Get important education news and analysis delivered straight to your inbox
When Alexandra Logue served as the chief academic officer of the City University of New York (CUNY) from 2008 to 2014, she discovered that her 25-college system was spending over $20 million a year on remedial classes. Nationwide, the cost of remedial education exceeded $1 billion annually; many colleges operated separate departments of “developmental education,” higher-education’s euphemistic jargon for non-credit catch-up classes. “Nobody could tell me if we were doing it the right way,” Logue said.
She suspected they weren’t. More than two-thirds of all community college students and 40 percent of undergraduates in four-year colleges had to start with at least one remedial class, according to a statistical report from the U.S. Department of Education. The majority of these students dropped out without degrees.
An experimental psychologist by training, Logue designed an experiment. She compared remedial math classes to the alternative of letting ill-prepared students proceed straight to a college course accompanied by extra help. The early results of her randomized control trial were so extraordinary that her study influenced not only CUNY in 2016 but also California lawmakers in 2017 to start phasing out remedial education in their state.
Over the seven years of Logue’s study, which took place at three of CUNY’s seven two-year community colleges, the results kept getting better. Students who started with college math were successfully passing the course at a fraction of the cost of remediation, getting their math requirements out of the way, earning their degrees faster and earning thousands more in the labor market. Many public colleges, from Nevada and Colorado to Connecticut and Tennessee, have followed suit, phasing out remedial ed.
Other data analyses have also shown benefits to bypassing remedial education, but this was one of the only real-life experiments, like a clinical trial, and so it carried a lot of weight. Most importantly, it studied math, often an insurmountable requirement for many students to complete their college degrees. This study has arguably been one of the most influential attempts to use experimental evidence to change how higher education operates and is now affecting the lives of millions of college students.
“It’s a great feeling of satisfaction,” said Logue, now a research professor at CUNY’s Graduate Center, “because it isn’t just CUNY. It’s across the country, using this really great evidence to help make things better.”
The third and final chapter of this long-term study was published in the January/February 2023 issue of the journal Educational Researcher, and as I pored over this body of research, I became confused about what it proved. The study could be seen as evidence against remedial education, but it could equally be seen as evidence for letting college students meet their math requirements without taking algebra.
The confusion stems from the study design. Instead of testing remedial versus college algebra, which would be a direct test of remedial education, the study compared remedial algebra to college statistics, a sort of apples to oranges comparison. In the experiment, CUNY randomly assigned almost 300 students who failed the algebra portion of a math placement test to an introductory statistics course. In tandem with this college class, students attended an extra two-hour workshop each week where a college classmate who had already passed the class tutored them. Researchers then compared what happened to these stats students with a similar group of almost 300 students who were sent to remedial algebra, the traditional first step for students who fail the algebra subtest. Logue had the same teachers teach sections of both courses – remedial algebra and college stats – so that no one could argue instructional quality was different. Also, only students who struggled with algebra, but not arithmetic, were part of this experiment; students with more severe math difficulties, as measured by the freshman placement test, weren’t asked to attempt the college course and were excluded from the control group.
By all measures, the students who went straight to college stats did better. More than half of the students who bypassed remedial algebra passed the stats class and earned college credit. Ultimately, these students finished their degrees a lot faster than those who started off in remedial algebra. They were 50 percent more likely to complete a two-year associate’s degree within three years and, according to the latest chapter of this seven-year study, they were twice as likely to transfer to a four-year institution and complete a bachelor’s degree within five years. Seven years after bypassing remedial ed, students were earning $4,600 more a year in the workplace, on average, than those who started in remedial math.
“What we can say is, for students who have been assigned to remediation, put them into statistics with extra help, and you will get a good result,” said Logue.
Some researchers argue that the shift to statistics might have made the difference.
“That switch from algebra to stats is a big one for a lot of students,” said Lindsay Daugherty, a senior policy researcher at the RAND Corporation who has studied remedial education and efforts to reform it. She said all the other studies that have looked at replacing remedial classes with college courses plus extra support haven’t produced better graduation rates. “This CUNY study is the only one,” said Daugherty.
