advert

JOURNAL OF RESEARCH IN NATIONAL DEVELOPMENT VOLUME 8 NO 1, JUNE, 2010


RISK FACTORS ON ESCALATION OF COMMITMENT BEHAVIOUR

 Olusola I. Akinbobola
Department of Behavioural Studies, Redeemer’s University
E-mail: solaakinbobola@yahoo.co.uk
and
 Benjamin O.Ehigie
Department of Psychology, University of Ibadan
E-mail:benosang@yahoo.com

                                                                     

 

Abstract

This study examined the predictability of risk propensity, risk norm and risk perception on escalation of commitment behaviour. An experimental study was conducted. Each participant was randomly assigned into one of two groups as personal responsibility was manipulated as (high/low). Participants were masters of business administration (MBA) students who were engaged in decision making in various organisations. Independent variables were measured with standardised instruments. Escalation of commitment behaviour was measured as the willingness to give further funding to marketing a product. Data collected were analysed using hierarchical regression. Result showed that risk propensity, risk norm and risk perception jointly predicted escalation of commitment behaviour (P< .05). Personal responsibility (P< .01) and risk norm (P< .05) independently contributed significantly to escalation of commitment behaviour. It is implied that management should integrate adequate norms into decision making process as prescribed appropriate behaviour.

 
Keywords: Risk propensity, risk norm, risk perception, escalation of commitment behaviour

 


Introduction
Individuals are the decision makers in the organization. The decision makers are not separate from the organisation becausetheyare answerable to the organization for the consequence of decisions they take. When such decision consequence is positive the decision is acceptable to both the decision makers and the organization. When the decision consequence is negative, decision makers in an attempt to prove to the organization that a negative decision consequence was really a correct decision may persist with the decision. When an individual is highly responsible for making an initial decision that leads to negative decision consequence, the individual may undergo risk, and therefore increase commitment to the failing initial decision. This is often referred to as escalation of commitment. Escalation of commitment involves risky decision making (Staw, 1997; Whyte, 1993).

Decision makers who are personally responsible for making initial decision that result in negative decision consequence may escalate commitment to save face with the organization. Furthermore, the need to obtain organizational reward such as payment of bonuses, accelerated or automatic promotion prospects and profit sharing may also lead to excessive risk taking (Ogunleye, 2001) by decision makers. As a result, organisations incur losses through risk taking (Udegbunam, 2000) on failing decision. Consequently, a large number of organisations may have decision fiasco and become distress because of excessive risk taking (Soyibo & Odusola, 2002; Udegbunam, 2001).

Janis (1989) and Whyte (1989) posited the role of risk related variables and their interrelationship in escalation situation. Risk taking may depend on an individual’s propensity to take the specific risk. Sitkin and Pablo (1992) predicted that the individual history of risk-taking is affected by the person’s risk propensity. Risk propensity describes an individual’s disposition and tendency to take or avoid risks, a relatively stable characteristic which can be modified through experience. Risk averse individuals overestimate the probability of loss which discourages risk taking whereas risk prone individuals underestimate the probability of loss which encourages risk taking.

Researchers, Sitkin and Weingart (1995) looked at other risk variables such as risk perception that may affect individual and organisations risk taking behaviour. Risk perception is the assessment that reflects the degree to which an individual perceives a particular situation as negative, as a threat and as out of control (Wong, 2005). Risk perception is defined as an individual’s assessment of how risky a situation is in terms of estimated probability of the degree of situational uncertainty, how controllable that uncertainty is and confidence in the estimate.

Sitkin and Weingart (1995) and Wong (2005) found that risk perception fully mediated the effects of risk propensity in risk taking behaviour. According to Wong (2005) there is a need to remove the direct effects of risk propensity on decision making behaviour. Hence a gap exists in literature to determine whether or not risk propensity is fully mediated by risk perception. Hence, there may be a missing link between risk propensity and risk perception.

