Understanding Decision Making During a Crisis: An Axiomatic Model of Cognitive Decision Choices

Abstract

This study investigated factors that influence the strategy process during a crisis within the chemical industry. It examined key organizational, environmental, and management factors—comprehensiveness, formalization, politicization, impact of the crisis, financial reporting—for their role in the strategy process during a crisis using regression analysis. The findings indicate that the strategy process during a crisis is influenced by several factors; specifically, politicization, formalization of the decision-making process, financial reporting, and the impact of the crisis. This study proposes an axiomatic model of cognitive decision making during a crisis. It suggests that decision making during a crisis is a complex problem-solving process contingent on several variables, which can be arranged on a scale with the proscriptive variables (variables that impede or hinder accommodation) at one end of the scale and supportive variables (variables that help advocate an organization’s position) at the other end of the scale, which when cross-joined with the advocacy/accommodation continuum yield a Cartesian product of communication options. Other implications and future areas for research are suggested.

Keywords

crisis decision making strategy contingency theory cognitive model

Cancel, Cameron, Sallot, and Mitrook (1997) argue dominant models of organization-public relationship fail to capture the complexity of the environment within which organizations function and propose the use of contingency theory of accommodation as a framework to understand the influence of multiple variables on an organization’s position (stance) in its interaction with its publics. They posit an organization’s position in its interaction with its multiple publics can be arranged on a continuum of advocacy accommodation. This study builds on the work of Cancel et al. (1997) and extends it to crisis management to understand which factors constrain decision making during a crisis. The implications of the use of contingency theory of accommodation into crisis communication strategy process research suggests that factors that influence any organization has to be understood as unique to that organization or more broadly to the industry to which the organization belongs. Furthermore, if crisis communication strategy used within a particular organization has to be understood as the organization’s own construct, it implies that these crisis communication strategies may not be directly translated to other organizations. It was within this context this study examined the crisis communication strategy process within the chemical industry. It was reasoned that there would be a greater consciousness and heightened awareness of the pitfalls of not planning for a crisis due to the hazardous nature of the products. Additionally, it was rationalized focusing on one particular industry would aid in generalizing the results since they operate in similar conditions and face the same threats and opportunities.

The crisis communication strategy development process is being defined as a decision-making process initiated in response to a crisis where decision makers analyze the situation to evaluate multiple organizational positions in order to minimize reputational damage and to maintain the organization’s competitiveness and/or its sustainability. Although an organizational crisis has been defined in various ways such as a situation that can potentially escalate in intensity, fall under close government or media scrutiny, jeopardize the current positive public image of an organization, or interfere with normal business operations, including hurting the bottom line (Fink, 1986), or a “major occurrence with a potentially negative outcome that affects an organization and its publics, products, services, or its good name and interrupts the organization’s normal flow of business” (Fearn-Banks, 2007, p. 6), the most significant and unifying characteristic of all crises is its potential for causing incalculable damage to an organization and its reputation. The purpose of this study is to investigate factors that influence the strategy development process during a crisis within the chemical industry.

Literature Review

Cancel et al. (1997) proposed a comprehensive list of 87 variables that could influence an organization’s position in its interaction with its publics. In a later study, Cancel, Mitrook, and Cameron (1999) found moderate support for variables such as organization size and business exposure to interest groups. It was observed that organizations producing consumer goods were more likely to be exposed to interest groups and, therefore, their position were more likely to driven by business exposure than organizations producing commercial and industrial products. The other variables that found support include access to the dominant coalition, characteristics of the individuals who make up the dominant coalition. Cancel et al. (1999) also found mixed support for age of the corporation, competition, social and political environment, politics, centralization of decision-making processes, reputation of the corporation, and corporation resources.

Further attempts by researchers Cameron, Cropp, and Reber (2001) to bring parsimony to the matrix of 86 variables identified by Cancel et al. (1999) resulted in the organization of these variables into a group called the proscriptive variables that could limit, prevent, preclude, or prohibit accommodation of a given public by an organization. These variables were identified as moral conviction, countermanding demands by multiple publics, regulatory constraints, jurisdictional issues, and legal constraints.

In a similar study, Reber and Cameron (2003) surveyed 91 top public relations practitioners to develop scales for five theoretical constructs: external threats, external public characteristics, organizational characteristics, public relations department characteristics, and dominant coalition characteristics in the context of crisis communications and organizational relationships. The findings of their study indicate that litigation could prove to be a major external threat to dialogue. Additionally, external public characteristics, such as size, credibility, commitment, and power, could serve as barriers to having a dialogue. Internal organizational characteristics that were found to impede dialogue were harmony among staff, past negative experiences, and the presence of influential in-house counsel. Furthermore, characteristics of public relations department such as prior experience in conflict management, seniority of the public relations practitioners predicted a willingness to negotiate or accommodate.

