” questionnaire to measure the relative propensity to take risks among members of a test group. This questionnaire has been extensively used by those interested in investigating the characteristics of risk taking. Wallach and Kogan used it to study differences in risk taking behavior between the sexes (1959) and between different age groups (1961). It is also a principal tool used to investigate the effect of group participation on the individual's risk taking propensity (Wallach, Kogan, and Bern, 1962; Kogan and Wallach, 1964 and 1967; and Belovicz, Finch, and Jones, 1968).

The management literature deals with risk in two areas. Decision theory holds that a manager makes decisions under either conditions of certainty, risk, or uncertainty. Here conditions of risk includes those instances where the manager has knowledge of only the probability of alternative outcomes, while conditions of uncertainty involve those situations where the manager possesses neither perfect nor probabilistic information about the possible occurrence of alternative outcomes. This approach generally stresses the need to quantify (if only subjectively) decision information, and it leads to a number of extremely valuable quantitative techniques to assist managerial decision making. Role theory, on the other hand, stresses the ambiguity and conflict inherent in various types of organizational positions, and the behavior expected of individuals filling those positions regardless of their personality.4 In particular, Kahn et. al. (1964) identifies three key determinants of role ambiguity as organizational complexity, rapid organizational change, and changing managerial philosophies. These are certainly key factors relating to the uncertainty inherent in the specific position and, as indicated earlier, they are particularly characteristic of a project manager's situation.

With some minor overlaps, the two fields deal with different aspects of the problem: psychological literature basically studies individual differences with regard to risk taking characteristics. The managerial literature generally deals with factors in the job situation which create uncertainy and risk in the specific managerial position. It is certainly not possible at this time to claim that a consistent relationship exists between the relative level of a manager's propensity to take risks and his success in positions with specific characteristics of uncertainty.


This research is an effort to combine the psychological and managerial approaches to risk taking to determine how the individual's propensity to take risks affects his job performance in a relatively risk-prone management position. The research was made possible by the availability of a large group of potential project managers at the University. Although all of these individuals were taking elective courses in operations management and network analysis, this sample population (n = 79) should not be confused with the usual college student test group. The majority of this sample consisted of individuals holding fulltime jobs managing various organizations and taking course work to increase their competence in specific areas. The mean age of this sample was 30.4 years (s = 6.2), much older than the usual student sample. The mean amount of full-time work experience was 11.1 years (s=6.0). A large portion of the sample (43%, n = 34) consisted of military officers subject to transfer into project management work. This is a real possibility, since the military does a great deal of project work. Those remaining individuals with full-time work experience were all considered to be among the ranks of “middle management,” from whom project managers are normally selected.5 Large portions of the sample had prior work experience with project bidding, contract management, and/or network analysis techniques. The work experience of this sample is summarized in Table 1. The point is that this sample is not a typical group of business managers, but is a fair representation of the population available to business and government organizations as a source of project managers.



Active Duty
With Network
With Contract
Management or
Project Bidding
83.5% (n=66) 43.0% (n=34) 35.4% (n=28) 24.1% (n=19)

The experiment consisted of requiring each member of the sample population to individually perform a computer-based project simulation using network analysis techniques. It should be emphasized that the simulation was an individual, not a group effort.

In brief, the computerized management game PERTSIM allows the participant to purchase information place bids, and then to supervise or manage a project within a totally simulated environment.6 Although the project contains a great amount of uncertainty, this uncertainty can be reduced by purchasing a variety of levels of information about each activity, thus bringing the estimates of the activity durations and costs closer to actual values. After analyzing this additional information, the participants submit project bids. As a result of this competitive bidding, each individual receives a probability of receiving the contract. This probability, based solely on the bids submitted, reflects the long-term aspects of project management and represents the individual's ability to bid competitively over several potential projects. It includes both the individual bid's absolute value and its relative rank among all other bids submitted. At the conclusion of the project, this probability is multiplied by the total project profit (or loss) to yield a total expected profit.

