A series of discussions - part II

Colleges and Universities A-Z University of California - Berkeley
1

<p>This week’s topic: graduation rate.</p>

<p>From the most recent Common Data Set, for the fall 1999 cohort:</p>

<p>58% graduated in 4 years or less
87% graduate in 6 years or less</p>

<p>Graduate rate is, in my opinion, one of the most important factors of an undergraduate education, as well as the undergraduate institution. Let’s dissect this dichotomy further:</p>

<li><p>High graduation rate benefits students. The worst possible scenario for a student would be to invest years of his time and money only to not graduate. For every Berkeley student, graduating should be the foremost priority, much like survival is our most primative instinct in nature. Thus, we must aim to get as close to 100% as possible.</p></li>
<li><p>High 4-year graduation rate benefits the institution. With continued funding cuts Berkeley must look for more ways to increase the resources/student ratio. One way we can achieve this is simply to reduce the number of students, by having more students graduating “on time” and not staying those extra years at the university and eat up resources like class seats.</p></li>
</ol>

<p>So what is the problem? Again it is two fold:</p>

<li><p>The 58% 4-year graduation rate, compared with 90+% for HYPS. The difference is simply too large. If Berkeley is to improve and try to match the level of the best undergrad programs in this country right now then I believe the 4-year graduation rate needs to shoot way up. It’s tough to defend Berkeley when someone comes and says “well, only about half the students graduate in 4 years.” </p></li>
<li><p>The large discrepancy between the 4-year graduation rate and the 6-year graduate rate. About 30% of the students take 1-2 years longer to graduate. Why is this? The colleges supposedly have “unit caps” so that after a certain number of units you must graduate, yet the fact remains that about 30% take 1-2 years longer to graduate. This is bad for the university for reasons stated above, and these students are taking limited seats away from classes that are needed by incoming students, which then may promote those students to graduate late. I understand that some like the college and want to stay, but it just creates a bad situation.</p></li>
</ol>

<p>So what can be done? Well, graduation rate is actually something I’m not as worried about, since it has been steady on the rise:</p>

<p><a href=“http://www.berkeley.edu/news/berkeleyan/1999/0317/grad.html[/url]”>http://www.berkeley.edu/news/berkeleyan/1999/0317/grad.html</a></p>

<p>

[quote]
Among students in the entering freshman class of 1992, 82.8 percent graduated within six years, the standard academic benchmark for graduation rates. This compares to 80.5 percent for 1991’s entering class, 80.2 for 1990’s and 74.9 for 1983’s.
/quote]</p>

<p><a href=“http://www.scienceblog.com/community/older/archives/G/ucb1084.html[/url]”>http://www.scienceblog.com/community/older/archives/G/ucb1084.html</a></p>

<p>

</p>

<p>And this trend needs to continue. Policies that encourage students to graduate faster should be encouraged. One such policy was mentioned in the Daily Cal:</p>

<p><a href=“http://www.dailycal.org/sharticle.php?id=17811[/url]”>http://www.dailycal.org/sharticle.php?id=17811</a></p>

<p>

</p>

<p>Other factors that may affect graduation rates:</p>

<li><p>Major. Students who double-major, change majors, or decide on a major late in the game tend to take longer to graduate (DRab, don’t ask me for evidence; this is from personal experience/logic and I think many would agree; it would take me too long to look this up). So one thing that can be done is help students to decide on a major early and map out a plan for the next four years or further into the future. This translates into better and more available advising, as well as advertising advising better.</p></li>
<li><p>Studying abroad. I have heard many times that one of the reasons for delayed graduation is studying abroad. I’m not too familiar with the process but perhaps someone could fill this in.</p></li>
<li><p>Availability of classes. I believe one reason is that some students simply didn’t plan well or for some reason, couldn’t/didn’t get into the courses needed to graduate. 4th years tend to get priority, and that’s a good start. But early planning can help alleviate this, which ties back to #1, and increasing supply of impacted courses, especially those needed to graduate, should be made a priority.</p></li>
</ol>

