Critical Incidents in Ways of Experiencing Ethical Engineering Practice

Instructional Strategies in Post-Secondary Engineering Ethics Education

Ethics education is an essential element of engineering education (Mitcham, 2009) and a required student outcome for accredited programs (ABET, n.d.). A recent systematic review of post-secondary ethics interventions in the US found that engineering ethics instructors tend to prioritize learning goals associated with ethical sensitivity and ethical reasoning (Hess & Fore, 2018). Martin, Conlon, and Bowe (2021a) offered a more expansive list of 11 possible learning goals for engineering ethics, including additional behavioral and affective-oriented learning goals, such as moral agency and action, moral character and virtuous development, and moral emotional development. Katz (2019) identified 18 goals of engineering ethics education, which included some of these goals but also more direct goals, such as “disaster avoidance,” “question decisions, practices, and processes,” and “ability to understand ethical issues beyond individual context” (p. 101). Thus, there are numerous goals that engineering ethics instructors aspire towards.

To meet ethics learning goals, the most common strategies for teaching ethics to engineers involve introducing case studies alongside codes, rules, or standards (Haws, 2001; Herkert, 2005; Hess & Fore, 2018; Rottmann & Reeve, 2020). As, the following review illustrates, these approaches, though valuable, often lack sufficient grounding in the factors that influence ethics development in practice, pointing to the need for more practice-based research. Thus in briefly describing these three strategies, we continuously consider, “To what extent do these strategies adequately reflect real-world contexts or experiences?”

Traditionally, as a part of a case study, students practice making individual judgments (Bairaktarova & Woodcock, 2015; Hess & Fore, 2018). For example, students might practice “drawing the line” (Harris, Davis, Pritchard, & Rabins, 1996, p. 94) between acceptable and unacceptable behaviors and resolving conflict between different values. Thus, students might consider tensions between individual, professional, and societal values or needs (Herkert, 2005). Martin et al. (2021b) offered a typology of ethics case studies which included five different scales for designing cases. Their taxonomy described variation based on micro-versus-macro contextual foci. For example, the “sphere” scale calls attention to “individual” versus “societal” issues, which align with micro and macro-contexts, respectively. Other scales included veracity (hypothetical versus factual), timeframe (historical versus real), duration (brief versus lengthy), and student role (predefined versus open). Importantly, other scholars have offered distinct case study taxonomies for engineering ethics instruction (Davis, 1999; Huff & Frey, 2005) that draw attention to numerous other facets.

While case studies are a common instructional strategy, they are oft-critiqued for their simplification of complex situations (Bucciarelli, 2008; Conlon & Zandvoort, 2011; Lynch & Kline, 2000). Lynch and Kline (2000), for example, critiqued case studies that have students analyze actions of individual moral agents, thus omitting a description or discussion regarding how organizational culture shapes practice. To counter that critique, Martin et al. (2021b) found that engineering instructors in Ireland preferred using “immersive case studies” which they defined as “realistic, experiential, relevant, engaging, provocative, facilitated properly, including various stakeholders, integrating ethical alongside technical or legal considerations, [and] based on real or real-time data and documentation.” This preference aligns with recommendations by Bairaktarova and Woodcock (2015) who indicated there is a need for “more complex scenarios with the involvement of practicing engineers who experience ethical issues first hand.” When case studies are sufficiently complex and when they sufficiently account for context, case studies can “promote discussion and alignment with practice” (p. 22). Thus, case studies may be deemed appropriate insofar as they sufficiently represent real-world complexity and engineers’ lived experiences, but research unpacking such real-world complexity remains limited.

This same limitation may apply to another common instructional strategy: having students engage with codes of ethics (Hess & Fore, 2018; Rottmann & Reeve, 2020). Codes of ethics offer a form of ethical guidance for engineers (Davis, 2006). Historically, public safety, health, and welfare have served as foundational ethical standards for engineering professionals (Davis, 1991). Thus, organizations such as the National Society of Professional Engineers (2019) codified these ethical obligations with canons like, “Engineers, in the fulfilment of their professional duties, shall hold paramount the safety, health, and welfare of the public.” Similarly, the first “professional obligation” in the BMES or Biomedical Engineering Society (n.d.) code of ethics stated that engineers shall “use their knowledge, skills, and abilities to enhance the safety, health, and welfare of the public.” While some authors have recognized limitations of codes of ethics (Riley & Lambrinidou, 2015; Smith, 2021), we follow Davis in acknowledging that codes can serve as a guide for practice by enabling engineers to act confidently and with the weight of their discipline behind them. While these codes may have emerged from and are designed to influence practice, in themselves, they often remain abstract rather than concrete.