Each participant is individually responsible for supervising the project through completion. Supervision consists of specifying the order and the relative crash level of the activities to be performed for a specified time period. The computer simulates the actual durations and costs for those activities completed or started during the specified time period. This information is given to the project manager, who revises his original plan in light of the new information and submits a revised plan for the next time period. This sequence is continued until the entire project has been completed. Although each of the items of information and the actual durations and costs are simulated, the same value is used for each individual selecting the same item of information or activity level. In essence, this eliminates randomness between individuals.

Two incentives were present to help insure the effective motivation of each participant: first, participants competed for maximum expected profits; second, the results influenced final course grades. Each individual managed the project for nearly a four-month period, while data concerning project performance was collected twice weekly. Near the end of the project data was collected from each participant concerning his method of arriving at project decisions. In addition, the Wallach and Kogan “Dilemmas of Choice” questionnaire was administered to determine the relative propensity to take risks among the sample population. This “Dilemmas of Choice” questionnaire was developed originally by Wallach and Kogan (1959). They demonstrated its validity in subsequent research.7

In an attempt to distinguish each individual's decision methodology, the following information was collected: (1) The total amount of information purchased as measured in total dollars spent, and the frequency with which information was requested; (2) The estimated amount of total study time spent in group effort; and (3) The time required each week to manage and control the project.

The simulation provided two measures of performance, total expected profits and total costs. The total expected profit is computed by subtracting the total costs from the dollar bid and multiplying the probability of receiving the contract. Since the total expected profit may be dominated by the single bid decision, it was felt that the total costs of the project might be more reflective of the individual's management and control skill. Ususally these two measures are closely related, but occasionally there are substantial differences due to errors in bid making.


The accompanying tables summarize the actual results of the study. Table II lists the simple correlation coefficients between the propensity to take risks and each of the factors analyzed. This study did not demonstrate the existence of any significant relationship between the measures of project performance and the subject's propensity to take risks.



Study Factors Risk* N**
Decision Methodology Factors a) Information Purchased (Total Dollars Spent) -.24580+ 77
b) Frequency of Requests (Total No. of Requests/ Project) -.06434 79
c) Group Effort (% of Total Time In Group Work) -.24556+ 69
d) Time Spent/Decision (Average Time in Minutes) -.19095++ 79
Performance Factors e) Total Costs (Dollar Cost of the Project) -.07517 77
f) Expected Profits (Dollar Profits from the Project) -.03985 77

*Risk Propensity is measured on a relative scale from 1 to 10. (mean = 5.71, standard deviation = 0.9).

**The number of responses (N) varies from the total sample size (79) due to Inaccurate or Incomplete survey responses.

+Significant at the .95 level.

++Significant at the .90 level.

However, with the exception of the frequency of requests, the results do demonstrate a significant relationship between the subject's propensity to take risks and his decision methodology.

The major concern of this study is to examine how the propensity to take risks influences decision-making behavior. Thus, risk is considered to be the independent variable, while the three significant factors of information purchase, group effort, and time spent per decision are assumed to be dependent in nature. Hence it was necessary to perform three separate linear regressions. Table III shows the detailed results of these regressions for the significant decision methodology factors.



Factor Intercept Slope F Significance
Information Purchased (Total Dollars Spent) 8924.04 -625.08 4.69 .95
Group Effort (% of Total Time in Group Work) 39.75 -4.06 4.17 .95
Time Spent/Decision (Average Time in Minutes) 131.28 -10.33 2.12 .90

*Risk, the Independent variable, is measured on a relative scale from 1 to 10 with 10 meaning high risk (mean = 5.71, standard deviation = 0.9).

Analysis and Conclusion

A project management situation was chosen as a basis for this research for several reasons. Project management is a high risk situation for a manager, involving many unknown factors which may impact the project's success. The increasing use of network analysis techniques for project work makes the success or failure of management's actions highly visible, a fact which both increases the risk level of the project manager's position and makes such a situation ideal for research purposes. Finally, a project manager has a higher level of responsibility for and control over activities that are a part of his project than most managers, and results can therefore be more directly related to his performance.

