Gender and Feedback in Medical Education

Background : More women are entering medical school and faculty positions, creating increased opportunities for interactions across gender during medical education. Little is known about how gender aﬀects attitudes towards and preferences around feedback. Objective : To explore diﬀerences in giving and receiving feedback between men and women in clinical education. Methods : An anonymous survey was distributed to 3 rd and 4 th year medical students, residents, and clinical faculty at West Virginia University School of Medicine at the start of the 2014-2015 academic year. Participants were surveyed about their attitudes, expectations, and knowledge about feedback in the clinical educational setting. Responses were mostly on a 5-point Likert scale. The authors used Fisher’s exact test (FET) and binary logistic regression to identify diﬀerences in responses by gender. Results : Of 762 potential respondents, 155 (20%) completed the survey. More women preferred to receive constructive feedback in written form (FET p = .05, OR = 4.18, p = .009). They also preferred to give feedback in written form (FET p = .04, OR 4.98, p = 0.003). However, a greater proportion of men preferred to give constructive feedback face-to-face (FET p = 0.003, OR NS), reported liking to give constructive feedback (OR = 4.40, p = 0.04), and felt comfortable giving constructive feedback (FET p = .03, OR = 10.21, p = 0.003). Conclusions : These ﬁndings suggest there are diﬀerences in the preferences and comfort men and women have for giving and receiving feedback. This has implications for approaches to clinical preceptor training around feedback.


Introduction
The importance of feedback for improving performance in clinical education is well established (Chowdhury & Kalu, 2004;Ende, 1983;Hewson & Little, 1998;Schartel, 2012). Moreover, there is evidence from other fields that men and women differ in the type of feedback they give, their preferences in how they like to receive feedback, and their response to feedback (Brewer, Socha, & Potter, 1996;Evans & Waring, 2011). For instance, male supervisors have been noted to give more specific corrective feedback than female supervisors (Brewer et al., 1996). Male student teachers have been shown to be more receptive of group feedback than female student teachers (Evans & Waring, 2011). There are now more opportunities for cross-gender interactions in medical education, with the proportion of women entering medical schools in the United States at nearly fifty percent (Diana M. Lautenberger, Dandar, Raezer, & Sloane, 2014). The proportion of women in faculty positions is also increasing although lower, near one-third (Diana M. Lautenberger et al., 2014). The medical education literature is sparse regarding gender differences in feedback, so it is unknown if the differences seen in the business and education literature are applicable in clinical settings. We sought to explore potential differences in attitudes and expectations between men and women in giving and receiving feedback in clinical education.

Setting and Participants
West Virginia University (WVU) School of Medicine 3 rd and 4 th year medical students (n=196), residents (n = 412), and clinical education faculty (n = 154) were notified of the opportunity to participate in the study via direct email with a secure link with one reminder from June to August 2014. Student electronic newsletters and posted flyers were also used to encourage participation. The total number of respondents was 155 (response rate of 20%), with a sample similar demographically to our population ( Table 1). The survey was administered anonymously via Qualtrics (a web-based software). A $25 gift card was offered to the first fifty respondents.

Survey Design
We designed the survey to explore gender differences in attitudes, expectations, and knowledge about feedback.
The survey included subject demographic information as well as fifteen statements on preferences when receiving feedback and twenty-four on giving feedback. The majority of questions required responses on a five point Likerttype scale, from "strongly agree" to "strongly disagree" or "never" to "always." We also included questions with the opportunity for free text responses. Some questions were based on the feedback portion of the Stanford Faculty Development Program Questionnaire (SFDPQ26), a validated instrument that assesses 7 areas of clinical teaching effectiveness, including feedback 8 . The survey was reviewed by one expert in feedback and another in survey methodology for both content and design. We piloted the survey with 5 participants, who provided feedback on readability, functionality, and clarity of the questions.
This study was reviewed by the WVU Institutional Review Board and approved for exempt status.

Analysis
We summarized and compared survey responses by gender and level of training using the Fisher exact test as well as binary logistic regression to adjust the responses for gender, age, training, and department. All statistical tests were

Receiving Feedback
There was a difference in the frequency with which male and female respondents report asking for feedback, with women asking more frequently (Always/Most of the time 94.1% versus 78.6%, p = 0.07; OR 5.61, p = 0.04). More female respondents, compared to males, also preferred to receive constructive feedback in written form than male respondents (Strongly agree 22.2% vs 7.5%, p = 0.05; OR = 4.18, p = 0.009).
There was no difference by gender in preference for receiving general or positive feedback face-to-face versus in written form, nor in a group versus individual setting.

Giving Feedback
Men, compared to women, more often preferred to give general ( We found no gender difference in response to whether feedback was altered based on the gender of the recipient, with only 27(16.5%) respondents agreeing that they made some alteration in the way they give feedback. However, of those who did report changing their approach, men were more comfortable than women in giving male recipients constructive feedback (76.5% vs 44.4%, p=0.05).

Discussion
We observed differences in how men and women preferred to give and receive feedback at one academic medical institution, with men preferring face-to-face contact while women preferred written comments, a difference that was most pronounced with constructive feedback. These differences persisted after controlling for age, stage of training, and department. Our respondents were more hesitant to give constructive feedback to female recipients and more comfortable giving constructive feedback to male recipients. Of those who admitted to changing their approach to feedback based on the gender of the recipient, they gave women more positive feedback and men more constructive feedback. These findings are similar to those previously described in the business literature (Brewer et al., 1996). To our knowledge, this is the first description of this in a medical education setting.
We also found that women asked for feedback more frequently than men. Previous studies have demonstrated that men and women differentially ask for feedback based on the gender composition of the group and the task at hand  (London, Larsen, & Thisted, 1999;Miller & Karakowsky, 2005). The clear preference that male respondents showed for giving feedback, especially constructive, face-to-face may be explained by their greater comfort in giving such feedback. Our results are consistent with previous studies indicating that men are more likely to give corrective feedback more frequently and sooner than female counterparts (Brewer et al., 1996). Best practices for feedback content indicate that a mix of positive/reinforcing and constructive feedback is optimal, and that receiving positive reinforcement alone does not improve performance as well as specific feedback on deficiencies (Boehler et al., 2006;Brinko, 1993).
The differences we found in the preference for written versus oral feedback between women and men needs to be further explored, but feedback may be better received if given in the preferred mode for the learner, as there is not clear evidence that one is superior to the other (Bing-You, Greenberg, Wiederman, & Smith, 1997;Elnicki, Layne, Ogden, & Morris, 1998;Veloski, Boex, Grasberger, Evans, & Wolfson, 2006). Knowing and acknowledging this may help with faculty development in encouraging a variety of modes of feedback or even asking the preference of the learner.
The generalizability of the findings of this study is limited by the survey design which could be influenced by recall bias as well as the possibility of sampling error due to the self-selected, small sample that may have already been biased towards the importance of feedback. This study also took place at a single institution, and local culture may play a role in attitudes, expectations, and skills related to feedback.

Conclusion
With more interactions occurring across gender between clinical faculty and learners, differences in how men and women give and receive feedback need to be further studied and addressed. There is a paucity of literature in medical education on the relationship between gender and attitudes, expectations, and skills around feedback. This study adds preliminary findings that will need to be explored with future investigations. All learners need clear, specific feedback both positive and constructive, and if the gender of the giver or recipient affects the type or content of the feedback, learners may not be receiving optimal information for future improvement.