Finally, ethical theories also play a role in ethics instruction, though findings vary regarding their use. For example, based on a literature review of ASEE papers between 1999–2001, Haws (2001) argued that theoretical grounding was another critical instructional strategy, albeit, he found it was often missing in the manuscripts he reviewed. Katz’s (2019) work on mental models in engineering education explored methods of teaching engineering ethics and, similar to Haws, there was a notable lack of presence of ethical theory. Conversely, Hess and Fore (2018) found that the use of ethical theories as instructional strategies were present in 42% of the articles they reviewed. While ethics codes can enable one to understand norms of the discipline, ethical theories can offer distinct perspectives of resolving ethical problems. Thus, theory has a role in ethics pedagogy, specifically for introducing new ways of thinking about ethical engineering to engineering students. Theories can help students interrogate ethical issues in ways that applying ethical codes alone will likely not (Riley & Lambrinidou, 2015). We argue that theories, like other instructional strategies, also require a deep understanding of practice in order to help students translate theories into meaningful action.

Regardless of which instructional strategy one uses, we argue that it is important to understand the factors that promote ethical engineering practice. In particular, since ethics instruction is intended to support student development, engineering instructors need a better understanding of what types of influences lead to changes in how engineers experience ethical practice. With such an understanding, instructors can design and implement ethical instructional strategies that align with important developmental factors derived from workplace experiences.

Potential Factors Influencing How Engineers Experience Ethics in Practice

While research on ethics in practice is limited, existing research does suggest some potential factors that influence how engineers experience ethics in practice: (1) engineering disciplinary culture; (2) organizational or company culture; and (3) individual attributes.

Disciplinary engineering culture influences how ethics is taught to engineers (Martin et al., 2021a) and is a potential influence on how engineers practice ethics in the workplace (Davis, 1991). While culture is a complex and much debated concept, we use the term here following Schein’s work, which defined the culture of a group as “a pattern of shared” and “taken-for-granted assumptions” that is apparent through “observable artifacts and shared espoused values, norms, and rules of behavior” (Schein, 2010, p. 32). Smith (2021) argued that codes of ethics provide “a clear window from which to observe the coevolution of shifting professional norms for engineers.” Thus, codes of ethics are both a common instructional strategy and a window into engineering culture. Davis (1991) argued that these codes inform what it means to be an engineering profession and, thus, provide “morally permissible” standards by which engineers ought to act. In Davis’s view, the codes offer guidance for practice, and thus serve as a cultural artifact that can influence how engineers act ethically in practice.

Similarly, organizational culture can influence compliance and ethical behavior for any individual within an organization (van Steenbergen, Ellemers, van Rooij, & Sokol, 2021). For example, with respect to engineering, Vaughan (2016) explored the challenger disaster and illustrated how organizational culture ultimately influenced decision-making in this case. Vaughan argued that NASA “normalized” high-risk technology through an early risk assessment wherein an “SRB [solid rocket booster] joint deviated from the performance predicted by the design” (p. 65). Yet, the company “negotiated the risk of the SRB joints” and then “normalized the deviant performance,” wherein what the team “first identified as a technical deviation was subsequently reinterpreted as within the norm for acceptable joint performance, then finally officially labeled as an acceptable risk” (p. 65). This eventually led to “decision-making patterns pertaining to technical components” wherein unacceptable risks were deemed acceptable. Thus, Vaughan highlighted how the organizational culture developed over time and, eventually, led to risk assessment judgments that were less conservative (and, as Vaughan argues, this less conservative decision-making trend was one primary reason the disaster occurred).

Finally, research suggests that ethical practice is also influenced by individual attributes such as moral judgment capacity (Borenstein, Drake, Kirkman, & Swann, 2010), character traits/virtues (Schmidt, 2014), or level of moral development (Gibbs, 2019; Rest, Bebeau, & Thoma, 2014). Moral judgment has historically been oft studied in organizations, and we use this as an example here. Moral judgment involves making decisions based on select frameworks or ethical theories (Van De Poel & Royakkers, 2011). One specific type of moral judgment involves risk assessment. As Whitbeck (2011) argued, risk assessment involves “a judgment about the acceptable level of risk” and “is implicit in any tradeoff of safety against other considerations” (p. 184). Thus, any event that involves risk assessment also involves moral judgment. Judgment, however, is, in part, personal, in the sense that it is based on many factors, including one’s “education and experience” (Davis, 2012, p. 792). Thus, individual judgment (like any individual attribute) can vary based on the diverse experiential histories of individuals.

