Case-Based Ethics Instruction: The Influence of Contextual and Individual Factors in Case Content on Ethical Decision-Making
- First Online:
- Cite this article as:
- Bagdasarov, Z., Thiel, C.E., Johnson, J.F. et al. Sci Eng Ethics (2013) 19: 1305. doi:10.1007/s11948-012-9414-3
- 570 Views
Cases have been employed across multiple disciplines, including ethics education, as effective pedagogical tools. However, the benefit of case-based learning in the ethics domain varies across cases, suggesting that not all cases are equal in terms of pedagogical value. Indeed, case content appears to influence the extent to which cases promote learning and transfer. Consistent with this argument, the current study explored the influences of contextual and personal factors embedded in case content on ethical decision-making. Cases were manipulated to include a clear description of the social context and the goals of the characters involved. Results indicated that social context, specifically the description of an autonomy-supportive environment, facilitated execution of sensemaking processes and resulted in greater decision ethicality. Implications for designing optimal cases and case-based training programs are discussed.
KeywordsCase-based instructionEthical decision-makingEthics casesTeaching ethics
The case-based approach (i.e., using factual or fictional scenarios exemplifying the issues at hand) to learning and instruction has been regarded as a highly valuable and effective method across multiple disciplines, among them law, medicine, and business (Kim et al. 2006; Rippin et al. 2002; Williams 1992). Case-based learning is touted for the positive effects it has on learners’ decision-making, critical thinking, and deductive and inductive reasoning skills (Falkenberg and Woiceshyn 2007; Kim et al. 2006; Menzel 2009). Case-based learning, consistent with these outcomes, has been extensively applied in ethics education, as it has proven to have clear benefits for individuals facing ill-defined, high-risk problem scenarios (Kolodner and Simpson 1989; Kolodner 1997). Despite the usefulness of cases in learning and instruction, literature is limited regarding appropriate case construction and content in order to ensure effectiveness of case-based learning. Particularly, no information exists in reference to case content that may influence case-based reasoning in learners of ethics. Although cases are highly prevalent in ethics education, we do not yet know whether characteristics of cases have differential effects on learning.
Ethical issues are multifaceted, dynamic, full of conflicting goals, and difficult to resolve (Werhane 2002). Case-based reasoning, an underlying function of case-based learning, aids in understanding ethical dilemmas by providing examples of similar situations along with methods that may be employed to resolve them (Kolodner 1992). Simply stated, “In case-based reasoning, a reasoner remembers a previous situation similar to the current one and uses that to solve the new problem” (Kolodner 1992, p. 4). This process is facilitated via sensemaking. Sensemaking is a complex cognitive process that enables ethical problem-solving through the identification of central features influencing an ethical decision, the integration of that information into a mental model, and the interpretation and application of that knowledge in making a final decision (Mumford et al. 2008; Sonenshein 2007). Mental model accuracy in sensemaking is contingent upon multiple pieces of information including case-based knowledge, accessed via case-based reasoning (Mumford et al. 2008). Given this, it is important to account for sensemaking when constructing cases involving problems with ethical implications because sensemaking can promote ethical decision-making (EDM) by facilitating case-based reasoning. However, this can only be accomplished if the key features of cases (e.g., critical causes and constraints) are clearly discerned and emphasized. Various studies have demonstrated the utility of sensemaking in EDM (Brock et al. 2008; Mumford et al. 2008; Sonenshein 2007; Thiel et al. 2011).
Case content can promote or hinder the facilitating mechanisms of case-based learning and sensemaking. Thus, it is important to consider the format and characteristics of the cases used in instruction, as they may be poorly constructed and inadequate for learning (Atkinson 2008; Falkenberg and Woiceshyn 2007; Herreid 1998; Kim et al. 2006). Though investigations into effective case construction and content are limited, several researchers have shown that “good” cases must be relevant to the reader, realistic, engaging, challenging, emotionally-arousing, and instructional (Atkinson 2008; Herreid 1998; Lynn 1999; Kim et al. 2006). To our knowledge, only three other studies have investigated the effect of case content on EDM (Harkrider et al. 2012; Johnson et al. 2012; Thiel et al. 2011). Our study aims to augment this effort by providing new information about case content attributes conducive to EDM.
