Development of culture-sensitive clinical teacher evaluation sheet in the Japanese context

Abstract

Aim: Many instruments for evaluating clinical teaching have been developed, albeit most in Western countries. This study aims to develop a validated cultural and local context sensitive instrument for clinical teachers in an East Asian setting (Japan), Japanese Clinical Teacher Evaluation Sheet (JaCTES).

Methods: A multicenter, cross-sectional evaluation study was conducted. We collected a total of 1368 questionnaires on 304 clinical teachers, completed by residents in 16 teaching hospitals. The construct validity was examined by conducting a factor analysis and using structural equation modeling (SEM). We also assessed the reliability using generalizability analysis and decision study.

Results: Exploratory factor analysis resulted in three-factor (role model, teaching activities, and accessibility) model including 18 items. Confirmatory factor analysis was performed, using SEM. The comparative fit index was 0.931 and the root mean square error of approximation was 0.087, meaning an acceptable goodness of fit for this model. To obtain a reliable dependability-coefficient of at least 0.70 or higher, 5–8 resident responses are necessary.

Discussion and conclusion: JaCTES is the first reported instrument with validity evidence of content and internal structure and high feasibility in Japan, an East Asian setting. Medical educators should be aware of the local context and cultural aspects in evaluating clinical teachers.

Introduction

Globalization is influencing medical education (Schwarz 2001; Stevens & Goulbourne 2012; Eltayeb 2015) and there is often a tendency to apply educational approaches that have been successful elsewhere in the own context, under the assumption that medical competence has universal elements. In the field of social sciences, globalization in education has been discussed in terms of two perspectives; global discourses of homogenization and local discourses of culture and context (Wong 2011). According to Holton, homogenization refers to convergence toward a common set of cultural traits and practices (Holton 2000). In the view of homogenization, all education systems should be standardized into a single universal model. But because of the domination of Western views on education, it is often seen as a form of “neocolonialism” (Ashcroft et al. 2006). This may lead to a selection bias with respect to educational development and research at a disadvantage to the non-Western views (Hodges & Segouin 2008; Frambach et al. 2012; Gosselin et al. 2015).

There have been many studies showing the impact of local cultural differences between Western and non-Western countries on higher education (Watkins & Biggs 2001; Apfelthaler et al. 2007; Wong 2011). In response, there has been a growing number of studies regarding cultural differences and context-specific issues, and locally-driven approaches are gaining increased attention in the domain of medical education (Ho et al. 2011, 2012; Wong 2011; Kikukawa et al. 2013; Pan et al. 2013; Gosselin et al. 2015). For example, the roles of the learners and teachers are dramatically different in different cultures, for example, because of hierarchical differences (Hofstede 2001), and the ways in which educational approaches work or do not work can only be understood through such cultural lenses (Hodges et al. 2009). For instance, the instrument for evaluating clinical teaching developed in the previous study has several differences when compared to instruments of Western origin (Kikukawa et al. 2014).

Table

Practice points

JaCTES is the first reported instrument in an East Asian setting. The validity of an instrument for evaluation of clinical teaching is influenced by local contexts and cultures. Medical educators should be aware of the influence of local contexts and cultural aspects on the process of evaluating clinical teachers.

A clinical teacher is regarded as having an important role in students’ training all over the world (Harden & Crosby 2000; Boonyasai et al. 2007; Sutkin et al. 2008). Effective teachers can promote students’ learning, optimize the learning potential of the workplace, and motivate students (Harden & Crosby 2000; Kisiel et al. 2010; Kikukawa et al. 2013).

Although there are many reports regarding effective teachers, again the majority has originated from Western contexts (Sutkin et al. 2008). Our previous study indicated that there are differences in perceptions of residents between Japan, East Asia on the one hand and Western countries on the other. Interestingly, in our previous study, medical knowledge was not identified as a theme by Japanese residents and clinical competence of teachers was not emphasized, even though these were the most commonly identified themes in Western countries. However, this result remains to be replicated by other studies (Kikukawa et al. 2013).

The evaluation of clinical teaching is essential for understanding this concept from many points of view, such as: improvement of clinical teaching, support and motivation for a teacher, annual performance appraisals, academic advancement and promotion, evolution of the curriculum, learning for the student, improved care for patients, and better educational programs for the institution (Copeland & Hewson 2000; Snell et al. 2000).

