Are College Remediation Rates SOLELY K-12 Schools' Responsibility?

It seems to me, from the outside looking in, that the reformy argument for the Common Core and its aligned tests, such as the PARCC, is largely based on the "problem" of college remediation rates.

JerseyCAN says the Common Core will help remediation rates. Community college presidents say Common Core and the PARCC will help close the "preparation gap." NJDOE officials say the Common Core and PARCC will "remedy" the problem of college remediation. The PARCC people themselves say aligning instruction with the test will help bring remediation rates down.

This is as good an example as I have seen of how the reformy mind works:

• College remediation rates are too high, because...
• High schools are graduating too many students who aren't really ready for college, therefore...
• We need to design K-12 instruction with the goal of "college readiness," which means...
• We must have tests like the PARCC which predict that readiness.

It's all so logical, isn't it? One point flows to the next...

But neither the premises nor the logical connections between them ever get challenged. No one ever stops to think:

• Are college remediation rates really unacceptably high?
• If they are, is it really the fault of K-12 schools that so many students aren't "college ready"?
• Is it really better for society if we withhold high school diplomas from those who aren't "college ready"?
• Are students really not aware of their own "college readiness"?

Carol Burris wrote a couple of posts this year for the Washington Post -- see here and here -- that shed light on the question of whether college remediation rates are all that bad. I want to explore her ideas and some others in future posts, but for right now, let's focus on this:

Can the blame for a lack of "college readiness" be put entirely on our K-12 system?

Burris starts her second post with an interesting anecdote:

I was standing on line at a local sandwich shop, when the young man behind me got my attention. He had graduated a few years ago from my high school, and he was anxious to share good news.

He beamed as he told me how well he was doing at Hofstra University. He was studying what he loved, and doing very well. He did not start out at Hofstra, however. Tom began at a local community college and then transferred to the four-year university.

I remember him as a great kid who struggled in math due to his learning disability. He was able to get through geometry, but math beyond that was too tough. After he told me about his success at Hofstra, I asked him how he did at the local community college where he began.

The young man said that after taking a test with, as he described it, “math I had not seen in years”, he was placed in two remedial courses. He got through them, and his family luckily did not mind spending tuition on classes for which he would get no credit. In English, he went right into a college level course.

I asked him if he felt more prepared for college math after having remediation. He laughed and described the credit0bearing course as “really not math at all…it was easier than high school.” Why this school insisted that he take two, non-credit bearing courses, only to then have him take a course he could have successfully completed without remediation, is something that neither he nor I understand. [emphasis mine]

This story highlights a real problem in making the claim about a lack of college readiness: there really isn't an objective definition of what "college readiness" actually is. I know there is a movement afoot to tie this to the SAT, but as Burris notes, colleges aren't even in agreement on the cut score that should constitute "readiness."

But let's set that aside and look at the highlighted quote: "math I had not seen in years." The implication here is that there was a time of at least a couple of years between this young man's last math course and his community college placement test. New York State requires 3 units (years) of math to earn a diploma. So it's reasonable here to think he had been out of high school at least a year before enrolling in community college.

Is this typical? Well...

This comes from the American Association of Community Colleges. 63 percent of community college students enrolled for credit are age 22 or older -- which means it is reasonable to assume that many of those students haven't had a high school math course for years.

I'm not a cognition and learning expert, so I won't claim to know the literature. There is at least some evidence, however, that memory of learning falls off quickly for those who don't achieve a high level of proficiency. But even if you do, you'll experience a sharp drop in memory early, followed by a period of stabilization. Of course, the issue is highly complex and outcomes depend on many factors.

As a music teacher, I know that it doesn't take long for facility at playing an instrument to drop off quickly, especially if you haven't achieved high levels of playing before (although there are other benefits to music training that have long-term benefits). This is almost certainly true in other domains. Learning is contextual, so if you lose the context, you almost certainly lose at least some of the learning. Given all this...

Is it at all reasonable to hold K-12 education responsible for all deficits in learning for those students who haven't been in high school for years?

You can't hold people or systems responsible for things that you can't attribute to them. Yes, I'll admit the K-12 school system can't shrug off it's responsibilities by claiming: "Hey, you won't remember any of this stuff anyway!"

But it is unreasonable to attribute a college student's math proficiency entirely to his K-12 school system if he hasn't been in that system for a long time. In other words: it's not necessarily your high school geometry teacher's fault if you need a refresher course in geometry years after you were in her classroom.

My point here is that even if we set aside highly subjective nature of college remediation rates, using them as "evidence" of K-12 failure is extremely problematic. We need much better research to show that insufficient instruction in high school and before is leading to the current community college remediation rate.

Once again, this is the problem with so much "reformy" thinking -- it is a victim of Yogi's Fallacy:

"We may be lost, but we're making good time!"

Before rushing off to "fix" our schools, we ought to think a lot harder about why outcomes are the way that they are, and whether our "fixes" will actually make things worse. More in a bit...


ADDING: Ajay in the comments points us to this post from Judith Scott-Clayton:

While remediation rates have risen slightly over time — to 22 percent of all first-time first-year students in 2003-4 from 18 percent in 1995-96, according to Department of Education statistics— the increases have been striking for students with strong high school grades. 

For students with high school grade-point averages between 3.5 and 4.0, remediation rates have more than doubled (see chart below). This is not a result of high school grade inflation – the percentage of students with G.P.A.’s in this range has not changed – but is consistent with increasingly ubiquitous placement testing.

Department of Education, National Center for Education Statistics, Beginning Postsecondary Students database; computation by N.C.E.S. QuickStats.

Screening seemingly prepared students for remediation is questionable for at least two reasons. First, the benefits of remediation are far from obvious: remediation has been referred to as the Bermuda Triangle of postsecondary education, because the majority of those who enter never make it out. 

Across several rigorous, quasi-experimental studies of the causal impact of remediation, only one found positive effects on college outcomes, while others found null to negative effects. 

Second, the tests commonly used to screen for college readiness are only weakly related to college outcomes, as two recent studies by the Community College Research Center show. (Disclosure: I am a senior research associate at the center and the author of one of these studies.) Some students manage to pass the tests even though they are not ready for college-level work, while even more who are ready for college-level work are kept out. 

My own research, using data from a large urban community college system with particularly high remediation rates, estimates that one in four students assigned to math remediation could have passed a college-level math course with a grade of B or better and one in three students assigned to English remediation could have passed freshman composition with a B or better.

Off hand, I think the GPA argument is a bit more complex: if we're getting more high-GPA students from lower-performing schools entering colleges, that could explain some of the increases in remediation.

But Scott-Clayton's larger point is solid: remediation is a questionable policy. And, again: how much of the "problem" of remediation is attributable to the K-12 system?

Big, knotty topic -- the sort of stuff we love here...