Each of the above potential influences may act in concert to influence how one might view or practice ethics. For example, organizations may espouse certain norms for ethical practice, but these may require interpretation based on one’s discipline and prior experiences. To this point, Vaughan (2016) argued that engineers in the Challenger case followed “practice rules” which were “operating standards consisting of numerous ad hoc judgments and assumptions that are grounded in evolving engineering practice” and “are experienced driven” (p. 201). Engineers in the Challenger case thus made ethical judgments by balancing risks and company norms, and this balancing act was informed by their individual attributes, such as their perceptions of what constitutes acceptable levels of risk or what they viewed as normal or appropriate behavior within their organizational context.

Like organizational norms, engineering codes of ethics by themselves do not always provide event-specific guidance. For example, in Smith’s study, “[engineers’] professional norms held them accountable to both their corporate employers or clients and the safety, health, and welfare of the public, providing no guidance about what to do when those two accountabilities conflict” (p. 165). Thus, interpretations of a code of ethics may vary by individuals. In the case of a code of ethics, for example, two engineers might reach a different conclusion regarding whether to prioritize one’s employer or society when such a prioritization is needed to move forward in a project.

Aligned with this understanding, Kim (2022) described the dynamic interplay between organizational culture and individual attributes in ethical practice of engineers. Based on a synthesis of the different theories of moral formation (including Haidt, Hoffman, and Kohlberg; see Gibbs, 2019 for an overview of these moral developmental theories), Kim argued that engineers can leverage environmental factors to inform or shape the way they or others practice ethics in their workplace. For example engineers who are in similar career stages and working in similar environments can behave differently, thus reflecting their individual differences.

Summary of Literature Review

There are numerous potential influences that inform how individuals experience ethical engineering practice in the workforce. Yet, there are relatively few studies and contextually rich examples specifically on ethical engineering which are explicitly utilized to inform current ways of teaching ethics to engineering students. Taken together, these challenges support the need for more detailed investigations of ethical development in engineering practice. Through such investigations, we can better align practices for teaching engineering ethics with workforce experiences. To this end, in this study we focus on one specific industry, health products, to address the research question, “What types of critical incidents influence engineers’ ways of experiencing ethical engineering practice?”

Positionality of authors

We designed this positionality statement based on the three themes offered by Hampton, Reeping, and Ozkan (2021), including (1): Disclosing Identities; (2) Disclosing Prior Experiences and Opportunities; and (3) Disclosing Journeys. Each author reflected on questions associated with each theme. Thereafter, we identified similarities and differences in author reflections.

We identified two primary similarities among authors: (1) research perspectives and (2) lack of workforce and health products experiences. First, we each attained (or, at the time of collecting and analyzing data, were currently pursuing) a PhD in Engineering Education from the same university. This training likely influenced our framing of the study scope, methodological strategies, presentation of findings, and direction of implications. To the latter point, we shared an express desire to improve engineering education strategies, both for ourselves and the broader community. Second, we each lacked experiences in the health products industry and had limited engineering workforce experiences. While this is a limitation, we also perceived it as a strength in that we were not overly biased as we did not bring preconceptions of what constitutes ethical engineering practice in this context nor what promotes changes in ethical practice. While we had limited experiences, we each felt that we brought connections and potentially transferable experiences to the study context, such as Hess and Kim who have previously co-taught or assisted teaching in biomedical engineering, Fila who participated in a biomedical engineering course as part of his undergraduate studies, and Kim who had prior training in chemical engineering and which she felt was linked to the biomedical applications.

We identified two primary differences among authors: (1) motivations for conducting this work and (2) demographic and cultural backgrounds. From the outset of the study, Hess expressed a primary desire to identify ways by which engineering ethics manifests both commonly and uniquely across disciplines; Kim expressed a desire for learning to conduct engineering education research, generally, with specific interests in engineering ethics and practice; and Fila served as a lead phenomenographer on the overall study and brought ambitions to apply phenomenography to a new context (health products) and phenomenon (ethical engineering practice). With respect to demographic and cultural backgrounds, two authors (Fila and Hess) were white males from the US and one author (Kim) was a female from South Korea; Fila and Hess were fluent in English whereas Kim was developing English fluency; the authors brought disciplinary backgrounds in civil engineering (Hess, BS & MS), chemical engineering (Kim, BS & MS), and electrical/computer engineering (Fila, BS & MS); and Hess and Fila were investigators on this project whereas Kim was pursuing a PhD and engaged in this work as a graduate research assistant and for graduate funding support.