Fishbein andAjzen (1975) in the theory of reasoned action and Ajzen (1998) in the theory of planned action proposed behaviour on the basis of three factors. These factors are personal attitude toward a behavior, the subjective norm regarding the behaviour and perceived degree of control over the behaviour. Behavioural theories state that behaviour is observable but some theories have propounded that behaviour can be preceded by intention (Ajzen & Fishbein, 1975).  Intention is in turn determined by the three factors namely attitude which is a global positive or negative evaluation regarding performing the behaviour. Subjective norm which is perceived social pressure for performing the behaviour and lastly perceived behavioural control which is the perceived ability to perform the behaviour.

Numerous studies provide support for the predictive validity of the theory of planned behaviour (Ajzen, 1998; 1991; Akinbobola, 2008; Armitage & Conner, 2001; Rivis, Sheeran & Armitage, 2006).  Studies also supported the mediating evidence of the theory of planned behavior (Conner & Abraham, 2001; Christian, Armitage & Abrahms, 2007).  It is important to understand that change in the three factors namely attitude, subjective norm or perceived control account for observed changes in intention and behaviour (Michie & Abraham 2004).

Rohrmann (2002) described risk propensity as attitude towards taking risk. Risk attitude is the individual’s belief or mindset toward taking or avoiding risk when deciding how to proceed in a situation. The definitions of risk propensity as proposed by Rohrmamn (2002) suggest that risk propensity may be understood as attitudes.  Risk perception is described as an individual’s perceived degree of control on how risky a situation is (Sitkin & Weingart, 1995). Risk perception is also related to perceived controllability of outcomes (Baird & Thomas, 1985).

The present study observed that risk propensity may represents Fishbein and Ajzen (1975) first factor which is personal attitude toward behaviour as also confirmed by Rohrmann (2002). Risk perception may represent Ajzen (1988) third factor which is perceived degree of control over the behaviour. There may be a link between risk propensity and risk perception as suggested by Wong (2005). The link may be risk norm representing Fishbein and Ajzen (1975) second factor which is subjective norm regarding behaviour. Subjective norm is an individual’s motivation to comply with the views of other people regarding the behaviour in question. The present study suggests, risk norm as subjective norm because risk norm has the characteristics of Fishbein and Ajzen (1975) subjective norm regarding behaviour.

The present study is proposing risk norm as a mediator between risk propensity and risk perception.  Risk norm refers to informal rule of prescribed appropriate behaviour, expectations of significant others that influences and compensates risk taking preferences of individuals. This explanation of risk norm implies that whether or not a person decides to take risk is based on norms and the expectation of significant others. In the oganisational setting, significant others consist of management and employers.  In this study, risk norms are informal guidelines that influence individuals to comply with employers and management prescribed behaviour in particular situation.    According to Staw (1981) organisational norms surrounding individuals and the effects of such norms may lead directly to increase in commitment to a course of action. Norms could be integrated into decision making (Fishbein & Ajzen, 1975) and viewed as one element of a prospective rational decision to commit resources (Staw, 1981).                        

Risk norm differs as it is practiced in various types of organisations. Organisations may influence decision makers risk taking behaviour through two different reward systems which are process-oriented approach and outcome-oriented approach. Process-oriented approach reward decision makers based solely upon the decision process, which if the decision makers followed the prescribed decision process, he or she is rewarded regardless of the outcome of the decision. Process-oriented approach fosters risk-seeking attitudes by effectively eliminating the personal risk exposure of the decision maker. An outcome-oriented approach rewards decision makers based on the outcome of their decisions, regardless of whether these decisions were arrived at by sound analysis or faulty judgment. The outcome-oriented approach induce the decision maker to become risk-averse because the outcomes have a direct effect on the personal wealth of the decision maker through receipt of organisational reward.

