There has been an explosion of articles on professionalism in medical education in recent years. In January 2006, we conducted a Medline search using the terms professionalism and medical education, and we observed a 10-fold increase in the volume of articles published during the last 15 years, from 28 between 1991 and 1995, to 98 between 1996 and 2000, to 273 articles between 2001 and 2005.
In 2005 alone, numerous articles appeared in the leading academic medicine journals covering wide-ranging topics such as professionalism teaching and learning,1–3 professional and unprofessional behaviors of students and interns,4–7 faculty development for teaching and assessing professionalism,8 assessment of professionalism through multisource feedback,9,10 assessment by peers,11,12 self-assessment,13 and assessment by patients.14
The publication of a number of recent reviews15,16 and an edited book17 clearly indicate that the assessment of professionalism is a hot topic within medical education. Much of this literature is of a technical nature, focusing on the reliability, validity, and practicality15 of assessment methods, and its primary focus has been on the assessment of observable behaviors.18
In his recent critical commentary on this popular topic of assessing professionalism, published as the final chapter of Measuring Medical Professionalism 17, Hafferty18 is right to stress the disparity between professionalism and professional behavior by asking the question: “Do we want physicians who are professional, or will we settle for physicians who can act in a professional manner?” An assortment of definitions for “professionalism” exist in the academic medicine literature; they commonly include various elements such as professional knowledge, attitudes, values, virtues, skills, and behaviors.19 Interestingly, the attitudinal components of professionalism have mostly been ignored in the professionalism assessment literature, resulting in educators focusing primarily on the assessment of behavior. However, by focusing on behavior alone, it is easy for educators to slip into the trap of viewing individuals’ actions as representative of their underlying attitudes, and vice versa—a problem that we discuss in this article.
In this viewpoint article, we aim to build on Hafferty’s18 critical commentary using theoretical insights from sociocognitive psychology. We begin by discussing the relationship between attitudes and behaviors and continue with a critical examination of the phenomenon of “faking it”20 (students faking professional behaviors for observers). This is contrasted with students being labeled “unprofessional” by observers because of the observers’ ignorance of important contextual circumstances affecting the students’ behavior. Taking into account these issues and the criticisms of professionalism assessment employing behavioral measures alone, we conclude with some practical implications for the future assessment of professionalism.
What Is the Relationship Between Attitudes and Behaviors?
An attitude can be defined as “a favorable or unfavorable evaluate reaction toward something or someone, exhibited in one’s beliefs, feelings, or intended behavior.”21 This definition implies that attitudes have multiple components, and multicomponent models of attitudes have traditionally included three aspects: affect, behavior, and cognition.22 Affect refers to feelings and the emotional component of attitudes, behavior refers to one’s overt actions, and cognition refers to an individual’s beliefs and thoughts.22 Although recent psychological research has focused on the two-component affect–cognition model to discuss attitudes,22 much social psychology research has centered on the relationship between attitudes and behaviors.23
Wallace et al23 conducted a meta-analysis of 797 studies published between 1937 and 2003 (with 1,001 effect sizes) to examine the relationship between attitudes and behavior. The studies examined a wide range of behaviors related to a number of lifestyle components, such as health, voting, education, career, finance, environment, prosocial behavior, religion, and social interaction. When the authors examined behaviors in relation to attitudes, they found that the mean attitude–behavior correlation was 0.41 when individuals experienced a mean amount of social pressure to carry out a specific behavior and when that behavior was of mean difficulty. However, when individuals experienced social pressure one standard deviation higher than the mean to perform a more difficult behavior (behavior one standard deviation above mean difficulty), the mean attitude–behavior correlation dropped to 0.30. This suggests that an individual’s attitudes only account for 9% of the variance in his or her behavior when strong social pressures exist to behave in a particular way and when the behavior itself is difficult to execute.
