Paper

When Do Opportunity Costs Count? The Impact of Vagueness, Project Completion

Stage, and Management Accounting Experience

Lisa Marie Victoravich University of Denver

ABSTRACT: Management accountants have recently migrated toward a business part- ner role, and as a result they often assist management with the decision-making pro- cess. Thus, it is imperative that they excel at identification of relevant information such as opportunity costs. This study experimentally tests the prediction that management accounting experience mitigates the tendency to ignore opportunity costs with respect to two factors: opportunity cost vagueness and project completion stage. This study also investigates whether attending to opportunity costs has an impact on project con- tinuance decisions. Results indicate that management accounting experience mitigates the effect of vague opportunity costs and project completion stage. It was also found that attention to opportunity costs acts as mediator and this in turn reduces the ten- dency to continue an existing project. This suggests that attending to opportunity costs influences decision-making and that it is likely to have an economic consequence.

Keywords: opportunity costs; vagueness; management accounting experience; project completion stage.

Data Availability: Contact the author.

INTRODUCTION

Opportunity costs are incurred whenever a decision-maker must choose between two or more courses of action; however, they are commonly overlooked by decision-makers. The opportunity cost concept is a fundamental component of classical economic theory and is

measured as the benefit forgone due to choosing an alternative course of action �Heymann and Bloom 1990�. In this study, I show that management accounting experience mitigates this dys- functional tendency to overlook opportunity costs as documented by prior studies �Becker et al. 1974; Neumann and Friedman 1978; Friedman and Neumann 1980; Hoskin 1983�. This suggests

I appreciate the helpful comments of Bud Fennema, Greg Gerard, Doug Stevens, Neil Charness, Bill Buslepp, participants at the Florida State University and University of Denver research workshops, and participants of the 2008 Midyear Meeting of the Management Accounting Section of the American Accounting Association. I also gratefully acknowledge members of the Institute of Management Accountants �IMA� for their willingness to participate in my study.

BEHAVIORAL RESEARCH IN ACCOUNTING American Accounting Association Vol. 22, No. 1 DOI: 10.2308/bria.2010.22.1.85 2010 pp. 85–108

Published Online: January 2010

85

that the tendency to ignore opportunity costs is not a permanent cognitive deficiency but rather something that can be unlearned. My study tests the strength of management accounting experi- ence in mitigating this tendency by introducing factors into the decision-making setting that should negate attention to opportunity costs. Finally, I show that attention to opportunity costs mediates the relationship between project continuance and the factors of interest and thus is consequential to financial decisions.

The factors examined in this study include opportunity cost vagueness and project completion stage. Prior research has typically presented opportunity costs in a precise manner, which is inconsistent with how they would appear in a realistic setting. Since opportunity costs are future- oriented, they are likely to be uncertain and therefore will be represented as an estimate. For example, a fixed asset might have a disposal value of an estimated amount plus or minus a range. It is also likely that decision-makers will consider whether additional resources should be allo- cated to projects at various stages of completion. For instance, a project that is under consideration for future resources may be at an initial stage of completion �e.g., 10 percent complete� or close to completion �e.g., 90 percent complete�.

Opportunity cost vagueness will likely reduce attention to opportunity costs, since decision- makers have been shown to avoid or discount vague situations �Camerer and Weber 1992� and information �Van Dijk and Zeelenberg 2003�, especially when feeling incompetent or unknowl- edgeable in a particular area �Heath and Tversky 1991; Goodie 2003�. Prior research has also shown that decision-makers have a natural tendency to finish what they started, especially when it is near to completion �Conlon and Garland 1993; Boehne and Paese 2000�. This bias toward project completion will likely reduce attention to opportunity costs when identifying relevant information for decisions regarding continuance of an in-progress project. It is important to attend to opportunity costs since failing to integrate them as relevant information will likely bias deci- sions toward continuing projects that are unprofitable.

The ability of management accounting experience to mitigate the effect of these factors on the tendency to attend to opportunity costs was tested by investigating decision-makers’ performance at a task highly relevant to the managerial accounting profession. This task consisted of a search for relevant information for use in a resource allocation analysis. Management accounting expe- rience and the task of searching for relevant information is of interest to accounting research, since a primary responsibility of management accountants is to provide management with sound advice about costs and benefits which affect business plans. Furthermore, over time management accoun- tants have transformed from serving the accounting profession as “bean counters” to “business partners,” and as a result they are spending more time involved with strategic and operational decision-making �Siegel 1999, 2000; Siegel et al. 2003�.

Despite having access to highly sophisticated accounting systems, some of the most important information is not reported by these accounting systems. Given that management accountants are gathering, analyzing, and interpreting information for use in organizational decision-making, it is important that they excel at these tasks. Thus, it is imperative that they develop the necessary knowledge that enables them to identify relevant information such as opportunity costs even in the presence of situational factors that may negate attention to this information.

This study also investigates the benefits of attending to opportunity costs for improving the overall quality of judgment and decision-making. I am particularly interested in whether the likelihood of allocating additional resources to an in-progress project is influenced by incorporat- ing opportunity costs as relevant information for use in a decision analysis. I investigate this cognitive process by examining attention to opportunity costs as a mediating variable between the independent variables of interest �opportunity cost vagueness and project completion stage� and decision-makers’ propensity to continue a project.

86 Victoravich

Behavioral Research In Accounting Volume 22, Number 1, 2010 American Accounting Association

Path analysis was used to investigate the causal relationship between the independent vari- ables �opportunity cost vagueness, project completion stage�, the moderating variable �manage- ment accounting experience in years�, the mediating variable �number of opportunity costs at- tended to�, and the dependent variable �project continuance�. A case-based scenario that describes the current progress and resources necessary to complete an internal logistics project was em- ployed. Participants were asked to perform an analysis of the project and decide whether it should be continued.

The first independent variable, opportunity cost vagueness, was operationalized by presenting the six embedded opportunity costs in the case as either precise or vague. Project completion stage was operationalized by varying whether the project was at an early stage of completion �10 percent complete� or late stage of completion �90 percent complete�. The moderating variable, management accounting experience, was measured as years of management accounting experience reported by participants. Data were collected from two groups of participants who had varying amounts of management accounting experience. These groups included members of the Institute of Management Accountants �IMA� and upper-level accounting major students at a large state university.

The remainder of this paper is organized as follows. In the next section, I discuss relevant literature and develop the hypotheses. This is followed by a discussion of the experimental design and the method used in testing the hypotheses. The final two sections discuss the study’s results and provide related conclusions and limitations.

THEORY AND HYPOTHESES Domain Experience, Knowledge, and Performance Relationship

Domain experience is the result of participation in events or tasks �e.g., capital project analy- sis� associated with a particular discipline �e.g., management accounting�, which facilitates the creation of knowledge stored in memory �Libby 1995; Vera-Muñoz et al. 2001�. Prior research has indicated that this knowledge gained from experience improves performance on judgment and decision-making tasks �Bonner et al. 1997; Nelson 1993; Nelson et al. 1995�. This is consistent with the Libby �1995� antecedents and consequences of knowledge model, which predicts that experience creates knowledge which in turn drives performance.

Instruction generates declarative knowledge and practice generates procedural knowledge �Anderson 1983�. Declarative knowledge is knowledge of facts �e.g., relevant information in- cludes outlay costs and opportunity costs�. Thus, students in a managerial or cost accounting course will have gained the basic declarative knowledge of what is considered relevant or irrel- evant information. Procedural knowledge is the use of declarative knowledge, which underlies the ability to perform a task �e.g., how to quantify, identify, and search for relevant information for use in decision-making�. Experienced management accountants such as a controller will have gained the procedural knowledge necessary to identify relevant information for use in decision-making tasks such as capital budgeting or forecasting. Thus, the key to excelling at a particular domain task is procedural knowledge developed through experience in a specialized domain �Anderson 2000; Herz and Schultz 1999�.

Findings from prior research indicate that knowledge gained from management accounting experience enables superior performance at identification of opportunity costs as relevant infor- mation for use in decision-making. Vera-Muñoz et al. �2001� found that individuals with high levels of management accounting experience identified more relevant information in the form of opportunity costs only when they chose a cash flow problem representation. However, prior research does not provide evidence as to whether management accounting experience and related knowledge mitigates the effect of factors that may hinder performance in terms of identifying

When Do Opportunity Costs Count? 87

Behavioral Research In Accounting Volume 22, Number 1, 2010 American Accounting Association

relevant information for use in decision-making. The current study investigates whether manage- ment accounting experience overcomes the effect of opportunity cost vagueness and project completion stage in relation to identifying opportunity costs as relevant information for use in resource allocation judgment and decision-making. Further, it investigates whether this increased attention to opportunity costs has a decisional impact in terms of influencing the likelihood of continuing an in-progress project.