Previous empirical work on risk factors (Sitkin & Weingart (1995) found that people high in risk propensity are more likely to perceive a situation as one of low risk and thus have a higher tendency to take risk compared with people low in risk propensity. According to Wong (2005) risk propensity was positively related to escalation of commitment, risk perception was negatively related to escalation of commitment and risk perception partially mediates the effects of risk propensity. These findings are consistent with Sitkin & Pablo (1992). Some studies Sitkin and Pablo (1992), Wong (2005) were interested in the top management team homogeneity on escalation. The present study will control for demographic factors.
The purpose of the present study is to determine the predictive ability of risk propensity, risk norm and risk perception on escalation of commitment behaviour. To fill the gap and determine whether or not risk propensity is fully mediated by risk perception.
Therefore the following hypotheses were stated:
Hypothesis
1.         Risk propensity, risk norm and risk perception will independently and jointly predict escalation of commitment behaviour.
2.         Risk norm will partially mediate the effect of risk propensity on risk perception.
Method
Design
 An experimental research was conducted.  The independent variables are risk propensity, risk perception and risk norm. Dependent variable is escalation of commitment behaviour. The dependent variable is escalation of commitment behaviour.
Participants
102 Masters of Business Administration(MBA) students of Lagos State University Ojoo, Lagos who participate in decision making in various organizations including service, engineering, marketing and financial institutions, following Wong (2005) suggestion for future research. Lagos is a commercial nerve centre in Nigeria. Participants voluntarily took part in the study 61 (59.8%) were males and 41 (40.2%) were females. Their mean age was 36.49 and standard deviation of 5.62 with age ranging from 25 to 51. The mean working experience is 9.26 and standard deviation of 6.02 with working experience ranging from 1 to 28 years.     Choice of participants was in line with Biyalagorsky, Boulding and Staelin (2001) pattern of study.
Instrument
The instrument used in this study is in a questionnaire form. Escalation of commitment behaviour was measured with a scenario modified from Arkes and Blumer (1985). Other studies (Conlon & Garland, 1993; Moon, 2001; Wong, 2005) used the scenario. Escalation of commitment behaviour was measured as the willingness to give further funding to marketing a product. The participants were asked to give willingness rating ranging between 0 percent (absolute no) and 100 percent (absolute yes). 
Risk propensity was measured with a 5 item scale developed by Sitkin and Weingart (1995). This scale asked participants to rate their tendency to make risky decision from 1 (very weak) to 7 (very strong). Wong (2005) reported (µ = 0.65). The present study reported Cronbach’s alpha of 0.74
Risk norm scale was constructed for the purpose of this research.  It originally had 7 items that described the influence of significant others on risk taking behavior. Following a thorough review of related literature, items for the scale were derived from risk literature including (Rohrmann, 2002; Wong, 2005). The item pool of 7 items was presented to psychologist as experts for content validity following Nunnally, (1978). The experts were requested to indicate whether or not they considered each item as measuring risk norm using a 5-point Likert type response ranging from strongly disagree (1) to strongly agree (5). 7 items that were above the average score of 3 were subjected to item-total correlation analysis resulting in the acceptance and selection of 7 items, with significant item-correlation ranging from 0.33 to 0.83 (Rust & Golombok, 1995).  All items loaded on one factor in a varimax rotation factor analysis conducted. These 7 items were therefore retained as valid items for measuring risk norm. The 7 items were subjected to reliability test and found to have a Cronbach alpha coefficient of 0.72. The scale was therefore accepted as reliable. This scale asked participants to rate influence of significant others, which is the influence of organisation on making risky decision from 1 (very low) to 7 (very high).   
Risk perception was measured by a 3 item scale developed by Sitkin and Weingart (1995). This scale asked participants their assessment of inherent risk of the current situation ranging from 1 (very low) to 7 (very high). Wong (2005) reported (µ = 0.81). The present study reported Cronbach’s alpha of 0.76
Variable for control and manipulative checks
Personal responsibility was measured with a 2 item scale developed by Conlon and Park (1987). Participants indicated the extent to which they feel responsible for making decision on giving further funding for the product, and for making a previous financial investment for the development of the product. Wong (2005) reported (µ = 0.72). The present study reported Cronbach alpha of 0.74
Procedure
 A pilot study was conducted to generate items for risk norm and to get psychometric properties of risk propensity, risk norm and risk perception. In the main study participants were assigned to high personal responsibility group and low personal responsibility group. Participants were assigned to groups through random sampling. The questionnaire consisted of three parts. Part 1 was a modified version of scenario from Arkes and Blumer (1985) case of the ‘blank radar plane’. A high personal responsibility version and low personal responsibility version of the scenario were constructed.  In the high responsibility group the participant was informed, “Last year you as a corporate financial executive made financial investment for the development of a new product X. You recently learnt the bad news that another company had already started marketing a similar product Y with a better design. You have been assigned to indicate your willingness to give further funding for the product X. What is the extent of your willingness to give further funding for the product X?”
In the low responsibility group the participant was informed,    “Last year the financial executive of the company, who is your colleague, made financial investment for the development of a new product X. You recently learnt the bad news that another company had already started marketing a similar product Y with a better design. Meanwhile the financial executive retired last year and you were assigned to indicate your willingness to give further funding for the product X. What is the extent of your willingness to give further funding for the product X?” Following previous studies (Conlon & Garland, 1993; Moon, 2001; Wong, 2005) the present study used willingness rating to indicate escalation of commitment behaviour.  In Part II, participants were asked to fill questionnaires assessing risk perception, risk norm and risk propensity.  Part III consisted of demographic variables of sex, age and work experience of participants which were analysed as control variables.
Result:
In Table 1 result of hierarchical regression analysis was computed with the predictor variables. The control variables of age, sex, working experience and personality responsibility were entered first. The control variables jointly and significantly accounted for variance on escalation of commitment behaviour. F (4, 94) = 13.85; P < 0.1. The independent contribution of each of the control variable showed that only personal responsibility contributes significantly to escalation of commitment behaviour (β = .59; P <. 01). The result of the beta weight showed that participants with high personal responsibility exhibit higher escalation of commitment behaviour than those with low personal responsibility.