Context of the Study and Justification

We recruited participants from the health products industry for three reasons. First, ethical issues in the industry are both relevant and prevalent, given their high visibility in the news (De Dobbelaer, Van Leuven, & Raeymaeckers, 2017), the potentially high positive and negative impacts resultant from ethical decision-making in this context (Lipworth, Kerridge, Morrell, Forsyth, & Jordens, 2015), and the necessity of adhering to industry regulations despite their oft-shifting nature (Blakely et al., 2022). Second, the focus on a single industry allows sufficient focus on ethical practice without undue variation across industries. Finally, the health products industry features a variety of engineering disciplines (see Table 1 below). We explored the health products industry by selecting participants from three different companies, each representing a different area of the health products industry (orthopedics, pharmaceuticals, and medical devices). By selecting three companies, we were able to explore a cross-section of the overall industry with variation in topic specialization, job role, and background. While we selected participants from this industry context, many participants brought forth experiences outside of health products. Thus, the critical incidents that we share in this study are primarily but not exclusively from health products.

Critical Incident Technique or CIT

We used CIT (Butterfield et al., 2005; Flanagan, 1954) to investigate the research question, “What types of critical incidents influence engineers’ ways of experiencing ethical engineering practice?” CIT focuses on building a “functional description of an activity” (Flanagan, 1954, p. 336) through expert or participant review of experiences or events relevant to that activity and by identifying critical incidents related to this activity. Here, the activity was changing one’s lived understanding or way of experiencing ethical engineering practice in the workplace. Critical incidents represented key events, contextual features, or experiences that influenced such ways of experiencing.

This study builds on a prior phenomenographic study which investigated variation in ways of experiencing a phenomenon, ethical engineering practice. The study was guided by a non-dualist ontology, a key aspect of phenomenography (Bowden & Green, 2005; Marton & Booth, 1997). In this context, a “way of experiencing” exists as a relationship between the individual and a phenomenon and it is informed by individual-phenomenon interactions, including what the individual brings to those interactions and the features of the phenomenon available throughout interactions. Thus, one’s way of experiencing ethical engineering practice is not simply what one understands about the phenomenon (e.g., organizational norms, codes) nor simply the features of their encounters (e.g., types of ethical situations, actions taken), but dialectic between the two. As ethical engineering practice represents a complex relationship between an individual and a phenomenon that evolves over time, critical incidents may take many forms. For example, in a similar study, critical incidents spanned features of environments, outcomes of encounters, and approaches taken by individuals (Fila & Hess, 2018).

In CIT, critical incidents are also noted as events and features that can be determined as critical by expert observers (Butterfield, Borgen, Amundson, & Maglio, 2005; Flanagan, 1954). Often, participants identify critical incidents directly, including incidents that they have experienced or have observed in others. Ways of experiencing are complex and typically best extracted by external analysts through in-depth interviews and rigorous, iterative analysis (Bowden & Green, 2005; Marton & Booth, 1997). As such, participants are not necessarily well-positioned to identify changes in ways of experiencing ethical engineering practice in themselves or others. Given these considerations, this study relies on extraction of critical incidents from interviews with participants.

The interviews in this study featured prominent discussions of experiences related to a phenomenon (ethical engineering practice), thorough discussion of conceptualizations of the phenomenon, and explorations of connections between one’s way of experiencing and conceptualizations. In the context of applying CIT, the “expert observer” was the team of researchers analyzing interviews. The tool that we utilized to observe critical incidents was the interview itself. The critical incidents were thus accounts detailed and synthesized within the interview, wherein participants thoroughly described experiences and encounters with ethical engineering practice.

Data Collection

To recruit study participants, we distributed a short survey inquiring into participants’ background (see Table 1 for requested demographic information), availability to participate in an interview between 60-90 minutes, and contact information. We distributed the survey through contacts representing one of three companies in the health products industry. Company foci included orthopedics, pharmaceuticals, and medical devices (41, 40, and 35 survey responses were recorded from each subsector, respectively).

Among the survey completers, we purposefully selected interview participants with a goal of (1) stratifying demographics and background based on survey responses (refer to Table 1) and (2) capitalizing on potentially important variables not captured via stratification. As our sampling was constrained by the survey respondents, demographic responses do not represent equal distribution across variables.