Sociocognitive models of behavioral prediction, such as the theory of planned behavior (TPB),24 have helped shed light on the complex relationship between attitudes and behaviors. The TPB suggests that an individual’s behavior is influenced by three sets of beliefs that combine to form a behavioral intention, which should then result in the target behavior if the individual has sufficient control to execute it. The three sets of beliefs are behavioral (beliefs about the consequences of the behavior), normative (beliefs about others’ expectations), and control (beliefs about features that may help or hinder the behavior). The more recent integrative model25 extends the TPB by suggesting that in addition to behavioral intention, environmental constraints and individuals’ skills and abilities also influence behavior. The key point from the sociocognitive psychology literature is that attitudes alone are poor predictors of behavior when external constraints, such as social pressure to behave a certain way and perceived difficulty of the behavior in question, are relatively strong23; arguably, these situational variables are particularly salient within medical practice.
How Acceptable Is It for Students to “Fake” Professional Behaviors?
When we follow codes of conduct or rituals of decorum, we are often just play acting, acting appropriately in outer conduct, irrespective of what is in our hearts.
—N. Sherman, British Journal of Educational Studies, 2005
Attitude strength—the certainty, importance, and centrality of our attitudes—has been found to moderate the attitude–behavior relationship, with stronger attitudes being more predictive of behavior than weaker attitudes, which are relatively unstable and, therefore, amenable to change.26 Consequently, it is reasonably straightforward for medical students (principally those with weaker attitudes) to behave in a professional manner, even if their underlying attitudes are inconsistent with professional behavior. In other words, it is perfectly possible for students to fake it,20 particularly when being observed, a phenomenon known as the Hawthorne effect.27
Indeed, in the first author’s own research exploring the acceptability of multisource feedback on students’ professional behaviors,9,28 students stated that they had play acted professional behaviors for their assessors to yield rewards such as praise or high grades. One of our first-year medical students stated: “We also know people outside and when we come into PBL [problem-based learning] class, people actually change and you can see they’re trying to tick that [top] box today.”28
Many medical educators hope that by constantly monitoring students’ professional behaviors, the students will eventually come to internalize appropriate attitudes in accordance with cognitive dissonance theory.29 This theory suggests that when we behave in a manner contrary to our attitudes, we experience dissonance, which we seek to reduce, either by changing our attitudes to match our behaviors or vice versa. Reflecting on the writings of Immanuel Kant, Sherman20 puts it another way: “decorum can, in some cases, change inner states.”
Further exploring the possible disconnect between attitude and behavior, Hafferty18 contrasts professional behavior with professionalism, suggesting that the former is superficial, whereas the latter is deep and intertwined with identity. He suggests that professional behavior arising from the latter is more authentic; he contrasts this authenticity with impression management, whereby individuals manage others’ impressions of them by behaving in a particular way.30 From sociological and moral philosophy standpoints, Hafferty18 and Sherman20 question the extent to which impression management is appropriate. Reflecting on Seneca’s treatise On Favors, Sherman20 argues that if a disparity exists between an individual’s inner virtues and his or her outer conduct, the morally suitable response is to behave appropriately: “Sometimes a posed face or a politeness that hides one’s real feelings may be the morally appropriate response.” Although Hafferty18 suggests that it may be sufficient for physicians to appear professional, his writing clearly reveals a discomfort with this state of affairs.
Frowe31 discusses the role of trust within educational relationships. He argues that when one attempts to control another’s behavior (e.g., through assessment), the need for trust between truster and trustee becomes superfluous, and the ability of individuals to use their own discretionary judgment is reduced, damaging the very essence of professionalism. By attempting to control medical students’ professionalism through behavioral assessment, it seems relatively easy to change their behaviors in specific contexts. However, we can never be entirely sure that those professional behaviors reflect underlying professional attitudes. So, from an assessment perspective, we may pass students who exhibit professional behaviors but who, nevertheless, have unethical attitudes. As we discuss next, the reverse is also true: we might fail students with unprofessional behaviors who possess ethical attitudes.
When Is Unprofessional Behavior Not Unprofessional?
Behavior can be defined in terms of four components: action, target, timing, and context.32 Although definitions of behavior must include action (and preferably the target), timing and context are frequently overlooked, despite their importance.32 The following definitions of physician behavior, both originating from the physicians’ charter on professionalism,33 concern honesty. The first definition specifies the action and the target of the behavior, but the second also includes timing and context:
Physicians must be honest [action] with their patients [target].