Of course, any instrument designed to assess the quality of education and in particular clinical teaching by physicians should have a good validity (Fluit et al. 2010). Establishing validity requires a clear view of the purpose of the measurement; an instrument cannot be valid in itself but is always valid only for a certain purpose. Therefore, validity must be seen as the hypothesized relationship between assessment information and conclusions about the validity of the results (Kene 2006). The sources of validity evidence thus contribute to accepting or rejecting this hypothesis (Downing 2003).

To help improve clinical teaching, this instrument would provide reliable and relevant feedback on clinical teachers’ strengths and weaknesses (Snell et al. 2000). Results from teaching evaluation instruments must be supported by a variety of validity evidence.

The American Psychological and Education Research Association (1999) published standards identifying five sources of validity evidence by:

1. Content: The relationship between a test’s content and the construct it is intended to measure.

2. Internal structure: The degree to which individual items within the instrument fit the underlying constructs.

3. Response process: Analyses of responses include the actions, strategies, and thought processes of individual responses or observers.

4. Relation to other variables: The relationship between scores and other variables relevant to the construct being measured.

5. Consequences: Evaluations are intended to have some desired effect.

This five category validity framework, formulated by Messick (1995), has been increasingly regarded by education and psychology researchers as the most comprehensive conceptualization of validity (Beckman et al. 2004). Moreover, experts emphasize the importance of incorporating these sources of evidence into clinical teaching evaluations (Fluit et al. 2010).

In light of the increasing awareness of cultural and local context sensitivity, our project aimed to develop a culturally and locally validated and sensitive instrument for the evaluation of the quality of clinical teachers in Japan to help improve clinical teaching (Japanese Clinical Teacher Evaluation Sheet: JaCTES). This study aims to assess the internal structure of the instrument which we developed with cultural sensitivity and content validity in the previous study. The internal structure is one aspect of five sources of validity evidence and defined as the degree to which individual items within the instrument fit the underlying constructs. To have more validity evidence (internal structure) of the instrument, this study investigated the construct of this instrument and its generalizability. A multicenter, cross-sectional evaluation study was performed so that the instrument can be applied in various educational settings of Japan, one of the East Asian countries.

Methods

Japanese culture

Although there are different definitions of what a culture entails, and Japanese culture has been discussed from a variety of perspectives (Varley 2000), Confucianism has been probably the most influential in shaping the behavior patterns and structures of communities such as the Japanese in East Asia (Park & Cho 1995). Confucianism, which was established by Confucius and his disciples, is a philosophy of human nature that considers proper human relationships as the basis of society. There are five virtues that one has to aim for: ren (benevolence, humanism), yi (integrity, uprightness), li (rite and propriety), chi (moral understanding, wisdom), and shin (trust). There are also dependent virtues (Yum 1988; Craig & Craig 2000). Fostered by Confucian values are the five principal relationships through which each person defines a sense of identity, duty, and responsibility. The five basic relationships are loyalty between ruler and subject (government and citizen), the closeness between father and son, distinction in duty between husband and wife, obedience to others between elders and youngsters, and mutual faith between friends (Kincaid 2013).

Influenced by Confucianism, it is often contended that East Asian countries like Japanese Korea and China are a hierarchical (high power distance) and collectivistic society. Power distance relates to the degree of inequality in power between a less powerful Individual (I) and a more powerful Other (O), in which I and O belong to the same social system (Mulder et al. 1971). Collectivism stands for a society in which people from birth onwards are integrated into strong, cohesive in-groups, which throughout people’s lifetime continue to protect them in exchange for unquestioning loyalty (Winfield et al. 2000; Hofstede 2001). Hofstede’s cultural investigation has also shown that Japanese culture is a high-power distance and collectivist culture (Hofstede 2001).

Setting

MK, KT, and YM asked 18 colleagues who are directors of residency programs, geographically dispersed from Hokkaido (Northern part of Japan) to Kagoshima (Southern part of Japan) to participate in this study. After a detailed explanation of the project, 15 program directors agreed to participate. In addition, one director asked another director to participate in this study and he approved. As a result, we conducted this study in 16 teaching hospitals (15 community hospitals and one university hospital). In 2004, a new 2-year postgraduate training program was launched. Typically students enter this program after having completed 6 years of undergraduate medical education. Successful completion of the postgraduate program leads to certification of residents’ clinical competence (Kozu 2006). The program entails mandatory rotations in internal medicine (more than 6 months), emergency medicine (more than 3 months) and community medicine (more than 1 month), and at least two rotations in the following specialties: surgery, anesthesiology, pediatrics, obstetrics and gynecology, and psychiatry. Under Japanese official postgraduate training program regulations, physicians have to have at least more than 7 years of clinical experience and take a Faculty Development Workshop certified by the Ministry of Health, Labour and Welfare before they are allowed to act as clinical teachers. In teaching teams, clinical teachers have a responsibility to teach the residents that rotate through their department.

