Nobody Majors in STEM to Fail

“How could he miss question two? That was a gimme.”

“There was a question just like that on the homework, and she never came to office hours.”

“He’s not taking the work seriously. He needs more maturity to succeed in this class.”

Faced with struggling students in their classes, many instructors search for explanations. It can be easy to find ways that students seemed to fall short of expectations. However, that perspective leaves out where students are coming from: what experiences they might have had before college, how those experiences are impacting their trajectories now, and how adjustments to college could better improve their outcomes.

In short, without considering the context, it’s easy to dehumanize students, and to think more about their failures than the ways things could have been designed to allow their success.

The goal of this article is to demonstrate that, indeed, “nobody majors in STEM to fail.” We will draw a line from middle school, through high school, to college; consider how those experiences might influence students’ college academic work; and provide concrete ideas for how to create better environments that support students’ successes and see them as humans. The aim is to see the highly varied and individual journeys that students take, and then draw those threads together with a special focus on the experiences of those who are underrepresented, such as Black and Latino students.

This article will closely parallel the author’s recent invited address at the Mathematical Association of America’s MathFest conference. It is, however, meant to be the start of a conversation, not the end; it will raise ideas and hopefully provoke more reflection, but it nevertheless represents the observations of just one person who continues to seek the best ways to support young people in mathematics.

Taking a Longitudinal View

Discussions at the college level often focus on where college students are at now without necessarily considering the contexts they came from. Hence, perhaps our most important contribution is illustrating how students’ backgrounds are crucial to designing effective interventions.

The observations here are drawn in part from the author’s work with Bridge to Enter Advanced Mathematics (BEAM⁠¹), a program that supports students in New York City and Los Angeles. BEAM works to create pathways for students from low-income and historically marginalized communities to become scientists, mathematicians, engineers, and computer scientists.

¹

www.beammath.org

✖

BEAM’s longitudinal program provides support from sixth grade through college graduation, including intensive enrichment classes and individual student advising and mentoring. This support gives its staff a holistic viewpoint on students’ journeys. For more about BEAM’s program, see Zah20.

The Middle and High School Experience

It’s common to break students down into groups and discuss them as monoliths: the need to improve, say, the experiences of Black students in STEM or of gender minorities. While it’s important to call out adverse impacts on large groups, this practice can also obscure diversity and individualism within each group. So here, we will take an opposite tack: while still focusing on the experiences of underrepresented minorities, we will explore their individual uniqueness and the wide variety of experiences they have growing up—experiences which, in turn, have a dramatic impact on their adjustment to and success in college.

Nonacademic experiences

Consider, for example, students’ families, home communities, and schools. We will paint a picture of different environments, and consider later how some of these contribute to college experiences.

The BEAM program has worked with highly engaged families pushing for every opportunity; families that prioritize education but see it as driven by school, focusing students on their schoolwork but not extracurriculars; families that don’t have time to support students and leave their child to their own initiative; immigrant families with a strong push to succeed but a lack of knowledge about the system; and more. Families might be close knit, or more individualistic; students may have responsibilities to their family while growing up (especially care of younger siblings), or be more independent. Students may have role models in the family who work in math and science, or they might not. Students might be the first in their family to go to college. All of that, of course, is just scratching the surface.

Similarly, students’ home communities might be large or small; close-knit or diffuse; more communal, or more individualistic; primarily immigrant or primarily multi-generational American; highly educated or not.

Schools exhibit many differences as well. Of course schools might be small, where each student gets a great deal of individual attention (often in contrast to college!) or large, where students tend to be more independent. Some schools are highly structured, where student expectations are clearly laid out for every minute of the day, which is especially common in some charter schools that serve Black and Latino students. Schools may have differing expectations for student achievement or college attendance and varying levels of support. (As just one example, there are an average of 408 students to one guidance counselor nationwide Ame, with huge variation by school. Of course, this ratio will have a much greater impact on a student whose family and community have less knowledge about college.)

Schools themselves serve very different populations of students, in part due to the segregation of housing and the school system overall. In the US, over two thirds of Black and Latino students attend schools that are more than 60% Black or Latino, while over two thirds of White and Asian students attend schools that are more than 60% White or Asian.⁠² These schools are still diverse along many metrics: a school that serves predominantly Black students might be serving multigenerational Americans, recent African or Caribbean immigrants, Black Latinos, as well as students with a variety of different religions and family backgrounds. In any case, school demographics have clear consequences for students’ feeling at home at different universities.