Three lead interviews engaged in data collection, including two authors (Hess and Fila) and a co-investigator on a grant associated with this study (Zoltowski). As another measure of consistency, a co-interviewer (Kim) participated in most interviews. To facilitate synchrony among interviewers, we performed multiple pilot interviews, with each lead interviewer leading at least one pilot interview and then debriefing as a team. Throughout the process of interviewing, the team continually met and discussed any discrepancies, challenges, or promising aspects.

The interview started with a background section intended to provide a brief history of the participant’s practice. The background section aimed to develop rapport between the interviewers and interviewee. Next, participants described experiences with ethics in their practice. Here, we asked participants to share specific experiences related to ethics in engineering. Most participants shared two to three experiences they had in their workplace and, through an intensive line of question, reflected on these experiences in detail.

Third, participants shared their conceptual understanding of “ethical engineering practice.” We asked a staring question about their conception and several follow-up questions to further unpack this conception. Notably, this is also the section where we systematically asked change-oriented questions, including, “How have your views of ethics in engineering changed over time?” and “What experiences do you believe contributed the most to your understanding of ethics in engineering?” Responses to conceptual questions either elicited new incidents or provided context for incidents described earlier in the interview.

Finally, the interview closed with a summative section, wherein participants shared concluding thoughts, chose a pseudonym, and asked interviewers questions.

The interviews were audio-recorded and later transcribed using a third-party software. These automated transcriptions were later checked for accuracy and anonymity by Kim.

Data Analysis

We started analysis by extracting critical incidents from interview transcripts. To identify incidents, we utilized CIT (Butterfield et al., 2005; Flanagan, 1954). One researcher reviewed all interview transcripts as a whole and identified potential critical incidents. This process was guided by criteria identified by Butterfield et al. (2005), including (1) antecedent information, which provided context for a critical experience, (2) detailed description of influencing factor(s), which illustrated an experience or series of experiences that influenced (i.e., changed or solidified) an individual’s way of experiencing ethical engineering practice, and (3) outcome(s) of the incident, which depicted at least one clear shift, refinement, or reaffirmation in one’s way of experiencing ethical engineering practice. These incidents ranged in detail from one paragraph to multiple pages.

After Kim identified potential critical incidents, Hess or Fila reviewed each potential incident. While reviewing, Hess and Fila identified incidents that were potentially weak, incidents they felt should be removed, or incidents that they felt could be divided into multiple incidents. After discussing suggestions, we chose to retain 135 critical incidents. Later, we reduced these incidents to workforce-related incidents alone, thus omitting 29 incidents from further analysis. Thus, 106 incidents became the final units of analysis for this study.

With these 106 critical incidents, we next performed content analysis (Krippendorff, 2012) to address our research question. One coder (Hess) initially focused on a subset of interviews and began grouping similar critical incidents without relying on existing literature or frameworks. As incidents were grouped together, the coder began narrating incident types, noting common factors present in incidents that influenced how one experienced ethics in their practice. Similar incidents were then grouped into broader categories.

Once coder 1 generated an initial draft of influence types and categories for 25 interviews, coders 2 and 3 (Kim and Fila) utilized the suggested categories and types as a deductive framework and coded incidents from these interviews without looking at coder 1’s application of codes. Next, all researchers discussed discrepancies between their coding of the 25 interviews through several meetings until they reached complete agreement for the grouping and description of incident types and categories. Finally, the research team repeated these steps with the remaining 18 interviews.

Thus, we employed an iterative coding procedure wherein the description of the categories and types evolved throughout the review, re-categorization of incidents in types and categories, and discussion processes. We developed agreement on the categorization of incidents extracted from all 43 interviews into incident types and categories.

In this study, we do not provide a synthesis of outcomes resultant from critical incidents. At least one outcome was a requirement of critical incident retention, but the incident types and categories reported in this study focus on influences. Nonetheless, in the results sections, we report exemplary incidents, and the incidents we describe generally contain outcomes resultant from individual critical incidents.