Participants

Between June 2013 and November 2015, residents evaluated their clinical teachers anonymously. Rotation periods varied from 4 to 32 weeks. Participation was voluntary and residents did not receive any incentive for their participation.

Steps taken to protect human participants

We asked the 16 teaching hospitals to choose clinical teachers from both mandatory and elective rotations. Each hospital de-identified the clinical teachers with a random number. First, we conducted a pilot study in two hospitals. In this process, we then asked seven residents to provide feedback on the clarity of formulation of the items and on the format of the questionnaire. Based on this feedback, minor revisions were made. Questionnaires with a written explanation were then sent out to the 16 teaching hospitals. Residents filled out the anonymous evaluation questionnaires. Residents also had the list of the combination of teachers’ name and the assignment number and filled out this instead of the teachers’ name. Through this process, we protected the confidentiality of the participants.

Residents evaluated three clinical teachers with whom they had the most interaction. We asked residents to evaluate their clinical teachers within 1 week of the end of each rotation. We saved the collected data on a password-protected computer. This study was approved by the Institutional Review Board for Clinical Research of Kyushu University, Fukuoka, Japan.

Instrument

First, we investigated the characteristics of a good clinical teacher as perceived by residents’ physicians through five focus group interviews (Kikukawa et al. 2013). We prepared a draft evaluation instrument based on the data from a literature search and the result of the earlier study, and developed an instrument for evaluating clinical teacher in Japan on content validity established through modified Delphi approach among three groups of stakeholders (clinical teachers, residents, and educational experts) (Kikukawa et al. 2014). The instrument has 25 items and has a five-point Likert scale (1: totally disagree; 2: disagree; 3: neutral; 4: agree; and 5: totally agree) (Appendix).

Analysis

First, we conducted exploratory factor analysis (EFA) to investigate factors of related variables using maximum likelihood estimation and promax rotation. In order to investigate the interpretability, we used Hatcher’s interpretability criteria in factor analysis; (1) At least three variables with a loading >0.5 per factor, (2) Variables loading on the same factors share the same conceptual meaning, (3) Variables loading on different factors appear to measure different constructs, which means that most variables load relatively high on only one factor and low on the other factors (Hatcher 1994; Schönrock-Adema et al. 2009).

Items with weak factor loadings (<0.5) were eliminated and the EFA was repeated. Using the subjects (clinical teachers), at least four forms were completed. To determine whether Hatcher’s interpretability criteria were satisfied, MK, TO, and SE discussed the factors and the number of factors for all relevant solutions until consensus was reached.

The analysis was performed using SPSS (SPSS 18 for Windows, SPSS Inc., Chicago, IL) for Windows.

Items were assigned to the factor on which their factor loading was the highest. The placement of each item was further discussed within the research group. To assess the fit of the structure obtained by the EFA, a confirmatory factor analysis (CFA) was performed. For CFA, we used structural equation modeling (SEM). The goodness-of-fit of the model was determined by examining the model chi-square, root mean square error of approximation (RMSEA), which examines residual error, comparative fit index (CFI). We used the fit indexes and criteria of CFI >0.9 and RMSEA <0.1. Through this analysis, the construct validity was examined. This analysis was performed using Amos18 software (Amos Development Corporation, Crawfordville, FL). We set 0.7 as a criterion of a minimum level of reliability.

Internal consistencies of the factors were also determined by calculating Cronbach’s alpha. Cronbach’s alpha >0.7 was considered satisfactory (Cronbach 1951).

In addition, we also used JMP11 to assess the reliability using generalizability analysis and a decision study. The evaluation design is I × T × R:T × I*T, where “I” denotes the evaluation items shown in Table 1 and “T” denotes teacher which is subject to be evaluated. The items are shown in Table 1. In the evaluation design, residents were nested within teacher (R:T) because the clinical teachers were evaluated by residents in 16 teaching hospitals. An interaction between I and T was set as variance component in this model. Consequently, four factors were set for the model. Note that the four factors are set as random effect since these factors were our scope of interest in terms of variance components as mentioned earlier. No fixed effect is assumed in this model. The subject (teachers) for whom at least four residents or more were used for the analysis. In this research, we calculated dependability coefficients (ϕ) instead of generalizability coefficient in order to evaluate the proposing models conservatively. Additionally, we are interested in the absolute score of the respondents, not the order of respondents.