Given the foregoing considerations, two case content variables (one contextual and one individual) were manipulated to investigate their effects on case-based reasoning. The setting, or social context (contextual factor), and the goals of the characters involved (individual factor) were embedded within case examples. We chose to examine social context over other contextual variables due to the overwhelming evidence suggesting that a rich description of a setting is a fundamental attribute of a well-constructed case (Atkinson 2008). With respect to the characters’ goal focus, we chose to concentrate on this individual factor because of the proposed importance of actors’ motives and intent within case studies (Atkinson 2008). Only via a rich description of characters’ goals can readers begin to understand the reasons for their chosen actions. Information about goals provides a level of detail that is needed for individuals to fully comprehend causes and is useful for integrating that information into their existing mental model to promote sensemaking.
Contextual and Individual Issues in Ethical Decision-Making
Numerous studies have provided theory and empirical evidence for the important role of contextual and individual elements in ethical dilemmas (Kurtines 1986; Mumford et al. 2007; Perri et al. 2009; Rabl 2011; Verbeke et al. 1996; Vergés 2010). Some of the more thoroughly researched contextual factors have included the ethical codes, ethical and organizational climate/culture, external environment, subjective norms, and training (O’Fallon and Butterfield 2005). The individual factors most commonly studied include gender, age, locus of control, Machiavellianism, nationality, religion, and organizational commitment (O’Fallon and Butterfield 2005). Because investigators agree that both contextual and individual factors can be highly influential when it comes to EDM, we investigated the impact of each using social context as a contextual factor and goal focus as an individual factor.
“Case studies live in the setting” (Atkinson 2008, p. 37). Descriptive information about the atmosphere, policies and procedures by which the organization operates, hierarchies, culture, climate, and politics all contribute to the developing setting in a case (Atkinson 2008; Urbanac 1998). These aspects of a case not only enhance its realism and clarity, but they also function to assist in the decision-making process (Atkinson 2008; Falkenberg and Woiceshyn 2007; Herreid 1998; Lundberg et al. 2001; Lynn 1999; Kim et al. 2006; Rippin et al. 2002). Despite this, inclusion of contextual factors in ethics cases has received very little attention.
Derived from self-determination theory (SDT; Deci and Ryan 1985; Ryan and Deci 2000), social context has been suggested to help satisfy three main psychological needs: Competence, autonomy, and relatedness. Specifically, SDT identified two types of social context, autonomy-supportive and controlling. Environments that are autonomy-supportive “minimize the salience of external incentives and threats, avoid controlling language, and acknowledge the learners’ frame of reference” (Vansteenkiste et al. 2004, p. 247). Whereas controlling environments “pressure people through use of incentives, deadlines, and punishments or through reliance on instructions that stress what people should do” (Vansteenkiste et al. 2004, p. 247). Numerous field and lab studies have provided empirical evidence for various learning benefits associated with the autonomy-supportive, as opposed to controlling, environment. Specifically, the autonomy-supportive social context has been shown to enhance test performance, task interest, persistence, in-depth information processing, and conceptual learning (Grolnick and Ryan 1987; Vansteenkiste et al. 2004, 2005). To date, these effects have only been observed when the social context is simulated, and thus have yet to be studied within case content.
Given that the sensemaking process is contingent upon in-depth processing of information (Mumford et al. 2008; Sonenshein 2007), we reasoned that sensemaking (i.e., constraint criticality and forecasting) would improve when cases presented information in an autonomy-supportive context and, in turn, result in the construction of more accurate mental models.
H1a: Description of an autonomy-supportive social context compared to controlling or no social context in cases will lead to enhanced recognition of critical constraints.
H1b: Description of an autonomy-supportive social context compared to controlling or no social context in cases will lead to improved forecasting of potential outcomes.
H2: Description of an autonomy-supportive social context compared to controlling or no social context in cases will lead to improved decision ethicality.