Physicians must ensure that patients [target] are completely and honestly informed [action] before [timing] the patient has consented to treatment and after treatment has occurred [context].
Although most would agree that behavioral definitions incorporating all four components are more comprehensive and therefore superior, assessment criteria of students’ professional behaviors commonly exclude timing and context, despite the context-dependent nature of behaviors. Indeed, many authors18,34,35 argue that behavioral assessment should not be divorced from context, as the following study illustrates.
Ginsburg et al36 conducted interviews with 30 clinicians after the clinicians had watched five videotaped scenarios of professionally challenging situations. They were asked what they thought students should and should not do in these situations, and they were also asked what they would do themselves. The authors found little agreement between clinicians. Ethical principles such as honesty were defined differently across clinicians and within clinicians across different scenarios, suggesting that dishonest behavior could be interpreted as unprofessional or not unprofessional depending on the context.
Using the same video scenarios, Ginsburg et al37 interviewed 18 fourth-year medical students to find out what they would do in each of the situations and why. The authors found that students considered their actions by reference to abstract or idealized principles, such as honesty; by reference to the implications for themselves, such as grades; and by reference to motivations unacknowledged by the profession, such as obedience to authority. Although this study illustrates students’ hypothetical rather than their actual behaviors, it serves as a reminder of one of the key contextual determinants of behavior already mentioned: social pressure.
A recent article38 published in the British Medical Journal (BMJ) demonstrated that of 702 intimate examinations conducted on anesthetized patients by medical students at one UK school, 166 (24%) were conducted without explicit consent, behavior that would be defined by most as unethical. So, should we label these students “unprofessional,” or should we argue that they have unprofessional attitudes on the basis of their behavior? Drawing on the classic social psychology experiments conducted by Stanley Milgram39 in the 1960s, we would recommend that they should not be labeled as unprofessional.
Among many similar comments, one student featured in the BMJ article explained: “I have never felt able to refuse a consultant, even though I have really felt very unhappy about it.”38 This comment is reminiscent of the tension vocalized by individuals participating in Milgram’s experiments on obedience to authority.39 In this research, participants were ordered by the experimenter to administer increasingly powerful electric shocks, up to 450 volts, to a fellow participant (a confederate of the experimenter). In the standard trial, 65% of participants obeyed the experimenter to the end of the experiment, even though some believed that they had actually killed the other participant.39 By altering the contextual conditions of the experiment, Milgram39 found different levels of obedience; for example, obedience decreased when the experimenter (the legitimate authority) was replaced with a clerk or was absent. Milgram39 also explored individual differences between those who obeyed and those who defied the experimenter and concluded, “often it is not so much the kind of person a man is as the kind of situation in which he finds himself that determines how he will act.”
The work outlined in this section suggests that we cannot simply judge students or their professionalism from just observing their behavior. What might look like unprofessional behavior from an observational perspective may not actually be unprofessional, if we take the time to talk to students about the context in which the behavior occurred. Furthermore, we may witness students engaging in unprofessional behavior against their will, purely because they have been told to act in that manner by a legitimate authority. What this means from an assessment perspective is that we may fail students who have demonstrated unprofessional behaviors, who nevertheless have professional attitudes. So, given the problems associated with behavioral assessment, how should we assess students’ professionalism?
How Should Professionalism Be Assessed?
Although we critique the assessment of professionalism based on behavior alone, we do not suggest avoiding assessment of students’ professional behaviors. Instead, we recommend that, rather like ethnography,40 we collect data about students’ professionalism using multiple and complementary methods, such as observation coupled with interviews, or “conversations with a purpose.”40
This proposal is not entirely new—Shiphra Ginsburg and her colleagues36,41 have already recommended that professionalism assessment incorporate students’ reasoning behind their actions, stating that we must ask students why they behaved in a particular manner. These researchers have focused on students’ reasoning behind professionally challenging events42 and have recommended ways of assessing students’ reasoning strategies, such as the Reflective Judgment Interview (RJI).43 Although a measure of students’ epistemological thinking might be extremely insightful,44 the RJI focuses on students’ abilities to reason through ill-structured problems and is therefore inappropriate to assess students’ reasoning behind their own professional and unprofessional behaviors immediately after events. So, how can we examine students’ explanations behind their behaviors in a practical way?