Explicit versus Implicit Opportunity Costs Although opportunity costs are not included in the calculation of accounting profits, oppor-

tunity costs are an intrinsic component of economic profits. Early research by Becker et al. �1974� found that despite the economic significance of opportunity costs, decision-makers placed more weight on outlay costs as a decision input than on opportunity costs. Subsequent studies found that individuals attended to a single opportunity cost in a decision analysis only if it was presented in an explicit manner or spelled out to them as an opportunity cost rather than in an implicit manner �Becker et al. 1974; Neumann and Friedman 1978; Friedman and Neumann 1980; Hoskin 1983; Northcraft and Neale 1986�.

The key difference between an explicit and implicit opportunity cost is that the former is presented in a more salient manner. For example, explicit opportunity costs in Becker et al. �1974�, Neumann and Friedman �1978�, and Friedman and Neumann �1980� were labeled “Forgone Profit” and the actual amount was provided alongside the outlay cost information given to par- ticipants �e.g., variable overhead $161,648, forgone profit $69,548�. Implicit opportunity costs were not specifically labeled as a forgone profit, but rather the actual amount was provided in a footnote below the outlay cost information �e.g., obsolete materials can be sold on the market, if not used in the project for $69,548�. Hoskin �1983� and Northcraft and Neale �1986� used a similar presentation such that explicit opportunity costs were clearly labeled and more salient while implicit opportunity costs were not clearly labeled and more peripheral.

Opportunity Cost Vagueness Early studies which investigated the impact of implicit versus explicit opportunity costs were

successful at identifying whether individuals who lack management accounting experience attend to opportunity costs if they were blatantly labeled as an opportunity cost. However, the studies failed to incorporate a key characteristic of opportunity costs in the real world, the lack of precision associated with a future dollar amount due to uncertainty. From a decisional standpoint an opportunity cost is often measured by looking ahead and is likely to be uncertain in amount. For example, the cost savings that are forgone if a piece of equipment used in production is not replaced is an estimated amount plus or minus an estimated range. The actual costing savings are not determinable until it is installed and are likely to depend on a variety of factors such as ease of use, employee productivity, and production volume.

Although there is a lack of consensus with respect to a single operational definition of vagueness,1 commonly cited definitions refer to vagueness as the result of missing information or lack of knowledge, which leads to uncertainty regarding a particular outcome �see Einhorn and

1 Although prior judgment and decision literature used vagueness and ambiguity interchangeably, in theory they are definitively different. The two can be distinguished by using the argument made in Budescu et al. �1988�. Ambiguity means that a statement or event can be interpreted in two or more different yet precise ways. An ambiguous statement might consist of asking an individual to select a “light” ball from an urn of black cork-filled balls and white lead-filled balls in which the probability of selecting a light ball is 80 percent. A vague statement or amount is one that is not clearly identified or cannot be understood precisely or exactly. The word vague is synonymous with imprecise, inaccu- rate, inexact, etc. Vagueness is the result of not knowing the probability distribution of an outcome �e.g., drawing a ball

88 Victoravich

Behavioral Research In Accounting Volume 22, Number 1, 2010 American Accounting Association

Hogarth 1985; Frisch and Baron 1988; Curley and Yates 1985�. Vagueness can be a common pitfall of real world situations because decision-makers do not have access to exact information about the costs and benefits of all available options. Instead, decision-makers must rely on inexact information. Prior research has not investigated whether opportunity cost vagueness reduces decision-makers’ tendency to attend to this relevant information.

Individuals have been shown to react to vagueness by avoiding vague situations. Ellsberg �1961� established the well-known Ellsberg Paradox by demonstrating that given the choice be- tween a vague and precise gamble, most individuals prefer the precise option. This vagueness aversion has been found to persist in a variety of contexts �see Camerer and Weber 1992�. Various theoretical explanations regarding aversion to vagueness exist including self-evaluation �Hamm and Bursztajn 1979; Roberts 1963�, evaluation by others �Ellsberg 1963; Gardenfors 1979; Toda and Shuford 1965�, and uncertainty avoidance �Curley et al. 1986�. These explanations do not make any predictions regarding the effect of decision-maker characteristics �e.g., experience� on the tendency to avoid vagueness.

Of interest to my study is the competence hypothesis which uses a decision-maker attribute, namely the feeling of competence, to explain aversion to vagueness. The competence hypothesis states that the willingness to bet on an uncertain event is a function of both the precision of a likelihood estimate and one’s general skill, knowledge, or understanding of the relevant context. Heath and Tversky �1991� found that in their area of competence or knowledge, individuals showed a preference to bet on their vague beliefs versus a lottery with a probability equal to their stated confidence. In a more recent study, Goodie �2003� found that the willingness to bet versus not bet at all sharply increased when the bet was related to participants’ area of competence or knowledge.

Based on these findings, it is likely that aversion to vague information is dependent on individuals’ experience and related knowledge which will influence the feeling of competence in a particular decision-making setting. Specifically, an insignificant amount of prior domain-specific experience and absence of a related knowledge base are likely to lead to the feeling of incompe- tence and in turn heighten the presence of vagueness. Inexperienced participants in a particular domain typically focus on the more superficial aspects of a problem �Chi et al. 1981; Gagne et al. 1993; Van De Wiel et al. 2000� instead of focusing on the relevant information that is necessary to solve the problem. For example, when solving a problem, novice physicists who lacked procedural knowledge focused on unimportant details �Wenk et al. 1997�. In a resource allocation decision, inexperienced individuals might focus on the vagueness of the available alternatives and will overlook the underlying importance of the alternatives to the decision being made. In fact, Van Dijk and Zeelenberg �2003� found that student decision-makers discounted vague information by treating it as insufficient or nonexistent when making resource allocation decisions.

Management Accounting Experience Friedman and Neumann �1980� was the first study to investigate if decision-maker experience

influences integration of opportunity costs in decision-making. This study found that both masters of business administration �M.B.A.� students and Certified Public Accountants �CPAs� showed a tendency to rely on a single opportunity cost if it was available at zero cost, meaning that they did not have to pay to obtain the information. If no information concerning an opportunity cost was

from an urn with an unidentified composition�, or when the outcome of an event is not known �e.g., may win between $200 and $400 if a green ball is drawn�. The former refers to probabilistic vagueness and the later to outcome vagueness.

When Do Opportunity Costs Count? 89

Behavioral Research In Accounting Volume 22, Number 1, 2010 American Accounting Association

present, participants did not consider the possibility of a forgone prospect and try to estimate an amount. No demographic information was given regarding work experience of the participants in the study.

Vera-Muñoz �1998� built on Friedman and Neumann �1980� and tested whether decision- maker knowledge and context �personal versus business� were influential on the inclusion of opportunity costs in decision-making. Consistent with expectations, she found that individuals with high levels of general accounting knowledge, represented by masters of accounting students, focused on historical costs since this is the primary focus of Generally Accepted Accounting Principles �GAAP�. This GAAP knowledge base interfered with individuals’ ability to attend to opportunity costs in a business context. Thus, individuals with high GAAP knowledge actually attended to fewer opportunity costs than individuals with low GAAP knowledge represented by M.B.A. students.

Extending Vera-Muñoz �1998�, Vera-Muñoz et al. �2001� investigated the role of decision- maker experience and choice of analysis format �cash flow versus earnings� on the tendency to integrate opportunity costs. She found that participants with more management accounting expe- rience integrated more opportunity costs only if they chose a cash flow-based format rather than choosing an earnings-based format. Notably, public accounting experience did not facilitate inte- gration of opportunity costs. These results were attributed to management accountants’ more extensive domain-specific experience, which often consists of engaging in forward-looking decision-making tasks such as financial planning, budgeting, and forecasting. This domain expe- rience led to the development of procedural knowledge, which facilitated the choice of an appro- priate problem representation and attention to opportunity costs.

Lending further credence that management accounting experience and related procedural knowledge will enable identification of opportunity costs, Bedard and Chi �1993� found that experienced individuals are better able to differentiate between information that is relevant and irrelevant. This is because experienced individuals exhibit a top-down approach to information acquisition using rules of thumb and structured mental checklists. This relevant information might include future-oriented outlay costs, cash inflows, expected returns, and opportunity costs. Con- sistent with these expectations, analysts performing a financial analysis task in Biggs �1984� demonstrated a highly structured search for information. Davis �1996� found that particular situ- ational experience improved auditors’ ability to select relevant information for making efficient, appropriate, control risk assessments.

