Comprehensive Exams: Uniform or Customized

As graduate students, passing comprehensive exams is one thing we all have to do to get the degree. I recently had quite a few conversations with colleagues on comprehensive exams. I have seen a range of different opinions on the goals and formats of comprehensive exams. I think this is something worth discussing here in a little more details.

What goals does a comprehensive exam try to achieve? I have seen a few different thoughts on the goal of a comprehensive exam. One thought is that a comprehensive exam checks the students’ mastery of fundamental knowledge needed to pursue their graduate degree. For example, one may argue that a student in ecology probably should have some knowledge of the fundamental population, community and ecosystem ecology, and should have some familiarity with classic and fundamental papers in ecology. In statistics, the qualifying exam specifically cover probability, mathematical statistics and theory of linear models. In this regard, the comprehensive exam is almost like a big exam covering all the fundamental required course. The goal of comprehensive exam is to ensure all students in the graduate problem master the same fundamental knowledge.

Another line of thought on comprehensive exam view it as a more advanced exam tailored to students’ particular research topic. The fundamental knowledge that every student in a particular program is supposed to know should be covered in required courses. If one passes the course, that’s sufficient evidence that he/she already gets the knowledge. The comprehensive exam does not need to be repetitive. Rather, the comprehensive exam should follow up on the existing courses and test student more specific knowledge related to one’s thesis/dissertation.

I can see both being valid arguments. One one hand, testing standard knowledge does not necessarily mean testing the same knowledge in the class. It can be more comprehensive and more advanced. For example, learning theory of linear model is very different than being able to use the theory to deal with real data. A comprehensive exam question could also link knowledge from multiple required courses to solve a problem. This sort of comprehension of fundamentals is not necessarily covered in required classes. Passing a class does not automatically make one capable of solving these problems. One the other hand, I see the utilization of tailoring the exam to one’s particular research interests and avoid repetition of materials one already know from classes. It provides further guidance to students, which can be more beneficial to one’s research since it is tailored.

In my opinion, I favor the opinion that a comprehensive exam should be on fundamentals and should be uniform. First, covering fundamentals is not the same as repeating required classes. But rather, it is an opportunity to make sure the students could link knowledge together and put them into use comprehensively. I don’t see it as a simple repeat of class materials if the exam is designed in this fashion. Second, it helps create more well-rounded scientists. Tailoring everything to the particular research suffers the danger of being too narrowly focused. Third, there are other opportunities to train students in the specifics of one’s research. The oral exam or even a dissertation proposal are more appropriate ways to serve that purpose. From a repetition perspective, I argue that comprehensive exam does not need to be a repeat of the oral exam and prospectus. Fourth, a comprehensive exam covering fundamentals make the expectation clear. In my mind, this is a more fair and rubout way to represent one’s mastery knowledge.

Different views on the contents of the comprehensive exam, either being uniform or being customized to specific research project, are natural consequences of different views on the goal of comprehensive exam. If one views comprehensive exam as a quality check of fundamental knowledge everyone in a particular problem should know, one may favor a uniform exam. But if one views the comprehensive exam as a test of knowledge specific for one’s research, a customized exam seems to be the only option.

You probably guess that I am in favor of a uniform comprehensive exam. First, a uniform comprehensive exam makes the expectation more clear and is fair to all students. What is viewed as a necessary knowledge to know will very likely depend on one’s expertise. If we leave it to each committee member or advisor to design exam according to his/her own view, it will undoubtedly reflect what he/she views as what constitute fundamental knowledge. While people probably share many common opinions on what a student should know, the chances that his/her own expertise influences the particulars will be high. As I have discussed before, we are all influenced by our own experience. Whether these are the things one should really know is debatable. An exam designed this way may put unreasonable expectation to students. Second, a uniform exam will likely involve more people in the design of the exam. As more opinions from different people are involved, chances that the exam realistically reflect the most important fundamental knowledge one should know is higher. Third, it is logistically easier and provide better timeline management for students. If a uniform exam is to be held regularly every year, it makes it easier for everyone to anticipate the schedule and guide students to better manage the steps one needs to finish for the degree. Plus, it gives the student an earlier chance to reflect their plans for the future. If you realize that graduate school is not the best thing to do for your future career, you want to know this earlier. You don’t want to know this when you are in your fourth or fifth year. Fifth, I think a uniform exam can be evaluated more objectively. A committee that evaluates all the exams from a program is by no means perfect. But it is certainly less likely to reflect the view of just one particular person. If I read a paper that plot everything against everything else, I will hate it. But this is not necessarily a bad approach. Others may like such empirical and exploratory approach of doing science.

Overall, I think a uniform exam can be more fair with more clear expectation. It can better help students prepare and move towards their degree. In my opinion, this is a better way to do comprehensive exam than a customized one.  I would like to hear your thoughts.


About Chao Song

I am a postdoctoral research associate at the Department of Fisheries and Wildlife, Michigan State University.
This entry was posted in Academia. Bookmark the permalink.

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