Table 1. Structural equation modeling of Japanese clinical teacher sheet.

Factors	Variances	Items	Standard coefficients	Cronbach’s Alpha	Ranking in Delphi study (out of 25)
Role Model	V1	Is a good role model for doctor–patient relationships.	0.835	0.882	3
	V2	Is a good role model for relationships with medical staff.	0.841		9
	V3	Shows social common sense.	0.815		23
	V4	Encourages residents to consider psychosocial problems.	0.736		17
Teaching activities	V5	Gives residents opportunities to practice.	0.597	0.938	8
	V6	Sets clear roles for residents.	0.77		19
	V7	Thinks speculatively about areas of uncertainty with residents.	0.819		13
	V8	Stimulates residents’ interest in learning and/or subject.	0.839		24
	V9	Shows enthusiasm for teaching.	0.767		5
	V10	Demonstrates the importance of safety.	0.682		10
	V11	Gives concrete indications about what should be improved.	0.809		6
	V12	Provides residents with practical advice for each problem at hand.	0.846		16
	V13	Explains clearly which aspects are important and why.	0.839		15
	V14	Displays reasoning processes.	0.805		4
Accessibility	V15	Is accessible.	0.814	0.899	7
	V16	Encourages residents to call him or her at any time for any reason.	0.805		20
	V17	Treats residents with respect.	0.815		2
	V18	I would like to work with this teacher again.	0.881		12

We determined the dependability coefficients (D-coefficient) of the ratings by estimating the number of residents’ ratings required for a reliable rating per individual clinical teacher. For the subset of 132 clinical teachers who received an evaluation from four residents or more, we calculated D-coefficients for each factor. D-coefficient is shown in Table 5.

Results

We collected a total of 1368 forms on 304 clinical teachers completed by residents between June 2013 and November 2015 in 16 teaching hospitals. Among these, 132 clinical teachers received four or more anonymous ratings. (Residents rotated through a variety of departments during the residency.)

EFA revealed that the three-factor solution was optimal with the best interpretability. Five items were removed based on their factor loading and leaving 20 items for further analysis. The final EFA resulted in the three-factor model including 18 items, which explained 66.1% of the total variance. We named them role model, teaching activities, and accessibility, respectively.

CFA was performed, using SEM (Table 1).

In Table 1, we indicate Cronbach’s alphas of the three factors and ranking of each item in Delphi study in the previous study.

The correlations between the three factors are shown in Table 2.

Table 2. The correlations between the three factors.

	Role model	Teaching activities	Accessibility
Role model	1	–	–
Teaching activities	0.867	1	–
Accessibility	0.880	0.849	1

The CFI was 0.931 and the RMSEA was 0.087, meaning acceptable goodness of fit for this model (Table 3).

Table 3. The goodness of fit.

CHIN	Df	p Value	CHIN/df	CFI	RMSEA
1217.227	132	0.00	9.221	0.931	0.087

Variance components associated with teachers were between 0.12 and 0.20, and those with residents within teachers vary per factor between 0.30 and 0.36. The results are shown in Table 4.

Table 4. Variance components of JaCTES for each factor.

Factor	Teacher (T)	Resident:Teacher (R:T)	Item	Teacher*Item
Role model	0.164	0.361	0.009	0.031
Teaching activities	0.127	0.309	0.010	0.027
Accessibility	0.2	0.300	0.001	0.013

Table 5. Generalizability (D-coefficients) per number of resident ratings of JaCTES for each factor.

No. of ratings	D-coefficients
	Role model	Teaching activities	Accessibility
1	0.27	0.26	0.36
2	0.43	0.41	0.53
3	0.52	0.5	0.63
4	0.59	0.57	0.69
5	0.63	0.62	0.73
6	0.67	0.66	0.77
7	0.7	0.69	0.79
8	0.72	0.72	0.81
9	0.74	0.74	0.83
10	0.76	0.76	0.84

Table 4 provides the D-coefficients per factor as a function of the number of resident response. To obtain a reliable D-coefficient of at least 0.70 of higher, 5–8 resident responses are necessary (Table 5).