²

This statistic was calculated based on data from Nata.

✖

Academic experiences

There are well-known disparities in academic performance across students from different backgrounds. For example, the National Assessment of Educational Progress (NAEP) reports that by 12th grade, only 24% of students score at the “proficient” level in mathematics. However, when we break it down by income, 33% of more affluent students (those not eligible for free- or reduced-price lunch) are rated proficient, while just 11% of low-income students (those who are eligible) score at proficient. The gap becomes even more stark at the “advanced” level of achievement: 4.79% compared to 0.69%. Natb

Most analyses stop with data like that—a simplistic look at “achievement gaps.” However, we should dig deeper to understand STEM preparation. Consider a typical student who might go on to a college STEM major. A common expectation might be that they did well in high school science and math classes, including (most likely) a calculus course, along with some high school exposure to physics, chemistry, and biology. Some communities also engage heavily with extracurricular work: camps, independent reading, competitions, computer programming, or independent learning (via books or even YouTube).⁠³

³

Although it does not address STEM, the enrichment spending gap is documented. See, for example, KMW11.

✖

Each of these expectations can be better understood with more digging. For example, consider the pathway to calculus. There are four years of standard high school mathematics, typically divided up as Algebra 1, Geometry, Algebra 2/Trigonometry, and Precalculus. To fit Calculus into a high school timeline, students must take Algebra 1 in eighth grade, or, less commonly, double up in math at some point during high school. Nationwide, 19% of students pass Algebra 1 in eighth grade, but not surprisingly, there are stark disparities when one disaggregates by race. While 29% of Asian students and 24% of White students pass Algebra 1 in eighth grade, only 14% of Latino students and 10% of Black students do so (see Table 1).

It is also worth noting that this data may understate the issues: Some studies have indicated that Algebra 1 courses at high-minority schools generally cover less content than the corresponding courses at schools that are not predominantly minority students; in other words, “passing Algebra 1” may have a different meaning at different schools. MRC20

Table 1.

Breakdown of the Algebra 1 pipeline in eighth grade, adapted from USDa.

Demographic Group	Percent enrolled in Algebra 1	Pass rate	Overall percentage passing Algebra 1
All students	24%	78%	19%
Asian students	40%	74%	29%
Black students	16%	65%	10%
Hispanic and Latino students	20%	72%	14%
White students	29%	85%	24%

There is no single cause for this opportunity gap, and again it helps to dig deeper. First of all, only 59% of middle schools offer Algebra 1,⁠⁴ and high-minority middle schools are precisely those that will have fewer or no seats in the class. USDb PRSM However, other causes for the gap may include bias (students may be coached out of advanced courses that they are ready for, or they may enroll but then have negative experiences); families who may not know the importance of taking Algebra 1; or students who may wish to stay with their friends or others who share their backgrounds, who may be disproportionately placed into eighth-grade math.

⁴

Although this statistic is slightly misleading: smaller middle schools are less likely to offer the course, and so overall 80% of students have access to Algebra 1.

✖

According to studies, prior academic preparation explains about 50% of the gap in Algebra 1 coursetaking. PRSM However, it is “turtles all the way down:” some of that lack of preparation is surely due to the same factors playing out in earlier grades.

These factors are present in later grades as well. Black and Latino students are less likely to attend a high school that offers calculus, and rates of passing the AP Calculus exam are correspondingly lower. Overall, 50% of high schools do not offer calculus, already a very high number that might be shocking to those in affluent communities. However, among those with high Black and Latino enrollment, that number goes up to 62% that do not offer calculus. In fact, 35% of high schools do not even offer precalculus USDa, making it even harder for a potential STEM major to succeed if their peers start college taking calculus or later courses. For a more detailed breakdown of calculus achievement, see Zah19.

In summary, by the time students arrive at college, they may have had different academic experiences, including starkly different levels of access to courses such as precalculus and calculus (to say nothing of enrichment experiences that also play a huge role in building problem solving skills). Although each student’s experience is different, on a societal level this lack of access disproportionately affects Black and Latino students.