In the second model, risk propensity was added to the equation to examine if it has direct effect on escalation of commitment behaviour. The result revealed that risk propensity along with the control variables jointly and significantly contributed F (5, 93) = 11.98; P < .01 to escalation of commitment behaviour. There is no significant independent contribution of risk propensity on escalation of commitment behaviour.

In the third model risk norm was added to the control variables and risk propensity to examine if risk norm mediates the effect of risk propensity. All jointly and significantly accounted for variance in escalation of commitment behaviour F (6, 92) = 11.03; P <. 01. The inclusion of risk norm led to 3 percent change in variance (∆R2 = .03; P <.05). Analysis of the independent contribution of each of the variables shows that risk norm (β = -.17; P < .05) contributes significantly and negatively to escalation of commitment behaviour. The beta weight result implies that participants who experience low risk norm exhibit higher escalation of commitment behaviour than those who experience high risk norm.

Although in the second model risk propensity did not have independent significant influence on escalation of commitment behaviour (β = -0.15; Pns). However in the third model risk propensity had a negative significant influence on escalation of commitment behaviour (β = -.19; P < .05). The result showed the impact of risk norm on risk propensity. In the fourth model, risk perception was entered to the equation to examine whether it mediated the effect of risk norm. The result of the hierarchical regression analysis reveals that risk perception along with other variable earlier entered which are demographic variables, risk propensity and risk norm, jointly and significantly contribute forty three percent (R2 = .43) variance in escalation of commitment behaviour F (7, 91) = 9.60; P < .01. The inclusion of risk perception alone resulted in one percent change in variance (∆ R2 = .01; pns) in escalation of commitment behaviour. The analysis of the independent contribution of each of the variable shows that risk perception (β = .08; pns) does not mediate the effect of risk norm.
Although risk propensity did not have an independent significantly contribution to escalation of commitment behavior in the hierarchical regression analysis (β = -0.15; Pns) but in Table 2 the bivariate correlation between risk propensity and escalation of commitment behaviour was negatively significant (r = -0.27; P<.01). The more a participant is inclined to take risk the less the participant exhibit escalation of commitment behaviour. The bivariate correlation also indicated a negative significant relationship between risk propensity and risk norm (r = -0.29; P< .01). The more a participant is inclined to take risk the less the influence of significant others or risk norm.
Discussion
The first hypothesis stated that risk propensity, risk norm and risk perception will independently and jointly predict escalation of commitment behaviour. The result revealed that risk propensity, risk norm and risk perception jointly predicted escalation of commitment behaviour. Personal responsibility and risk norm independently predicted escalation of commitment behaviour. Personal responsibility positively and significantly predicted escalation of commitment behaviour. Those who have high personal responsibility exhibit escalation of commitment behaviour more than those who have low personal responsibility. This result supported those of some other researchers such as Staw (1976) and Wong (2005).
Risk norm negatively and significantly predicted escalation of commitment behavior. The result confirms Staw 1981 findings.  Those with high risk norm exhibit lower escalation of commitment behaviour than those with low risk norm. Those with high risk norm, have influence of an organization with process-oriented approach which reward decision makers based solely upon the decision process, regardless of the outcome of the decision.  Process-oriented approach fosters risk-seeking attitudes by effectively eliminating the personal risk exposure of the decision maker. The decision maker will not need to make irrational commitment even when given a negative decision feedback for an initial decision made.