Shin, Cameron, and Cropp (2006) grouped the 86 contingent variables identified by Cancel et al. (1997) into 12 factors across two dimensions—external and internal. They identified the five external factors as external threats, industry environment, public relationships, political/social/cultural environment, and public power. They grouped top management characteristics, organization development, organization structure, public relations department independence, public relations department governance, individual characteristics and individual capabilities as internal factors. However, external publics, organization’s characteristics, public relations department characteristics, and individual characteristics were observed to group under more than one factor.

In further elaboration of the contingency theory of accommodation, Jin and Cameron (2006) developed scales to measure an organization’s accommodative position. They identified two factors, action-based accommodation and qualified-rhetoric-mixed accommodation, as a means to measure an organization’s level of accommodation of its publics in a given situation. However, the scale developed in this study operationalizes an organization’s accommodative position in its relationship with its publics in noncrisis situations.

In further tests of the validity of the accommodative positions outlined in an earlier study in 2006, Jin and Cameron (2007) examined the effect of threat type (internal vs. external) and threat duration (long vs. short) on cognition (perceived demands and organizational resources), affective (emotional valence and arousal), and conation (action-based accommodative and qualified-rhetoric-mixed-accommodative stances). The findings indicate that external threats led to more cognitive appraisal, more intense affective responses, and more qualified rhetoric. However, the threat type did not affect valence. Overall, external and long-term threat combination led to higher situational demands appraisal and more intensive emotional arousal.

In a similar study, Hwang and Cameron (2008) studied how publics of an organization estimate an organization’s position based on perceived leadership styles and perceived severity of crisis and the interaction between the two variables. It was found that public expected more advocacy strategies when the perceived leadership style was autocratic, neutral when the perceived leadership style was transactional, and more accommodative strategies when the leadership style was perceived to be transformational or democratic.

A review of the literature above suggests that researchers have investigated several factors in the context of organization-public relationship and have either developed taxonomy of variables for parsimony or to understand their role in the selection of crisis management strategies. This study empirically tests the following variables—comprehensiveness, politicization, formalization of the decision processes, financial reporting, and impact of the crisis—to determine if the strategy development process during a crisis is dependent on them based on the contingency theory.

Comprehensiveness

Simon (1957) was the first to suggest the use of comprehensiveness in decision making as a means to deal with complex and uncertain situations. Comprehensiveness has been defined as the extent to which organizations attempt to be exhaustive in making and integrating strategic decisions (Fredrickson & Mitchell, 1984).

Since a crisis is marked by a high degree of complexity and uncertainty, it was reasoned that decision makers would be comprehensive in their decision making. On the other hand, due to the unpredictable and often immediate nature of a crisis, decision makers often have to take swift, yet informed decisions in order to survive, which might preclude comprehensiveness. However, it is argued organizations that proactively scan the environment have an advantage in decision making during a crisis as it accelerates the process of cognition of decision makers, which can help in making faster decisions with greater success. Since comprehensiveness is a means to deal with the paradox of decision making in a complex situation beset with uncertainties and constraints of resources, such as time and money, it is proposed:

Hypothesis 1: Decision makers will be comprehensive in their analysis of information during the strategy development process in a crisis.

Politicization

Strategic decision-making processes within organizations are inherently political in nature, in which they involve resolution through the transaction of power. Often, power within organizations is exercised through actions such as negotiation, coalition formation, lobbying, cooptation, withholding information, and controlling agendas (Pettigrew, 1973; Pfeffer, 1981). However, decision making during conflicts within organizations can also be open and forthright discussion based on all the available information without the covert use of power to enhance influence (Eisenhardt & Bourgeois, 1988). Although different scholars have characterized the nature of politicization to be either the covert or overt use of power, all of them refer to the presence of strong disagreements among members of an organization and resolution through use of power.

Crisis situations are typically dynamic and volatile in nature often accompanied by the stress to make quick decisions. It is argued that the decision makers might have to balance competing or conflicting interests of the internal and external publics involved in the crisis. In such a situation, making decisions that protect the reputation of the organization and simultaneously secures the interests of the affected publics might create strong disagreements. These disagreements would then require that the decision makers use power to resolve them. It is hypothesized:

Hypothesis 2: Decision makers will exercise their power to resolve disagreements in the strategy development process during a crisis.