Given little or no experience, decision-makers will ignore vague opportunity costs due to a propensity to focus on the vagueness of information. Management accounting experience is ex- pected to mitigate the effect of opportunity cost vagueness. First, management accounting expe- rience will enable development of a knowledge structure that facilitates identification of opportu- nity costs as relevant information regardless if the opportunity costs are precise or vague. Second, this experience and related knowledge will overcome the feeling of incompetence that is present when individuals have little or no experience, which in turn will decrease the effect of opportunity cost vagueness. This suggests that decision-makers will attend to fewer opportunity costs when presented in a vague manner, and this effect is mitigated by management accounting experience.

H1a: Decision-makers will attend to fewer opportunity costs when they are presented in a vague rather than precise manner.

H1b: Management accounting experience decreases the effect of vagueness on attention to opportunity costs.

Project Completion Stage Escalation of commitment is defined as a situation where a decision-maker commits addi-

tional resources to a losing course of action because of prior investment �Staw 1976; Staw and Fox

90 Victoravich

Behavioral Research In Accounting Volume 22, Number 1, 2010 American Accounting Association

1977; Staw and Ross 1978�. A common explanation for escalation of commitment is the comple- tion effect, which predicts that decision-makers continue to invest in a failing project as they draw close to project completion. Several studies �Conlon and Garland 1993; Boehne and Paese 2000; Moon 2001� have supported this contention by finding that decision-makers are more likely to continue a project that is nearly complete �e.g., 90 percent complete� versus far from complete �e.g., 10 percent complete�.

Based on the aforementioned studies it is clear that project completion stage influences the decision to continue or discontinue a project, yet it is unknown whether project completion stage influences the search for information to be used in the decision-making process. Specifically, my study investigates how project completion stage influences the tendency to attend to relevant opportunity cost information. The tendency to ignore relevant opportunity cost information may be useful in explaining resource allocation judgment and decision-making �e.g., how it attributes to escalation of commitment�. When evaluating an in-progress project, the stage of project completion is an irrelevant cue and all costs and benefits associated with continuing and discon- tinuing the project should be considered regardless of proximity to completion.

There is considerable evidence that irrelevant cues inappropriately influence the decision- making of both inexperienced and experienced individuals. A literature review by Gaeth and Shanteau �1984� identified several studies documenting the effects of irrelevant information on judgments. For example, irrelevant information influenced evaluations by school administrators �Rice 1975�, soil classification judges �Gaeth and Shanteau 1984�, and human resource officers �Nagy 1981; Haefner 1977�. Findings from these studies suggest that decision-makers may have difficultly ignoring information which is irrelevant for the task at hand. Findings from research on financial decision-making have also found that inexperienced decision-makers tend to search for confirmatory information �Anderson 1988; Bouwman et al. 1987�. To the contrary, experienced financial professionals tend to acquire information by searching for contradictory information �Anderson 1988; Bouwman et al. 1987�.

Management Accounting Experience Findings from prior research indicate that even experienced decision-makers are influenced by

irrelevant cues; domain-specific management accounting experience should mitigate the effect of a nearly complete project on the tendency to ignore opportunity costs. As previously stated, management accountants are often required to perform analyses �e.g., capital project analyses� which require identification of relevant cash inflows and outflows such as increases in revenue, maintenance costs, and the salvage value of project-specific assets �Roehl-Anderson and Bragg 2005�. Practice performing these tasks will likely lead to development of procedural knowledge and facilitate identification of opportunity costs as relevant information.

Not only will prior practice and development of a related knowledge base facilitate identifi- cation of opportunity costs as relevant information in general, it will lead to problem-solving strategies which successfully identify opportunity costs regardless of the irrelevant cue of project completion stage. First, consistent with finance professionals in Biggs �1984�, experienced decision-makers are likely to exhibit a well structured manner of identifying relevant information. Second, this structured search for information is likely to include use of a mental checklist �Bedard and Chi 1993; Bouwman et al. 1987�. Third, when faced with a problem the reasoning process of experienced decision-makers tends to be automated and thus they are less likely to get hung up on surface features of a problem �Shiffrin and Schneider 1977; Shanteau 1992�. Overall, a structured search for information, use of a mental checklist, and automated reasoning are tools developed through experience which will enable decision-makers to identify opportunity costs regardless of a project’s proximity to completion.

When Do Opportunity Costs Count? 91

Behavioral Research In Accounting Volume 22, Number 1, 2010 American Accounting Association

Based on the preceding discussion, the presence of a nearly complete project is likely to decrease attention to opportunity costs in the face of an insignificant amount of management accounting experience. This is because decision-makers who lack experience are likely to get hung up on surface features such as a project’s nearness to completion and tend to focus on confirming rather than disconfirming information. The experience gained in the domain of management ac- counting is likely to mitigate the effect of project completion stage on decision-makers’ tendency to attend to opportunity costs. This suggests decision-makers will attend to fewer opportunity costs when project completion stage is high and this effect is mitigated by management accounting experience.

H2a: Decision-makers will attend to fewer opportunity costs when project completion stage is high rather than low.

H2b: Management accounting experience decreases the effect of project completion stage on attention to opportunity costs.

Attention to Opportunity Costs and Project Continuance Identification of opportunity costs as relevant information is important because inclusion in an

analysis is likely to influence judgments and decisions. Specifically, it is important to attend to opportunity costs because failing to include them as relevant information is likely to bias in favor of continuing projects that are unprofitable. Prior research has yet to investigate the cognitive process by which attention to opportunity costs influences judgment and decision-making and whether it has a decisional consequence. If attending to opportunity costs does not impact judg- ment and decision-making, then the argument regarding the significance of attending to opportu- nity costs and including them in a decision analysis has little or no merit.

It is a logical argument that attention to opportunity costs will influence judgment and decision-making since it should bring other available prospects to the attention of the decision- maker. Underlying this argument is the key importance of the opportunity cost concept such that it forces decision-makers to consider all alternatives, and it is only by assessing opportunity costs that a decision-maker is able to determine the true cost of any course of action �Ezzamel and Hart 1987�.

By attending to opportunity costs, decision-makers will become aware of alternative courses of action that will boost profitability and the tendency to continue an ongoing project will de- crease. As the number of opportunity costs identified in a decision analysis increases, the associ- ated cost of not taking the alternative course of action will increase. Individuals who do not attend to opportunity costs will not be receptive to the cost associated with forgoing the alternative course of action and in turn will remain focused on the ongoing project and favor continuance.

H3: Attention to opportunity costs decreases the likelihood of continuing a project.

Both opportunity cost vagueness and project completion stage are predicted to influence attention to opportunity costs. Due to bringing decision-makers’ attention to alternative prospects, attention to opportunity costs is predicted to reduce the tendency to continue a project. This suggests that attention to opportunity costs mediates the causal sequence between the independent variables, opportunity cost vagueness and project completion stage, and the dependent variable, likelihood of continuing a project �Robins and Greenland 1992�.

Based on the causal model, attention to opportunity costs is expected to mediate opportunity cost vagueness and project continuance, and attention to opportunity costs is expected to mediate project completion stage and project continuance �See Figure 1�.

92 Victoravich

Behavioral Research In Accounting Volume 22, Number 1, 2010 American Accounting Association

EXPERIMENT AND METHODOLOGY Participants and Administration of Experiment

To examine how management accounting experience influences attention to opportunity costs, two groups of participants were used in the study’s experiment. The first group consists of pro- fessionals who were recruited based on their membership in the Institute of Management Accoun- tants �IMA�. The domain of management accounting is unlike the auditing domain because it is not possible to identify a task that individuals at a particular rank regularly perform �e.g., an audit senior performing an internal control evaluation�. Due to this characteristic of the management accounting domain, members of the IMA were identified as a group of professional participants who are expected to have practice performing forward-looking resource allocation-based tasks such as forecasting or capital budgeting.

As summarized in Table 1, professional participants reported having an average of 3.7 years of public accounting experience, 8.6 years of management accounting experience, and 7.6 years of general resource allocation experience. Table 1 also reports experience in years with respect to specific tasks that are commonly performed by management accountants. Fifty �51 percent� of the experienced participants were CPAs and 63 �64 percent� were Certified Management Accountants �CMAs�.