<p>Of course there is always the issue of those who flunk out completely. I don’t see an end to weeders anytime soon (although I would like to), so right now I think a better focus would be to decrease the large gap between 4-year and 6-year graduation rates. If we do that, I think the university and the students will be in a much better situation.</p>

2

<p>There is also the issue of people transferring OUT of Berkeley. I would divide this into 2 categories.</p>

<ul>
<li>Those who transfer out of Berkeley to a 'better' school. For example, one of my old friend's boyfriend (now her fiance) transferred from Berkeley to Stanford, and this may have been a very good move for him because he later got into the MD/PhD program at Harvard. But that begs the question of why couldn't Berkeley retain a budding superstar like that? Why did he feel that he had to leave Berkeley for Stanford? Or, conversely, why can't Berkeley get superstars to transfer INTO Berkeley?<br></li>
</ul>

<p>*Those who leave Berkeley because of poor fit or because they can't hack it at Berkeley. For example, I have read about other Berkeley students who transferred to places like Cal Poly or even to San Jose State because they found Berkeley to just be too darn difficult. But then that begs the question of why Berkeley is even admitting students in the first place if they are going to fit poorly or are unable to do the work? By not admitting these students, Berkeley would free up resources for other students who are actually going to graduate. I believe a proper statistical data analysis would be able to discover correlating attributes for those students who are most likely to experience difficulty and thus transfer out, and Berkeley could then simply admit fewer students with these attributes. There is little sense in Berkeley expending resources on students who aren't going to graduate.</p>

3

<p>So students who transfer out are counted as those who do not graduate in the graduation rates? How do they calculate this?</p>

4

<p>I don’t think graduation rate is much of a problem. I don’t think that it is “one of the most important factors of an undergraduate education,” but maybe I misunderstand this statement. I am not sure if graduation should be a student’s foremost priority, either.</p>

<p>
[quote]
1. The 58% 4-year graduation rate, compared with 90+% for HYPS. The difference is simply too large. If Berkeley is to improve and try to match the level of the best undergrad programs in this country right now then I believe the 4-year graduation rate needs to shoot way up. It's tough to defend Berkeley when someone comes and says "well, only about half the students graduate in 4 years."

[/quote]
</p>

<p>I think your data might be off, or perhaps my data is off. According to the Princeton Review, Harvard’s 4 year graduation rate is 83%, while the 6 year graduation rate is 92%. As to Berkeley, 58% sounds like significantly MORE THAN half to me. Additionally, Berkeley has instituted MANY things in order to increase graduation rate, such as an overly severe unit cap and better advising via the Finding Your Way program (in L and S as of this year). According to my calculations from Stanford’s common data set, their four year graduation rate doesn’t seem so hot. Now, Princeton seems good, but the other two, not so much. Maybe you have some more easily accessible data, or something which is probably more accurate than the Princeton review?</p>

<p><a href="http://www.stanford.edu/home/statistics/%5B/url%5D"&gt;http://www.stanford.edu/home/statistics/&lt;/a>
<a href="http://www.princetonreview.com%5B/url%5D"&gt;www.princetonreview.com&lt;/a&gt;&lt;/p>

<p>THE UNIT CAPS, as far as I can tell, are fairly NEW!!! Their effects will be seen in the coming years.</p>

<p>People need time to take classes and explore things given that SO MANY of the things at Berkeley were completely unavailable to them in high school. Vic, of course double-major, change majors, or decide on a major late make it take longer for people to graduate. This is obvious. I ask you for evidence for things that you should provide evidence for, not something which probably has very hard to find evidence and is pretty obvious if you know enough people. </p>

<p>Surely more planning would be good, but you can’t force a student to know what he or she wants to study and eventually do after college, merely help them along the way. The process seems to me primarily self-reflective. It is difficult to cause students to become self-reflective. New programs, such as the Finding Your Way program (started this year), are a good start to better advising.</p>

