4 thoughts on “How can the NSF evaluate the quality of a PI’s mentorship when a grant proposal is submitted?

  1. This is definitely a tough problem. The issue with understanding the quality of a PI’s mentoring is that only the graduate students can truly give an accurate account. If the PI is outstanding, graduate students are more than happy to brag about their boss. However, in the case of a sub-par mentor, it is a difficult decision to hurt your boss’s chances at being awarded a grant and even more troublesome to your future if they know if was because of a negative mentoring review.

    Finding some way to get genuine feedback from graduate students is key. I think the process needs to maintain anonymity and be out of the PI’s hands. The latter is necessary because I have seen PIs take full control over where and how they are nominated for awards. For example, PIs will tell students to nominate them for mentoring awards and edit the recommendation letters before they are submitted.

    Liked by 3 people

  2. Here is an idea for how to evaluate graduate mentorship by getting feedback from mentees, while protecting the anonymity of the mentees’ responses. Survey data could be obtained from all former graduate student mentees of each professor who mentors graduate students in each graduate program, on the overall quality of their mentorship, and possibly on specific characteristics of their mentorship. Each graduate student would complete this survey at the time when they exit their graduate program (one copy of the survey for each official mentor they had while they were in the graduate program). These survey data would then be aggregated across all mentors in each graduate program and overall summary statistics (showing the average and the range/variability of the data, but not including any individual data points) would be generated for the mentorship quality in each graduate program as a whole.

    Summarizing the data at the level of the entire graduate program, rather than the individual mentor, would result in large enough sets of respondents to protect against any risk of the responses of individuals being identifiable. And collecting the data from former, rather than current, mentees would elicit more honest responses, since at that time, the respondent’s own funding situation is no longer partly a function of their graduate mentor’s funding, and the respondent’s degree has been secured. This approach would also have the expected effect of promoting collaborative efforts by the faculty of each graduate program to institute the necessary policies to ensure consistently high quality mentorship experiences for all of their graduate students. Ideally, this summarized survey data on mentee perceptions of the mentorship in each graduate program, would not only be used by NSF, and other agencies, when making funding decisions, but would also be used by universities and graduate programs in their efforts at continuous improvement, and would be made publicly available for prospective graduate students to consider when selecting a graduate program.

    Liked by 3 people

  3. In my opinion, much of the quality of students’ mentorship experiences is determined by the context in which the mentorship occurs, as created by the policies and structure of the graduate program and the university. So, one approach would be to evaluate whether the graduate program has the types of policies in place that promote high quality mentorship experiences for all students. For example, grant applicants could be asked whether entering students in their graduate program are consistently provided with adequate time and opportunity to make an informed choice of mentor (such as through a period of lab rotations or extensive provision of information on the mentorship styles of each professor, i.e mentorship philosophy statements, list of names and contact information of all former mentees, etc.). They could be asked whether students have the opportunity to form clear, mutual agreements with their mentor before committing to the lab (such as through a written and signed mentorship agreement or Individual Development Plan). They could be asked what policies the program has to facilitate quick and easy changes of lab for students with valid reasons why they need to change (such as a transitional period of lab rotations and a policy of no tolerance for retribution). They could also be asked whether the grading of student research is unilaterally controlled by the same professor who benefits from the student’s labor, or whether a grading system is in place that avoids this conflict of interest. Mentorship is much more likely to work out well when programs ensure that students have real choice, information and agreements they can rely on, and fair evaluation.

    Liked by 2 people

Leave a comment

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s