Behavioral explanations are social acts, so conversational models should be applied to behavioral assessments.31 Students should be asked to explain both their professional and unprofessional behaviors in informal and formal settings, from informal conversations, to formal focus-group discussions and reflective essays. Telling stories is arguably the fundamental building block of medical practice,45 so a narrative approach46,47 to behavioral explanation seems appropriate. If, for example, an instructor observed a student communicating with a patient inappropriately, the instructor could ask the student to tell the story of what happened. Analogous to physicians asking patients open questions at the start of medical consultations, this conversational approach should allow students to tell their stories, thus revealing the students’ behavioral explanations without prompting. However, open questions do not always result in rich and detailed stories, so medical educators must be prepared to ask students probing questions to help them think more deeply and comprehensively about their behaviors. These questions can be guided by the sociocognitive psychology literature on behavioral explanation.48
Malle’s48 folk-conceptual theory of behavioral explanation differs from the models of behavioral prediction24,25 previously outlined in this article in that it examines aspects of how people make sense of behavioral events (both their own and others’) and includes the social function of behavioral explanation. Briefly, Malle’s theory attends to three levels of behavioral explanation: conceptual, psychological, and linguistic. The conceptual level involves three modes of explaining intentional behavior (reasons, causal history of reasons, and enabling factors) and one mode for explaining unintentional behavior (causes).48 The psychological level examines processes that guide explanations of behavior and includes, for example, pragmatic goals, such as audience design (tailoring an explanation to a particular audience).48 The final level identifies the linguistic forms that individuals employ to articulate their explanations.48 This model may provide a useful framework for asking students questions to probe their behavioral explanations (Figure 1).
Malle’s theory can help us explore the content of individuals’ explanations, but it also reminds us to think about how explanations are articulated and to consider phenomena like audience design and impression management. For example, omitting mental state markers such as “thought,” “felt,” “liked,” or “wanted” from behavioral explanations makes the explanations sound more objective and true.48 The unmarked form, “I had no choice, the consultant told me to do it” sounds more factual than the marked form, “I thought I had no choice, the consultant asked me to do it.” By listening for marked and unmarked forms, we can gauge the extent to which students employ strategies to distance themselves from their actions. Not only can we challenge students’ explanations with further questions such as “What exactly did the consultant say?” and “Why did you think you had no choice?” we can also offer students remediation to increase their sense of empowerment in making choices about their behaviors. Ultimately, by attending to the linguistic level of behavioral explanation, medical educators can be aware of issues, such as students giving socially desirable answers, and thus can avoid accepting students’ behavioral explanations at face value.
Although these are general suggestions based on sociocognitive models,48 these questions and students’ responses should help assessors unpick any disparity between students’ behaviors and their attitudes, arguably leading to a fairer assessment of students’ professionalism. As well as being advantageous from a summative perspective, such conversations are also formative. Not only will conversation help students to reflect critically on their behaviors (and the broader social context of their behaviors), but it should also help students to develop their professional behaviors in the future. Without such conversation, we are arguably doing our students a disservice.
Sociocognitive psychology shows us that there is often a mismatch between behaviors and attitudes. As a result, we must be wary of making assumptions about students’ professionalism on the basis of observed behavior alone. Not only can students fake professional behaviors, but those with professional attitudes can at times behave in a manner that may be considered unprofessional. Therefore, we need assessment methods that capture both behaviors and attitudes—observation coupled with conversations guided by sociocognitive models. The TPB sheds insight into the different aspects occurring around the time of behaviors, whereas Malle’s theory may help us disentangle the various post hoc explanations of those behaviors. Not only will multimethod approaches to assessment of professional behavior result in fairer summative assessments, but they will also lead to formative learning experiences for students. The future challenge for medical educators is how we can assess conversations reliably. Perhaps we could capitalize on the emerging literature on reflective portfolios,49 using similar criteria and processes to assess students’ ability to reflect critically on their behaviors, thus revealing their attitudes. We hope this viewpoint article will stimulate further debate on the assessment of such conversation.
