20 April 2008

Flunking Out

A high percentage of Colorado high school students drop out. Many who do graduate don't go to college. Many who start college don't finish.

Too Much Respect?

The Denver Post today focused on the low percentage of students who graduate from public colleges in the state, and on the impact on ethnicity and preparation for college retention. Two big barriers to college are somewhat counter intuitive: excessive respect for authority and aversion to debt:

Misinformation runs rampant. One parent thought students had to pay for all of college upfront. Others are gripped by fear about taking out even a modest loan to pay for books or tuition.

"I'm changing my students' minds, but I'm also changing my parents' minds," Romero said. "I have to say, 'A student loan is not always a bad thing. It's a personal investment.' "

Students are coached to fight and advocate for themselves. They are taught to question professors about a bad grade and push for help understanding a tough math assignment.

"They come from traditional families where they don't question authority," she said.

The numbers present a less convincing case that ethnicity is important in retention rates. But the system does have a high tolerance for failure.

Retention Numbers

At the state's more selective institutions (shown with enrollment), CU-Boulder (31399), CSU-Ft. Collins (26723) and the School of Mines (4056), about one in three students fail to graduate within six years.

At the University of Colorado-Colorado Springs (8583), and the University of Northern Colorado (12981), fewer than half of students graduate in six years.

At Adams State (4899), CSU-Pueblo (6205), Fort Lewis (3907), Mesa State (5938), University of Colorado-Denver (19766) and Western State (2094), only about one out of three students graduate in six years.

At Metropolitan State in Denver (20761), only one out of four graduate in six years.

The situation in community colleges is even more dismal. At Morgan Community College, only 30% are still in college a year after they start. The Community College of Aurora, the Community College of Denver, Trinidad State Junior College, Otero Junior College, Front Range Community College and Colorado Northwestern Community College all lose more than half of their students in the first year.


Six year graduation rates are hardly a strict measure of student retention for college students who pursue full or near full time programs. Almost all of the other four year colleges mentioned fit a more traditional model than Metro State to some degree.

At Metro State, where a large share of the student body is part-time or taking classes on a continuing education basis, rather than genuinely working towards degrees, a six year graduation rate measure may distort the numbers, although it is worth noting that a large share of part-time college students pursuing four year college degrees never complete their degrees in any case.

One year retention is a very reasonable measure for community college retention.


Any way you cut it, a large share of the students who start college at public colleges in Colorado will never finish.

Few people deny that spending significant sums of money to allow residents of Colorado to graduate from four year colleges and earn degrees at community colleges, despite limited financial means, through a government subsidy, is a worthwhile end. It is good for Colorado's economy and it advances the ideal of equality of opportunity more effectively than anti-discrimination laws.

The harder question is whether the state is doing anyone any favors by admitting huge numbers of students who will ultimately drop out.

The Argument For The Status Quo

There are essentially two arguments for the status quo.

First, some people believe firmly that attending college is a good investment in individual human development, even if those individuals never earn degrees.

Second, many people believe that giving people the opportunity to sink or swim in an actual college environment is preferable to making that decision for them through a selective college admissions process.

Certainly, every year, some people who get into college despite poor graduates and test scores manage to graduate, while some people who are well academically prepared for college fail. Moreover, there will always be a gray area. Where should the cutoff be? A 10% chance of success? A 30% chance of success? A 50% chance of success? A 75% chance of success? A 90% chance of success? Many of the nation's law schools have 95% or better retention rates, and almost all have at least a 75%-80% retention rate. In contrast, no four year public college in Colorado has better than a 68% retention rate. Why should the different levels of education be treated so differently?

Also, it is worth noting that students on the college admission margins are disproportionately minorities. Even though academically weak students graduate less often than academically strong students, some consistently small percentage of them do graduate. Those academically weak students who nevertheless are admitted to college and graduate are disproportionately minorities and dramatize the costs of a selective admissions system that prevents people from even attempting college if they are too academically weak.

Also, no all college dropouts are created equal. The value of college is probably greatest for students who almost graduate, at a greater cost to the high education system. The value of college is probably least for students who drop out after only a semester or two, particular if, as is the case in most academically weaker students, this happens in a less expensive per student community college or non-flag ship four year college.

The Argument Against The Status Quo

There is some merit to both arguments, but there are strong arguments that this is no way to run a higher education system as well.

We can't know precisely who will graduate and who will drop out, but people who do poorly on traditional college admissions measures, like high school GPA, the mix of courses taken in high school, extracurricular involvement, ACT scores and SAT scores are meaningful predictors of whether or not someone will graduate from college. Academically strong applicants are much more likely to graduate than academically weak applicants.

Also, students who drop out of college tend to be the students who were failing academically. While, attending some college courses may add value to particular students, students who are academically weak to the point that they drop out are least likely to be benefiting from the courses that they are taking. The large percentage of students taking remedial courses in the state's community colleges and less selective four year colleges are a symptom of this problem.

The current system's open enrollment system doesn't do a great job of moving minority students towards graduation.

Pikes Peak [Community College] loses about 63 percent of black males after the first semester and about 82 percent after the first year.

To a meaningful extent, our secondary and higher education system is a time consuming and expensive test that sorts people by ability. For employers, the sorting value often exceeds the value of what is actually learned in high school and college.

We know from overwhelming data that students who graduate from high school do better economically than students who drop out of high school, that students who attend some college after graduating do better economically than students who merely graduate from high school, and that students who graduate from college do better economically than students who attend college and then drop out.

At face value, this seems to show that higher education adds value. But it isn't at all clear how much of the increased economic success of those who have some college comes from attending college, per se. Much of the economic value of education is a mere sorting effect.