Formalization

Formalization of the decision-making processes is argued to improve managerial decision making as it influences the flow of information between the different levels of hierarchy. In addition, it contextualizes the nature of managerial interactions, thus influencing strategic decisions (Miller, 1987). Formalization of decision-making processes also clearly delineates the roles and the responsibilities of the decision makers, which enable quicker decision making. Formalization in the decision-making process within organizations is achieved through formal organizational policies and guidelines. It is hypothesized decision making during a crisis will take place within the formal system of an organization for efficient and effective decision making as it delineates the roles and responsibilities and facilitates quicker decision making. It is posited:

Hypothesis 3: Decision makers will follow formal policies and guidelines in the decision-making process during a crisis.

Financial Reporting

Research suggests that financial reporting within organizations is influenced by the magnitude of the impact of a decision. Dutton (1986) found that during a crisis, organizations are more likely to use financial reporting as a means to explain and justify the financial liabilities of the situation. Since a crisis requires a significant investment of money to deal with financial liabilities, it is invariably expected to affect the financial health of an organization. Since a crisis is a low-probability and a high-impact event, it is argued that decision makers will consider financial reporting during a crisis as a means to explain financial liabilities. The following hypothesis is proposed:

Hypothesis 4: Decision makers will consider financial reporting when developing crisis response strategies.

Severity of the Crisis

Coombs (2007) examined the severity of the crisis in his situational crisis communication theory model. Although Coombs examined the severity of the crisis as a moderating factor in the selection of the crisis communication strategies, he does so within the framework of attribution of crisis responsibility and damage to reputation. It can therefore be subjective. For instance, if the external publics of an organization wrongly attribute the financial downfall of an organization to the unfavorable external economic conditions, when in fact the organization suffered financial losses due to poor management decisions, then the public is likely to assign lesser responsibility to the organization per se. In such an instance, this would result in lower reputational damage and the development of strategies that might be suboptimal.

Additionally, the impact of the crisis often extends beyond damage to the reputation of the organization and may cause significant financial loss, loss of human lives, and environmental damage. Product harm crises in general, and product recalls in particular, have the potential to damage an organization’s reputation that can lead to revenue and market share losses (Laufer & Coombs, 2006; Rhee & Haunschild, 2006; Van Heerde, Helsen, & Dekimpe, 2007). Additionally, Langford (2009) noted that environmental crises stick in the public mind and have a greater impact on the public opinion than other corporate crises, which are quickly forgotten. It is therefore, argued that the decision makers will formulate strategies that consider the severity of the crisis on its impact on the organization’s financial position, damage to the environment, loss of human lives, process interruptions, loss to plant and machinery during a crisis. The following hypothesis is proposed:

Hypothesis 5: Decision makers will consider the severity of the crisis when developing crisis response strategies.

Methodology

This study used surveys to collect data from corporations within the chemical industry. By combining directories of trade associations, a sampling frame of 179 chemical corporations was created. The chemical organizations were chosen as the sampling frame because the available directory listed the chemical organizations and not their individual plant sites. Therefore, it seemed reasonable and appropriate to choose the main organization as the sampling frame. The individual chemical plant or site was the sampling unit. Surveys were mailed or e-mailed to the plant or site managers of larger chemical corporations and to the CEOs and vice presidents of smaller organizations, as they were structured differently. Additionally, they were deemed to be the one who were most closely involved with the decision-making process during a crisis. In addition, the surveys were also mailed to the corporate communications managers of the organizations. The survey respondents were mailed a gift certificate worth $15 as an incentive to complete and return the survey.

In all, 425 surveys were mailed or e-mailed. A follow-up e-mail was sent 3 weeks later to those who were e-mailed a request to complete the survey. Eighty-one responses were received from the survey participants. Of the 81 responses received, 11 were incomplete. Therefore, there were only 70 usable survey responses for a response rate of 16.5%.

Although selection of the correct sample size is critical for the investigation of any research, the appropriate sample size is dependent on several factors, such as the purpose of the study, the population size, and the risk the researcher is willing to accept, in addition, to the confidence level, the confidence interval, and the variability in the given population. Though several formulae and published tables are available for calculating the sample size, situations, however, may exist where the use of sample size tables and formulae may not provide the intended results (Bartlett, Kotrlik, & Higgins, 2001). Multiple regression analysis is one such case. In the case of multiple regression analysis, the ratio of observation to the independent (predictor) variables should not fall below 5 (Hair, Anderson, Tatham, & Black, 1995). Some researchers, however, recommend a more conservative estimate of 10 observations for each independent (predictor) variable (Halinski & Feldt, 1970). Due to the challenges faced in reaching the decision makers in the chemical industry, the ratio of observations to independent variables was chosen to arrive at the sample size. This study examined five independent/predictor variables; using the conservative estimate of 10 observations required for each predictor variable, the sample size of 70 exceeds the required sample size. The results obtained in this study can be deemed reliable because it still meets the minimum sample size requirements for conducting regression analyses.

