사회적 네트워크와 교수개발코스가 임상강사Supervisor의 의학교육혁신 도입에 어떠한 영향을 주는가? (Acad Med, 2013)

사회적 네트워크와 교수개발코스가 임상강사Supervisor의 의학교육혁신 도입에 어떠한 영향을 주는가? (Acad Med, 2013)

How Do Social Networks and Faculty Development Courses Affect Clinical Supervisors’ Adoption of a Medical Education Innovation? An Exploratory Study

Erik Jippes, PhD, Yvonne Steinert, PhD, Jan Pols, MD, PhD, Marjolein C. Achterkamp, PhD,

Jo M.L. van Engelen, PhD, and Paul L.P. Brand, MD, PhD

 

경험학습experiential learning 에 대한 이론적 프레임워크는 실제 데이터에 근거하여 널리 인정되고 지지를 받아왔으며, 이에 따라 대부분의 교수개발을 위한 노력은 워크숍/코드/세미나 등의 형태를 띄고 있다.

As the theoretical framework of experiential learning has been widely accepted and supported by empirical data,1 most faculty development efforts in medicine take the form of workshops, courses, and seminars.2

공식적인, 구조화된 교수개발 활동을 넘어선 추가적인 요인을 탐구해야 함을 제시하며, 이 요인들이 어떻게 clinical supervisors (이하 CS)가 지식과 skill을 교육에 적용하고, 혹은 정말 적용하는지 여부에 영향을 주는지를 탐구해야 한다.

These findings suggest that we should explore additional factors—beyond formal, structured faculty development activities—that can affect how and whether clinical supervisors transfer knowledge and skills into their teaching practices.

경영분야의 연구나 헬스케어 테크놀로지의 도입과정에 대한 연구를 보면, 사회적 네트워크가 그러한 요인 중 하나임을 보여준다. 사회적 네트워크는 '제한된 set of actors 사이의 관계'라고 정의될 수 있다. 사회적 네트워크는 혁신의 잠재적 adopter들이 혁신에 대해 의사소통하고, 건설하고, 협상하는 채널로서 기능하며, 그것의 생소함novelty를 줄여주고, 잠재적 성과에 대한 불확실성을 줄여준다.

Research on business/management4 and the implementation of health care technologies5 suggests that social networks may be one such additional factor—as well as a promising avenue of research in determining the degree to which individuals adopt innovations. A social network can be defined as the relationships between a finite set of actors.6 Social networks function as channels through which potential adopters of an innovation can communicate, construct, and negotiate that innovation, thereby reducing its novelty and their uncertainty regarding its potential outcomes.4,5

개개인(혹은 집단) 사이의 관계의 패턴에 대한 원칙이 개개인(혹은 집단) 자체의 특성보다 성과에 더 큰 영향을 미친다는 원칙에 근거하여, 사회적 네트워크 분석가들은 혁신의 수용adoption에 있어서 개인 간 connection의 영향을 연구한 바 있다.

On the basis of the principle that the pattern of relationships among individuals (or groups) has greater influence on outcomes than do the attributes of the individuals (or groups) themselves, social network analysts examine the effects of connections between individuals on the adoption of innovations.7

사회적 네트워크 분석의 핵심적 variable은 개개인이 다른 사람들과 그 네트워크 상에서 얼마나 연결되어있는지이며 이것을 actor centrality라고 한다. 기업체에서 centrality가 높은 사람은 승진의 가능성이 유의하게 더 높다. 또한 학생의 네트워크 내에서의 centrality는 학습을 얼마나 증기는지와 학업적 성공 모두에 관련되어 있다.

A key variable in social network analysis is the extent to which an individual is connected to other actors in the network, referred to as actor centrality.6 In business organizations, individuals with high centrality are significantly more likely than others to be promoted.8 Further, students’ centrality in their networks is associated with their enjoyment of learning and their academic success.7

네덜란드 PGME 프로그램에서 CS들에게 최근 교육적 혁신을 도입할 것이 권장되었고, 그 혁신은 S&C피드백이다. 네덜란드의 레지던트 프로그램은 2004년 전까지만 해도 CS의 피드백이 비구조화된, 종종 모욕적인 형태로 전달되었다.

