These days more four-year colleges are dropping their SAT and ACT requirements. Pandemic-era disruptions have played a role, but so has the broader worry that a reliance on standardized admissions exams may exacerbate disparities among students of different races and income levels.
Similar concerns have prompted community colleges to reconsider their own versions of entrance assessments: the placement tests they’ve traditionally used to determine whether students are “ready” for college-level learning or should instead start their pursuit of higher education in remedial math or English classes.
“It’s definitely part of the same conversation: How much of an emphasis do we put on a single test?” says Elisabeth Barnett, a senior research scholar at the Community College Research Center. “It was developed to be an equalizer. That’s not what we are seeing these days.”
Although remedial—also called developmental—education is intended to prepare students for advanced courses, research shows that it often blocks student progress instead. Only a fraction of community college students who start in developmental courses ever earn a degree, and many never even make it into college-level classes. Meanwhile, these students are paying for courses that rarely provide them with college credit. These negative effects fall disproportionately on low-income students and those from some racial minority groups.
“The bottom line has been that the types of developmental education in place for many years have not done a very good job of helping students ultimately complete degrees and credentials,” Barnett says.
So over the past few years, researchers, educators and policymakers have been working to reform community college developmental education. Moving away from standardized tests as the sole predictor of success has been one key tactic.
Two new, large-scale studies—one from New York, the other from California—show that reduced reliance on placement exams increases student access to college-level courses. But the findings also reveal that not all college faculty are eager to let go of these tests or to admit more students of varied readiness levels into their courses.
The “remedial mindset” that some professors have toward students whom they suspect are not fully prepared for college may hinder institution-level reform efforts, says Katie Hern, executive director of the California Acceleration Project and an English professor at Skyline College.
“Deficiency-oriented views of students are really common,” Hern says. “It’s a very long-held paradigm that is sort of hard to shake.”
The Math-English Divide
In October, the Center for the Analysis of Postsecondary Readiness published the results of a several-year study of nearly 13,000 students at seven community colleges in the State University of New York system. To determine placement in college-level or remedial courses, the students were randomly assigned either to a control group that used standardized exam scores—the normal procedure—or to an intervention group that used an alternative method: an algorithm.
For each student in the latter group, the algorithm drew on several pieces of data to make its decision. It considered placement exam scores but also took into account information from each individual’s high school records, including GPA and the type of diploma earned. It did not use demographic data.
The algorithm worked. It “bumped up” a significant share of students into college-level English and math courses compared to how they would have been placed by exam scores alone. And these students were 8 to 10 percentage points more likely to complete a college-level math or English course within three semesters. The results were especially promising for getting more women into college-level math classes and supporting women, Black students and low-income students in completing a college-level English course.
Meanwhile, the algorithm did bump some students down into developmental courses. Those bumped students were 8 to 10 percentage points less likely to complete a college-level math or English course within three semesters, meaning they would have done better if allowed to enroll right into college-level classes.
But the results were more impressive for English classes than math classes. That may have to do with how professors from those departments influenced the intervention.
When it comes to determining which students are assigned to remedial courses, “traditionally faculty do have a strong say,” says Barnett, who co-authored the study. So to develop the algorithm, researchers studied historical data at each college about student pass rates, then consulted with professors about what probable pass rate levels they were willing to accept. The theory was that allowing more students to go directly into college-level courses may lower overall pass rates but also boost overall access.
Departments were divided about how liberal the algorithm should be with placement. English professors “were more convinced the algorithm would do a good job,” Barnett says, while the math professors were “concerned whether the algorithm would do a good job.”
That difference in attitude showed up in the study results. Among students assessed by the algorithm, 44 percent were “bumped up” into college-level courses in English, compared to 16 percent bumped up into college-level courses in math. Gains in class enrollment and completion lasted three semesters for English, but only one semester for math.
Allowing math faculty to influence the selection algorithm meant “very few students got access to transferable college math,” says Hern, who was not involved with the study. “The impact of the project was really constrained by how conservative the faculty wanted to be about allowing students into transferable college-level courses.”