Other experiences

While we have been focusing on more systemic factors, students of color are likely to encounter explicit racism as well as implicit bias. Depending on their experiences, they may have felt marginalized, out of place at certain institutions, or been led to question their belonging. A full discussion of these factors would exceed the space available here as well as the expertise of the author, but it is important not to minimize these factors as students navigate college.

Similarly, many students may have experienced housing insecurity, food insecurity, medical insecurity, violence, or other traumatic challenges. Again, a full discussion exceeds the author’s expertise, but these challenges are disproportionately faced by groups that are also underrepresented in math.

The College Experience

What actionable insights can we draw from students’ varied backgrounds? It can feel overwhelming to try to take those experiences into account once students arrive in a college classroom. In this section, we translate that challenge into some actionable insights.

However, it’s important to emphasize that this section provides examples, not a recipe. Our goal is to develop skills to understand students on an individual level, and thus to better support their success, but how exactly to do so depends on the institution, the students being served, and many other factors.

Sometimes, people cite the uniformity of experience as a sign of equality: A STEM major in college may be hard, but “everybody goes through it.” This article is to some extent a refutation of that idea: the ways in which STEM majors are hard have disproportionate impact on some people based on their past experiences. Changes to the STEM major experience, such as those in this article, can mitigate the disproportionate impact.

We explore this disproportionate impact through a series of questions that show how past experiences impact future ones.

Why do students drop STEM majors?

This is clearly a complex question, and not one we will come close to answering in full. Instead, we’ll consider some explanations to provide a framework for additional reflection.

Many of the reasons may seem familiar: low grades; less preparation before college; inadequate guidance on how to navigate college/STEM majors; and the overall environment within a STEM major. Any of these can contribute to students’ choice to leave before finishing their degree.

However, each of these can also have disproportionate impact on students depending on their backgrounds. Consider:

•

Low grades will have a psychological impact on every student. However, some students may have family members or older siblings who completed STEM majors, and can put those grades in context, while other students will compare college grades to high school grades and conclude that they’re out of their depth. Additionally, students who already feel different or isolated may find that lower grades further depress their senses of self worth and belonging.

•

Becoming aware of gaps in their academic preparation will affect all students. However, some students will have more access to resources such as private tutors, peers, or friends who recommend additional resources. Conversely, students with fewer financial means will have work obligations to pay for college that prevent dedicating time to learning prerequisite material.

•

All students find navigating STEM majors difficult, and the guidance provided by colleges is often inadequate. However, while some students again have family members or peers (even older graduates of their high school) to go to for advice, others may be largely on their own, especially if their social group does not have many STEM majors.

•

All students experience a change of culture when they go to college. However, some students may find that culture change to be more difficult: there may be fewer peers who share their background, or there may be more of a difference between the expectations and norms in college as compared to their home environment. This is to say nothing of racism or bias that students of color may face.

Some of these challenges are systemic: the way that courses and prerequisites work or the composition of a college class are all about the systems that students find themselves in. However, that is not a reason to give up on making change; any individual can implement changes locally that will impact the student experience, the culture, the advising that students get, or even how grades are communicated to students. (How much can a discussion at office hours change a student’s perception of their midterm grade?)

However, it is important to remember that each student experience is different. BEAM has had a number of students of color go on to STEM majors at a huge variety of colleges, from community colleges to the Ivy League. Some of these students attended high school with strong preparation; some with weaker preparation. Some students had been in high schools or summer programs where their peers were predominantly White or Asian; others had not. Some had strong networks of college advising, or family members who were engineers. Each student’s story is unique, and it is critical not to make assumptions.

For example, one BEAM student was a young Latino man, who had gone to a strong high school, taken Calculus there, and gotten an A in the course. He then attended a selective public college. However, Calculus 2 proved to be difficult. His past school had been a highly structured charter school; college offered much more freedom, and relied much more on self-study. Although he worked hard, he did not have study skills, he could not afford the textbook (and instead went to the library), and he often worked from home, which was crowded and loud. He felt like his own work was key to his success, and so he did not seek outside help. When COVID hit, he could no longer access the library and he failed his Calculus 2 course. While some obstacles listed above played a role (such as the impact of low grades, or cultural expectations about the college environment and seeking help), others, such as preparation gaps, did not.

The next question asks about a remedy that might have helped this young man: what could have inspired him to seek help?

Why don’t students come to office hours?