The only other randomized control trial of remedial education is Daugherty’s Texas experiment to replace remedial English courses with college courses plus extra support. Going straight to college courses helped more students earn college credits in English but that didn’t help them get through college. Dropout rates were the same for students in both the remedial and the “corequisite” courses, as the college plus extra help version is often called.
“We know that the way that we did it before with these standalone [remedial] courses was not helping students, and most states and colleges have made a change and are moving towards corequisites,” said Daugherty. “But the evidence does not suggest that these corequisite courses are the magic potion that is going to change completion and persistence. It’s going to take a lot more and a lot of other support.”
What we don’t know from this study is how to help students who are behind in math learn college algebra, a course that is similar to intermediate high school algebra, which remains a requirement for many business, health and engineering majors. All the students in this landmark CUNY study had intended to major in non-STEM fields that didn’t require algebra, such as criminal justice and the humanities, and for which college statistics would fulfill their math requirements.
Logue originally sought to conduct a simpler, cleaner study of only algebra, comparing the remedial prerequisite to the college course plus tutoring support. But she ran into problems with the algebra faculty. (There were too many different versions of college algebra for different majors and across different colleges at CUNY, each covering different topics, she said, and it was impossible to test one version of a basic college algebra course.) Meanwhile, the statistics department was open to the experiment and their introductory courses were very similar from professor to professor.
It’s unclear from this study how essential the weekly tutoring sessions were to helping students pass the statistics course. The experiment didn’t test whether students could pass the normal college stats class without peer tutoring.
The good news is that the switch from remedial algebra to college stats didn’t seem to harm anyone. Indeed, the students in the statistics group were just as likely to complete advanced math courses, along the algebra-to-calculus track, as students who started with remedial algebra, according to co-author Daniel Douglas, director of social science research at Trinity College in Hartford, Connecticut, who led the data analysis. In the final number crunching, the stats students were just as likely to complete math-intensive degrees that required college algebra. Starting with stats didn’t thwart students from changing their minds about their majors and returning to an algebra-to-calculus track, Douglas said.
The bad news is that a lot of community college students still fell through the cracks. Although there was a 50 percent boost to the number of students who completed an associate’s degree within three years, only a quarter of the statistics students hit this milestone. Almost three-quarters didn’t. And though bypassing math remediation and heading straight to college stats led to a 100 percent increase in the number of bachelor’s degrees, only 14 percent of the statistics students earned a four-year degree.
The main benefit of allowing students to bypass remedial classes is speed, according to Douglas. Over the course of seven years, the students who started in remedial algebra eventually caught up and hit many of the same milestones as the students who started with statistics. “At the end of our data collection in the fall of 2020, their degree completion – the elementary algebra group and the stats group – they’re not that different,” said Douglas. As those students enter the workforce and gain experience, it’s quite possible that their wages will catch up too.
A CUNY spokesperson told me that their college system stopped placing new students into remedial classes in the fall of 2022. For students who are behind in math, there are now “corequisite” math classes, where the extra support is more costly and differs from the tutoring that was tested in this study I am writing about here. Now the college-level course is two hours longer each week, blurring the lines between regular instruction and extra help support, and entirely taught by instructors, not peer tutors. Many instructors who used to teach remedial courses now teach these corequisite courses.
For students who are significantly behind — struggling not only in algebra, but also in basic arithmetic — CUNY operates a separate pre-college program, called CUNY Start, where students take only remedial classes. These students haven’t yet matriculated at the college and don’t pay tuition, and so CUNY doesn’t count them as students. And the numbers of students in this pre-college remedial program had been swelling before the pandemic.*
Students did better in these newer pre-college remedial classes than those who took traditional remedial classes, according to a separate 2021 study that Logue was also involved in. But these students aren’t necessarily doing better in college and earning more credits, unless they get a lot more advising and counseling support during their college years. Helping more young adults get through college isn’t going to be easy or cheap.