Discussion

In our study into the structure of our instrument, a three-factor model was extracted, lending support for the validity of our instrument for its cultural context of an East Asian setting. In addition, we found that 5–8 ratings were enough to obtain sufficient generalizable ratings. Also, the items we used were the same as the ones receiving the highest rankings in our previous Delphi study, indicating that the 18 items were supported by both contents and construct validity evidence (see Table 1).

In this study, three domains have been highlighted in the previously developed instrument with the consensus among stakeholders, which implies that these three factors signify the crucial components in Japanese clinical settings. We will discuss the reasons why these factors emerged in terms of Confucianism lens mainly.

First, “role model” was extracted as one domain of the instrument. The main items of “role model” are associated with a relationship with others and social norm. Confucianism emphasizes the importance of relationships between others as a member of a society (Yum 1988). Under the Confucianism heritage, the elders are in charge of setting a moral standard and demonstrating benevolence toward their younger subordinates (Pratt et al. 1999) as duty and responsibility. In the field of education, teachers are expected to make an effort to pursue ren, yi, li, chi, and shin. Instead, the youngsters should show their elders their due respect. Pratt et al. (1999) stated that teachers and learners are enacting these roles through social scripts which can only be understood through observation. In this light, we understand the factor “role model” to be particularly highlighted in this instrument as the resonance of this cultural aspect. Although the concept of role model has also been reported in Western teaching instruments (Beckman et al. 2003; Stalmeijer et al. 2010). For example, “was a good role model of a caring doctor” is seen as a part of “learning climate” factor of Mayo Teaching Evaluation Form (MTEF-28) (Beckman et al. 2003). However, under Confucianism, a teacher is regarded not as part of a learning climate, but as an embodied goal. Hofstede (2001) pointed out that many of Asian countries are more socially oriented, and teachers assume the role of gurus who transfer personal wisdom in their respective countries. We assume that there are different perspectives on the role of a clinical teacher as modeling between Western countries and Japan among stakeholders. Further studies are required to investigate these assumptions.

Second, “accessibility” is an individual domain. In Japanese schools, influenced by Confucianism, students display a different way of learning compared to Western schools. If they do ask questions, learners may be perceived as suggesting that the teacher did not teach the material well enough (Yum 1988). By this Japanese propensity, residents often hesitate to ask questions of clinical teachers, and, yet, novice doctors will sometimes need to ask questions about their patients in the clinical workplace because of patient safety or for their own learning. As a result, residents often experience huge stress in a clinical workplace. Within the culture and Japanese context, we suggest that “accessibility” has emerged from this instrument. In fact, residents regarded this theme as an important factor of “good” clinical teacher in our previous qualitative study (Kikukawa et al. 2013).

To the best of our knowledge, this is the first report of a validated instrument for clinical teachers in Asian countries. But here again, we need to be careful to not make the “one size fits all” approach mistake, though, because the assumption that an instrument or educational approach developed in one context will automatically work in all contexts and situations is being increasingly criticized (Pratt et al. 2001). We suggest that our domains were emphasized more by the needs of stakeholders in an East Asian, Japanese, cultural context. But there is a broader context regarding the validity of this instrument. Any instrument focusing on the perception of the quality of clinical teaching requires a clear perspective of the interaction between the instrument and its user. So, the instrument per se cannot be valid but can only obtain its validity in the hands of the user. As the user’s interpretation, the sense-making is highly influenced by their cultural upbringing; their cultural upbringing needs to be incorporated in the way the instrument supports the user in their judgments or evaluations. This means that in a globalizing world replication of “validated” instruments and the simple transplantation across cultures is likely to be suboptimal and that a mere translation and re-translation approach is not sufficient to ensure cross-cultural and cross-linguistic validity. Adaptability of education to the local situation and research to promote a deep understanding of the learning is more important than pure replication (Cianciolo et al. 2013). Therefore, we hope that our findings have helped to shed more light on the importance of considering the local context and cultural background in medical education.

This study was conducted in only one country, Japan, and its generalization to other East Asian countries, influenced by Confucianism, may be limited. Further studies may be needed to examine this question. We do not suggest that these studies should replicate our specific items or factor structure, but that they should equally shed light on what kind of adaptability is needed, even between countries with near similar philosophies on learning. As such, a finer granularity of our understanding of the influence of cultural context on education would be obtained.

Also, our interpretations are only from the Confucianism view. It is argued that Japanese culture is influenced by other religions, such as Buddhism. Here too, a replication of the study to better understand the possibilities for adaptability in such a case would be needed.