Office hours can be a critical tool for student success. They provide a link between student and instructor; allow the student to get customized help; and help create connections that can lead to letters of recommendation or information about future opportunities. However, attending them can be difficult for students. Going can be intimidating, especially if the student doesn’t already know the instructor; the timing may not work in a student’s schedule; a student may have poor time management and so may not do problem sets before office hours; and students may be worried that by going, they’ll reveal that they “don’t really understand the class.”

As before, though, each of these concerns may have a disproportionate impact on certain students. For example:

•

Office hours can be intimidating for every student, but for students who never met a professor before college (because there were none in their family or network), or whose parents never graduated college, a professor can seem much more intimidating.

•

Making time for office hours, similarly, can be hard; any student may have a conflict with another class or obligation. However, some students may have additional family obligations such as caring for younger siblings, may work to pay their way through college, or may live farther away from campus to save money.

•

Time management is similarly a challenge for everyone, and especially people in their late teens or early 20s who are often in college. However, the challenge can be much greater if someone’s high school was not very challenging (and hence did not require learning good time management skills), or if it was highly structured, where students had very little flexibility on how they spent their time. The adjustment to the freedom of college is not the same for everyone.

•

Office hours can feel risky, because no student wants to demonstrate that they “don’t understand” the topics in class. However, that worry can be heightened for someone who already “sticks out” (perhaps because they’re one of few students of color), or who is the first in their family to go to college, or who is more intimidated by professors than their peers.

It is worth lingering on that last point. Even at BEAM, which is a highly supportive program that works with students beginning when they are 11 or 12 years old, students fear showcasing a lack of knowledge. The implication of the BEAM program (that it is “advanced” mathematics) leads students to worry that they don’t belong, and to hide places where they struggle in school math classes. We see this especially with male-identifying students; in any case, it may be a significant struggle for students to admit to having difficulty.

Fortunately, professors can take action to mitigate some of this disproportionate impact. For example, opening up to students and sharing their own stories can help to demystify “the professor” and put students at ease. Similarly, professors can communicate that their goal is student growth, and publicly share their own stories of struggle (or those of students from past years) to normalize the challenges students face.

A second approach is to ensure that all students come to office hours at least once. Professors can make repeated invitations in class, demystifying office hours by explaining what happens and why it is beneficial. Students can also be required to come to office hours at least once in the first month. In this way, students will have broken the ice and may be more likely to come in the future.

Finally, individual follow up can make a positive impact on students. A professor might personally email students who struggled on the first quiz or exam, sharing positive expectations and inviting them to come into office hours to discuss some of the problems. While it does not take much time to write such a message, it can spur a student to take the step to come in!

Why do students struggle in intro courses?

Introductory courses are often one of a students’ biggest barriers. In fact, about half of students who drop out of college do so in their first year.

Examples of challenges that students might face in their introductory classes include their overall adjustment to college; lacking access to adequate academic support; and a significant jump in the sophistication of the work students are now expected to do.

Also as before, there will be a disproportionate impact on some students. Adjusting to college is more difficult for many students for reasons discussed above, such as less family experience with college, coming from different communities, or feeling different from other students. Financial circumstances and social networks can impact access to academic support. Rather than rehashing these topics, we will consider the jump in the sophistication of work that often comes with going to college.

The “Algebra 2/Trigonometry” Regents Examination is the most sophisticated math exam required to graduate in New York State. In a review of the August 2016 exam New, the author picked out the following question as the “hardest:”

“Using the identity $\sin ^2 \theta + \cos ^2 \theta = 1$, find the value of $\tan \theta$, to the nearest hundredth, if $\cos \theta$ is $-0.7$ and $\theta$ is in Quadrant II.”

Of course, even the typical college calculus exam question is generally much more difficult than this. For students who are used to the most difficult questions both telling you how to do the problem (“Using the identity $\sin ^2 \theta + \cos ^2 \theta = 1$”) and being relatively simple, it can be a big jump to the typical exam question requiring justification or developing a solution method without guidance. This emphasizes again how a good high school calculus course (to which there is uneven access) can help students succeed even without skipping college calculus.

Where, then, do students get access to more sophisticated problem solving that prepares them for college? Sometimes a good high school will offer it as part of their classes, although more often, students may do other enrichment programs or independent study (such as online). Of course, access to those resources is uneven. Many students are told to simply do well in their school classes, and that will be enough—unfortunately, those students are often the most likely to attend high schools that offer less of an education in problem solving.