Clinical supervisors in postgraduate medical education (PGME) programs in the Netherlands were recently encouraged to implement an educational innovation— structured and constructive (S&C) feedback—in their teaching of residents. In Dutch residency programs before 2004, feedback from clinical supervisors to residents, was given in an unstructured and sometimes derogatory manner.9,10

2004년 CCMS는 모든 네덜란드의 전문과목학회에 대해 PGME프로그램을 CanMEDS에 기반하여 revise할 것을 요구하는 legal directive를 발표하였다.  

In 2004, the Royal Dutch Medical Association–Central College for Medical Specialists (CCMS)11 issued a legal directive requiring all medical specialist societies in the Netherlands to revise their PGME programs on the basis of the Canadian Medical Education Directions for Specialists (CanMEDS) framework of core competencies,12

추가로 CCMS는 S&C피드백을 "펜들톤 법칙Pendleton's rule"에 따라 전달할 것을 권고하였다. 

In addition, CCMS11 recommended introducing S&C feedback that follows “Pendleton’s rules”13:

(1) Feedback should be structured,

(2) clinical supervisors should give residents opportunities to express their opinions,

(3) clinical supervisors should provide positive comments,

(4) clinical supervisors should provide specific comments regarding areas for improvement, and

(5) clinical supervisors should provide feedback in a “safe” way.

방법

Method

연구대상자

Study participants

두 가지 이유로 다양한 대학과 병원의 전공과목을 조사함 

We chose to incorporate a medical specialty of each discipline type and a mix of university and general hospitals for two reasons.

• 첫째, 전공에 따라 사회적 네트워크가 다를 수 있다 (주로 혼자 일하는 마취과, 주로 함께 일하는 소아과)
  First, physicians’ social networks may differ according to the characteristics of their specialties (e.g., anesthesiologists primarily work independently, whereas pediatricians work more closely together).
• 둘째, 병원의 환경(세팅)이 커뮤니케이션 구조에 영향을 줄 수 있다.
  Second, the hospital setting may influence physicians’ communication structures (e.g., faculty members working in smaller teams in general hospitals may have more frequent contact with one another than do their counterparts in university hospitals).

설문지와 혁신도입 점수

Study questionnaires and innovation adoption scores

We sent two e-mail reminders and stopped collecting data four weeks after the initial invitation was extended to the unit/team.

4년간 수집하였고, 과에 따라서 도입이 빠른 과와 늦은 과가 있었기 때문

We collected study data across a four- year period because we wanted to ensure that the clinical supervisors of the participating teams were at the same stage of implementing S&C feedback to residents. Some specialties—such as pediatrics14—were early adopters of this educational innovation and were therefore included early in the data collection period, whereas others—such as radiology17—began to implement the innovation later and therefore were included late in the data collection period.

'도입'에 대한 Rogers의 정의를 사용하였다. 따라서 S&C피드백 도입의 '여부'를 묻는 대신 도입의 '정도'를 물었다.

We adapted Rogers’4 definition of adoption—“the decision to make full use of an innovation as the best course of action available”—to the context of our study. The questionnaires therefore asked clinical supervisors and residents to assess the degree to which the supervisor had adopted S&C feedback rather than asking them to indicate whether the supervisor had or had not adopted the innovation.

레지던트 설문

Resident questionnaire.

The five-item resident questionnaire asked residents to assess the nature of the S&C feedback given by their clinical supervisors in the six months before the questionnaire was administered (see Supplemental Digital Appendix 1, http://links.lww. com/ACADMED/A116). Residents were provided with a list of the clinical supervisors in their unit and were asked to rate each supervisor on items worded according to the five components of Pendleton’s rules13 (e.g., “The clinical supervisor provides feedback in a structured way,” “The clinical supervisor gives residents the opportunity to express their opinion”).

Only clinical supervisors who were assessed by at least two residents were included in the data analysis; the supervisor’s mean resident-assessed innovation adoption score was used as the dependent variable in analyses.

CS 설문

Clinical supervisor questionnaire.

Clinical supervisors completed a similar, 10-item, Web-based questionnaire (see Supplemental Digital Appendix 2, http://links.lww.com/ACADMED/ A116). Supervisors rated their own adoption of the five components13 of the S&C feedback innovation (e.g., “I provide feedback in a structured way,” “I give residents the opportunity to express their opinion”) on a five- point Likert scale ranging from “totally disagree” (scored as 1) to “totally agree” (scored as 5).