As discussed previously, individual or dispositional factors also have a pronounced effect on EDM. Thus, considering the characters in cases, along with their personalities and motivations, is of importance. Atkinson (2008) was adamant that “students of ethics should think about motive and intent of the actors in their case studies” (p. 38). When readers are provided information about the motive behind a character’s actions, the reader can better understand the reasons for that character’s approach to the problem. Describing a character’s goal provides a level of detail that is necessary for the reader to ascertain the key causes and constraints of the situation. This information is then used to improve the reader’s mental model, which in turn leads to improved sensemaking. It is essential, then, for characters’ goals, which are likely to be competing in ethical dilemmas, to be presented in a clear and realistic manner within a case.
Based on regulatory focus theory (Higgins 1997, 1998), goals can take the form of striving for positive outcomes (promotion focus) or avoiding negative outcomes (prevention focus). “People with a promotion focus are concerned with the presence or absence of positive outcomes. They regulate their behavior towards advancement, aspirations, and accomplishments” (Florack and Hartmann 2006, p. 626). While, “people with prevention focus are concerned with the absence or presence of negative outcomes. They orient their behavior towards protection, safety, and responsibilities” (Florack and Hartmann 2006, p. 626). Regulatory focus has consistently been linked to decision-making, such that individuals focusing on prevention goals tend to be risk-averse, more careful and make fewer mistakes (Crowe and Higgins 1997; Florack and Hartmann 2006; Förster et al. 2003). Nevertheless, very few studies have investigated regulatory focus and EDM. Two such studies have shown that individuals with a promotion goal focus were more likely to accept unethical behavior on the part of others or behave unethically themselves (De Bock and Van Kenhove 2010; Gino and Margolis 2011).
H3a: Description of a prevention goal focus compared to promotion goal focus or no goal focus in cases will lead to enhanced recognition of critical constraints.
H3b: Description of a prevention goal focus compared to promotion goal focus or no goal focus in cases will lead to improved forecasting.
H4: Description of a prevention goal focus compared to promotion goal focus or no goal focus in cases will lead to improved decision ethicality.
Participants were graduate students taking part in the Responsible Conduct of Research (RCR) ethics training course at the University of Oklahoma. One hundred and twenty two adults volunteered to participate in this study. Participants were recruited from various academic fields, including social (28.6 %), biological (6 %), engineering (53.6 %), health (2.4 %) sciences, as well as scholarship fields (9.5 %) (e.g., history, modern languages, philosophy, religious studies). Those volunteering were 58 % male with an average age of 26, SD = 3.99.
Design and Procedure
A 2 × 2 between-subjects design with a separate control group was employed in this study. The influences of social context (autonomy-supportive vs. controlling) and goal focus (promotion vs. prevention) on ethical decision-making were assessed via manipulated case content.
Participants were recruited to take part in this from an ethics training course. Completing the study materials was required as part of the training program, but inclusion of their data in this research was completely voluntary. Each participant signed a consent form prior to the start of the 2-day training program.
The study was incorporated into the training program on the second day, lasting a total of 1.5 h. Study materials were divided into two separate packets, with manipulated cases and low-fidelity questions in the first packet, and knowledge and reaction measures in the second. Presentation of cases and low-fidelity tasks was counterbalanced within conditions. Following the completion of packet 1, each participant returned the materials back to the study administrator and then received the second packet. Participants were asked to complete questions associated with each of the two cases, as well as a general reaction measure to the case content and the activities. Participants were not allowed to refer back to contents of packet 1. All materials were returned to the study administrator before resuming the regular training course.
Two cases involving an ethical dilemma were modified for use in this study. Each case incorporated ethical issues such as data fabrication and falsification, plagiarism, mentor–mentee relationship, and bid and contract practices. The cases, entitled Big Pharma and Side Business, were comparable with regard to structure and characters. These cases were realistically complex and each involved an individual in a subordinate position such as a graduate student and one superior (i.e., professor, laboratory director, or advisor) conducting biomedical research in a research institution. Case content for both cases varied across conditions with respect to the study variables (i.e. main character’s goal focus and social context). However, the case content manipulations were consistent across the two cases presented to a single participant. Each case was no longer than two pages.