One hundred four IMA members attending IMA-sponsored events participated in the study. Five participants’ responses were not usable as they were incomplete. Eighty-six usable responses were obtained via onsite administration by the researcher at one regional conference, two monthly chapter meetings, and one continuing professional education session. Participants were not given any time constraints. The remaining responses were not obtained via onsite administration because there was not enough time available at a monthly IMA chapter meeting to formally conduct the

FIGURE 1 Theoretical Model

Opportunity Cost Vagueness (X1)

Project Completion Stage (X2)

(+) a

Number of Opportunity Costs (M)

Project Continuance (Y)

H1a (-)

H2a (-)

Management Accounting Experience (W)

H1b (+)

H2b (+)

H3 (-)

a This path was not part of the study’s hypotheses; however, it was established in prior research (see Vera- Muñoz et al. 2001). Its inclusion ensures the completion of the model and the associated causal sequence.

When Do Opportunity Costs Count? 93

Behavioral Research In Accounting Volume 22, Number 1, 2010 American Accounting Association

experiment.2 Thirty packets were handed out by the chapter’s president containing the experimen- tal instrument and envelopes addressed to the researcher. Of the 30 packets, 13 usable responses were received for a response rate of 43 percent. On average, this group completed the case in 44 minutes.

The second group of participants consists of 116 students recruited from a large state univer- sity who were expected to have little or no management accounting experience. As expected, these participants reported minimal amounts of professional experience �see Table 1 for detailed statis- tics�. This group included upper-level accounting major students enrolled in a cost accounting course that had completed several economics and accounting courses. This coursework was ex- pected to provide them with the declarative knowledge necessary to perform the experimental task. The case was administered during regular class time and they were given extra credit for participation. This group was also not given any time constraints and on average completed the case in 57 minutes.

Case Materials The study’s experiment consisted of a single case-based scenario in which a regional grocery

store �Fresh Foods, Inc.� was developing an internal logistics project. Participants were asked to assume the role of the company’s recently appointed Internal Investment Project Supervisor and perform a cash flow analysis to determine whether the logistics project should be continued or discontinued. The recently appointed verbiage was used to avoid creation of a sponsorship bias toward the logistics project �see Chenhall and Morris 1991�. Sample case materials are provided in the Appendix. As described in Table 2, the experiment had two phases: an experimental phase �Phase I� and a post-experimental phase �Phase II�.

2 There was no significant difference between the dependent variables of the participants who completed the task onsite versus on their own time �number of opportunity costs attended to, t � 0.65, p � 0.72, two-tailed; continuance judgment, t � 0.46, p � 0.65, two-tailed; continuance decision, t � 0.41, p � 0.72, two-tailed�.

TABLE 1

Participant Experience (in years)

Professional (n � 99)

Non-Professional (n � 116)

Mean (SD) Range

Mean (SD) Range

Public Accounting 3.7 �5.0�

�0–27.0� 0.02 �0.07�

�0–1.0�

Management Accounting 8.6 �8.1�

�0–31.0� 0.01 �0.02�

�0–2.0�

Resource Allocation 7.6 �6.8�

�0–31.0� None Reported

None Reported

Capital Budgeting 6.3 �6.5�

�0–30.0� None Reported

None Reported

Analysis of Capital Projects 4.0 �5.6�

�0–29.5� None Reported

None Reported

Project Management 5.3 �6.4�

�0–2.0� 0.01 �0.14�

�0–1.5�

94 Victoravich

Behavioral Research In Accounting Volume 22, Number 1, 2010 American Accounting Association

Independent Variables There are two independent variables, opportunity cost vagueness and project completion

stage, which were manipulated between subjects in all cells. Vagueness of opportunity costs was presented at two levels �precise and vague� and included both probabilistic and outcome vagueness.3 Probabilistically vague opportunity costs were presented as having a likelihood of occurrence �e.g., there is an 85 percent to 90 percent chance that fixed assets will be resold for $170,000�. Outcome vague opportunity costs were presented as an interval �e.g., annual return on marketable securities will be 8 percent � 4 percent�. Both types of vagueness were included for completeness purposes as prior research has identified them as equally important �Ho et al. 2001, 2002; Kuhn and Budescu 1996�.

Project completion stage was manipulated at two levels using a percentage �low as 10 percent complete and high as 90 percent complete�. This is consistent with previous completion effect studies �see Conlon and Garland 1993; Garland and Conlon 1998; Boehne and Paese 2000�.

Moderating Variable Since management accounting experience is expected to reduce the effect of opportunity cost

vagueness and project completion stage, it is included in the causal model as a moderating

3 Opportunity costs that exhibited outcome vagueness and probabilistic vagueness were attended to by 62 percent and 59.5 percent of the study’s participants, respectively. The difference was not statistically significant �t � 0.81, p � 0.42, two-tailed�, suggesting that there was no difference in the tendency to attend to opportunity costs that are probabilistic or outcome vague. This is consistent with related research that has found individuals respond similarly to both types of vagueness �Ho et al. 2001, 2002; Kuhn and Budescu 1996�.

TABLE 2

Overview of Experimental Phases

Phase I: Experimental Item Description

Instructions �pp. 1–2� Details about assumed role, task, company, and table of contents.

Background Information �p. 3� Description of company, logistics project, and resource allocation request.

Internal Memorandum �p. 4� Request from company’s CEO to make a continuance decision. Summary Report �pp. 5–6� Financial and nonfinancial information about project funding,

current progress, additional outlay, third party logistics costs, and in-house logistics costs.

Participant Responses �pp. 7–8� Asked to perform cash flow analysis, continuation judgment and decision, and summary of any nonfinancial factors considered.

Phase II: Post-Experimental Item Description

Manipulation Check �p. 1� Indicate project completion condition. Background Questionnaire �pp. 1–2� Demographic, academic, work experience, difficulty, and

realism questions. Multiple Choice Questions �pp. 3–8� Opportunity cost knowledge and analytical ability questions.a

a Questions were used to obtain a measure of participants’ analytical ability and knowledge of the opportunity cost concept for use as control variables in the causal model.

When Do Opportunity Costs Count? 95

Behavioral Research In Accounting Volume 22, Number 1, 2010 American Accounting Association

variable. Management accounting experience is operationalized using years of experience reported by both participant groups, students and IMA members, who reported varying amounts of man- agement accounting experience.

Mediating Variable Number of opportunity costs attended to in participants’ cash flow analyses is expected to act

as a mediator between the independent variables �opportunity cost vagueness and project comple- tion stage� and the dependent variable �project continuance�. The number of opportunity costs varies from zero to six. The six opportunity costs which were incurred if the logistics project was continued include: �1� profit from the initiation of an internal product line, �2� return on market- able securities, �3� disposal proceeds from semi-trucks, �4� sublease revenue on warehouse, �5� staffing revenue from manager, and �6� cost of goods sold savings due to implementation of a Just-in-Time inventory system.

Participants were given two separate sheets of paper labeled “continue” and “discontinue” with a blank line at the bottom of each page labeled total cash outflow �inflow�. This was provided in order to facilitate the organization of their cash flow analyses. An opportunity cost was consid- ered “attended to” if it was included in participants’ calculations for the analysis regarding whether the logistics division should be continued or discontinued. For example, if the resale value of the trucks appeared as an item in the total cash flows associated with continuing the project, it was counted as one opportunity cost.

The number of opportunity costs attended to in participants’ analyses was identified by two independent coders who were blind to the experimental conditions. Coders were given both pieces of paper for each participant which were labeled “continue” and “discontinue” as discussed above. As well, they were compensated and walked through a practice case to ensure they were familiar with the mechanics of their assigned task. The coders agreed 94.5 percent of the time and inter- rater reliability was assessed with the Kappa Coefficient �Kappa � 0.927, p � 0.01�. Discrepan- cies were resolved via discussion between the two coders and were the result of a coder overlook- ing an opportunity cost and thus not including it in their count for a participant.

Dependent Variable The dependent variable is participant’s likelihood judgment concerning continuance of the

project.4 After the participants performed their analyses regarding continuance or discontinuance of the logistics division project, they were asked to make a judgment and decision about continu- ing the project. For the project continuance judgment, participants were asked to place an X on the number of an 11-point scale that best represents the likelihood they would continue the logistics project with endpoints labeled extremely unlikely and extremely likely. As well, participants were asked to mark an X next to their favored course of action in terms of continuing or discontinuing the project.

Control Variables Data were also collected on five control variables that could affect participants’ performance

on the study’s experimental task. The first four control variables—analytical ability, knowledge of the opportunity cost concept, perceived realism of the task, and perceived difficulty of the task— were included based on arguments made in prior studies that the variables may affect attention to opportunity costs �see Vera-Muñoz 1998; Vera-Muñoz et al. 2001�. The first control was analytical

4 The causal model was also run with project continuance measured as a dichotomous variable with continue � 1 and discontinue � 0. The results of the model were similar to those with project continuance measured on an 11-point scale.