Credibility and Trustworthiness Procedures

As critical incident technique has moved away from its positivist roots, at least nine methods of checking credibility and trustworthiness of critical incident studies have been developed to ensure critical incidents are authentic to and robustly represent the domain of study (Butterfield et al., 2005). These methods, or what Butterfield et al. (2005) called “checks,” include the following:

1. Extracting critical incidents from data by an independent expert

2. Cross-checking critical incidents and incident categorization by participants

3. Categorizing subsets of critical incidents by independent judge(s)

4. Ensuring saturation of incident categories

5. Reviewing incident categories by field experts

6. Ensuring sufficient participation rate in each incident category

7. Scrutinizing results based on extant literature and theory

8. Ensuring accurate data recording/transcription

9. Checking interview fidelity by independent expert(s)

We employed eight of these nine methods. We omitted check number two, participant cross-checking, as checking critical incident categories and specific incidents was infeasible for every participant. Due to the nature of the target domain (changes in one’s way of experiencing ethical engineering practice), it was more relevant to have the research analysts immersed in the data ensure the relevance of the incident categories and check the fit between participants’ critical incidents and their appropriate classification into incident types and categories.

We generally met the criteria for each check set forth by Butterfield. One exception was to number six, wherein Butterfield et al. (2005) indicated that 25% of participants ought to share incidents associated with a category. For context, in this study, we present both incident categories and incident types, wherein we grouped multiple types within each category. At the category level, Categories 1, 2, and 3 meet the threshold. In contrast, Category 4 (Ethical Failures) included 20% of respondents and Category 5 (Mentorship) included 16% of respondents. The incident types nested within categories generally did not meet this threshold. Thus, the categories that we present in this may be deemed more trustworthy than the individual incident types. Nonetheless, we present incident types as they offer what we feel is important specificity and they convey the variation of influences associated with each category.

In addition to the traditional checks of CIT credibility identified above, we employed two additional methods: pragmatic validation (Walther, Sochacka, & Kellam, 2013), communicative validation (Walther et al., 2013), and persistent team approach. To achieve pragmatic validation, we employed a consistent lens throughout analysis by focusing on the practical use of the findings in engineering education. To achieve communicative validation, we elicited feedback from members of our research team and advisory board. Moreover, we reported early results from a subset of participants and received feedback from the engineering education community (Kim, Fila, & Hess, 2020). With respect to persistent team approach, we continuously and iteratively tasked three researchers with coding data independently. We were able to collaboratively reach agreement among all coders for any discrepancies and the identification and categorization of incidents thus represents shared agreement across all three authors.

Category 1: Cultural Immersion

This category includes incidents wherein participants learned from working in and observing multiple facets of a culture. We identified and grouped four incident types here: (1) Workplace cultures, (2) Role immersions, (3) Cumulative experiences, and (4) Cultural comparisons.

Category 2: Interpersonal Encounters

This category included incidents wherein engineers learned from interactions and relationships with others in the workplace. This category included five incident types: (1) Questionable behavioral observations; (2) Difficult collaborations; (3) Distinct perspective observations; and (4) Non-belonging encounters.

Category 3: Ethical Actions

This category represents engineers responding to challenging ethical situations. The four incident types that we grouped to this category included: (1) Steadfast actions, (2) Uncertain actions, (3) Inconvenient actions, and (4) Critical questions.

Category 4: Ethical Failures

Incident types in this category reflected participants grappling with an unexpected event that led to (1) Personal failures or (2) Technical failures.

Category 5: Mentorship Events

Incidents in this category found individuals learning from mentoring or provisions of guidance through (1) Receiving mentorship, (2) Providing mentorship, or (3) Managing employees.

Cultural Immersions

Our findings indicated that contextualized and immersive experiences in workplace cultures were the most common type of critical incident described by participants. Many authors have critiqued the decontextualized nature of instructional approaches in engineering ethics education, particularly when approaches oversimplify the complexity of practice (Bucciarelli, 2008; Conlon & Zandvoort, 2011; Lynch & Kline, 2000). The pervasiveness of the workplace cultures incident type highlights the import of guiding students to become aware of the influence of context and culture undergirding their learning experiences. Thus, in engineering ethics education, it is important to design and implement instructional strategies that authentically integrate the sociotechnical complexities of engineering practice and help students become aware of the potential influence of overarching cultures on their ethical views and practices (Cech, 2014; Martin et al., 2021a).

Engineering faculty members’ mental models suggest that workplace experiences are a primary domain where students learn ethics (Katz, 2019). Thus, engineering courses and programs can promote student awareness of the import of workplace cultures by encouraging internship and co-op experiences and ensuring these experiences provide opportunities for critical incidents to manifest. For example, in alignment with the role immersions incident type, universities might encourage students to participate in co-ops/internships across different branches of an organization (e.g., regulatory versus validation versus manufacturing). These role shifts will also position students to (1) compare cultures by enabling reflection on the sub-cultures in a single company and across organizational branches and (2) develop cumulative experiences which can help them become more confident in their future ethical engineering practice.