Social Context Case Content
Social context was manipulated within each case to portray either an autonomy-supportive or a controlling work environment. The characters in each case were described as functioning in each of these environments and having to deal with unethical issues. An autonomy-supportive social context was presented as a comfortable working environment in which the advisor trusted his students, avoided controlling language, minimized salience of external incentives, and did not set deadlines. One example of this environment in the Big Pharma case was the following: “In fact, Dr. Davis prided himself on cultivating an atmosphere in his lab for students requiring little guidance and direction. He valued those students willing to take initiative and made it very clear to all incoming lab members that he will never take time to remind them of any deadlines, punish, or threaten them in any way.”
Contrary to the autonomy-supportive context, the controlling environment was defined by exerting pressure on subordinates via deadlines, potential punishments, strict guidelines, and lack of autonomy. One such example from the Big Pharma case reads as follows: “It is no secret, however, that all members of the Davis lab are constantly under pressure to perform, and that Dr. Davis is not the easiest man to work with. He also has very high standards for performance in his lab, including the requirement to publish at least 2 papers a year. In fact, Robin has heard from other students that he is known to throw people out of his lab for failure to perform according to his standards.” Please see Appendix 1 for a complete example of a manipulated case.
Goal Focus Case Content
Goal focus was manipulated in each of the two cases to include either a promotion or a prevention goal. The two types of goals were assigned to the main character in each case faced with an ethical dilemma. A character motivated by a promotion goal was portrayed as one with a desperate need to attain positive outcomes, succeed, make her supervisor proud by her accomplishments, and acquire as much knowledge as possible while working in the laboratory. In the Side Business case, for instance, the main character was falling behind in graduate school because her advisor required her to help him manage his side project. Her promotion goal was presented in the case in the following manner: “She figures that the great applied experience she is receiving can only bolster her future career aspirations and resolves to work harder in order to achieve her ultimate career goals.”
Conversely, a character assigned a prevention goal in a case was depicted as someone willing to do anything to prevent failure, circumvent angering her supervisor, and avoid losses at all cost. To provide an example of how the same case was manipulated to include prevention goal content, the following was developed: “Landing a spot in Dr. Michaels’ lab was a dream come true for Susan. She thoroughly admired his work and has learned a lot over the last 4 years. She resolved long ago to do whatever necessary to prevent disappointing Dr. Michaels and worked hard in order to avoid failure.”
The comparison control group received the same cases, but devoid of all social context and goal information.
EDM Transfer Task
Two low-fidelity simulation cases provided the bases for assessing the transfer of case-based knowledge. Each low-fidelity case contained one primary ethical dilemma, and was approximately 1–2 pages long. Participants were asked to take the role of the main character in each case and answer questions in the first person. One of the low-fidelity simulations, entitled Tight Schedule, described the woes of a graduate student torn between meeting her personal and professional goals (see Appendix 2). The other low-fidelity simulation, labeled Friendswood City Council, depicted a conflicted building contractor caught in a brawl between two city-council members. Following each case were a series of seven questions tapping utilization of sensemaking processes critical to EDM (Mumford et al. 2008). The questions required each participant to identify the main ethical dilemma, causes, constraints, and challenges, while also forecasting possible outcomes and making a final decision. Responses to these questions were later coded by trained raters for presence of sensemaking processes.
Coding Variables and Procedure
The variables coded included recognition of critical constraints, forecast quality, and overall decision ethicality. Each of these variables is further discussed below.
Recognition of Critical Constraints
Constraint criticality was assessed by four trained raters to identify the most crucial constraints acting on ethical decision-making. The extent to which the identified constraints were a hindrance, related to the ethical dilemma at hand, and of importance to the final decision were assessed on a 5-point Likert scale (1 = few critical constraints identified and 5 = many critical constraints identified). Interrater reliability, demonstrated through the intra-class correlation coefficient (ICC), or the extent to which multiple raters assigned a similar score to the same participant, was high, .84.
The extent to which respondents’ forecasts included thorough, detailed, complex, and thought-out predictions of potential outcomes was assessed using a 5-point Likert scale (1 = low quality and 5 = high quality). The interrater reliability (ICC) was .89.
Overall ethicality of respondents’ final decisions to the ethical dilemma was evaluated based on a 5-point Likert scale (1 = low ethicality and 5 = high ethicality). The final decision was judged to be of high ethicality if it exhibited the following characteristics: (1) regard for the welfare of others, (2) attention to personal responsibilities, and (3) adherence to/knowledge of social obligations. Interrater reliability for this variable was high, .85.