<p>Sakky, can data analysis indicate who’s going to be more likely to miss home and want to return closer because of family, close friends or a romantic relationship? Sure, do the data analysis (which isn’t going to happen anytime soon), but I think that the improvements would be fairly minimal. No school perfectly predicts who will do well in the environment and who will not, just as students do not perfectly predict where they will do well and where they will not, or how they feel about a place once they get there. The reality seems far less mechanically predictable than the golden data mining makes it sound. Perhaps I am just unfamiliar with the process and success of data analysis, but it doesn’t seem like solution which will solve much. But this way of approaching Berkeley I do not like. Is Berkeley a degree mill, and should it be?</p>

<p>
[quote]
So students who transfer out are counted as those who do not graduate in the graduation rates? How do they calculate this?

[/quote]
</p>

<p>Yes. They are included in the overall number of students, but never end up graduation, so are not included in the 4, 6, overall, or any other graduation rate. As far as I know, this is how schools are supposed to calculate graduation rate.</p>

5

<p>A minor correction brought to light by DRab, I wrote:</p>

<p>
[quote]
1. The 58% 4-year graduation rate, compared with 90+% for HYPS.

[/quote]
</p>

<p>I actually only recall seeing 4-year graduation rates for Harvard and Princeton in the past, and I remembered them as slightly over 90%. I assumed that Yale and Stanford would have similar figures. A closer look at the most Common Data Set showed:</p>

<p>Harvard: Not available on CDS</p>

<p>Yale: 88.5% in 4 years or less
96% in 6 years or less</p>

<p><a href="http://www.yale.edu/oir/cds.pdf%5B/url%5D"&gt;http://www.yale.edu/oir/cds.pdf&lt;/a&gt;&lt;/p>

<p>Princeton: 90% graduated in 4 years or less
97% graduated in 6 years or less</p>

<p><a href="http://registrar1.princeton.edu/data/common/cds2005.pdf%5B/url%5D"&gt;http://registrar1.princeton.edu/data/common/cds2005.pdf&lt;/a&gt;&lt;/p>

<p>Stanford: 76% graduated in 4 years or less
94% graduated in 6 years or less</p>

<p><a href="http://www.stanford.edu/home/statistics/#enrollment%5B/url%5D"&gt;http://www.stanford.edu/home/statistics/#enrollment&lt;/a&gt;&lt;/p>

<p>I'm not sure why there is such a large discrepancy between Stanford numbers and Princeton/Yale numbers, but in any case, there is still a huge gap between the percentage of Berkeley students graduating in 4 years and the percentage of Princeton/Yale/Stanford students graduating in 4 years.</p>

6

<p>
[quote]
I'm not sure why there is such a large discrepancy between Stanford numbers and Princeton/Yale numbers

[/quote]
</p>

<p>Athletics, in part.</p>

<p>
[quote]
there is still a huge gap between the percentage of Berkeley students graduating in 4 years and the percentage of Princeton/Yale/Stanford students graduating in 4 years.

[/quote]
</p>

<p>There is, and I think it is partially explained by grading practices and athletics at Berkeley, although I think other factors are relevant.</p>

7

<p>
[quote]
THE UNIT CAPS, as far as I can tell, are fairly NEW!!! Their effects will be seen in the coming years.

[/quote]
</p>

<p>The advisors have not hinted that the unit caps are anything new but you could be right. I haven't looked extensively into unit caps. But like I said, Berkeley's graduation rate is improving and there have been steps taken. I am just saying we should continue to encourage policies to drive graduation rates up because as of right now, we're still lagging far behind.</p>

<p>
[quote]
People need time to take classes and explore things given that SO MANY of the things at Berkeley were completely unavailable to them in high school. Vic, of course double-major, change majors, or decide on a major late make it take longer for people to graduate. This is obvious.