The survey questionnaire used a 5-point Likert-type scale to explore the factors that influence strategy development within the chemical industry. The response categories in the survey ranged from “very unlikely” to “very likely.” In addition, the respondents had the opportunity to rate the importance of an item on a 5-point Likert-type scale scored as “1” for least important and “5” for most important. The constructs were based on previously validated scales developed by Papadakis, Lioukas, and Chambers (1998). Comprehensiveness was based on Fredrickson’s (1984) rationality/comprehensiveness dimension. The “Comprehensiveness” scale consisted of three items: have detailed action plans, assign responsibility to a group of people or committee, seek outside assistance. The “Formalization of Decision-Processes” scale consisted of two items: have formal goals and objectives; consider formal policies or guidelines. It was based on the scale developed by King (1975) and Stein (1980). The “Politicization” scale consisted of two items: negotiate between different departments or groups and consider external resistance or pressure from outside groups. This scale was based on Pettigrew (1973), Mintzberg, Raisinghani, and Théorêt (1976) and Hickson, Butler, Cray, Mallory, and Wilson (1986). The “Financial Reporting” scale was based on King (1975), Stein (1980), Marsh, Barwise, Thomas, and Wesley (1988) and consisted of two items: consider proforma financial statements; consider historical financial data such as operating margin, net profit margin, return on equity. The scale, “Severity of the Crisis” consisted of nine items: process interruptions, damage to plant and machinery, loss of life, damage to the environment, harmful to human health, damage to the reputation, financial loss, impact on market share, threat to information technology/other secure business systems and information. The dependent variable strategy development was operationalized as “Schedule formal or informal meetings to analyze the situation and come up with possible solutions.”

It is not unusual for researchers sometimes to come across situations when poor-quality items have to be removed from a limited item pool, resulting in scales with two items. Eisinga, Grotenhuis, and Pelzer (2013) addressed the concern of how to best estimate reliability in such situations with a two-item scale. They state that more items lead to a better construct representation and the primary way to make measures more reliable is to increase the number of items. However, if research design or circumstances dictate that the scale has two possibly congeneric items, then it is best to report the Spearman-Brown reliability estimate as a measure of reliability.

Cronbach’s alpha is an accurate estimate of reliability under rather restrictive assumptions (Eisinga et al., 2013). And as such, coefficient alpha almost always underestimates true reliability, sometimes rather substantially (Revelle & Zinbarg, 2009; Sijtsma, 2009). For two-item measures, the Spearman-Brown coefficient is never lower than coefficient alpha and almost always higher. It is also, on average, less biased, especially if the correlation between the items is relatively strong (Eisinga et al., 2013). Furthermore, as the correlation between the two congeneric items increases, the underestimation of coefficient alpha remains substantial even if the interitem relationship is rather strong. The underestimation by coefficient alpha is, on average, larger than the misestimation by the Spearman-Brown statistic. Hence, the most appropriate reliability statistic for a two-item scale is the Spearman-Brown coefficient (Eisinga et al., 2013).

For a reliability coefficient to accurately reflect the true reliability of a two-item scale, the measure should be parallel. Spearman-Brown coefficient assumes that the two items are parallel (the means and the variances of the items are the same) and as such, the interitem correlation represents the correlation between one-half of the test with the other half, that is, the split-half reliability of the scale (Hulin, Netemeyer, & Cudeck, 2001). This split-half reliability when converted into a reliability coefficient for a two-item scale using Spearman-Brown formula is equivalent to standardized coefficient alpha based on standardized items.

To test the reliability of the items in the scale, Spearman-Brown coefficient was calculated using SPSS. The Spearman-Brown coefficient for the scale Comprehensiveness was .778, Financial Reporting was .927, Impact of the Crisis was .774, Formalization was .746, and Politicization was .696.

The regression model being tested in this study includes the dependent variable, “Strategy Development” and the independent variables “Comprehensiveness,” “Formalization,” “Politicization,” “Financial Performance,” and “Severity of the Crisis.”