The above variables were coded within responses to the low-fidelity simulation questions. Four senior-level graduate students in Psychology served as raters for this study. All raters were required to undergo 20 h of training, as well as practice coding the specified variables according to the official coding manual. All raters were given a number of example responses to practice and calibrate their ratings. Several rater meetings took place prior to the final approval from the experimenter to code the participants’ responses.
Three 2-way between-subjects analyses of variance (ANOVAs) were conducted in order to assess the influences of social context and goal focus content information with respect to the three dependent variables: Constraint criticality, forecast quality, and overall decision ethicality. Comparison tests were also conducted in order to discern the nature of the differences between each level of the independent variables.
Correlation matrix for all EDM transfer task variables
1. Constraint criticality
2. Forecast quality
3. Decision ethicality
There was no main effect of goal information on recognition of constraint criticality, F(1,117) = .04, p = .84, and no interaction effect, F(1,117) = 1.51, p = .22. Thus, Hypothesis 3a was not supported.
A 2 × 2 between-subjects ANOVA was conducted to assess the influence of social context and goal focus content on forecasting quality. There was a significant main effect for social context, F(1,117) = 4.67, p < .05, η2 = .04. Comparison procedures specified that those who were given autonomy-supportive social context information within a case produced better quality forecasts (M = 3.05, SD = .12) than those who received controlling (M = 2.67, SD = .13) or no social context information (M = 2.62, SD = .21). There was not a significant difference in forecast quality between those receiving controlling and no social context information. These findings were consistent with Hypothesis 1b.
Hypothesis 2 stated that those receiving information about autonomy-supportive as opposed to controlling or no social context would exhibit improved decision ethicality. Decision ethicality did, in fact, vary by social context condition, F(1,117) = 3.51, p < .05, η2 = .05. Comparison analyses revealed a significant increase in decision ethicality for those who received the autonomy-supportive context (M = 3.26, SD = .74) compared to those who received no social context information (M = 2.83, SD = .81). There was not a significant difference in decision ethicality for those receiving the controlling social context (M = 2.98, SD = .72) and no social context, and a marginal difference between autonomy-supportive and controlling social context (p = .06).
Finally, it was hypothesized that participants receiving prevention goal information would develop more ethical decisions than those receiving promotion or no goal information. This hypothesis was not supported, F(1,117) = .01, p = .91. There was no significant interaction effect, F(1,117) = .04, p = .84.
The intention of this study was to examine case-based learning in light of case content attributes. Specifically, we explored the influences of social context and goal focus case information with reference to sensemaking processes—critical constraints and forecasting. As a consequence, we expected to observe changes in ethical decision-making.
The findings augmented a very limited research area and provided new insight into optimal case construction for case-based instruction. For example, we found that describing an autonomy-supportive social context in a case example led to improved recognition of critical constraints and better quality forecasts, culminating in greater decision ethicality. There are several potential explanations for the influence of autonomy-supportive social context on execution of sensemaking processes and EDM. One interpretation is derived directly from the predictions of self-determination theory (Ryan and Deci 2000). As discussed earlier, SDT suggests that the social context can facilitate satisfaction of three main psychological needs, competence, autonomy, and relatedness, with autonomy-supportive social context dealing directly with promotion of the need for autonomy. Thus, it is plausible that describing the setting of the ethical dilemma in a way that aligned with learners’ basic psychological needs led to improved execution of sensemaking processes and in turn, ethical decision-making (Vansteenkiste et al. 2004). Another reasonable explanation for these findings is that reading about an autonomy-supportive environment stimulated perceived internal locus of causality in the learners, whereas the controlling context is known to bring about feelings of lack of control (Grolnick and Ryan 1987; Vansteenkiste et al. 2005). Producing feelings of external locus of causality in our participants may have undermined their interest in the task and led to poor integration of material (Grolnick and Ryan 1987). Finally, previous research shows that autonomy-supportive environments produce increased interest and task involvement (Vansteenkiste et al. 2004). Perhaps reading about a pleasant climate resulted in heightened interest in the case content and the subsequent task, leading to better overall performance on the low-fidelity task. Readers may have also experienced more positive affect when reading about an autonomous working environment, resulting in better decision-making. In fact, numerous studies have indicated the powerful effect of positive affect on decision-making and problem-solving (Estrada et al. 1994; Isen 2001; Nygren et al. 1996). One such example comes from the work of Estrada et al. (1994).