[/quote]
</p>

<p>Good. I'm glad it's obvious to you. Sometimes you like to question things which seem perfectly obvious to me.</p>

<p>On this topic, I wanted to add something else, which is that Berkeley should promote more exploration of majors by limiting major prerequisites. With many students undecided, and many more gunning for more than one major due to impacted majors, more and more students are preparing for more than one major, i.e. taking more than one major's prerequisites. This is not only very difficult but severly impairs the student's ability to explore outside his/her intended major. So I think some creative management on major prerequisites are needed here. Like I said, one option is to reduce the number of major prerequisites (does Haas really need all those prereqs AND 7 breadth? If students coming from L&S will eventually complete 7 breadth courses anyway. Just combine the two i.e. students do not have to complete 7 breadth before entering Haas. Rather, have the Haas breadth and the L&S breadth be one and the same, and require that no matter which college the student is in, for the breadth to be completed before graduation). Another option is to tweak prereqs so there are more overlapping prereqs. I'm sure (I'm hoping) others will have more ideas.</p>

8

<p>
[quote]
The advisors have not hinted that the unit caps are anything new but you could be right. I haven't looked extensively into unit caps. But like I said, Berkeley's graduation rate is improving and there have been steps taken. I am just saying we should continue to encourage policies to drive graduation rates up because as of right now, we're still lagging far behind.

[/quote]
</p>

<p>
[quote]
Note: The unit cap on these majors applies to freshmen admitted to UC Berkeley FL 04 or later and to transfer students admitted FL 06 or later. Students admitted to UC Berkeley prior to these dates are not subject to the unit cap.</p>

<p>If you are planning to declare a capped major, be sure also to prepare for an alternative major that is not capped.

[/quote]
</p>

<p><a href="http://ls-advise.berkeley.edu/choosingmajor/definition.html%5B/url%5D"&gt;http://ls-advise.berkeley.edu/choosingmajor/definition.html&lt;/a&gt;&lt;/p>

<p>I am uncomfortable with altering policy too much after this giant change has been instituted but the effects thus far unobservable. I think we should see how current policies actually play out to a further extent prior to changing too much more.</p>

9

<p>
[quote]
Sakky, can data analysis indicate who’s going to be more likely to miss home and want to return closer because of family, close friends or a romantic relationship? Sure, do the data analysis (which isn’t going to happen anytime soon), but I think that the improvements would be fairly minimal. No school perfectly predicts who will do well in the environment and who will not, just as students do not perfectly predict where they will do well and where they will not, or how they feel about a place once they get there. The reality seems far less mechanically predictable than the golden data mining makes it sound. Perhaps I am just unfamiliar with the process and success of data analysis, but it doesn’t seem like solution which will solve much.

[/quote]
</p>

<p>When car insurance companies can calculate how high of premiums you have to pay based on a host of prior data regarding your driving record, your demographic (men pay more than women, young men pay much more than everybody else), where you live (certain parts of the country are more prone to accidents or car theft), what kind of car you drive (sportscars are more likely to end up in accidents), what sorts of safety features you have on your car, and so on, I am fairly confident about the robustness of data mining as a statistical tool. In this particular case, I am not interested in determining causality. I don't care WHY a particular attribute is correlated with a higher dropout rate. All I care about is that the correlation exists. </p>

<p>The question of causality would be part of a second-order analysis that would attempt to make changes in Berkeley to ameliorate the number of dropouts that occur. For example, if it is determined that students who come from a certain kind of high school tend to drop out because they receive inadequate counseling, then providing more counseling to those students is in order. But if Berkeley doesn't want to take this step, then an easier way to solve the problem would be for Berkeley to simply not admit those students in the first place. Like I've always said, there is little to be gained from either the students' or from Berkeley's standpoint to bring in students who aren't going to graduate. The students waste their time, and Berkeley wastes its resources. </p>

<p>No doubt we are not going to reach a 100% graduation rate by this method. I am not asking for that. But that doesn't mean that you don't try to optimize. Just like Toyota doesn't ever reach a perfect 100% yield and quality figure in its auto plants, but it always strives to get there. </p>