*Clarification: This paragraph has been modified to reflect that the CUNY Start program dates to 2009 and the number of students in it grew during the 2010s. Enrollment in CUNY Start has decreased in recent years, mirroring the general drop in enrollment at community colleges. An earlier version implied that the CUNY Start program was new and that the number of students in it is still increasing.
This story about remedial math in college was written by Jill Barshay and produced by The Hechinger Report, a nonprofit, independent news organization focused on inequality and innovation in education. Sign up for Proof Points and other Hechinger newsletters.
At The Hechinger Report, we publish thoughtful letters from readers that contribute to the ongoing discussion about the education topics we cover. Please read our guidelines for more information. We will not consider letters that do not contain a full name and valid email address. You may submit news tips or ideas here without a full name, but not letters.
By submitting your name, you grant us permission to publish it with your letter. We will never publish your email address. You must fill out all fields to submit a letter.
I was a math major and did some grad school in physics, but i tutored people in math during undergrad. Statistics is much simpler because for most people because they’re using real things, situations people conceptually understand, like exponential growth. Almost all the concepts even the complicated ones can be explained with a drawing.
The place that people break down in algebra is polar coordinates. They just cannot figure it out. There’s no fun way to learn it you just have to use it enough until it’s natural and you don’t have to think about it. (There are really good videos on YouTube for this now, for example a video of a line going in a circle and showing how it draws a sine wave.)
Once you remember the unit circle, you have to build an unnatural intuition in your brain for converting between coordinate systems. I don’t think this is easy but that type of thinking is really important for high-level math.
There’s so many assignments that we had where we would go to the library and sit together and we would just stare at the same problem for hours and have no idea how to do it, not even how to start. And these would be proofs where you can’t start a problem and then learn something from whatever strategy you chose not working. So instead you just stare at it and then you go home and you sleep and you wake up and suddenly you just get it. It’s just happens, that your brain understands and you can just do it as if it was easy.
Trying to explain the satisfaction, the rush of epiphany to people is impossible. You can’t get it unless you’ve experienced it.
But the core issue is that people do not like to have to be logical. They don’t want to be originally constrained by rationality. I could go into why I think this is becoming a huge problem for society but suffice to say that the reason that they can’t do math isn’t because they’re incapable, it’s because they don’t like being forced to do something where there’s only one right way to do it and you have to for years just sit there be told how to do something and just do it. It took thousands of years to get to the point where we had calculus and now it seems obvious that you could use numbers that way. And the way we got to calculus was logic. Mathematicians used to fight over whether or not numbers were like arbitrary 😂
The point is parents don’t know how to teach their kids logic anymore. In one sense I lucked out and was born into a family that’s almost exclusively engineers. But in other sense that has me navigating society very difficult because it’s listen to see someone say something that is not internally consistent and have to treat that with seriousness.
I think that some universities will use any excuse to move students ahead without making them take quantitative (read: difficult) courses, even or especially developmental ones. The skills used in statistics, and the ones we see being stressed here, are good skills to have on their own, but the skills needed for algebra and the other math used to measure things can be completely different. As professors we need to know that students can think methodically and apply themselves to learning linked concepts. The students need to build confidence with numbers, as well. With statistics, we find out whether students can interpret a histogram, but often that requires little to no knowledge of how median, mean and/or standard deviation come together to summarize knowledge. How does one teach standard deviation when students don’t understand basic algebra? How can we teach testing of hypotheses without understanding what goes into a t-statistic (as a ratio)? It’s teaching definitions, at best. Most of us try to build on that math knowledge at upper levels, or with our graduate students, and it’s tough enough trying to teach people how to measure things without taking away their foundation of math learning to start with. Finally, I suspect that many of the people promoting the “jump” approach 1) don’t have tenure, and 2) don’t teach quantitative courses regularly. The folks without tenure are pressured to move students through without regard to student outcomes (understanding, not grades). And that’s how we end up having this debate.
The two-hour weekly tutoring workshop that the Stats students received, but the remedial Algebra students did not, feels rather glossed over in this article. Couldn’t it be the case that this study is a vindication of one-on-one tutoring, rather than a discrediting of remedial Algebra?
Submit a letter