That’s not a fluke, she says: “You see that everywhere—that math faculty are much more conservative about reform and much slower to act.”
Math professor reluctance to allow more students in college-level courses comes in part from the fact that they tend to view their discipline as sequential, Barnett says, fearing that “if students don’t have the prerequisite content, they’re not able to master the next level.”
There is a strong belief among many professors that all students need to succeed in intermediate algebra in order to progress, agrees Olga Rodriguez, a research fellow at the Public Policy Institute of California Higher Education Center. Yet she also cites another barrier to broadening access to college-level math: teaching style.
In English courses, making curricula and instruction more culturally relevant to diverse students is a relatively “straightforward” task, Rodriguez says, whereas in math, “the traditional lecture-based approach where students practice problems and take tests and sit in lectures is the standard. The shift in pedagogy there is that much more challenging.”
Changing Policy—and Pedagogy
How might the process of teaching and learning look different if we truly believe every student is highly capable, and if we believe that their success is our responsibility?
California has been at the forefront of reforming remedial higher education, and a study published in November from the Public Policy Institute of California shows how the state’s sweeping policy change has produced significant improvements in student outcomes.
California law AB 705 required the state’s community colleges to totally flip their approach to developmental education by fall 2019. Previously, students had to prove they were ready for college-level courses, primarily by taking placement tests. Now, credit-bearing classes are students’ default option, unless institutions can prove that students are not ready and would benefit from remedial education.
“As a result of this law, the system leaders essentially eliminated the use of the standardized placement test for placement purposes at all colleges across the system,” says Rodriguez, who co-authored the study.
That’s led to a big shift in access to credit-bearing English and math classes. In fall 2019:
- 96 percent of students who took an English course for the first time enrolled in college-level composition, compared to 38 percent who had access in fall 2015.
- 78 percent of first-time math students took college-level math, compared to 21 percent who had access in fall 2015.
- Students completed credit-bearing English and math courses at more than double the rate of students four years prior. That translates to tens of thousands of students.
These largely positive changes for students have presented professors with new challenges, however. Interviews conducted for the study with faculty at community colleges revealed that some feel student failure is “more visible” now that students with a wider range of preparation levels are showing up in their classrooms. This has lowered morale for some professors, who feel newly responsible for students they may never have encountered under the old system. And among instructors who are supportive of the policy change, the report says, some “identified changing the ‘faculty mindset’ as a challenge to implementing developmental education reforms.”
Success did not touch all students in California equally. The report shows that completion of college-level courses rose about 20 to 25 percentage points for all racial and ethnic groups, but there’s still some gaps for Latino and African American students, and African Americans remain “substantially underrepresented, especially in math.”
The findings about persisting inequities and the role professors play in alleviating or exacerbating them was a key topic of conversation among panelists who discussed the study results in a webinar held Nov. 20.
“What we need to truly achieve not just equitable placement but equitable outcomes is an educational philosophy that all students can learn at any level and all students can achieve high academic outcomes,” said Aisha Lowe, vice chancellor of educational services at the California Community Colleges Chancellor’s Office, during the event. “How might the process of teaching and learning look different if we truly believe every student is highly capable, and if we believe that their success is our responsibility?”
Among the many reasons why some groups of students may do worse in college-level courses, panelists identified several factors that professors directly influence, including classroom culture, curricular choices and the interactions instructors have with students. Attitude matters, too: Panelists cited research showing that faculty beliefs about intelligence predict how well racial minority students do in STEM classes.
When digging into classroom-level data, “the variation in success rates across instructors is shocking. It is not uncommon to see success rates that vary between 20 to 90 percent for the same course at the same college,” said Myra Snell, co-founder of California Acceleration Project and a math professor at Los Medanos College. “After we remove the structural barriers—and we need to do that—we need to start the hard work of supporting faculty in safe ways to investigate the classroom-level issues.”
Lowe agreed, adding that professors need more preparation to be able to provide individualized instruction to students of all readiness levels.
“We train brilliant individuals who know their content well, and then we just stick them in the classroom,” Lowe said. “There’s a lot that we need to do to really support faculty in understanding the teaching and learning process.”