Other items on the questionnaire measured the clinical supervisor’s communication structures to allow us to analyze the supervisor’s social network connections to determine his or her centrality and asked whether the supervisor had attended a Teach- the-Teacher course (both independent variables, described below).

교육자교육 과정 참석

Teach-the-Teacher course attendance (independent variable)

Teach-the-Teacher training took the form of a two-day course on how to apply adult learning principles in clinical teaching19; approximately 70% of the course time was devoted to providing S&C feedback. Teach-the-Teacher courses were designed by medical schools/universities according to national government guidelines20,21 and were conducted by certified trainers and accredited educational institutes in the Netherlands. For the purposes of this study, we considered these courses to be comparable. The courses employed various methods to provide instruction on S&C feedback, such as interactive discussions, role-play, and mini-lectures. During role-play activities, participants gave S&C feedback to colleagues who had completed five-minute teaching sessions; this was followed by debriefings with other participants on how well participants had applied Pendleton’s rules.13

In this study, we asked clinical supervisors to indicate on their questionnaire whether they had attended a Teach-the- Teacher course during the past three years so that we could examine whether course attendance had any impact on their adoption of the S&C feedback innovation.

사회적 네트워크 분석(Centrality 점수)

Social network analysis: Centrality score (independent variable)

We used a “full roster” design for social network analysis.6 Following standard practice, we provided each clinical supervisor with a list of their fellow, departmental clinical supervisors. In the Web-based questionnaire, we asked each supervisor to rate the intensity of his or her communication with each colleague “in the past half year about the introduction of innovations, new methods or procedures, or new developments related to the work situation” using a six-point scale of “never” (scored as 1), “less than once a month” (2), “more than once a month” (3), “weekly” (4), “daily” (5), or “more than once daily” (6).

To calculate supervisor centrality, we created an undirected dichotomous matrix by recoding responses as follows: 1 (never) and 2 (less than once per month) were recoded as “0,” indicating no communication between the individuals, whereas 3 (weekly) through 6 (more than once daily) were recoded as “1,” indicating a communication relationship between the individuals.

통계적 분석

Statistical analysis

First, we assessed the effects of Teach-the- Teacher course attendance and supervisor centrality on innovation adoption scores using t tests and correlation analyses. Subsequently, we analyzed the independent effects of Teach-the- Teacher course attendance and supervisor centrality on innovation adoption scores after adjusting for age, gender, and attitude toward the S&C feedback innovation. To account for the nested structure of the data (individuals within teams), we used two-level hierarchical linear modeling22 (statistical software program MLwiN version 2.17, University of Bristol).

결과

Results

응답자

Study respondents

 

 

단변수 분석

Univariate analyses

Teach-the-Teacher course attendance and centrality score.

Supplemental Digital Figure 1 (http://links.lww.com/ ACADMED/A116) presents a visual representation of a social network of clinical supervisors in a participating department, including the centrality of individual faculty members and their resident-assessed innovation adoption scores.

Control variables.

 

 

 

다변수 분석

Multivariate analyses

After we adjusted for these control variables, Teach-the- Teacher course attendance was weakly but significantly related to the supervisor’s self-assessed innovation adoption score (P = .001; explained variance = 11.49% on the individual level) but not to the resident-assessed innovation adoption score (P = .371). In contrast, clinical supervisor centrality was significantly related to innovation adoption, both as rated by residents (P = .023; explained variance = 4.15%) and by the supervisors themselves (P = .024; explained variance = 9.29%).

 

고찰

Discussion

비록 교육자교육 코스 참여여부가 혁신의 도입에 대한 자기평가점수에는 유의한 영향이 있었지만, 레지던트가 CS의 혁신도입에 대해서 평가할 때는 영향이 없었다. 반대로, CS의 Centrality는 유의하게 혁신 도입에 대한 자기평가과 레지던트평가에 영향을 주었다. 이 관계는 연령/성별/S&C피드백에 대한 태도를 통제하였을 때에도 마찬가지였다.

Although Teach-the-Teacher course participation was significantly related to self-assessed innovation adoption scores, it had no effect on residents’ ratings of their supervisors’ adoption of the innovation. In contrast, the clinical supervisor’s centrality within his or her department’s social network was significantly related to both self- assessed and resident-assessed innovation adoption scores. These associations remained significant after we adjusted for supervisor age, gender, and attitude toward the S&C feedback innovation, 

개개인이 사회적 네트워크에 얼마나 연결되어있느냐가 비지니스 혁신의 도입에 주된 영향을 준다는 것은 보여진 바 있다. 최근의 연구결과에 따르면 교수개발 프로그램에 대해서 네트워크 접근을 사용하는 것이 그러한 프로그램의 성공에 중요하다는 것을 보여준다.