Another finding emerging from this study showed that while the effect of goals was not as pronounced as that of social context, motivating people with goals aimed at preventing negative outcomes for themselves and others resulted in higher quality forecasts. There are a few possible explanations for this finding. As exemplified by previous research, prevention-focused individuals are very inflexible, risk-avoidant, careful, and vigilant (DeBock 2010; Higgins 1997, 1998). It is possible that the description of the prevention-focused character within a case example prompted a tendency in the reader to avoid future uncertainty by generating better, more thorough, future forecasts. It may also be that a description of the prevention goal put the readers into a more critical framework, one that promoted better evaluation of the ethical case scenario. Thus, the critical evaluation led to better integration of the case principles as evidenced by better future forecasts.
These findings have important theoretical implications, as well as implications for case-based ethics training development. One of the most important reasons for conducting a study of this nature was to investigate the significance of case content in case-based teaching and learning. In general, this is an under-researched topic that warrants more attention. While some information exists about what makes a “good” case, few researchers have explored the features of case content necessary for optimal effectiveness. This study provides empirical evidence suggesting that the setting in which the ethical dilemma unveils itself is not only a key aspect of cases, but when described to be supportive of characters’ need for autonomy, the case has capacity to promote ethical decision-making. This means that if case content is aligned with readers’ basic psychological needs, one can expect better integration of case attributes into case-based reasoning schema. Another implication refers to our finding regarding individual factors in case content. Specifically, certain attributes of case characters (i.e., goal focus) can prime critical analysis of cases, thus facilitating case-based learning. To supplement this, our results indicate that not only are certain types of case information critical for case-based learning (i.e., social context and goal focus), but the nature of those variables (i.e., autonomy-supportive environment and prevention goal focus) influences the extent to which the content is more or less effective. Finally, we also showed that readers are aware and discerning of even the most subtle contextual factors. This is an important implication because it further supports the powerful influence of case content.
These results have important implications for designing optimal cases for ethics training. Specifically, these findings suggest that ethics education courses should be concerned with the level of realism and expression in their case examples. The social environment/setting of the dilemma should be clearly portrayed to the reader in order to promote engagement and interest. It is also important to provide some indication of the character’s intent or motive. Such details will help the reader discern key factors and causes, leading to the construction of a more accurate mental model and promoting the sensemaking process.
There were a few identifiable limitations to the current study. First, generalizing these results to a wider population is problematic because the findings were based on a small, graduate student sample. We would, however, like to point out that our study dealt with cognitive outcome variables, making the representativeness issue less relevant. Nevertheless, the study should be replicated with a larger, more professionally representative sample.
Another limitation is that we only manipulated one type of contextual factor (social context) and one personal element (goal focus). Whether the social context is autonomous or controlling is just one characteristic of the social environment. Similarly, the goals people pursue are just one individual factor; many others have yet to be examined. Future studies should attempt to incorporate each previously identified contextual and individual factor (for a review of an extensive list of such factors see O’Fallon and Butterfield 2005) influencing ethical decision-making into case examples.
It is also possible that individual differences may have affected the perception and use of case material. Participants’ own prevention or promotion focus could have influenced their sensitivity to the prevention and promotion goals we embedded in the case examples. This should be controlled for in future replications of this study.
The final limitation refers to the measurement of our dependent variables, which relied on expert raters. Generating ratings is inherently subjective and thus may be criticized. However, the raters underwent thorough training and the interrater agreement coefficients were quite high, indicating a reliable scoring methodology.
With this study we have added to a very limited, yet exceedingly important, area of research. The findings demonstrated the need for a rich and realistic description of the social context (setting) in cases, with the autonomy-supportive environment resulting in pronounced effects on sensemaking processes and ethical decision-making. This information will, presumably, be of use to ethics case developers and others employing case-based instruction.
This research was supported by grant # 0931539 from the National Science Foundation. The project tile is: “Case-Based Reasoning and Ethics Instruction: Content and Processing Exercises for Effective Education.”