Results

The results indicate that the decision makers are likely to consider financial statements (5-point scale; M = 2.92, SD = 1.20); consider financial data (M = 2.87, SD = 1.23); will have formal goals and objectives (M = 4.37, SD = 0.543); likely to consider formal company policies and guidelines (M = 4.40, SD = 0.824); likely to consider the impact of the crisis on: loss of human lives (M = 4.94, SD = 0.482); human health (M = 4.75, SD = 0.533); damage to the environment (M = 4.67, SD = 0.586), damage to the reputation (M = 4.12, SD = 0.832), and financial loss (M = 3.61, SD = 0.943); process interruptions (M = 3.74, SD = 1.010); damage to the plant and machinery (M = 3.91, SD = 0.836); negotiate between different departments/groups (M = 3.29, SD = 1.105); consider external resistance (M = 3.63, SD = 0.935) as important in the strategy development process during a crisis. Table 1 summarizes the results of the descriptive analysis of the variables that predict strategy development during a crisis.

Table 1.

Descriptive Analysis of Variables Predicting Strategy Development.

Item	M	SD
Have detailed action plans	4.46	0.755
Assign responsibility to a committee	4.49	0.847
Seeks outside assistance	3.77	1.09
Consider financial statements	2.92	1.20
Consider historical financial data (e.g., operating margin, profit margin, return on equity)	2.87	1.23
Has formal goals and objectives	4.37	0.543
Consider formal company policies and guidelines	4.40	0.824
Negotiate between groups or departments	3.29	1.105
Consider external resistance from outside groups	3.6.3	0.935
Consider organization’s responsibility for the crisis in the eyes of the public	4.09	0.812
Considers loss of life	4.94	0.482
Process interruptions	3.74	1.010
Damage to plant and machinery	3.91	0.836
Damage to the environment	4.67	0.586
Harmful to human health	4.75	0.553
Damage to reputation	4.75	0.553
Financial loss	3.61	0.943
Impact on market share	3.36	1.03
Threat to secure business systems an and information technology	3.13	1.16

In the regression model, five predictors—comprehensiveness, formalization, politicization, financial reporting, and impact of the crisis—were regressed on the dependent variable. The results of regression analysis for the model, F(24.57) = 0.8404, p < .0001, significantly predicts the variance in the dependent variable strategy development. The overall model was able to explain 84.04% of the variance (R² = .8404). Predictor variables formalization/standardization (p < .000), politicization (p < .000), financial reporting (p < .000), and impact of the crisis (p < .000) were found to be significant predictors of the strategy development process during a crisis. Table 2 summarizes the results of linear regression analyses examining the predictors of the strategy development during a crisis.

Table 2.

Summary of Regression Analysis of Factors That Predict Strategy Development During a Crisis.

	β	β	Standard error	F value	p Value
Comprehensiveness	0.294	0.392	0.119	2.471	.232
Formalization and Standardization	0.030	0.040	0.121	8.91	.0001***
Politicization	0.062	0.109	0.075	3.95	.0011***
Financial Reporting	0.026	0.050	0.113	6.46	.0001***
Impact of the Crisis	0.280	0.443	0.108	6.72	.0001***
R ²	.8404
Model F(24.57)	0.0001

p < .05 (two-tailed test). **p < .01 (two-tailed test). ***p < .001 (two-tailed test).

The findings of this study indicate that Hypotheses 2, 3, 4, and 5 were supported. Comprehensiveness was not found to be a predictor of the strategy development process during a crisis. Thus, Hypothesis 1 was not supported. Since a crisis can seldom be anticipated, decision makers have limited time to be exhaustive in their generation and evaluation of options. In addition, decisions have to be made quickly. Therefore, it is not unusual that Hypothesis 1 was not supported.

Discussion and Conclusion

The findings indicate that decision makers use negotiation as a means to exercise their power to resolve disagreements when faced with pressures from external groups. As is often the case, the responsibility for managing the crisis is assigned to a team that searches for ready-made solutions or designs custom solutions (Hale, Hale, & Dulek, 2006). Sometimes, the solutions under consideration maybe selected based on group bargaining, when individual solutions may not be acceptable to the group. In some cases, a higher authority might accept or reject the choice when politicization adversely affects the climate and the quality of the decision-making process. This may explain why decision makers may have to negotiate during a crisis. In addition, when a crisis comes under the scrutiny of the public, or the media, or the government, it is very likely that the attention from the media and the public concern or outrage may put the decision maker under pressure to take certain decisions. At other times, decision makers may face resistance from activist public whose interests may have been affected by the crisis. In such situations when the decision makers are compelled to interact with these groups to resolve the issue, negotiation and use of power may be the only tools to resolve the crisis. This study introduces for the first time the notion of politics in the decision-making process during a crisis. Since the crisis communication literature does not recognize the fact that decision making involves a set of people, there is no reference of negotiation or bargaining techniques. However, Grunig and Hunt (1984) alluded to negotiation and compromise through their two-way symmetrical model to resolve conflicts that arise between an organization and its external publics.

