Never be afraid to get dirty, but be sufficiently sure-footed to avoid the abyss of contamination.


It started in elementary school. Every few months, I would bring home a card with grades marking how well I had done in a variety of subjects. My parents would sign it, and I’d take it back to school. As I got into middle school, I started applying a meaning to these grades. There was a certain threshold I wanted to meet defined by these grades, and if I succeeded in meeting the threshold, it would be visible on the report card. These grades were based on regular exams, homeworks, and reports, and these markers allowed me to adapt my studying habits and behaviors to control for the type of grades I wanted. The system worked through college.

Graduate school introduced me to a totally different game. Research took precedence and classes were not emphasized as much. Unlike classes, research did not have an organized system of homework, exams, or reports. Their replacements: weekly meetings, talks, and papers gave only a vague indicator of how one was performing, and the opportunity for each was far more infrequent than their classwork counterparts.

There does exist a system to evaluate and control performance. It comes in the form of an annual student progress review, in a sense similar to annual performance reviews employees get at companies. The review is an opportunity for the student and advisor to review the student’s performance over the course of the year and suggest skills that might benefit from further development. While a nominal score is assigned to the student’s performance, it is the exercise of engaging in discussion with one’s advisor that allows a student to understand what directions might help that student improve as a researcher.

A grading system is useful when there are concrete skills that can be tested either via some type type of exercise, whether that exercise involves writing, problem solving, or creating a presentation. With classwork, one usually expects that a homework problem has a solution or there is a persuasive argument one can make in writing an essay. Research does not benefit from this certainty. Research directions may lead to dead ends, and success is not always guaranteed a priori. Thus, having a system in which a student gets advice on how to improve as a researcher without as much of an emphasis on meeting predefined benchmarks may actually be more useful. At the very least, it helps the student prepare to deal with a life that won’t hand back a report card every few months.


2 responses

  1. Kush

    Does this system exist at Berkeley or anywhere else, or is it just an idea that you feel would be a good addition?

    April 16, 2007 at 7:11 pm

  2. Krish

    I’d like to claim credit for the idea, but Berkeley EECS had one in place since I started. There are many departments on campus that don’t have one, though.

    April 16, 2007 at 7:22 pm

Leave a Reply

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

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

Google+ photo

You are commenting using your Google+ 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 )


Connecting to %s