Engineering courses and programs may also apply the above considerations by intentionally fostering student reflection on sub-cultures in and across their institution. Feister, Zoltowski, Buzzanell, Zhu, and Oakes (2014) found that student groups across four institutions approached ethical reasoning in different ways, albeit, in large alignment with their respective institutional norms and values. Similarly, Martin et al. (2021a) argued that “the beliefs and practices of individual instructors are impacted by institutional measures and policies set by accrediting bodies, as well as by the cultural milieu in which they were educated or currently teach” (p. 59). Thus, instructors might prompt students to reflect on how the sub-cultures they experience within and across their organization, including at the institutional, departmental, and course levels, influence their ways of thinking about ethics.

Engineering courses and programs can also merge incident types nested within Cultural Immersions with common engineering ethics instructional strategies. For example, case studies have been a common strategy for teaching engineering ethics for decades (Haws, 2001; Herkert, 2000; Hess & Fore, 2018). Instructors may prefer more complex and immersive case studies (Martin et al., 2021b), but, when lacking industry experience, engineering faculty may feel unable to effectively include ethics into engineering curriclum (Bielefeldt, Polmear, Knight, Swan, & Canney, 2018). Thus, faculty tend to teach ethics in alignment with their own beliefs (Martin et al., 2021a) or mental models (Katz, 2019). Hence, integrating real-world complexity in sufficient ways can be a challenge, particularly if faculty bring limited workforce experiences. We suggest that instructors employ the thick description of engineers’ experiences with ethics (i.e., the incidents in this study) to design and implement engineering ethics case studies. We propose different ways to facilitate the design of such case studies below.

Instructors may build on role immersions by employing role-play within case studies, a common strategy utilized in the delivery of engineering ethics case studies already (Boudreau et al., 2016; Brummel, Gunsalus, Anderson, & Loui, 2010; Prince, 2006). Similarly, instructors may directly share the critical incidents with students, prompt students to form groups of two or three, task individual students to take on the role of a primary stakeholder, and then task students to reason through how their stakeholders might collaborate through the issue. If instructors find incidents in this study are not applicable to their instructional contexts, they could follow our same research procedures to elicit workforce incidents from practitioners in their discipline. Other scholars have extracted incidents and used them to inform ethical teaching. For example, Rottman and Reeve (2020) extracted incidents associated with equity and ethics among practitioners and designed pedagogy therefrom. Instructors might even have practitioners share their critical incidents via stories directly with students.

Role-play can be time-intensive to design and implement, and thus related but less intensive “participatory” approaches to case studies, wherein students imagine stakeholder perspectives (Huff & Frey, 2005), can be utilized. To this end, in two contexts (a graduate-level biomedical engineering ethics course and a multidisciplinary engineering ethics courses) we have developed and implemented a brief (i.e., around five-minute) participatory case based on Iowa’s personal failure incident. We provided a one-paragraph summary of the case, emphasized that it was based on a workforce critical incident, and prompted students to imagine what they would do if they were in Iowa’s situation. Students seemed to find the framing of the case (i.e., a real-world critical incident with tensions between personal and organizational values) salient and quickly became engaged.

While role-play may facilitate a greater depth of engagement with individual stakeholders, participatory case studies and perspective-taking activities can facilitate a greater breadth of engagement with stakeholder perspectives or values. With either instructional strategy, by embedding multiple role-play or participatory experiences across a program or course, instructors will help students imaginatively experience a multitude of ethical issues, thereby providing students with cumulative experiences.

Cultural comparisons also manifested when individuals compared cultural norms over time. To this end, instructors may prompt students to reflect on their personal views of ethics and compare how these views have shifted during their collegiate career. Such reflections might also be captured in a way to generate assessment evidence that can guide formative programmatic decision-making. Such assessment might also be beneficial if assessment data indicates shifts in unanticipated or undesirable directions. For example, if institutions find that learners tend to disconnect ethical engineering from public concerns (Cech, 2014), and if this is an ethical value that the institution aspires to cultivate, the institution can then identify important time points that require refinement. As many instructors have uncertainty regarding how best to address the ABET ethics criterion (Barry & Ohland, 2012), such assessment and resultant improvements would also be invaluable for accreditation.