The findings also reveal that decision making during a crisis follows a formal process. Most organizations have formal systems for lateral and hierarchical communication. Formalization in the decision-making process during a crisis is achieved through the creation of crisis management teams. The crisis management teams helps in contracting the decision-making process (Coombs, 1999; Laye, 2002), which helps in increasing the speed with which decisions are made. Pearson and Clair (1998) and Witt and Morgan (2002) argue that organizations experience greater success when crisis management rests with a crisis management team rather than with an individual.

Furthermore, the findings of this study indicate that decision makers consider financial reporting when making decisions during a crisis. A crisis generally requires commitment of significant financial resources either to take corrective actions, pay financial compensation, or to manage contingencies like litigation expenses. Since a crisis generally requires a significant commitment of financial resources, it appears logical that decision makers consider financial reporting. The findings of this study support Dutton’s (1986) results, which suggest that during a crisis, decision makers employ increased levels of resources, enhance control over issue resolution, and increase the level of issue-related explanation. Financial reporting may apparently help decision makers in explaining the financial results to the stockholders and others affected by the success of the organization. This finding is also in line with Papadakis et al.’s (1998) study, which found that situations perceived as crises are associated with more financial reporting activities. Increased financial reporting may serve as a means to exercise greater control during a crisis. In addition, since a crisis involves significant investment of resources, incurring significant financial losses is a strong possibility and financial reporting may provide decision makers an avenue to account for that. Additionally, since financial reporting is used for accounting for events or issues that occur infrequently, it is reasonable to understand their use during a crisis.

Furthermore, the impact of the crisis on damage to the environment, loss of human lives, financial situation, harm to human health, threat to secure business systems and information technology in addition to the reputation of the organization was found to play a significant role in the development of crisis response strategies. It is argued that decision makers are likely to approach a crisis that causes process interruptions and damage to the plant and machinery differently from a crisis that results in loss of human lives or significant damage to the environment. This implies that the severity of a crisis will affect the process and the content of the strategy. Based on the results of this study, one can argue that a crisis that has irreversible effects (such as loss of human life, irreparable damage to the environment), is long-term, and hurts the viability of an organization will affect the decision-making process during a crisis. On the other hand, less severe crisis (such as process interruptions, damage to plant, and machinery) that are reversible and have a comparatively short-term impact will also affect the decision-making process but the content of the strategy will be different, signifying that different impacts will result in different strategies and strategic decision-making processes.

This study examined the severity of the crisis not only on reputation of the organization but also on a wider range of factors that include internal organizational measures such as process interruptions (damage to plants and machinery, financial loss) or external environmental factors (impact on market share, damage to the environment, damage to the reputation). In addition, it suggests that severity of the crisis should be measured in terms of its long-term effects, the irreversibility of the effects, and finally, whether it affects the viability of the organization itself. This study demonstrates that the impact of a crisis is more widespread than reputational damage alone and affects the strategy development process.

Interestingly but not surprisingly, strategy development during a crisis was not found to be contingent comprehensiveness in decision making. Time constraints compel the decision makers to choose the most satisfactory option instead of the most optimal solution. Furthermore, stable environments allow decision makers to be more exhaustive in their search for options as compared with dynamic environment. This explains why comprehensiveness was not found to be a predictor of decision making during a crisis.

Contributions of This Study

This study proposes an axiomatic model of cognitive decision making during a crisis. Based on the findings of this study and the work of previous researchers (Cameron et al., 2001; Cancel et al., 1997; Cancel et al., 1999; Hwang & Cameron, 2008; Jin & Cameron, 2007; Reber & Cameron, 2003; Shin et al., 2006), it suggests that decision making during a crisis is a complex problem-solving process contingent on several variables, which can be arranged on a scale. It is proposed that the proscriptive variables, variables that impede or hinder accommodation (Cameron et al., 2001), can be arranged at one end of the scale and supportive variables (variables that help advocate an organization’s position) at the other end of the scale. In any crisis, for example, an organization’s positive reputation in the past, corporate resources might provide support in repairing or restoring its relationship with its affected stakeholders. The axiomatic model of decision making during a crisis suggests that if these two sets of variables were cross-joined or cross-paired, then the total number of decision alternatives available to a decision maker during a crisis can be thought of as a Cartesian product of the contingent variables and continuum of accommodative/advocacy stance or position (Figure 1). A Cartesian product is being proposed because there is a cross-pairing or cross-joining of two sets of variables and the outcome of the cross-pairing is the product of the relationships between the two subsets, which is representative of a Cartesian product.