96 Victoravich

Behavioral Research In Accounting Volume 22, Number 1, 2010 American Accounting Association

ability and was measured using eight questions from prior Graduate Record Exams �GRE�. The second control, knowledge of the opportunity cost concept, was also measured using eight multiple-choice questions which were designed to assess participants’ knowledge of the concept. The third and fourth controls are participants’ perceived difficulty and realism of the task, which were measured on 11-point Likert scales.

The fifth control, participants’ age, was included to control differences among the participants due to maturity, social class �money made�, and the number of decisions made inside and outside of the workplace, and how these differences may affect attention to opportunity costs and project continuance. Age was used since each of these factors in general is likely to vary directly with age.

RESULTS Manipulation Check

To ensure participants were aware of the percent complete manipulation, they were asked to indicate whether the logistics project was 10 percent or 90 percent complete in the post- experimental questionnaire. All of the participants correctly identified their experimental condition suggesting that the project completion manipulation was successful. A manipulation check was not included for opportunity cost vagueness because of an inability to elicit participants’ awareness of the vague presentation of the six opportunity costs using a manipulation check-related question. Furthermore, as hypothesized, inexperienced participants were expected to ignore vague opportu- nity costs due to the imprecision of the information. Thus, if asked they would not be aware if they attended to the vagueness of the information.

Descriptive Statistics

Mediating Variable Descriptive statistics regarding the mediating variable, number of opportunity costs attended

to by experimental condition, are reported in Table 3. Participants are classified by range of experience in years and for participants as a whole. The mean number of opportunity costs attended to by participants was 3.87 �� � 2.20�.

Dependent Variable Descriptive statistics for the dependent variable, project continuance judgments, are reported

in Table 4 for participants classified by range of experience in years and for participants as a whole. The mean project continuance judgment is 4.86 out of 11 �� � 2.16�.

Statistical Model The study’s model is estimated with structural equation modeling �SEM� using maximum

likelihood estimation via AMOS software. This method is used in addition to the Sobel Test �Sobel 1982� and as an alternative to the traditional Baron and Kenny �1986� method via regression analysis because it is more superior in terms of controlling measurement error �Holmbeck 1997; Hoyle and Kenny 1999; Kline 1998�. As well, it enables an alternative way to explore the medi- ated effect in terms of bootstrapping �Preacher and Hayes 2004; Kline 1998; Hoyle and Kenny 1999�. Bootstrapping is a nonparametric method that can be used to test hypotheses. It does not assume normality �Efron and Tibshirani 1993; Mooney and Duval 1993� and allows for a more powerful test �Preacher and Hayes 2004�.

Model Fit Model fit is assessed for the tested model using common fit indices including the Chi-square

��2� test statistic, Comparative Fit Index �CFI�, and the Root Mean Square Error of Approximation

When Do Opportunity Costs Count? 97

Behavioral Research In Accounting Volume 22, Number 1, 2010 American Accounting Association

�RMSEA�. The model had a good fit to the data ��2 = 0.90, p � 0.64; CMINdf � 0.41; CFI � 0.98; RMSEA � 0.02�. A conclusion of good fit is derived on the basis that the �2 statistic is insignificant �Joreskog 1969�, the CMINdf ratio is less than 5 �Wheaton et al. 1977�, the CFI is greater than 0.95 �Hu and Bentler 1999�, and the RMSEA is less than 0.06 �Hu and Bentler 1999; Kaplan 2000�. Results are presented graphically in Figure 2.

TABLE 3

Descriptive Statistics: Mediating Variable Mean Number of Opportunity Costs Attended Toa

Panel A: Range of Experience (Years) and Opportunity Cost Vagueness Range of Experience Precise Vague Total

0 to 1 year 3.48 �2.17�

n � 57

1.87 �1.90�

n � 64

2.68 �2.01�

n � 121 1 to 5 years 5.20

�1.30� n � 15

4.91 �1.51�

n � 18

4.88 �1.41�

n � 33 5 to 10 years 4.95

�0.80� n � 12

5.27 �1.01�

n � 15

5.04 �0.91�

n � 27 10 to 15 years 4.50

�1.11� n � 6

5.33 �0.82� n � 7

4.92 �0.58�

n � 13 � 15 years 5.10

�0.88� n � 9

5.45 �0.69�

n � 12

5.29 �0.78�

n � 21 Total 3.94

�1.99� n � 99

3.07 �2.30�

n � 116

3.87 �2.39�

n � 215

Panel B: Range of Experience (Years) and Project Completion Stage Range of Experience 10% 90% Total

0 to 1 year 3.42 �2.15�

n � 63

1.97 �1.99�

n � 58

2.68 �2.01�

n � 121 1 to 5 years 5.25

�1.16� n � 15

4.75 �1.67�

n � 18

4.88 �1.41�

n � 33 5 to 10 years 4.80

�0.92� n � 13

5.21 �0.89�

n � 14

5.04 �0.91�

n � 27 10 to 15 years 5.23

�0.70� n � 5

4.53 �1.37� n � 8

4.92 �0.58�

n � 13 � 15 years 5.58

�0.51� n � 9

4.89 �0.93�

n � 12

5.29 �0.78�

n � 21 Total 4.13

�1.96� n � 105

2.96 �2.23�

n � 110

3.87 �2.39�

n � 215

Standard deviation in parentheses. a Mean number of opportunity costs participants attended to when performing project continuance cash flow analysis

�minimum possible is zero and maximum is six�.

98 Victoravich

Behavioral Research In Accounting Volume 22, Number 1, 2010 American Accounting Association

Control Variables An analysis of the control variables indicates that analytical ability, knowledge of the oppor-

tunity cost concept, and perceived realism of the task are not associated with attention to oppor- tunity costs in the causal model �see Table 5, Panel B�. Perceived difficulty of the task is nega-

TABLE 4

Descriptive Statistics: Dependent Variable Project Continuance Judgmenta

Panel A: Range of Experience (Years) and Opportunity Cost Vagueness Range of Experience Precise Vague Total

0 to 1 year 5.00 �2.68�

n � 57

5.92 �2.73�

n � 64

5.46 �2.73�

n � 121 1 to 5 years 3.80

�1.92� n � 15

3.64 �2.50�

n � 18

4.34 �2.27�

n � 33 5 to 10 years 4.40

�2.29� n � 12

3.73 �2.15�

n � 15

3.85 �1.85�

n � 27 10 to 15 years 3.70

�2.89� n � 6

4.33 �1.51� n � 7

4.09 �3.51�

n � 13 � 15 years 3.69

�1.26� n � 9

3.09 �1.64�

n � 12

3.38 �1.47�

n � 21 Total 4.61

�2.47� n � 99

5.10 �2.72�

n � 116 n � 215

Panel B: Range of Experience (Years) and Project Completion Stage Range of Experience 10 Percent 90 Percent Total

0 to 1 year 5.34 �2.74�

n � 63

5.57 �2.73�

n � 58

5.46 �2.73�

n � 121 1 to 5 years 3.63

�2.56� n � 15

3.75 �2.12�

n � 18

4.34 �2.27�

n � 33 5 to 10 years 4.00

�2.05� n � 13

3.76 �1.76�

n � 14

3.85 �1.85�

n � 27 10 to 15 years 4.60

�2.23� n � 5

3.50 �2.07� n � 8

4.09 �3.51�

n � 13 � 15 years 3.17

�1.64� n � 9

3.66 �1.23�

n � 12

3.38 �1.47�

n � 21 Total 5.13

�2.56� n � 105

4.93 �2.59�

n � 110 n � 215

Standard deviation in parentheses. a Mean project continuance judgment made on an 11-point scale.

When Do Opportunity Costs Count? 99

Behavioral Research In Accounting Volume 22, Number 1, 2010 American Accounting Association

tively associated with attention to opportunity costs �� � �0.19, p � 0.01�. Age is positively associated with attention to opportunity costs �� � 0.35, p � 0.01� and negatively associated with project continuance �� � �0.14, p � 0.05�.

Hypothesis 1a Hypothesis 1a predicts that fewer opportunity costs will be attended to when presented in a

vague manner rather than a precise manner. The negative path loading �� � �0.24, p � 0.01� between opportunity cost vagueness and the number of opportunity costs attended to provides support for H1a. See Figure 2 for the tested model and Table 5 Panel A for detailed statistics.