In the case of this study, each subject was given the same base knowledge level at the same relative point in time concerning both network analysis in general and the PERTSIM simulation. Controlling the information flow in this way effectively prevented differences in knowledge of the situation from affecting the study's results.

A significant relationship was demonstrated between the propensity to take risks and the methods used to make decisions. Those with less propensity to take risks tend to work more in groups, purchase more information, and/or spend more time making individual decisions. Note that the same individual does not necessarily do all three. This is substantiated by the correlation matrix presented in Table IV, showing no significant relationships among the individual decision factors. Risk does not appear to be a dominant factor in the actual decisions made, as shown by the regression slopes in Table III. Nevertheless, the individual's propensity to take risks has been shown to be influential in determining the methods he uses to arrive at decisions.

The three methods of making decisions studied here can actually be subjectively related by considering them all as simply methods of obtaining additional information about the specific decisions required in this simulated environment. Individual subjects were apparently using varying combinations of these approaches in an effort to remain within their own tolerance levels for frustration.8: that is, to reduce the level of risk inherent with the situation to a level they could be “comfortable” with.

This analysis also provides a possible explanation for the failure to demonstrate any relation between performance and the propensity to take risks. By obtaining additional information and reducing the level of risk, each subject was actually adjusting the situation to fit his own preferred risk level. With each subject thus operating within his own risk tolerance, no consistent relationship between performance and risk propensity could be expected. The study thus supports the basic thesis that project managers vary in their willingness and ability to accept and deal with different levels of managerial risk.

The differences between the simulated project environment used in this study and an actual project environment may well be critical here. This specific study demonstrates no relationship between the project manager's risk propensity and project success, and thus provides no justification for selecting project managers on the basis of the individual's risk taking propensity. In the business environment, however, the project manager's ability to adjust the level of risk to his own desires is much more limited in this simulation. In this respect the sample used in the study was biased toward potential project managers. In the business environment the project manager who cannot adjust the risk to a level with which he is relatively comfortable has the choice of functioning under conditions of frustration or withdrawing (“quitting” or changing jobs in some other way). Thus project managers with relatively low risk propensities may be “forced” to work under conditions of frustration, conditions in which they are unlikely to perform at maximum efficiency. It can be argued, then, that a relationship between the manager's propensity to take risks and his performance should exist as he reaches and exceeds his level of frustration in trying to adjust the risk in the project situation to a level which he can accept and deal with. Empirical support for this view awaits an analysis of working project managers and the success of their projects.



Study Factors Information Purchased Group Effort Time Spent per Decision
a) Information Purchased (Total Dollars Spent) -- .10087* .18958*
b) Group Effort (% of Total Time in Group Work) -- .04312*
c) Time Spent/Decision (Average Time in Minutes) --

*Not significant

Incidentally, it is obvious that the higher the frustration level at which the project manager is forced to work, the greater the chances that he will withdraw. Here, then, is support for the existence of a natural process of selection leading to a concentration of those managers with a higher propensity to take risks in the more risk-prone project management positions. This thesis also requires confirmation through a detailed study of working project managers.

In summary, this research represents only a first step in understanding the role of risk taking in managerial situations. Yet it is evident that this aspect of an individual may ultimately help explain why different managers use different techniques for solving similar problems. It is possible that a lower propensity to take risks is the principal reason many project managers demand a costly and “excessive” information base, or tend to delegate decisions to committees. It is also possible that an individual with a high propensity to take risks would be a better choice for manager of a project requiring quick, decisive, authoritative actions on a tight schedule. Much research remains to be done before this study and such conclusions can be substantiated. To the extent they are so, however, they imply that forcing a manager into someone else's preconceived mold of a “best” organization design for a specific task is likely to increase his frustration and hence reduce his effectiveness. Instead, managers should be allowed to develop organizations tailored to providing the kind and quantity of information they need as individuals to make efficient decisions. This in turn implies the need for much greater flexibility on the part of most firms than exists even in the most progressive project and matrix organizations of today.