Most high school dropouts were failing in high school (and had disciplinary problems) before dropping out. They drop out because they are less intelligent or less diligent (or both) than students who graduate from high school.

Students who apply to and are admitted to a college program are, on average, more intelligent and diligent than students who never even attempt college.

Success in college is meaningfully related to the intelligence and diligence that the graduating students had when they started, and success in college course work is closely tied to college completion.

There are much cheaper ways to sort people by ability if that is all that one wants to do. Particularly in the case of people who attend college and drop out, it isn't at all clear how much of the added economic value of college attendance comes from the sorting involved, and how much comes from what students who drop out of college learn in college. Partial evidence that the sorting value of college is more important than the education value for these students is that students who earn associates degrees do not fair better economically than students who drop out of four year programs.

If indeed college doesn't add much value to the average student who drops out of college, then the logical conclusion is that we should radically overhaul the way that we run our higher education system. We should be far more selective in college admissions, pretty much across the board, to free up resources now spent to subsidize students who flunk out or drop out on students with a greater chance of actually graduating.

One could easily imagine that tightened admissions standards would reduce the number of students graduating from public colleges in Colorado only negligibly, while reducing the number of students enrolled in institutions of higher education in the state by 25% or more. Indeed, the improved student-teacher ratios and higher educational resources per student might improve the value added by the system to the students who remain in it.

The removal of a huge swath of students who are likely to fail from the system could also change the prevailing attitudes on college campuses. A culture in which failure is the norm could be replaced by one in which success is expected. It would also encourage professors to develop less of a sink or swim attitude towards students, impossible in a context where most students are doomed to fail even if a professor makes herculean efforts to support them.

On the other hand, perhaps students who drop out, through their tuition dollars, are actually subsidizing their more academically successful peers. In a college lecture class, the cost of teaching a class with 80 students instead of 60 is modest, but the increased revenue per faculty member is great. Indeed, this kind of subsidy may explain why non-flag ship four year colleges are cheaper per student than flagship colleges, and why community colleges are cheaper per student than non-flag ship four year colleges.

Even if academically weak students do subsidize better students, one has to ask if the state is doing anyone any favors by perpetuating that system. Would academically weak students who are just barely strong enough academically to be admitted a non-flagship four year college or community college be better off with more work experience and income from working, instead of sitting in classes where they fail? If it fair to pay for the education of academically strong students on the backs of academically weak students?

Middle ground?

One approach would be to tighten admissions standards across the board in the four year college system, while leaving the community college system on its existing opens admission basis.

If we are to have system based upon weeding out students based upon actual performance in college, rather traditional admissions criteria, community college is the least expensive place for that experimentation for students and the state alike.

Enrollments at the state's four year colleges would be reduced. Many students who would have attended CU or CSU or the School of Mines in prior years, would be squeezed to less prestigious public four year colleges. Many students who would have attended non-flagship four year colleges would be squeezed into the community college system. Four year colleges statewide would have the smaller classes and greater climate of success that would help first generation college students succeed.

Community college enrollments would surge. But students who would have been admitted to non-flagship four year colleges in prior years would have a more meaingful chance of sticking it out for at least a year or two, than existing community college students. Indeed, these students would probably start to dominant second year classes in community colleges. This might increase academic standards in community colleges and also might greatly increase the number of students who end up transferring to four year college.

Today's article in the Denver Post doesn't provide a figure, but the number of students transferring from community college to four year colleges is not a large percentage of either community college students or of four year college students. But the numbers show that those community college students who do persist and transfer to a four year college do just as well as students who begin their studies at a four year college.

A recent study provided these figures:

The Community College Transfer Study, completed . . . by CU’s Boulder, Denver and Colorado Springs campuses, confirmed that the academic performance of Colorado community college transfer students from 13 state system community colleges and two local district community colleges is a “record of success.” The study considered the performance and relative success of 18,635 students who entered the three CU general campuses as undergraduates between the summer of 1995 through the spring of 1997. The review included 2,333 community college transfer students, 6,253 transfers from mostly four-year institutions and 10,049 first-time CU freshmen. The study considered first term and degree grade point average as well as the percentage of transfer and native students graduating. . . .

* Community college students transferring to the Boulder campus graduate with an average GPA essentially the same as those of native students;
*Community college transfer students to the Colorado Springs campus have higher first-term GPAs and graduate with GPAs nearly identical to other students. Additionally, community college transfer students graduate at a higher rate than do native students;
*At UCD, 51 percent of community college transfer students graduate compared to 40 percent of native students.

Perhaps six year graduatation rates at flagship schools could increase from two-thirds to three-quarters or four-fifths. Perhaps at non-flagship four year colleges, six year graduation rates could increase to the two-thirds that are the norm at flagship schools now. Perhaps at community colleges, two-thirds of students could make it to a second year, and many would complete associate degrees, at least.

There is no evidence that those would managed to transfer would be worse off, and those who dropped out of community college would still be better off than those who started at a more expensive four year college (or might get, at least, an associate degree rather than simply an unresolved fizzling out of their educational experience).

The total number of bachelor's degrees might not change much, or more even fall slightly, but far fewer Coloradans would be marched on a path to failure to may discourage them from ever pursuing education again.


Dave Barnes said...

This is a PARENTAL issue.
I remember when I moved to Colorado in 1980 and was shocked at the attitudes of the "Oakies" that I met here about schooling. Coming from the white-bread suburbs of Boston, it was a huge shock.
In the suburbs of Boston, education was everything. Massive emphasis in high school on graduating and going off to higher education (trade school or college).
In greater Denver, it was "who cares".

Andrew Oh-Willeke said...

I don't doubt that what you experienced, but I also think that casting that experience as bad parenting is neither constructive nor a fair description of what is going on.

Maybe I'll post on what I think about that when I get a moment.