Hypothesis 1b Hypothesis 1b predicts that management accounting experience mitigates the effect of vague-

ness on the tendency to attend to opportunity costs and thus acts as a moderator. To test for moderation, the significance of the path from the interaction between management accounting experience �in years� and opportunity cost vagueness to the number of opportunity costs attended to was examined �Hopwood 2007�. The significant path �� � 0.23, p � 0.01� between the interaction term �vagueness management accounting experience� and number of opportunity costs attended to suggests that management accounting experience moderates the effect of oppor- tunity cost vagueness, which supports H1b.

FIGURE 2 Tested Model

Opportunity Cost Vagueness (X1)

Percent Complete (X2)

(0.16*)a

Number of Opportunity Costs (M)

Project Continuance (Y)

H1a (-0.24**)

H2a (-0.29**)

(H3 -0.18*)

Management Accounting Experience (W)

H1b (0.23**)

H2b (0.12**)

(0.06)

(0.03)

*, ** Indicate p < 0.05 and p < 0.01, respectively. a The model was also run with project continuance measured as a dichotomous variable, with continuance = 1

and discontinue = 0. The results of the model were similar to those with the 11-point dependent variable. b Results indicate that number of opportunity costs mediated the relationship between project vagueness and

project continuance, and project completion stage and project continuance. Note: Dashed lines are used to represent the mediational relationship between X, M, and Y.

100 Victoravich

Behavioral Research In Accounting Volume 22, Number 1, 2010 American Accounting Association

Hypothesis 2a Hypothesis 2a predicts that fewer opportunity costs will be attended to when project comple-

tion stage is high rather than low. The negative path loading �� � �0.29, p � 0.01� between project completion stage and the number of opportunity costs attended to provides support for H2a.

Hypothesis 2b Hypothesis 2b predicts that management accounting experience mitigates the effect of project

completion stage on the tendency to attend to opportunity costs and thus acts as a moderator. In

TABLE 5

Standardized Path Estimates and 95 Percent Confidence Intervals (CI) for Models Predicting Project Continuance with Number of Opportunity Costs as a Mediator

Panel A: Main Variables

Model Paths (�)

Estimatea Hypothesis Predicted

Sign

MA Experience → Opportunity Costs

0.16* — —

Vagueness → Opportunity Costs

�0.24** H1a ���

�Vagueness MA Experience� → Opportunity Costs

0.23** H1b � �

Percent Complete → Opportunity Costs

�0.29** H2a ���

�Percent Complete MA Experience� → Opportunity Costs

0.12* H2b � �

Opportunity Costs → Project Continuance

�0.18* H3 ���

Vagueness → Project Continuance

0.06 — —

Percent Complete → Project Continuance

0.03 — —

Panel B: Control Variables Model Paths (Control Variables)

(�) Estimatea Hypothesis

Predicted Sign

Analytical Ability → Opportunity Costs

0.01 — —

Knowledge of Opportunity Cost Concept → Opportunity Costs

0.07 — —

Perceived Difficulty of Task → Opportunity Costs

�0.19** — —

Perceived Realism of Task → Opportunity Costs

�0.04 — —

Age → Opportunity Costs 0.35** — — Age → Project Continuance �0.14* — —

*, ** p � 0.05, and p � 0.01, respectively. a Bootstrapped confidence intervals are used as an additional means of testing the significance of the maximum likelihood

path estimates. All significance per the confidence intervals is consistent except for Age → Project Continuance. This suggests the path is not significant.

When Do Opportunity Costs Count? 101

Behavioral Research In Accounting Volume 22, Number 1, 2010 American Accounting Association

testing for moderation, the path from the interaction term between management accounting expe- rience �in years� and project completion stage to the number of opportunity costs attended to was examined for significance �Hopwood 2007�. The significant path �� � 0.12, p � 0.05� between the interaction term �project completion management accounting experience� and number of opportunity costs attended to suggests that management accounting experience moderates the effect of opportunity cost vagueness, which supports H2b.

Hypothesis 3 Hypothesis 3 predicts that attention to opportunity costs decreases the likelihood of continu-

ing an ongoing project. This prediction is supported as indicated by the negative and significant path coefficient �� � �0.18, p � 0.05� between number of opportunity costs attended to and project continuance.

Tests for Mediation Since attention to opportunity costs is expected to act as a mediator between each independent

variable �vagueness and project completion stage� and the dependent variable �project continu- ance�, tests were conducted to determine whether no, partial, or full mediation was present using procedures outlined in Shrout and Bolger �2002�.5 The steps are similar to those outlined in Baron and Kenny �1986�; however, the use of path analysis provides bootstrapped confidence intervals that are used in addition to traditional tests of significance to provide more evidence regarding hypothesized relationships between variables. As well, the condition requiring a relationship be- tween the independent variable to the dependent variable is relaxed due to the possibility of suppression by a competing process �Kenny et al. 1998�. This argument has been supported by several researchers �see Preacher and Hayes 2004; Collins et al. 1998; MacKinnon 2000�.

The first condition, that X �opportunity cost vagueness� is related to Y �project continuance�, is not met. This is not a concern since it has been suggested that this condition be relaxed as discussed above. The second condition, that vagueness is related to the number of opportunity costs �M�, is met �� � �0.24, p � 0.01�. The third condition, that attention to opportunity costs is related to project continuance when vagueness is included in the model, is also met �� � �0.18, p � 0.05�. The fourth condition, that vagueness is related to project continuance in the presence of attention to opportunity costs, is not met �� � 0.06, p � 0.65�. The failure to meet the fourth condition suggests that complete mediation is present; thus, there is no direct effect of vagueness on project continuance after controlling for attention to opportunity costs. As well, the significance of the Sobel Test �z � 2.50, p � 0.05�, which directly tests the presence of mediation by assessing whether the effect of X on Y is significantly reduced upon the addition of a mediator to the model, suggests complete mediation is present �Preacher and Hayes 2004�.

Overall, these results suggest that the relationship between vagueness and project continuance is completely mediated by attention to opportunity costs, and any effect of vagueness on project continuance is indirect via attention to opportunity costs. This also suggests that participants are less likely to continue a project if more opportunity costs are attended to even after controlling for opportunity cost vagueness.

In testing whether the number of opportunity costs attended to acts as a mediator between project completion stage and project continuance, the first condition of a significant relationship between X �project completion stage� and Y �project continuance� was not met. Based on the

5 Since there was a presence of a moderator and a mediator, necessary procedures were performed to ensure that moderated mediation was not present. The condition of a significant interaction between the moderator �management accounting experience� and mediator �number of opportunity costs attended to� in the dependent variable equation was not met. This suggests the absence of moderated mediation �Preacher et al. 2007�.

102 Victoravich

Behavioral Research In Accounting Volume 22, Number 1, 2010 American Accounting Association

argument made by Kenny et al. �1998�, this condition was relaxed and the second condition was tested. The significant path loading from attention to opportunity costs to project completion stage �� � �0.29, p � 0.01� suggests the second condition is met. The third condition, that attention to opportunity costs is related to project continuance in the presence of project completion stage, is also met �� � �0.18, p � 0.05�. The fourth condition, that project completion stage is related to project continuance in the presence of attention to opportunity costs, is not met �� � 0.03, p � 0.67�, indicating the presence of complete mediation. This finding indicates that there is no direct effect of project completion stage on project continuance after controlling for attention to oppor- tunity costs. Thus, attention to opportunity costs completely mediates the relationship between project completion stage and project continuance. The significance of the Sobel Test �z � 3.27, p � 0.01� also indicates complete mediation is present.

Overall, this finding suggests that the relationship between project completion stage and project continuance is completely mediated by attention to opportunity costs, and any effect of completion stage on project continuance is indirect via attention to opportunity costs. This also suggests that attention to opportunity costs reduces the likelihood of continuing a project even after controlling for project completing stage.

SUMMARY AND CONCLUSIONS This study demonstrates that opportunity cost vagueness and a nearly complete project exac-

erbate the tendency to discount opportunity costs in the absence of significant levels of manage- ment accounting experience. Results from the tested causal model demonstrate that management accounting experience mitigates the effect of these factors. This suggests that through on-the-job experience, management accountants have acquired the necessary procedural knowledge to search for and identify relevant opportunity cost information when making resource allocation decisions, even in the face of situational factors that may negate attention to opportunity costs.

The study also finds that attention to opportunity costs acts as a mediator between the two factors of interest, opportunity cost vagueness and project completion stage, and the likelihood of continuing an ongoing project. There was a negative relationship between attention to opportunity costs and the tendency to continue an ongoing project, suggesting that attending to opportunity costs highlights the cost of forgoing alternative courses of action. Thus, as indicated by the mediated relationship, considering opportunity costs has an economic decision consequence.