The second hypothesis stated that risk norm will partially mediate the effect of risk propensity on risk perception. The hypothesis is not confirmed.  Wong (2005) suggestion is not supported as risk norm did not partially mediate effect of risk propensity on risk perception and escalation of commitment behaviour. Result of the bivariate correlation analysis highlighted negative relationship between risk propensity and risk norm and between risk propensity and escalation of commitment behaviour. Risk propensity is an attitude and people have different attitude toward taking risk, while some people may be risk seeking some people may be avoiding risk. When participants reported high level of risk propensity, that is, the more a participant was inclined to take risk, they reported less risk norm or less influence to comply with prescribed behaviour in particular situation. Also when participants reported high level of risk propensity, the less they exhibited escalation of commitment behaviour.
Risk norm has implication for escalation of commitment. It is implied that management should integrate process-oriented approach into decision making process as prescribed appropriate behaviour.
 
References
Ajzen, I. (1991) The theory of planned behavior. Organizational behaviour and Human Decision Processes  50, 179 – 211.

Ajzen, I. (1998) Attitudes Personality and behavior. Milton keybney UK; Open University  Press.

Akinbobola, O. I. (2008). Leadership style, personality factors, decision responsibility and consequence as predictors of escalation of commitment in corporate investment decision. A Ph.D Thesis submitted to Faculty of the Social Sciences, Department of Psychology, University of Ibadan, Ibadan, Nigeria.            

Arkes, H., & Blumer, C. (1985) The psychology of sunk costs. Organisational Behaviour and Human Decision Processes, 35, 124- 149.

Armitage, C.T., & Conner, M. (2001) Efficiency of the theory of planned behaviour: A meta-analytic review. British Journal of Social Psychology 40, 471 – 499.

Baird, I.S. & Thomas, H. (1985) Toward a contingency model of strategic risk taking. Academy of Management Review 10, 230 -243.

Christian, J., Armitage, C.J., & Abrahms, D. (2007) Evidence that theory of planned variables mediate the effects of sociodemographic variables on homeless peoples’s participation in service programmes. Journal of Health Psychology 12, (5) 805-817.

Conlon, D. E., & Garland, H. (1993) The role of project completion information in resource allocation decisions. Academy of Mangement Journal, 36, 402- 413.

Conlon, E.J., & Park, J.M. (1987) Information requests in the context of escalation. Journal of Applied Psychology 72, 344-350.

Conner, M., & Abraham, C. (2001) Conscientiosness and the theory of planned behavior: Toward a more complete model of the antecedents of intentions and behavior. Personality and Social Psychology Bulletin, 27, (11), 1547-1561

Fishbein, M., & Ajzen, I. (1975) Belief, attitude, intention and behavior. Reading, MA: Addison-Wesley.

Janis, I. (1989) Crucial decisions: Leadership in Policy Making and Crisis Management. New York: Free Press.

Michie, S., & Abraham, C. (2004) Intervention to change health behaviours: Evidence- based or evidence inspired? Psychology and Health 19, 29-49.

Moon, H. (2001) The two faces of conscientiousness: Duty and achievement striving in escalation of commitment dilemma. Journal of Applied Psychology, 86. 533 -540.

Nunnally, J. C. (1978) Psychometry Theory (2nd Ed) New York: McGraw-Hill.

Ogunleye, G.A. (2001) The Causes of Bank Failures and Persistent distress in the Banking Industry. Paper presented at the public hearing on developments in the banking system. Conveyed by Nigeria House of Representatives.