Figure 1.

Axiomatic model of cognitive decision making during a crisis.

The decision maker then chooses a stance or position based on a criterion or criteria, for resolving the crisis. The criterion could be to either protect and restore the reputation of the organization or its relationship with its stakeholder. The criterion or criteria will evidently depend on the impact of the crisis on the organization. In summary, the axiomatic model of cognitive decision making during a crisis can be deconstructed or decomposed into two components: the decision situation and the decision process. The decision situation comprises the Cartesian product of contingency variables and the advocacy/accommodation continuum. The decision process involves identifying the criterion or criteria for selecting the most optimal stance or position from the Cartesian sets. A decision maker may choose more than one set of strategic communication option to resolve the crisis.

Furthermore, this study suggests that proscriptive and supportive variables have to be understood as variables unique to an organization or the corporation; that is, a variable may be proscriptive for one organization but prove to be supportive for another organization. For instance, external industry environment may prove to be conducive or supportive of an accommodative stance in one type of organization, whereas it might preclude accommodation in another organization. It is, therefore, vital to identify accurately the complete list of the proscriptive and the supportive variables of an organization.

Limitations of the Study and Directions for Further Research

One of the drawbacks of this study was its sample size. However, given the nature of the study and the difficulty in gaining access to decision makers within a highly diverse industry, achieving a larger sample size was extremely challenging. Since this is an exploratory study, the findings still have important implications for future research within the crisis strategy development process. Several extensions, both methodological and substantive, are suggested to comprehend fully the strategy process within organizations during a crisis. Additional factors such as organizational structure and ownership of an organization, whether it is government-owned or privately owned, should be examined to see how they affect the strategy process during a crisis.

Further research involving interviews with decision makers within organizations needs to be conducted to test the generalizability of the proposed decision-making model during a crisis. Since this study was done in an industry highly vulnerable to accidents, it is possible that they are probably more concerned with the strategy process during a crisis whether and how contextual factors differ with different industries.

Footnotes

Author’s Note

I confirm that the findings reported in the article are original and the article is not under review elsewhere and has not been published previously.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biography

Tulika Varma is an assistant professor in the Department of Communication at the University of North Florida in Jacksonville, Florida.

References

Bartlett

J. E.

Kotrlik

J. W.

Higgins

C. C.

(2001). Organizational research: Determining appropriate sample size in survey research. Information Technology, Learning, and Performance Journal, 19(1), 43-50.

Cameron

G. T.

Cropp

Reber

B. H.

(2001). Getting past platitudes: Factors limiting accommodation in public relations. Journal of Communication Management, 5, 242-261.

Cancel

A. E.

Cameron

G. T.

Sallot

L. M.

Mitrook

M. A.

(1997). It depends: A contingency theory of accommodation in public relations. Journal of Public Relations Research, 9, 31-63.

Cancel

A. E.

Mitrook

M. A.

Cameron

G. T.

(1999). Testing the contingency theory of accommodation in public relations. Public Relations Review, 25, 171-197.

Coombs

W. T.

(1999). Information and compassion in crisis responses: A test of their effects. Journal of Public Relations Research, 11(2), 125-142.

Coombs

W. T.

(2007). Protecting organization reputations during a crisis: The development and application of situational crisis communication theory. Corporate Reputation Review, 10, 1-14.

Dutton

J. E.

(1986). The processing of crisis and non-crisis issues. Journal of Management Studies, 23, 501-517.

Eisenhardt

K. M.

Bourgeois

L. J.

(1988). Politics of strategic decision making in high-velocity environments: Toward a midrange theory. Academy of Management Journal, 31, 737-770.

Eisinga

Grotenhuis

M. T.

Pelzer

(2013). The reliability of a two-item scale: Pearson, Cronbach, or Spearman-Brown? International Journal of Public Health, 58, 637-642.

10.

Fearn-Banks

(2007). Crisis communications: A casebook approach. New York, NY: Routledge.

11.

Fink

(1986). Crisis management: Planning for the inevitable. New York, NY: American Management Association.

12.

Fredrickson

J. W.

(1984). The comprehensiveness of strategic decision processes: Extension, observations, and future directions. Academy of Management Journal, 27, 445-466.

13.

Fredrickson

J. W.

Mitchell

T. R.

(1984). Strategic decision processes: Comprehensiveness and performance in an industry with an unstable environment. Academy of Management Journal, 27, 399-423.

14.

Grunig

J. E.

Hunt

(1984). Managing public relations. New York, NY: Holt, Rinehart and Winston.

15.