1 David L. Wilemon and John P. Cicero, “The Project Manager - Anomalies and ambiguities”, Academy of Management Journal, Vol. 13, No. 2 (September, 1970), 277-279.

2 For example, Wallach, Kogan, and Bern have reported that, “… persons with stronger risk-taking proclivities tend to become more influential in a group than a person who is more conservative.” Michael A. Wallach, Nathan Kogan, and Daryl J. Bern, “Group Influence on Individual Risk Taking,” Journal of Abnormal and Social Psychology, Vo. 65, No. 2 (August, 1962), p. 85. These and similar results are also reported in references [10] and [19].

3 Alan C. Filley and Robert J. House, Managerial Process and Organizational Behavior (Glenview, Illinois: Scott, Foresman and Company, 1969), 115.

4 Joseph L. Massie, Essentials of Management (Englewood Cliffs, New Jersey: Prentice-Hall, Inc., 1971), p. 74.

5 John M. Stewart, “Making Project Management Work,” Business Horizons, Vol. 8, No. 3 (Fall, 1965), p. 61.

6 Swanson, Lloyd A. and Pazer, Harold L., PERTSIM: Test and Simulation, Scranton, Pennsylvania, International Textbook Company, 1969.

7 See references [9], [10], [19], [20], and [21]. For a complete copy of the “Dilemmas of Choice” questionnaire, see Nathan Kogan and Michael A. Wallach, Risk Taking: A Study in Cognition and Personality (New York: Holt, Rinehart and Winston, 1964), pp. 256-261.

8 Frustration, in this instance, is used as defined by Sayles and Strauss: “When people are put under too much pressure they become frustrated . . . they react in strange ways that tend to reduce the effectiveness of the organization in its main task of getting out production”. Leonard R. Sayles & George Strauss. Human Behavior in Organizations (London, England: Prentice-Hall International, Inc., 1966), p. 139.


1. Atkinson, John W., “Motivational Determinents of Risk-Taking Behavior,” Psychological Review, Vol. 64, No. 6 (1957), pp. 359-372.

2. Barkin, Stephen R., The Use of Management Games in Organizational and Behavioral Research. A Pamphlet of the Working Paper Series (MISRC-WP-71-03) Prepared by the Management Information System Research Center. Minneapolis, Minnesota: Graduate School of Business Administration, University of Minnesota, July, 1971.

3. Belovicz, Meyer W., Finch, Frederick E., and Jones, Halsey, “Do Groups Make Riskier Decisions than Individuals, “Academy of Management Proceedings, 28th Annual Meeting (December 26-28, 1968). pp. 73-85.

4. Coombs, C. H. and Pruitt, D. G., “Components of Risk in Decision Making: Probability and Variances Preferences,” Journal of Experimental Psychology, Vol. 60, No. 5 (November, 1960), pp. 265-277.

5. Filley, Alan C. and House, Robert J., Managerial Process and Organizational Behavior, Glenview Illinois: Scott, Foresman and Company, 1969.

6. Gray, Clifford, “Performance as a Criterion Variable in Measuring Business Game Success: An Experiment with a Multiple Objective Performance Model.” A Research Paper Presented at the 2nd Annual Southeastern Regional Conference of the American Institute for Decision Sciences, April 27-28, 1972.

7. Kahn, Robert L., “Stress from 9 to 5,” Psychology Today. Vol. 3, No. 4 (September, 1969), pp. 34-38.

8. Kahn, Robert L., et. al., Organization Stress: Studies in Role Conflict and Ambiguity, (New York: John Wiley & Sons, Inc., 1964).

9. Kogan, Nathan, and Wallach, Michael A., “Risk Taking as a Function of the Situation, the Person, and the Group,” New Directions in Psychology III, ed. G. Mandler, et. al., (New York: Holt, Rinehart and Winston, 1967).