The failure to include opportunity costs as relevant information can be costly since it may cause decision-makers to disregard information that may be suggestive of a more favorable course of action. This in turn can lead professionals such as management accountants to recommend an unfavorable course of action to other business partners within the organization. Additionally, opportunity costs can be used by management accountants to highlight the overall benefit of a recommended decision.

This issue should not only be of interest to practicing management accountants but should also be of interest to managerial accounting instructors. Lessons in the area of relevant information for decision-making often present students with opportunity cost information that is precise, which is inconsistent with how the information would likely appear in an actual decision-making setting. Students are also rarely presented with situations in which they must overlook decision setting characteristics �e.g., nearly complete project� and focus on how the information provided can be used to make a particular decision. Instructors might consider more realistic case-based instruction which would provide students with practice on how to search for and process relevant information for decision-making.

This study is subject to at least two limitations. First, although the case materials were developed to be as realistic as possible, the task and information presentation may differ from

When Do Opportunity Costs Count? 103

Behavioral Research In Accounting Volume 22, Number 1, 2010 American Accounting Association

what decision-makers are faced with in the “real world.” For example, all relevant information might not be available in summarized form. Further, in order to obtain strong internal validity the case was absent various qualitative factors �e.g., current economic conditions�.

A second limitation that is inherent to a study that compares the differential performance of individuals who vary on the basis of experience is that other external factors may be correlated with experience and may drive the results. Although I included several variables to control for these factors, it is impossible to control for every factor and/or to develop a perfectly reliable control. Beyond these limitations, the results of my study suggest that management accounting experience enables the ability to integrate opportunity costs as relevant information despite a vague presentation and a nearly complete project. As well, attention to opportunity costs does in fact influence judgment and decision-making and thus is likely to have an economic consequence.

APPENDIX EXPERIMENTAL CASE

Fresh Foods, Inc., is a mid-sized grocery store chain with locations throughout the Southeast United States. The company’s headquarters are located in Orlando, FL. Although they carry nonperishable foods, they are known for their large variety of organic and non-organic fresh fruit, vegetables, and healthy baked goods.

Fresh Foods has decided to implement an in-house logistics division which would completely replace their third-party distributor. The logistics division project was initiated by Mr. James London, Fresh Foods’ prior Internal Investment Project Supervisor. Unfortunately, Mr. London was diagnosed with a serious illness and had to resign. After much careful consideration, the Board of Directors has appointed you as Mr. London’s successor. Since the logistics division project has not been given adequate attention in the last few weeks, it is imperative that you review the progress of the project and ensure it will be beneficial to Fresh Foods.

Step 1: As Fresh Foods’ Internal Investment Project Supervisor you must perform a cash flow analysis regarding the logistics division project. If the project is continued, it will be allocated additional funds required for completion. Please carefully consider all of the previ- ous information presented to you when performing your analysis and when making your judgments and decisions. (Note: At this point, detailed information was provided to allow the participants to perform the cash flow analysis. The experimental manipulations as described in the body of the paper were embedded in this detailed information section).

Step 2: Please provide answers to the following while referring to your cash flow analysis.

1. On the scale below please place a mark on the number that best represents the likelihood that you would continue the in-house logistics project.

Extremely Unlikely

Extremely Likely

0 1 2 3 4 5 6 7 8 9 10 Neutral

2. If you had to make an absolute decision you would decide to (Please mark an X next to your decision)._____ Discontinue logistics project. _____ Continue the logistics project.

104 Victoravich

Behavioral Research In Accounting Volume 22, Number 1, 2010 American Accounting Association

REFERENCES

Anderson, J. R. 1983. A spreading activation theory of memory. Journal of Verbal Learning and Verbal Behavior 22: 261–295.

——–. 2000. Cognitive Psychology and Its Implications. 5th edition. New York, NY: W.H. Freeman. Anderson, M. 1988. A comparative analysis of information search and evaluation behavior of professional

and nonprofessional financial analysts. Accounting, Organizations and Society 13: 431–446. Baron, R. M., and D. A. Kenny. 1986. The moderator-mediator variable distinction in social psychological

research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psy- chology 51: 1173–1182.

Becker, S. W., J. Ronen, and G. H. Sorter. 1974. Opportunity costs—An experimental approach. Journal of Accounting Research 12: 317–329.

Bedard, J. C., and M. T. H. Chi. 1993. Expertise in auditing. Auditing: A Journal of Practice & Theory 12 �Supplement�: 21–45.

Biggs, S. F. 1984. Financial analysts’ information search in the assessment of corporate earnings power. Accounting, Organizations and Society 9: 313–324.

Boehne, D. M., and P. W. Paese. 2000. Deciding whether to complete or terminate an unfinished project: A strong test of the project completion hypothesis. Organizational Behavior and Human Decision Pro- cesses 81: 178–194.

Bonner, S. E., R. Libby, and M. W. Nelson. 1997. Audit category knowledge as a precondition to learning from experience. Accounting, Organizations and Society 22: 387–410.

Bouwman, M. J., P. A. Frishkoff, and P. Frishkoff. 1987. How do financial analysts make decisions? A process model of the investment screening decision. Accounting, Organizations and Society 12: 1–29.

Budescu, D. V., S. W. Weinberg, and T. S. Wallsten. 1988. Decisions based on numerically and verbally expressed uncertainties. Journal of Experimental Psychology: Human Perception and Performance 14: 281–294.

Camerer, C., and M. Weber. 1992. Recent developments in modeling preferences: Uncertainty and ambigu- ity. Journal of Risk and Uncertainty 5: 325–370.

Chenhall, R., and D. Morris. 1991. The effect of cognitive style and sponsorship bias on the treatment of opportunity costs in resource allocation decisions. Accounting, Organizations and Society 16: 27–46.

Chi, M. T. H., P. J. Feltovich, and R. Glaser. 1981. Categorization and representation of physics problems by experts and novices. Cognitive Science 5: 121–152.

Collins, L. M., J. W. Graham, and B. P. Flaherty. 1998. An alternative framework for defining mediation. Multivariate Behavioral Research 33: 295–312.

Conlon, D. E., and H. Garland. 1993. The role of project completion information in resource allocation decisions. Academy of Management Journal 36: 402–413.

Curley, S. P., and F. J. Yates. 1985. The center and range of the probability interval as factors affecting ambiguity preferences. Organizational Behavior and Human Decision Processes 36: 273–287.

——–, ——–, and R. A. Abrams. 1986. Psychological sources of ambiguity avoidance. Organizational Behavior and Human Decision Processes 38: 230–256.

Davis, J. T. 1996. Experience and auditors’ selection of relevant information for preliminary control risk assessments. Auditing: A Journal of Practice & Theory 15: 16–37.

Efron, B., and R. J. Tibshirani. 1993. An Introduction to the Bootstrap. Boca Raton, FL: Chapman and Hall. Einhorn, H. J., and R. M. Hogarth. 1985. A contrast/surprise model for updating beliefs. Working Paper,

University of Chicago. Ellsberg, D. 1961. Risk, ambiguity and the savage axioms. The Quarterly Journal of Economics 75: 643–

669. ——–. 1963. Risk, ambiguity and the savage axioms. The Quarterly Journal of Economics 77: 336–342

Reply. Ezzamel, M., and H. Hart. 1987. Advanced Management Accounting—An Organizational Emphasis. London,

U.K.: Cassell Educational Limited. Friedman, L. A., and B. R. Neumann. 1980. The effects of opportunity costs on project investment decisions:

A replication and extension. Journal of Accounting Research 18: 407–419.

When Do Opportunity Costs Count? 105

Behavioral Research In Accounting Volume 22, Number 1, 2010 American Accounting Association

Frisch, D., and J. Baron. 1988. Ambiguity and rationality. Journal of Behavioral Decision Making 1: 149–157.

Gaeth, G. J., and J. Shanteau. 1984. Reducing the influence of irrelevant information on experienced decision makers. Organizational Behavior and Human Decision Processes 33: 263–282.

Gagne, E., C. Yekovich, and F. Yekovich. 1993. The Cognitive Psychology of School Learning. New York, NY: Harper Collins College Publishers.

Gardenfors, P. 1979. Forecasts, decisions, and uncertain probabilities. Erkenntnis 14: 159–181. Garland, H., and D. E. Conlon. 1998. Too close to quit: The role of project completion in maintaining

commitment. Journal of Applied Social Psychology 28: 2025–2048. Goodie, A. S. 2003. The effects of control on betting: Paradoxical betting on items of high confidence with

low value. Journal of Experimental Psychology: Learning, Memory, and Cognition 29: 598–610. Haefner, J. E. 1977. Race, age, sex, and competence as factors in employment selection of the disadvantaged.

