The Influence of Education Research on Math Ed Policy

If you’re a math policy nerd like me, you have probably noticed that colleges’ approach to students with weak math preparation has shifted radically over the past couple of decades. When I began working in education philanthropy about 20 years ago, California community colleges were placing as many as 90 percent of their students into remedial math sequences of one to four courses.

Today, many college systems are phasing out remedial courses, largely in response to a decade or so of research. A growing body of evidence has shown that remedial education is not the way to increase success for students deemed unprepared in math. Instead of relying on remedial courses and traditional placement tests, colleges are increasingly placing these students into college-level courses with appropriate supports, such as corequisite courses. 

Research study after research study after research study has demonstrated that such an approach increases—sometimes dramatically—the proportion of students who complete their math requirements. 

This is all great news. But it raises more questions, which new research is beginning to answer. 

Question 1: Taking lower-level math courses in high school puts students at a distinct disadvantage for future academic success. If placing students into nonremedial courses is a successful strategy in college, can placing ill-prepared high school students into grade-level classes keep those students on track? 

An intriguing new study by a team from Stanford University offers a preliminary yes. 

The authors analyzed a detracking initiative in a California school district that was designed to benefit ninth graders who were not deemed ready for Algebra I. Many of these students normally would have been placed into a remedial course.

The initiative provided teachers with supplemental professional development focused on fostering academic conversation and assessing student understanding. The teachers also benefited from an extra planning period, four coaching days per semester, and participation in a professional learning community. 

Academically, the students deemed to be below grade level were 22 percentage points more likely to complete geometry and 14 percentage points more likely to earn Algebra II credit than control group students taking remedial math. Assignment to the initiative was also associated with large test score gains. 

Question 2: From postsecondary research, we know a key reason that students placed into lengthy remedial course sequences often don’t complete them. It’s not that the students are unable to complete the courses. Rather, a substantial portion of students who complete individual courses don’t continue to the next course in the sequence. (See this, this, and this.) But what do we know about why this happens?

The same Stanford researchers shed new light on this question by looking at enrollment patterns of high school students in a single district. They found that underprepared students who took Algebra I as part of the detracking initiative were absent much less often than those in the control group. They were also less likely to leave the district. 

Researchers surmised that, in the remedial context, “Isolation with a homogeneous, low-achieving peer group, stigma, and weakened academic expectation reduce student engagement.” On the contrary, students in the detracking initiative demonstrated “heightened levels of belonging and satisfaction for an academically vulnerable population.” 

It stands to reason that a similar effect may play out with college remedial courses, though that is harder to research outside of compulsory education. 

Question 3: It’s one thing to say that students shouldn’t have to repeat remedial versions of courses they took in high school. But it’s quite another to say that students can simply skip over courses, such as Algebra II, that they never took in the first place. Wouldn’t students who didn’t take high school Algebra II but want to take calculus benefit from a remedial algebra course? 

It sounds like a sensible idea, but research doesn’t confirm that. When the Research and Planning Group for California Community Colleges (RP Group) studied this question in 2021, they found that, yes, students who had taken Algebra II in high school performed better in college math than students who hadn’t. At the same time, assigning students without Algebra II on their transcripts to remedial algebra was associated with lower success rates than assigning these students to a college-level course. 

Question 4: Underprepared students are more likely to complete their math requirements when placed into college-level, as opposed to remedial, courses. But students pursuing STEM degrees often face a series of math prerequisite courses before they can enroll in Calculus I. Just Equations found last year that some universities place the least prepared students into as many as three or four courses prior to calculus. Would these students be more likely to succeed if they started in calculus? 

The RP Group recently researched this question about college-level math prerequisites. They found that, again due to the effects of attrition, students who begin their sequence in Calculus I, rather than a prerequisite such as college algebra or precalculus, were far more successful. Surprisingly, this was true even for students who had not taken Algebra II. Under a new law, AB 1705, California colleges may not place students into such prerequisites unless they are highly unlikely to succeed in a calculus course without the prerequisite. According to the study, the researchers could not find a category of students who were more likely to complete calculus if they started in a prerequisite course. 

Most of these findings are preliminary, and all of them raise even more questions. For example, do the findings on calculus prerequisites change if math grades are considered? And, can any pedagogical strategies increase the effectiveness of calculus prerequisites? 

Two things are clear: Changes to mathematics placement were long overdue, and more research is needed.