The degree to which an individual in a social network is connected to other individuals in the network (actor centrality) has been shown to have a major influence on that individual’s adoption of business innovations.4,23 A recent publication suggested that taking a network approach toward faculty development programs in medicine might be important in determining the success of such programs.24

우리의 앞선 연구와 일관되게, 우리는 CS의 centrality가 레지던트가 CS에 대해서 혁신도입수준을 판단할 때 유의한 기여를 한다는 것을 찾아내었다. 이번 연구에서 우리는 CS의 centrality가 혁신도입에 대한 CS의 자기평가에도 비슷한 유의한 영향을 준다는 것을 알아내었다. 그러나 교육자교육 코스 참석은 supervisor의 자기평가에만 관련되어 있었다.

In concordance with our earlier work,16 we found that clinical supervisor centrality contributed significantly to the residents’ assessment of the supervisor’s innovation adoption. In the current study, we found that supervisor centrality was similarly significantly related to the supervisor’s self-assessment of innovation adoption; however, Teach-the-Teacher course attendance was related only to the supervisor’s self-assessment.

두 번째 결과는 이전 연구결과와 연결지을 수 있는데, 교수자의 행동에 교수개발 코스가 미치는 영향이 제한적이라는 사실이다. 그러나 이것이 그러한 워크숍이나 코스를 없애abandones야 한다는 것을 의미하진 않으며, 어떤 정보를 집단에 퍼트리는데 효과적인 방법이기 때문이다. 그러나 이번 연구 결과에 기반하여 우리는 교수개발 코스가 상호작용활동을 높은 비율로 포함할 것을 권고하고자 하는데, 그 이유는 (1)능동적 참여가 참여자의 지식과 스킬의 retention에 도움이 되기 때문이며, (2)이러한 상호작용이 참여자간 social network 구조를 활성화시킬 수 있기 때문이다. 이러한 사회적 네트워크에 참여engage하는 것은 장기간의, 포괄적comprehensive 교수개발프로그램이 더 효과적인지를 설명하는데 도움이 된다.

The latter finding is in line with previous studies2,3 showing that the effect of a faculty development course—in this case, Teach-the-Teacher training—on teaching behavior may be limited. This does not mean that such educational workshops and courses should be abandoned, especially as they are effective methods of disseminating information to groups. On the basis of the results of this study, however, we recommend that faculty development courses include a high proportion of interactive exercises (e.g., role-play, discussions) because (1) active participation is likely to improve participants’ retention of knowledge and skills25 and (2) these interactions may activate social network structures among participants. Such engagement of social network structures may also help explain why long-term, comprehensive faculty development programs appear to be more successful than isolated workshop- based interventions.3

최근의 연구는 교수개발에서 social practice의 역할을 강조하고 있으며, 교수개발이 실천공동체 형성에 중요하다는 것을 보여준다. 우리의 연구결과는 레지던트가 교육자교육 훈련 프로그램에 참석하지 않은 CS에 대한 평가에 유의한 차이가 없었음을 보여주며, 이것은 CS가 S&C피드백을 전달하는 방법을 다른 경로를 통해서도 배웠을 것임을 암시한다. 우리의 결과는 동료 CS와의 사회적 네트워크가 '다른 경로'일 것임을 제안한다. 그러나 explained variance가 상대적으로 낮기 때문에, 다른 사회적 네트워크 소스가 관련되었을 수 있다.

Recent articles have highlighted the role of social practices in faculty development and the importance of faculty development in building communities of practice.26,27 Our finding that there were no significant differences in residents’ assessments of clinical supervisors who had and who had not attended Teach-the-Teacher training suggests that clinical supervisors may have learned from other sources how to incorporate S&C feedback into their teaching practices. Our results suggest that social networks of peer clinical supervisors could have provided this input; however, considering that our explained variance is relatively low, other social network sources may be involved (e.g., network connections with medical educators).