Notes on contributors

Makoto Kikukawa, MD, MMEd, is a Lecturer of Department of Medical Education, Kyushu University, Japan.

Renee E. Stalmeijer, MSc, PhD, is an Assistant Professor of the Faculty of Health, Medicine and Life Sciences, Department of Educational Development and Research, Maastricht University, The Netherlands.

Tomoya Okubo, PhD is an Assistant Professor of the Department of Test Analysis and Evaluation, National Center for University Entrance Examinations, Japan.

Kikuko Taketomi, RN, MPH, PhD, is a Research Fellow of the Center for Medical Education, Hokkaido University, Japan.

Sei Emura, MD, PhD is an Associate Professor of Centre for Graduate Medical Education Development and Research, Saga University Hospital, Japan.

Yasushi Miyata, MD, PhD is a Professor of Department of Primary Care and Community Health, Aichi Medical University School of Medicine, Japan.

Motofumi Yoshida, MD, PhD, is a Professor of Office of Medical Education, International University of Health and Welfare, Japan.

Lambert Schuwirth, MD, PhD, is a Professor of Medical Education in the Prideaux Centre for Research in Health Professions Education, School of Medicine, Flinders University, Australia.

Albert J.J.A. Scherpbier, MD, PhD is a Professor of Medical Education and Dean Faculty of Health, Medicine, and Life Sciences, Maastricht University, The Netherlands.

Glossary

Culture, Evaluation, Confucianism: Confucianism was established by Confucius and his disciples, is a philosophy of human nature that considers proper human relationships as the basis of society. Fostered by Confucian values are the five principal relationships through which each person defines a sense of identity, duty, and responsibility. The five basic relationships are loyalty between ruler and subject (government and citizen), the closeness between father and son, distinction in duty between husband and wife, obedience to others between elders and youngsters, and mutual faith between friends.

Kincaid DL. 2013. Communication theory: eastern and western perspectives. Orlando, FL: Academic Press.

Acknowledgements

We thank the residents, the clinical teachers who participated in this study. The authors also wish to thank Hiroshi Otani, Hideki Yamada, Shiori Kakudou, Kazumi Yamamoto, Hiroaki Kawashiri, Hiroaki Nishioka, Yukihiro Yamaguchi, Kei Mukohara, Akira Ohya, Keiichi Matsuda, Kouichirou Sada, Osamu Nakano, Yoshimoto Inoue, Kazutoyo Tanaka, Akiteru Takamura, Takumi Daino, and Takayuki Hirabayashi for supporting us with the data collection. We are also grateful for the contributions of Sakai Iwasaki and all the staff in Saga University Hospital, Sunagawa City Medical Center, Ebetsu City Hospital, Kin-ikyo Chuo Hospital, Chidoribashi Hospital, Kagoshima Seikyo Hospital, Tachikawa Sogo Hospital, Komenoyama Hospital, National Hospital Organization Nagoya Medical Center, Mimihara General Hospital, Kenwakai Otemachi Hospital, Kobe City Medical Center General Hospital, Nabari City Hospital, National Hospital Organization Nagasaki Medical Center, Saiseikai Fukuoka General Hospital and Kyoto Min-iren Chuo Hospital.

Disclosure statement

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the article.

Additional information

Funding

This work was supported by JSPS KAKENHI [grant Number JP15K15170].

Appendix

Table

1. Shows enthusiasm for teaching.

2. Is accessible.

3. Encourages residents to call him or her at any time for any reason.

4. Displays reasoning processes.

5. Provides sufficient support.

6. Stimulates residents’ interest in learning and/or subject.

7. Treats residents with respect.

8. Actively involves residents in patient care.

9. Sets clear roles for residents.

10. Stimulates residents to think critically when solving a problem.

11. Is a good role model for relationships with medical staff.

12. Is a good role model for doctor–patient relationships.

13. Is a good clinical supervisor at all times.

14. Gives residents opportunities to practice.

15. Does not pretend to know all things.

16. Demonstrates the importance of safety.

17. Thinks speculatively about areas of uncertainty with residents.

18. Explains clearly which aspects are important and why.

19. Gives concrete indications about what should be improved.

20. Provides residents with practical advice for each problem at hand.

21. Encourages residents to consider psychosocial problems.

22. Teaches residents how to conduct clinical research.

23. Discusses clear training goals with residents during rotations.

24. Shows social common sense.

25. I would like to work with this teacher again.