There are no easy answers to these challenges. However, professors can be explicit about the kind of problem solving necessary, normalize struggle, and prepare students psychologically for a jump from what they are used to. It is also possible to provide more scaffolding, such as earlier problems that guide students through the process of problem solving. Regardless, course design and good teaching is especially important in introductory classes. Finally, if resources are available, advocacy within colleges or universities to offer additional preparation, perhaps in parallel with existing courses, can make a difference for student success.

How does the culture of STEM affect students?

The people who do math are a diverse group, and there is no one culture to mathematics. However, there are cultural trends that are common among STEM communities. For example, a focus on inquiry, problem solving, and understanding, or an enjoyment of puzzles. Mathematical communities also often (but not always!) show support to those who are less social or may not be neurotypical.

However, it’s also the case that the dominant culture in mathematics may be foreign to students. Moreover, the culture of mathematics is often individualistic and competitive,⁠⁵ which may be very different from what they’re used to at home.

⁵

Thanks to Dr. Nicole M. Joseph for introducing me to these terms.

✖

While there is not space to fully address this claim here, it’s not hard to see competitive trends in the classroom. The student in class who raises their hand primarily to show off their knowledge, or the student who scoffs at another for not having learned something yet, are common tropes that can imply a competitive mindset.

As before, individualism and competitiveness also disproportionately impact some students. For example, those who have less preparation, less of a sense of belonging, or who feel less of a connection to others who do STEM may be more negatively impacted by a sense of competition. So, then, how do we address individualism and competitiveness?

Of course, it is best to address these broadly, but even within a classroom there are changes that can make a significant difference. For example, taking the time to explicitly welcome students from varied communities to the class, or setting a mindset based around growth in the class (rather than absolute achievement) can make a positive impact for students. Setting wait times before students can answer questions (to allow more students to develop their answers) will similarly help.

It may not be intuitive, but one common classroom technique that encourages a competitive mindset is the way that grades are often curved.⁠⁶ Test grades are often curved by ordering them, looking for bunches, and setting the boundaries between As, Bs, and Cs depending on where there are gaps. While this is easy, it sets students against each other and does not make sense to them: their grades are not based on what they know, but how they compare to others in the class. A grading system where there are absolute measures corresponding to each letter grade that are based on understanding of the course material (rather than how peers do) emphasizes that students must achieve at a certain level, rather than in competition with others. An ‘A’ will really mean “I fully understood the material” and not “I did better than most other people.”

⁶

This is also cited directly by students as a problem in STEM courses; see, for example, SH19.

✖

Naturally, these solutions are only the beginning of addressing the disproportionate impact of the culture of mathematics and the classroom. Nonetheless, they can begin to make a difference.

Structural racism

Although the focus here is on individuals, it is important to emphasize that many of the issues discussed above are more likely to impact students from certain backgrounds: racial or ethnic minorities, first-generation college students, and those from low-income families. Systemic racism refers to society-wide practices that advantage particular groups based on race (even if they are on their face neutral); the examples here, then, are examples of how systemic racism works at the college level, although the disproportionate impact does not only affect people based on race. On its face, we say “college is hard for everybody.” In reality, it is harder for some than for others because of the overall design of the system.

That said, BEAM’s students have encountered not just systemic racism but direct, individual racism. For example, a student at a highly selective high school was called the n-word by her peers. While this article has focused on addressable challenges, it does not intend to minimize the adverse experiences students may have with direct racism.

Conclusions and Next Steps

In the work to better support students from marginalized backgrounds to achieve STEM degrees, it is important to balance expectations. Understanding students’ backgrounds and designing an approach to mitigate the disproportionate impact of existing structures can make a positive difference in students’ experiences—even if doing so will not magically solve all the inequities that persist. Seeing students as people and supporting their aspirations can improve their experience—even if there remain other negative factors in their environment.

And yet, despite the magnitude of the challenges, there are many steps that individuals and departments can take to make a genuine positive impact. It is possible to create a healthier environment that views students as individuals who are each trying their best to succeed. Nobody majors in STEM to fail; many students do find success, and more will do so with targeted support.

Questions of equity remain core to the future of society; work to address those questions will have rich and important benefits. Those of us educating future STEM professionals have an important role to play if we commit ourselves to the goal.