교육자교육 참석여부가 혁신의 도입에 대한 CS의 자기평가만 영향을 주었다는 점도 흥미롭다. CS는 자기자신의 적응적행동adoptive behavior를 과대평가했을 수 있다. 이는 성인학습자가 스스로의 임상적, 교육적 역량을 잘 평가하지 못한다는 결과와도 일관된 것이다. 반대로 레지던트들은 CS의 적응적행동을 과소평가했을 수 있는데, 왜냐하면 피드백이 주어졌을 때 그것이 피드백이라고 인지하지 못했거나 '교육'과 '피드백'을 혼동했을 수 있기 때문이다.

It is also interesting to note that clinical supervisors were significantly more likely to consider themselves to be successful adopters of S&C feedback than were their residents, as the effects of Teach- the-Teacher course participation on the adoption of the educational innovation was demonstrated only in the supervisors’ self-assessments and not in the residents’ assessments of their supervisors. Supervisors may have overrated their own adoptive behavior; this is consistent with others’ findings that adult learners often perform poorly in assessing their own clinical or educational competence.28 Alternatively, residents may have underrated supervisors’ adoptive behavior as they may not have recognized feedback when it was given or may have confused feedback with teaching.29,30

우리는 세 가지 잠재적 confounding factor를 조사하였다.

We examined the influence of three potential confounding factors in this study: age, attitude toward the S&C feedback innovation, and gender.

• CS는 연령이 증가할수록 적응적행동이 떨어졌다. 나이 든 교수가 새로운 스킬을 배우거나 적용하는 능력이 떨어졌기 때문이라는 주장도 가능하지만, 우리는 다른 해석도 고려되어야 한다고 생각한다. (1)레지던트들이 젊은 CS를 더 쉽게 identify했을 수 있다 (2)젊은 CS는 그들이 받은 수련을 고려했을 때 나이 든 CS보다 S&C피드백에 더 친숙할 수 있다.
  First, we found that clinical supervisors were less likely to show adoptive behavior (as measured by residents’ assessments) with increasing age. Although it could be argued that older faculty members are less able to learn or less willing to adopt new skills,31 we believe two other explanations should be considered: (1) Residents may have been more likely to identify with younger clinical supervisors and therefore rated them more highly, and (2) younger faculty may have been more familiar than older faculty with S&C feedback because younger faculty’s medical training may have been more oriented toward this innovation.
• 둘째, S&C피드백에 긍정적인 태도를 가진 CS일수록 혁신을 더 잘 도입한다.
  Second, we were not surprised to find that clinical supervisors with more positive attitudes toward S&C feedback were more likely to adopt the innovation, according to their self-assessments. This finding is in agreement with previous research on attitude and innovation adoption in health care.18
• 셋째, 성별에 대해서, 레지던트들은 남자 CS에게 더 높은 점수를 주었다. multivariate 분석에서 단지 borderline significance만을 보였다는 점에서 더 높은 점수는 남자가 사회적네트워크상 더 강력한 embeddedness를 보였기 때문일 수 있다. 남성의 centrality가 높은 것이 혁신에 친숙해지거나 혁신을 도입할 기회를 높여 주었을 것이다.
  Third, with regard to gender, residents gave higher innovation adoption scores to male supervisors than to female supervisors. Our finding that gender was only of borderline significance in the multivariate analysis (Table 2) suggests that the higher scores given to the men may be partly due to the stronger embeddedness of male than female clinical supervisors in their respective social networks. Men’s higher centrality may have given them more opportunities to become acquainted with, and adopt, the innovation.

우리의 결과는 개개인이 사회적 네트워크상에서 중심적 지위를 지닐수록 사회적 자본을 획득에 유리할 것임을 보여준다. 메타분석에서 높은 centrality인 경우...

Our results on centrality confirm research showing that individuals may gain social capital benefits from holding central positions in their social networks. A meta-analysis32 of eight business studies found that individuals with high centrality were likely...

• 리더가 되거나 to emerge as leaders,
• 팀 퍼포먼스에 더 만족하거나 to be more satisfied with team performance, and
• 과제해결을 도입 또는 개발하는데 더 참여하거나 to participate more in developing and implementing task solutions.

 

다른 연구에서는 centrality가 workplace performance를 예측하였고, 높은 centrality는 피고용인이 그 자리를 유지할 가능성을 높여주었다. 광고와 공공관계기관에서 centrality는 혁신의 도입에 가장 유의한 예측인자였다. 더 나아가 centrality가 높을수록 새로운 아이디어를 도입할 기회가 더 많을 것이고 이 새로운 아이디어를 도입하는데 필요한 자원을 끌어오는데avail 기회가 많을 것이다.