The Journal of Applied Psychology 62: 199–202. Hamm, R. M., and H. Bursztajn. 1979. A medical version of a decision paradox: Is it still a paradox? Paper

presented at First Annual Society for Medical Decision Making, Cincinnati, OH. Heath, C., and A. Tversky. 1991. Preference and belief: Ambiguity and competence in choice under uncer-

tainty. Journal of Risk and Uncertainty 4: 5–28. Herz, P. J., and J. J. Schultz. 1999. The role of procedural and declarative knowledge in performing account-

ing tasks. Behavioral Research in Accounting 11: 29–44. Heymann, H., and R. Bloom. 1990. Opportunity Cost in Finance and Accounting. Westport, CT: Quorum

Books. Ho, J. L., L. R. Keller, and P. Keltyka. 2001. Managers’ variance investigation decisions: An experimental

examination of probabilistic and outcome ambiguity. Journal of Behavioral Decision Making 14: 257–278.

——–, ——–, and ——–. 2002. Effects of outcome and probabilistic ambiguity on managerial choices. Journal of Risk and Uncertainty 24: 47–74.

Holmbeck, G. N. 1997. Toward terminological, conceptual, and statistical clarity in the study of mediators and moderators: Examples from the child clinical and pediatric psychology literatures. Journal of Consulting and Clinical Psychology 65: 599–610.

Hopwood, C. J. 2007. Moderation and mediation in structural equation modeling: Applications for early intervention research. Journal of Early Intervention 29: 262–272.

Hoskin, R. 1983. Opportunity cost and behavior. Journal of Accounting Research 21: 78–95. Hoyle, R. H., and D. A. Kenny. 1999. Statistical power and tests of mediation. In Statistical Strategies for

Small Sample Research, edited by Hoyle, R. H. Newbury Park, CA: Sage. Hu, L. T., and P. M. Bentler. 1999. Cutoff criteria for fit indexes in covariance structure analysis: Conven-

tional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal 6: 1–55.

Joreskog, K. G. 1969. A general approach to confirmatory factor analysis. Psychometrika 34: 183–202. Kaplan, D. 2000. Structural Equation Modeling: Foundations and Extensions. Thousand Oaks, CA: Sage. Kenny, D. A., D. A. Kashy, and N. Bolger. 1998. Data analysis in social psychology. In The Handbook of

Social Psychology, edited by Gilbert, D., Fiske, S., and Lindzey, G. Boston, MA: McGraw-Hill. Kline, R. B. 1998. Principles and Practice of Structural Equation Modeling. New York, NY: Guilford. Kuhn, K. M., and D. V. Budescu. 1996. The relative importance of probabilities, outcomes, and vagueness in

hazard risk decisions. Organizational Behavior and Human Decision Processes 68: 301–317. Libby, R. 1995. The role of knowledge and memory in audit judgment. In Judgment and Decision Making

Research in Accounting and Auditing, edited by Ashton, R. A., and Ashton, A. H. New York, NY: Cambridge University Press.

MacKinnon, D. P. 2000. Contrasts in multiple mediator models. In Multivariate Applications in Substance Use Research, edited by Rose, J. S., Chassin, L., Presson, C. C., and Sherman, S. J. Mahwah, NJ: Erlbaum.

Moon, H. 2001. Looking forward and looking back: Integrating completion and sunk cost effects within an escalation-of-commitment progress decision. The Journal of Applied Psychology 86: 104–113.

106 Victoravich

Behavioral Research In Accounting Volume 22, Number 1, 2010 American Accounting Association

Mooney, C. Z., and R. D. Duval. 1993. Bootstrapping: A Nonparametric Approach to Statistical Inference. Newbury Park, CA: Sage.

Nagy, G. 1981. How are personnel selections made? An analysis of decision strategies in a simulated personnel selection task. Unpublished doctoral dissertation, Kansas State University.

Nelson, M. W. 1993. The effects of error frequency and accounting knowledge on error diagnosis in analyti- cal review. The Accounting Review 68: 804–824.

——–, R. Libby, and S. E. Bonner. 1995. Knowledge structure and the estimation of conditional probabilities in audit planning. The Accounting Review 70: 27–47.

Neumann, B. R., and L. A. Friedman. 1978. Opportunity costs: Further evidence through an experimental replication. Journal of Accounting Research 16: 400–410.

Northcraft, G., and M. Neale. 1986. Opportunity costs and the framing of resource allocation decisions. Organizational Behavior and Human Decision Processes 37: 348–356.

Preacher, K. J., and A. F. Hayes. 2004. SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behavior Research Methods, Instruments, & Computers 36: 717–731.

Preacher, K. J., D. D. Rucker, and A. F. Hayes. 2007. Addressing moderated mediation hypotheses: Theory, methods, and prescriptions. Multivariate Behavioral Research 42: 185–227.

Rice, M. F. 1975. Influence of irrelevant biographical information in teacher evaluation. Journal of Educa- tional Psychology 67: 658–662.

Roberts, H. V. 1963. Risk, ambiguity, and the savage axioms. The Quarterly Journal of Economics 75: 690–694 comment.

Robins, J. M., and S. Greenland. 1992. Identifiability and exchangeability for direct and indirect effects. Epidemiology (Cambridge, Mass.) 3: 143–155.

Roehl-Anderson, J. R., and S. M. Bragg. 2005. Controller’s Function: The Work of the Managerial Accoun- tant. New York, NY: John Wiley and Sons.

Shanteau, J. 1992. Competence in experts: The role of task characteristics. Organizational Behavior and Human Decision Processes 53: 252–266.

Shiffrin, R. M., and W. Schneider. 1977. Controlled and automatic human information processing: II. Per- ceptual learning, automatic attending, and a general theory. Psychological Review 84: 127–190.

Shrout, P. E., and N. Bolger. 2002. Mediation in experimental and nonexperimental studies: New procedures and recommendations. Psychological Methods 7: 422–445.

Siegel, G. 1999. Counting more, counting less: The new role of management accountants. Strategic Finance 81: 20–21.

——–. 2000. What do management accountants do? Strategic Finance 82: 85–86. ——–, J. Sorensen, and S. Richtermeyer. 2003. Being a business partner. Strategic Finance 85: 38–43. Sobel, M. E. 1982. Asymptotic confidence intervals for indirect effects in structural equation models. In

Sociological Methodology, edited by Leinhardt, S., 290–312. San Francisco, CA: Jossey-Bass. Staw, B. M. 1976. Knee deep in the big muddy: A study of escalating commitment to a course of action.

Organizational Behavior and Human Performance 16: 27–44. ——–, and F. V. Fox. 1977. Escalation: The determinants of commitment to a chosen course of action.

Human Relations 30: 431–450. ——–, and J. Ross. 1978. Commitment to a policy decision: A multi-theoretical perspective. Administrative

Science Quarterly 23: 40–64. Toda, M., and E. H. Shuford. 1965. Utility, induced utilities, and small worlds. Behavioral Science 19:

368–372. Van De Wiel, M. W. J., H. P. A. Boshuizen, and H. G. Schmidt. 2000. Knowledge restructuring in expertise

development: Evidence from pathophysiological representations of clinical cases by students and physicians. The European Journal of Cognitive Psychology 12: 323–355.

Van Dijk, E., and M. Zeelenberg. 2003. The discounting of ambiguous information in economic decision- making. Journal of Behavioral Decision Making 16: 341–352.

Vera-Muñoz, S. C. 1998. The effects of accounting knowledge and context on the omission of opportunity costs in resource allocation decisions. The Accounting Review 73: 47–72.

——–, W. R. Kinney, and S. E. Bonner. 2001. The effects of domain experience and task presentation format on accountants’ information relevance assurance. The Accounting Review 76: 405–429.

When Do Opportunity Costs Count? 107

Behavioral Research In Accounting Volume 22, Number 1, 2010 American Accounting Association

Wenk, L., R. Dufresne, W. Gerace, W. Leonard, and J. Mestre. 1997. Student-Active Science: Models of Innovation in College Science Teaching. Orlando, FL: Saunders College Publisher.

Wheaton, B., B. Muthen, D. F. Alwin, and G. F. Summers. 1977. Assessing reliability and stability in panel models. In Sociology Methodology, edited by Heise, D. R. San Francisco, CA: Jossey-Bass.

108 Victoravich

Behavioral Research In Accounting Volume 22, Number 1, 2010 American Accounting Association

Copyright of Behavioral Research in Accounting is the property of American Accounting Association and its

content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's

express written permission. However, users may print, download, or email articles for individual use.