10. Kogan, Nathan and Wallach, Michael A., Risk-Taking: A Study in Cognition and Personality (New York: Holt, Rinehart and Winston, 1964).

11. Lewin, Arie Y. and Weber, Wesley L., “Management Game Teams in Education and Organization Research: An Experiment in Risk Taking,” Academy of Management Journal, Vol. 12, No. 1 (March, 1969), pp. 49-58.

12. Massie, Joseph L., Essentials of Management, 2nd ed., (Englewood Cliffs, New Jersey: Prentice-Hall, Inc., 1971).

13. McGregor, Douglas, The Professional Manager, eds. Caroline McGregor and Warren G. Bennis, (New York: McGraw-Hill Book Co., 1967).

14. Philippatos, George C. and Moscato, Donald R., “Laboratory Experiments with Large-Size Business Games.” A Research Paper Presented at the 2nd Annual Southeastern Regional Conference of the American Institute for Decision Sciences, April 27-28, 1972.

15. Roberts, John M., et. al., “Action Styles and Management Game Performance: An Exploratory Consideration,” Naval War College Review, XXIV, No. 10 (June, 1972), pp. 65-82.

16. Sayles, Leonard R. and Strauss, George, Human Behavior in Organizations (London, England: Prentice-Hall International, Inc., 1966).

17. Scodel, Alvin, “Value Orientations and Preference for a Minimax Strategy,” Journal of Psychology, Vol. 52 (1961), pp. 55-61.

18. Stewart, John M., “Making Project Management Work,” Business Horizons, Vol. 8, No. 3 (Fall, 1965), pp. 54-68.

19. Swanson, Lloyd A. and Pazer, Harold L., PERTSIM: Text and Simulation, (Scranton, Pennsylvania: International Textbook Company, 1969).

20. Wallach, Michael A., Kogan, Nathan and Bem, Daryl J., “Group Influence on Individual Risk Taking,” Journal of Abnormal and Social Psychology, Vol. 65, No. 2 (August, 1962), pp. 75-86.

21. Wallach, Michael A. and Kogan, Nathan, “Aspects of Judgement and Decision Making: Interrelationships and Changes with Age,” Behavioral Science, Vol. 6 (1961), pp. 23-36.

22. , “Sex Differences and Judgement Processes,” Journal of Personality, XXVII (March-December, 1959), pp. 555-564.

23. Wilemon, David L. and Cicero, John P., “The Project Manager - Anomalies and Ambiguities,” Academy of Management Journal, Vol. 13, No. 3 (September, 1970), pp. 269-282.



Related Content

  • PMI White Paper

    The Impact of the COVID-19 Crisis on Project Business member content open

    By Project Management Institute | Project Business Foundation This is a report on the results of a survey jointly conducted by PMI and the Project Business Foundation. The intention was to replace observations and opinions with reliable data. The mission was…

  • Project Management Journal

    Multiphase assessment of project risk interdependencies member content locked

    By Hwang, Wenli | Hsiao, Bo | Chen, Houn-Gee | Chern, Ching-Chin Project risks evolve dynamically, so variations in risk influences during the life cycle of an information system development project require analyses to devise risk management strategies cost…

  • PMI Global Congress—North America

    TAFT (this ain't freakin' Texas) member content open

    By Fournet, Bill a case study into a failure to manage assumptions

  • A model to develop and use risk contingency reserve member content open

    By Shrivastava, Narendra K. This paper describes a simple model that project managers and risk team members can implement immediately to begin using risk contingency reserve in their projects. Using a risk register and an…

  • Project Management Journal

    A new decision making model for subcontractor selection and its order allocation member content locked

    By Abbasianjahromi, Hamidreza | Rajaie, Hossein | Shakeri, Eghbal | Chokan, Farzad All experts agree on the importance of subcontracting. The high impact of subcontractors on the construction process means that the selection of subcontractors is a sensitive activity. Previous…