Other studies showed that centrality independently predicted individuals’ workplace performance33 and that high centrality increased the likelihood of employees remaining in their positions.34 In a study of an advertising and public relations agency, centrality was found to be the most significant predictor for involvement in innovation.23 Furthermore, individuals who are more central have more opportunities to be introduced to new ideas, to avail themselves of the necessary resources for implementing these new ideas, and to adopt innovations.5

Strengths and limitations

Conclusion

 

 

16 Jippes E, Achterkamp MC, Brand PL, Kiewiet DJ, Pols J, van Engelen JM. Disseminating educational innovations in health care practice: Training versus social networks. Soc Sci Med. 2010;70:1509–1517.

18 García-Goñi M, Maroto A, Rubalcaba L. Innovation and motivation in public health professionals. Health Policy. 2007;84:344– 358. 

19 Mann KV. Theoretical perspectives in medical education: Past experience and future possibilities. Med Educ. 2011;45:60– 68.

4 Rogers EM. Diffusion of preventive innovations. Addict Behav. 2002;27:989–993.

5 Greenhalgh T, Robert G, Macfarlane F, Bate P, Kyriakidou O. Diffusion of innovations in service organizations: Systematic review and recommendations. Milbank Q. 2004;82:581– 629.

7 Baldwin T, Bedell M, Johnson J. The social fabric of a team-based MBA program: Network effects on student satisfaction and performance. Acad Man J. 1997;40:1369.

24 Baker L, Reeves S, Egan-Lee E, Leslie K, Silver I. The ties that bind: A network approach to creating a programme in faculty development. Med Educ. 2010;44:132–139.

26 Steinert Y. Faculty development: From workshops to communities of practice. Med Teach. 2010;32:425–428.

27 Steinert Y, Macdonald ME, Boillat M, et al. Faculty development: If you build it, they will come. Med Educ. 2010;44:900–907.

29 Shute VJ. Focus on formative feedback. Rev Educ Res. 2008;78:153–189.

---

Acad Med. 2013 Mar;88(3):398-404. doi: 10.1097/ACM.0b013e318280d9db.

How do social networks and faculty development courses affect clinical supervisors' adoption of amedical education innovation? An exploratory study.

Jippes E1, Steinert Y, Pols J, Achterkamp MC, van Engelen JM, Brand PL.

Author information

• 1Center for Medical Imaging-North East Netherlands, University Medical Center Groningen, University of Groningen, Groningen, The Netherlands. e.jippes@umcg.nl

Abstract

PURPOSE:

To examine the impact of social networks and a two-day faculty development course on clinical supervisors' adoption of an educational innovation.

METHOD:

During 2007-2010, 571 residents and 613 clinical supervisors in four specialties in the Netherlands were invited to complete a Web-based questionnaire. Residents rated their clinical supervisors' adoption of an educational innovation, the use of structured and constructive (S&C) feedback. Clinical supervisors self-assessed their adoption of this innovation and rated their communication intensity with other clinical supervisors in their department. For each supervisor, a centrality score was calculated, representing the extent to which the supervisor was connected to departmental colleagues. The authors analyzed the effects of supervisor centrality and participation in a two-day Teach-the-Teacher course on the degree of innovation adoption using hierarchical linear modeling, adjusting for age, gender, and attitude toward the S&C feedback innovation.

RESULTS:

Respondents included 370 (60%) supervisors and 357 (63%) residents. Although Teach-the-Teacher course participation (n=172; 46.5%) was significantly related to supervisors' self-assessments of adoption (P=.001), it had no effect on residents' assessments of supervisors' adoption (P=.371). Supervisor centrality was significantly related to innovation adoption in both residents' assessments (P=.023) and supervisors' self-assessments (P=.024).

CONCLUSIONS:

A clinical supervisor's social network may be as important as faculty development course participation in determining whether the supervisor adopts an educational innovation. Faculty development initiatives should use faculty members' social networksto improve the adoption of educational innovations and help build and maintain communities of practice.

PMID:

 

23348089

 

DOI:

 

10.1097/ACM.0b013e318280d9db

[PubMed - indexed for MEDLINE]