The authors would like to thank their colleague Dr. Jeffrey Bishop, Principal Lecturer in Medical Ethics and Law, Peninsula Medical School, for his feedback on an earlier version of this manuscript.
1 O’Toole TP, Kathuria N, Mishra M, Schukart D. Teaching professionalism within a community context: perspectives from a national demonstration project. Acad Med. 2005;80:339–343.
2 Hilton SR, Slotnick HB. Proto-professionalism: how professionalisation occurs across the continuum of medical education. Med Educ. 2005;39:58–65.
3 McNair RP. The case for educating health care students in professionalism as the core content of interprofessional education. Med Educ. 2005;39:456–464.
4 Eggly S, Brennan S, Wiese-Rometsch W. “Once when I was on call …,” theory versus reality in training for professionalism. Acad Med. 2005;80:371–375.
5 Teherani A, Hodgson CS, Banach M, Papadakis MA. Domains of unprofessional behavior during medical school associated with future disciplinary action by a state medical board. Acad Med. 2005;80(10 suppl):S17–S20.
6 Ginsburg S, Kachan N, Lingard L. Before the white coat: perceptions of professional lapses in the pre-clerkship. Med Educ. 2005;39:12–19.
7 Stern DT, Frohna AZ, Gruppen LD. The prediction of professional behaviour. Med Educ. 2005;39:75–82.
8 Steinert Y, Cruess S, Cruess R, Snell L. Faculty development for teaching and evaluating professionalism: from programme design to curriculum change. Med Educ. 2005;39:127–136.
9 Rees C, Shepherd M. The acceptability of 360-degree judgements as a method of assessing undergraduate medical students’ personal and professional behaviours. Med Educ. 2005;39:49–57.
10 Frohna A, Stern D. The nature of qualitative comments in evaluating professionalism. Med Educ. 2005;39:763–768.
11 Shue CK, Arnold L, Stern DT. Maximizing participation in peer assessment of professionalism: the students speak. Acad Med. 2005;80(10 suppl):S1–S5.
12 Dannefer EF, Henson LC, Bierer SB, et al. Peer assessment of professional competence. Med Educ. 2005;39:713–722.
13 Rees C, Shepherd M. Students’ and assessors’ attitudes towards students’ self-assessment of their personal and professional behaviours. Med Educ. 2005;39:30–39.
14 van Zanten M, Boulet JR, Norcini JJ, McKinley D. Using a standardised patient assessment to measure professional attributes. Med Educ. 2005;39:20–29.
15 Veloski JJ, Fields SK, Boex JR, Blank LL. Measuring professionalism: a review of studies with instruments reported in the literature between 1982 and 2002. Acad Med. 2005;80:366–370.
16 Lynch D, Surdyk P, Eiser A. Assessing professionalism: a review of the literature. Med Teach. 2004;26:366–373.
17 Stern DT. Measuring Medical Professionalism. New York: Oxford University Press; 2006.
18 Hafferty F. Measuring professionalism: a commentary. In: Stern DT, ed. Measuring Medical Professionalism. New York: Oxford University Press; 2006:281–306.
19 Arnold L, Stern DT. What is medical professionalism? In: Stern DT, ed. Measuring Medical Professionalism. New York: Oxford University Press; 2006:15–37.
20 Sherman N. Of manners and morals. Br J Educ Stud. 2005;53:272–289.
21 Myers DG. Social Psychology. Boston: McGraw-Hill College; 1999.
22 Verplanken B, Hofstee G, Janssen HJW. Accessibility of affective versus cognitive components of attitudes. Eur J Soc Psychol. 1998;28:23–35.
23 Wallace DS, Paulson RM, Lord CG, Bond CF. Which behaviours do attitudes predict? Meta-analyzing the effects of social pressure and perceived difficulty. Rev Gen Psychol. 2005;9:214–227.