Hair

Anderson

R. E.

Tatham

R. L.

Black

W. C.

(1995). Multivariate data analysis. Upper Saddle River, NJ: Prentice Hall.

16.

Hale

J. E.

Hale

D. P.

Dulek

R. E.

(2006). Decision processes during crisis responses: An exploratory investigation. Journal of Managerial Issue, 18, 310-320.

17.

Halinski

R. S.

Feldt

L. S.

(1970). The selection of variables in multiple regression analysis. Journal of Educational Measurement, 7, 151-157.

18.

Hickson

D. J.

Butler

R. J.

Cray

Mallory

G. R.

Wilson

D. C.

(1986). Top decisions: Strategic decision making in organizations. San Francisco, CA: Jossey-Bass.

19.

Hulin

Netemeyer

Cudeck

(2001). Can a reliability coefficient be too high? Journal of Consumer Psychology, 10, 55-58.

20.

Hwang

Cameron

G. T.

(2008). Public’s expectation about an organization’s stance in crisis communication based on perceived leadership and perceived severity of threats. Public Relations Review, 34, 70-73.

21.

Jin

Cameron

G. T.

(2006). Scale development for measuring stance as degree of accommodation. Public Relations Review, 32, 423-425.

22.

Jin

Cameron

G. T.

(2007). The effects of threat type and duration on public relations practitioner’s cognitive, affective, and conative responses in crisis situations. Journal of Public Relations Research, 19, 255-281.

23.

King

P. F.

(1975). An investigation of the process of large-scale capital investment decision making in diversified hierarchical organizations (Unpublished doctoral dissertation). Cambridge University, London, England.

24.

Langford

(2009). Crisis public relations management. In Tench

Yeoman

(Eds.), Exploring public relations (pp. 385-408). Harlow, England: Pearson.

25.

Laufer

Coombs

W. T.

(2006). How should a company respond to product harm crisis? The role of corporate reputation and consumer-based cues. Business Horizons, 49, 379-385.

26.

Laye

(2002). Avoiding disaster: How to keep your business going when catastrophe strikes. Hoboken, NJ: John Wiley.

27.

Marsh

Barwise

Thomas

Wesley

(1988). Managing strategic investment decisions in large diversified companies. London, England: Centre for Business Strategy, London Business School.

28.

Miller

(1987). Strategic industrial relations and human resource management: Distinction, definition, and recognition. Journal of Management Studies, 24, 347-361.

29.

Mintzberg

Raisinghani

Théorêt

(1976). The structure of “unstructured” decision processes. Administrative Science Quarterly, 21(2), 246-275.

30.

Papadakis

V. M.

Lioukas

Chambers

(1998). Strategic decision-making processes: The role of management and context. Strategic Management Journal, 19, 115-147.

31.

Pearson

C. M.

Clair

J. A.

(1998). Reframing crisis management. Academy of Management Review, 23, 59-76.

32.

Pettigrew

A. M.

(1973). The politics of organizational decision-making. London, England: Tavistock.

33.

Pfeffer

(1981). Power in organizations. Marshfield, MA: Pitman.

34.

Reber

Cameron

G. T.

(2003). Measuring contingencies: Using scales to measure public relations practitioner limits to accommodation. Journalism & Mass Communication Quarterly, 80, 431-446.

35.

Revelle

Zinbarg

R. E.

(2009). Coefficients alpha, beta, omega, and the glb: Comments on Sijtsma. Psychometrika, 74, 145-154.

36.

Rhee

Haunschild

(2006). The liability of good reputation: A study of product recalls in the U.S. automobile industry. Organization Science, 17, 101-117.

37.

Shin

J. H.

Cameron

G. T.

Cropp

(2006). Occam’s Razor in the contingency theory: A national survey on 86 contingent variables. Public Relations Review, 32, 282-286.

38.

Sijtsma

(2009). On the use, the misuse, and the very limited usefulness of Cronbach’s alpha. Psychometrika, 74, 107-120.

39.

Simon

H. A.

(1957). Models of man, social and rational: Mathematical essays on rational human behavior in a social setting. New York, NY: John Wiley.

40.

Stein

(1980). Contextual influence on strategic decision methods (Doctoral dissertation). Available from ProQuest Dissertations and Theses database. (No. AAI8018615)

41.

Van Heerde

Helsen

H. K.

Dekimpe

M. G.

(2007). The impact of a product-harm crisis on marketing effectiveness. Marketing Science, 26, 230-245.

42.

Witt

J. L.

Morgan

(2002). Stronger in the broken places: Nine lessons for turning crisis into triumphs. New York, NY: Henry Holt.