<p>
</p>
<p>Let me put it to you this way. According to the EECS department’s official guidelines, the typical grade assigned for lower-division courses should be around 2.7 (hence a B-), with basic prereq CS courses having a typical grade of 2.5 (slightly above a C+). </p>
<p>[Grading</a> Guidelines for Undergraduate Courses | EECS at UC Berkeley](<a href=“http://www.eecs.berkeley.edu/Policies/ugrad.grading.shtml]Grading”>http://www.eecs.berkeley.edu/Policies/ugrad.grading.shtml)</p>
<p>How many humanities or social science department can say the same with regards to the grading of their lower-division courses? Is there even one such department? </p>
<p>
</p>
<p>Yet subjectivity is a concept entirely separable from the grading. Just because a topic is subjective doesn’t mean that you have to assign high grades to everybody. You could just as easily assign low grades to everybody. </p>
<p>Consider an example. The Berkeley humanities PhD programs are - just like any world-class humanities PhD programs - notoriously difficult to finish, despite the fact that PhD humanities work is surely even more subjective than is undergrad work. You just can’t just write any old essay as a dissertation and then simply expect that Berkeley will grant you a PhD in English, no matter how valid you may think the arguments to be. Some of those students spend over a decade crafting a dissertation and still cannot produce something to the satisfaction of their dissertation committee. That’s why such a large percentage of PhD humanities students never earn the PhD at all. The committee’s attitude is not: “Well, since humanities research is highly subjective in nature such that we can’t really tell whether you’re right or wrong, we’re just going to grant you a PhD degree by default”, but rather entirely the opposite: the default is not to grant you the degree. </p>
<p>Now obviously I’m not expecting undergrads to perform at the level of PhD students. But surely it is possible to take some of the skepticism that humanities faculty apply to the graduate students and direct it towards the undergrads as well. If a humanities graduate program can be made to be grueling, why not an undergrad program as well? </p>
<p>
</p>
<p>Then let me tell you a story. I know a guy who knew the exam material well and after some silly mistakes, scored in the mid-80’s % on a particular exam… and was devastated to the point of depression. Why? Because the mean of the exam was a 95%, and the course was a weeder with a harsh curve which basically meant that, according to the curve, he basically failed the exam. He knew the vast majority of the material on the exam. But that didn’t matter, because grading was determined not by how much you know, but rather your scorerelative to everybody else’s score. </p>
<p>That’s why, given that weeding must take place anyway, I actually support having a 40-60% average. After all, the alternative is to have straightforward exams where the averages would centralize around 80-90%, such that a few silly mistakes would push you towards a failing grade according to the weeding curve. If you must flunk and weed people out, better to flunk out those who don’t actually understand the material rather than those who just make silly mistakes. </p>
<p>But the far better question is, why do you even need to weed people out in the first place? </p>
<p>More specifically, why do you have to weed out certain majors such as engineering, but not others? If weeding out was truly so valuable as a quality-control mechanism, then why not weed out every major? Why doesn’t, say, American Studies have a gateway weeder course?</p>