Interpersonal Encounters

Interpersonal Encounters found participants observing questionable behaviors, engaging in difficult collaborations, observing distinct perspectives, and experiencing non-belonging events. Each of these incident types can be harnessed in engineering curriculums by prompting student engagement with how professionals’ experience ethical engineering practice. Like the considerations described in the preceding section, one can literally observe ethical engineering practice through co-ops or through real-world case study materials (Huff & Frey, 2005) or instructors can design case studies to foster collaborative decision-making (Loui, 2000). To ensure that such collaborative acts require difficult collaborations, instructors can imbue dissonance by requiring that students attend to challenging or non-normative facets of a case. In tandem, instructors may extend the non-belonging encounters by encouraging students to consider the perspective of a minority in a case study. For example, to continue the personal failure incident shared by Iowa, instructors might task students to consider how an environmentalist in a company which deprioritizes environmental impacts might respond when the environment is deprioritized. Alternatively, building on the Angela non-belonging encounter, companies might prompt students to consider how a female engineer might respond to a client’s request for support from a male engineer or, conversely, how a male engineer may respond to a request for a female engineer.

Acting Ethically

This category highlights the import of creating opportunities for students to behave or act ethically. Many extant instructional modalities require students to act and thus provide ripe opportunities to replicate incident types identified in this category. Embedding ethical frameworks and reflection in extant experiential curricular components, such as capstone design or service-learning, can be a relatively straightforward modality for embedding ethical action in engineering curriculum. Within these events, instructors might make space for students to reflect on ethical experiences before-action, in-action, or even on-action (Schön, 1987). Instructors might even encourage such reflection-on-ethical-action outside of the curriculum by encouraging students to first engage in an extra-curricular engineering organization and second reflect on extra-curricular engineering experiences. To this point, Wittig (2013) found that students who engaged in Engineers Without Borders found it to be a more influential ethical learning experience compared to course-based engineering curricula.

Finally, similar to above, instructors might also encourage students to engage in co-op or internship and reflect on the ethical implications of their work, including how they can behave more ethically. Howland, Kim, and Jesiek (2022) found such work experiences provide critical opportunities for students to learn engineering ethics even without specific guidance. We expect that with more experiences to act ethically, coupled with targeted guidance and mentorship, students will be able to learn how to practice ethical engineering more effectively.

Ethical Failures

Ethical Failures found participants grappling with technical systems that performed in unexpected ways (thus experiencing technical failure) or engaging in behaviors that did not align with one’s personal values (thus experiencing personal failure). No participants discussed extreme failure experiences, and few incidents involved severe misconduct, which is perhaps due to the sensitive nature of such types of failure and participants’ reluctance or inability to share such sensitive incidents. Nonetheless, the few instances of failure that participants described seemed to have a powerful influence on participant’s ways of experiencing, thus supporting calls for focusing on or fostering ethical failure in engineering classrooms (Colby & Sullivan, 2008; Eyler, 2018). Connecting back to earlier suggestions, instructors may consider utilizing role plays or participatory case-based instruction to help students take on practitioner perspectives and imagine how they might fail and then react in these scenarios (Huff & Frey, 2005).

Mentorship Events

Finally, Mentorship Events found participants receiving mentorship, providing mentorship, or struggling to identify how best to manage employees. Building on the receiving mentorship incident type, engineering courses and programs might purposefully design strategies to help students communicate ethical challenges and receive guidance in safe ways and at critical junctures. Engineering curriculums offer no shortage of potential mentors or mentees, but we encourage instructors to harness these potential roles by considering students as both potential mentors and mentees. While such mentoring structures can be informal, instructors seeking to promote and celebrate more widespread mentoring opportunities could emulate extant peer mentoring ethics initiatives such as the Ethics Bowl (Cruz, Frey, & Sanchez, 2004).

Limitations and Future Work

The ultimate motivation for this work was to better align post-secondary engineering ethics instruction with workplace experiences. Additional levels of analyses with these data can continue to support such efforts. For example, in the future we intend to identify resultant outcomes of critical incidents and map the outcomes to incident types and categories. We also recognize that this study was situated within the health products industry. Additional research is needed to identify transferability of findings to other industrial contexts and to other engineering disciplines not represented in our sample. Third, we did not account for differences by demographic factors. While the recruitment procedures sought participant diversity, a focus on how critical incidents vary along select demographic factors was beyond the scope of this study. Finally, we observed that incidents often overlapped, but we did not systematically code for such overlap. For example, interpersonal considerations, individual actions, failure experiences, and mentorship events seemed to be common aspects throughout the incident types associated with Cultural Immersions. Future research is needed to better understand and empirically support category-category and type-type interconnections.