24 Ajzen I. The theory of planned behaviour. Organ Behav Hum Decis Process. 1991;50:179–211.
25 Fishbein M, Hennessy M, Yzer M, Douglas J. Can we explain why some people do and some people do not act on their intentions? Psychol Health Med. 2003;8:3–18.
26 Holland RW, Verplanken B, van Knippenberg A. On the nature of attitude-behaviour relations: the strong guide, the weak follow. Eur J Soc Psychol. 2002;32:869–876.
27 Mays N, Pope C. Qualitative research: observational methods in health care settings. BMJ. 1995;311:182–184.
28 Rees CE. Assessing Students’ Professionalism and Humanism at a New Medical School: The Views and Experiences of Students and Assessors [master’s thesis]. Exeter, UK: University of Exeter; 2003.
29 Festinger L. A Theory of Cognitive Dissonance. Stanford, CA: Stanford University Press; 1957.
30 Goffman E. The Presentation of Self in Everyday Life. Garden City, NY: Doubleday; 1959.
31 Frowe I. Professional trust. Br J Educ Stud. 2005;53:34–53.
32 Sutton S. Determinants of health-related behaviours: theoretical and methodological issues. In: Sutton S, Baum A, Johnston M, eds. The Sage Handbook of Health Psychology. London: Sage; 2004:94–126.
33 ABIM Foundation, ACP-ASIM Foundation, European Federation of Internal Medicine. Medical professionalism in the new millennium: a physician’ charter. Lancet. 2002;359:520–522.
34 Arnold L. Assessing professional behavior: yesterday, today, and tomorrow. Acad Med. 2002;77:502–515.
35 Ginsburg S, Regehr G, Hatala R, et al. Context, conflict and resolution. A new conceptual framework for evaluating professionalism. Acad Med. 2000;75:S6–S11.
36 Ginsburg S, Regehr G, Lingard L. Basing the evaluation of professionalism on observable behaviors: a cautionary tale. Acad Med. 2004;79:S1–S4.
37 Ginsburg S, Regehr G, Lingard L. The disavowed curriculum: understanding students’ reasoning in professionally challenging situations. J Gen Intern Med. 2003;18:1015–1022.
38 Coldicott Y, Pope C, Roberts C. The ethics of intimate examinations – teaching tomorrow’s doctors. BMJ. 2003;326:97–101.
39 Milgram S. Obedience to Authority: An Experimental View. London: Pinter & Martin; 2005.
40 Atkinson P, Pugsley L. Making sense of ethnography and medical education. Med Educ. 2005;39:228–234.
41 Ginsburg S, Lingard L. Using reflection and rhetoric to understand professional behaviors. In: Stern DT, ed. Measuring Medical Professionalism. New York: Oxford University Press; 2006:195–212.
42 Ginsburg S, Regehr G, Lingard L. To be and not to be: the paradox of the emerging professional stance. Med Educ. 2003;37:350–357.
43 Wood PK. A secondary analysis of claims regarding the reflective judgement interview: internal consistency, sequentiality and intra-individual differences in ill-structured problem solving. In: Smart J, ed. Higher Education: Handbook of Theory and Research. Bronx, NY: Agathon, 1997;243–312.
44 Knight LV, Mattick K. “When I first came here, I thought medicine was black and white”: making sense of medical students’ ways of knowing. Soc Sci Med. 2006;63:1084–1096.
45 McLellan MF. Literature and medicine: physician-writers. Lancet. 1997;349:564–567.
46 Reissman CK. Narrative Analysis. Sage: London; 1993.
47 Bleakley A. Stories as data, data as stories: making sense of narrative inquiry in clinical education. Med Educ. 2005;39:534–540.
48 Malle BF. How the Mind Explains Behavior: Folk Explanations, Meaning and Social Interaction. Cambridge, MA: MIT Press; 2004.
49 Rees C, Sheard C. The reliability of assessment criteria for undergraduate medical students’ communication skills portfolios: the Nottingham experience. Med Educ. 2004;38:138–144.