<p>
[quote]
Harvard: Not available on CDS

[/quote]
</p>

<p>About 87% for 4-year graduation. Available here.</p>

<p><a href="http://nces.ed.gov/ipeds/cool/screen.aspx?screenId=60%5B/url%5D"&gt;http://nces.ed.gov/ipeds/cool/screen.aspx?screenId=60&lt;/a&gt;&lt;/p>

<p>
[quote]
Additionally, Berkeley has instituted MANY things in order to increase graduation rate, such as an overly severe unit cap

[/quote]
</p>

<p>Actually, a unit cap of some sort has existed for at least 15 years, probably more. I believe the cap is now much stricter than it used to be, but a cap of some sort still existed in the past. </p>

<p>
[quote]
On this topic, I wanted to add something else, which is that Berkeley should promote more exploration of majors by limiting major prerequisites.

[/quote]
</p>

<p>This is a fine idea. I would also propose providing a system of waiver exams to allow people to skip over various prereqs, especially those prereqs for which no AP exam exists. For example, I knew a guy who was a math whiz and, on his own time during the summer, had already mastered math all the way up to Real Analysis. So why would Berkeley force him to take multivariable calculus and linear algebra? If he could prove that he could do well on the final exams of both classes (and I'm sure he could have), then he should have been able to skip over those classes and go right to upper-division math. </p>

<p>Another method is to make prereqs 'soft' in that they just become guidelines but are not binding, such that if you have to skip a prereq and still do well in later classes, you can just have the prereq waived later on. For example, I know one guy who studied Chemical Enginering at Berkeley and did well despite never having taken Chem1B (because he had problems in getting into the lab). But he took all of the upper-division chemistry courses required for ChemE and got A's in all of these courses. So clearly he didn't really need Chem 1B, and he later on got it waived. But why not institute this as an actual policy? The truth is, plenty of prereqs aren't really true prereqs in the sense that you don't really need to know all of the stuff in that course in order to move on to later courses, and what you do need to know can often be obtained via self-study. For example, you don't really need to know very much of Chem1B in order to do well in upper-division chemistry courses like 112, 120, or 104. Chem112 is a case in point as Chem112 is basically just a faster version of Chem 3, because it's all just OChem, yet Chem3 doesn't require 1B. </p>

<p>Another change I would propose is lifting the requirement that AP exams will be counted if you took those exams only before you enrolled at Berkeley. Why only before? What's so bad about a student receiving AP credit for an AP exam he took while he was already a matriculated student? </p>

<p><a href="http://ls-advise.berkeley.edu/faq/ap.html%5B/url%5D"&gt;http://ls-advise.berkeley.edu/faq/ap.html&lt;/a&gt;&lt;/p>

<p>I would also similarly ask why can't AP exams be used to fulfill breadth requirements. Why not? I can perhaps understand not wanting to grant breadth to somebody who only scored a 3 on the AP (as a 3 is rather mediocre), but if somebody scored a 5 in something, I have no problem in granting that person some breadth credits. For example, if you're completely fluent in a foreign language such that you can easily swing a 5 on the foreign language AP, then I think it's safe to say that you're a pretty broad person and so you should be able to get away with taking fewer actual breadth courses. I would couple that with the fact that some foreign language AP's are not recognized by Berkeley (i.e. the Chinese and Japanese language AP exams), and many languages don't even have a corresponding AP exam (i.e. Hindi/Urdu, Vietnamese, Korean, Tagalog, etc.) despite being fairly popular languages within the Berkeley student community. For these students, I would say that if they can pass some sort of a waiver exam proving their fluency, they should be able to get credit for some breadth courses. For example, I think it's ridiculous that a student who is fluent in Chinese (i.e. was actually born and raised in China/Taiwan) might still feel compelled to formally take the basic Chinese courses just to fulfill breadth requirements.</p>