Rivis, A., Sheeran, P., & Armitage, C.J. (2006) Augmenting the theory of planned behaviour with the prototype/willingness model: predictive validity of actor versus abstainer prototypes for adolescents’ health-protective and health-risk intentions. British Journal of Health Psychology 11, (3) 483-500.

Rohrmann,  B. (2002) Risk attitude scales: concepts and questionnaires. Retrieved February 18, 2009 for http://www.rohrmsannresearch.net/pdfs/rphrmann-ras-report-pdf.

Rust, J., & Golombok, S. (1995) Modern Psychometric: The Science of Psychological Assessment. New York. Routledge.

Sitkin, S.B., & Pablo A.L.  (1992) Reconceptualising the determinants of risk-taking behaviour. Academy of Management Review 17, 9 – 39.
 
Sitkin, S.B., & Weingart L.R. (1995) Determinants of risky decision – behaviour: A test of mediating role of risk perceptions and propensity. Academy of Management Journal, 38, 1537 – 1592.

Soyibo, A., & Odusola, A.F. (2002) Financial Sector Soundness: Conceptual Issues Conference Proceedings: Enhancing Financial Soundness in Nigeria, Abuja, Central Bank of Nigeria’s Second Monetary Policy Conference.

Staw, B. M. (1976).  Knee-deep in the big muddy:  A study of escalating commitment to a chosen course of action.  Organisational Behaviour and Human Performance, 16, 27 - 44.

Staw, B. M. (1981) The escalation of commitment of a course of action.  Academy of Management Review 6, 577-587.

Staw, B. M. (1997) The escalation of commitment: An update and appraisal, In. Shapira, Z. (Ed) Organisational Decision Making. (pp. 191-125) New York: Cambridge University Press.

Udegbunam, R. I. (2000) Financial Distress and performance differences among commercial bank in Nigeria. A multivariate analysis. The Quarterly Journal of Finance 15, (2).

Udegbunam, R.I. (2001) Examining the internal factors determining the disparity in loan performance across the commercial banks in Nigeria. Central Bank of Nigeria Economic Review 30 (2), 32-120.

Whyte, G. (1989) Groupthink reconsidered . Academy of Management Review, 14, 40-56

Whyte, G. (1993) Escalating Commitment in individual and group decision making: A prospect theory approach. Organisational Behaviour and Human Decision Processes 54, 430-445.

Wong, K. F. E. (2005). The role of risk in making decisions. Applied Psychology: An International Review, 54(4), 584-607.


Table 1: Hierarchical regression of escalation of commitment behaviour on risk propensity, risk norm and risk perception

 

Variables

Model I
Beta                        

Model II
Beta                

Model III
Beta             

Model III
Beta             

Step 1:
Control Variables

 

 

 

 

 

 

Personal Responsibility
Work experience
Sex
Age

0.59**
-0.17
0.01
0.10

 

0.56**
-0.16
-0.01
0.08

 

0.56**
-0.19
-0.03
0.08

0.55**
-0.18
-0.01
0.95

Step 2:

 

 

 

 

 

 

Risk propensity

 

 

-0.15

 

-0.19*

-0.21*

Step 3:

 

 

 

 

 

 

Risk norm

 

 

 

 

-0.17*

-0.16*

Step 4:
Risk perception

 

 

 

 

 

 

0.08

 

F
R
R (square)
Change in R (Square)
Adjusted R (Square)
Df
SE

 

13.85**
0.61
0.37
0.37**
0.34
4/94
23.61

 

 

11.98**
0.63
0.39
0.02
0.36
5/93
23.34

 

 

11.03**
0.65
0.42
0.03*
0.38
6/92
22.95

 

9.60**
0.65
0.43
0.01
0.38
7/91
22.94

**P < .01         *P< .05
Table 2: Zero order correlation matrix showing relationship between independent variables and the dependent variable

 

1

2

3

Mean

Standard deviation

1. Risk propensity

 

 

 

11.89

2.66

2. Risk norm

-0.29**

 

 

22.67

3.02

3. Risk perception

0.13

-0.12

 

20.61

6.12

4. Escalation of commitment behaviour

-0.27**

-0.11

0.14

60.59

29.51

                                    **P < .01         *P< .05