No More "Sink or Swim": Incorporating Subgroup Accountability into the Higher Education Act

For too long, college has been a “sink or swim” environment where historically underserved student populations are not provided with the systemic opportunities and supports necessary to succeed. This mentality has led to major equity gaps that are not improving over time. Black and Latinx adults are almost half as likely to get a postsecondary degree as their white peers, and a recent Third Way analysis found that Pell-receiving students have graduation rates that are on average 18 percent lower than their non-Pell peers.¹ These gaps have remained stagnant or grown larger over the last twenty years.

Although there is still a great deal of work to do to close equity gaps in K-12 education, the recent reauthorization of the federal government’s K-12 education law, the Every Student Succeeds Act (ESSA), as well as older laws like No Child Left Behind (NCLB), provide valuable lessons learned that should be used to inform efforts to increase accountability in the Higher Education Act (HEA).² ESSA, importantly, reaffirmed the role of the federal government in ensuring the transparency of student outcomes and holding educational institutions accountable for providing equitable resources and success for underserved student population groups. It’s time for policymakers to step up and apply that same principle to federal higher education policy.

In K-12 education policymaking and practice, it is generally not accepted to point at specific populations of students as a rationale for poor institutional outcomes. Instead, it is acknowledged that with fiscal support, strong leadership, and effort, all schools have the ability to help all of their students succeed. In higher education, however, this argument is all too common. In the next reauthorization of the HEA, more must be asked of institutions to close gaps in access, persistence, and completion for students of color, low-income students, and students with disabilities.

The Problem

Equity outcomes at postsecondary institutions vary greatly.

Analysis of existing data shows that institutions with similar student populations and selectivity can have very different outcomes for historically and currently underserved student groups. For example, Middle Tennessee State University’s graduation rate for Black students is more than double that of Eastern Michigan University, even though the institutions are equally selective and share similar enrollment rates for Black students.³ And at the University of San Francisco, Latinx students graduate at a rate of 72 percent, compared to a 45 percent Latinx graduation rate at Hofstra University, a peer institution.⁴ Demographics do not have to be destiny, but colleges and universities have to do their part to help students succeed.

Institutions are not held accountable for equity.

Currently, the very limited accountability measures in the HEA only hold institutions accountable for their overall student outcomes, and the data used by the federal government to determine whether institutions are serving students well is based on averages. Using averages for accountability masks outcomes for many of the most vulnerable student populations. This differs greatly from K-12 education law, where outcome metrics are disaggregated by groups of students who have been historically disadvantaged, and schools and districts are held accountable for the outcome of each and all student groups. 

The federal government makes a massive (and worthy) investment in higher education each year, including $122.5 billion in grants, loans, and work-study, and over $30 billion in tax credits.⁵ Federal sources make up 16 percent of public college and university budgets, even excluding student loans and federal tax credits.⁶ At the same time, the federal government holds K-12 schools strongly accountable for outcomes of at-risk student population groups, even though federal spending makes up only 8.5 percent of K-12 school budgets.⁷ Although most federal postsecondary student aid is targeted to low-income students, that’s where much of the resemblance to other pieces of federal education legislation stops for higher education. Simply put, higher education receives more federal funding with less oversight of outcomes for the most vulnerable student populations.

College is more important than ever.

Ensuring equitable access to higher education is more critical now than ever. Individuals with a postsecondary degree or credential earn $1 million more over the course of their lifetime than their counterparts with a high school diploma or less.⁸ And from an economic perspective, the need for workers with a postsecondary credential or degree will only increase. When the HEA was first written, the demand for workers to have a postsecondary credential was much lower. In 1973, eight years after the HEA was signed into law, only 28 percent of jobs required postsecondary education and training. However, by 2020, that number will increase to 64 percent.⁹ Beyond simple economics, a college degree benefits students and society as a whole. College graduates face lower unemployment, earn more, pay more in taxes, are healthier, and are more civically engaged.¹⁰

Significant gaps exist for underserved student groups.

The significant inequities that currently exist across the higher education system when it comes to access, completion, and affordability only serve to further amplify economic inequality. The problem starts with lack of access to quality postsecondary education opportunities. For example, in 2015, 70 percent of White recent high school graduates enrolled in postsecondary education, compared to only 55 percent of Black students.¹¹ At the same time, students of color are more likely to be enrolled in less-selective, open-access, and for-profit higher education institutions, which often lead to lower levels of completion, post-enrollment employment, and loan repayment.¹² And children from families in the top one percent are 77 times more likely to attend more elite colleges compared to the children from families in the bottom quintile.¹³

Major equity gaps also exist in college completion. While 60 percent of high-income students have received a bachelor’s degree or higher, only 14 percent of their low-income peers show the same result.¹⁴ And just as of last year, 55 percent of White adults had attained an associate’s degree or higher, while Black and Latinx adults’ attainment rates sat at 35 and 28 percent, respectively. These gaps have only grown wider in the last thirty years.¹⁵ Higher education outcomes for students with disabilities are also extremely concerning. For instance, one study showed that only about a third of students with a disability who enrolled at a four-year university had graduated, even eight years after enrolling.¹⁶

The college pipeline is changing.

The disparity in access to and completion of higher education for students of color, low-income students, and students with disabilities is a matter of concern that must be addressed, and fast. By 2025, the percentage of Latinx high school graduates is expected to increase by 43 percent, while the percentage of White high school graduates will drop 14 percent by 2031.¹⁷ The high school graduation rate for Asian and Pacific Islander students is also projected to increase, while the rate for Black students is expected to hold relatively steady over the coming decades. Latinx students already constitute the majority of high school graduates in many Western states such as Arizona, California, Nevada, New Mexico, and Texas. And in some Southern states where minority population growth is rising quickly, including Florida, Georgia, and Mississippi, the majority of students will be students of color by 2020.¹⁸ The population of the United States is shifting rapidly, and higher education institutions (and the policy governing them) must keep up with this new reality of students in the education pipeline.

Even as a postsecondary education becomes more important and the student pipeline becomes increasingly diverse, there are widening gaps in access and outcomes for students of color, low-income students, and students with disabilities. Policymakers have the opportunity to ensure more American students can attain a high-quality postsecondary education by amending the HEA to hold institutions accountable for equity gaps.

A Framework for Subgroup Accountability in the Higher Education Act

Over the past several years, it has become clear that there is strong bipartisan support for increasing accountability in the HEA to provide more students with access to high-quality colleges and universities and to ensure good stewardship of limited federal funding. However, given the vastly inequitable higher education outcomes for underserved students, policymakers should ensure that a reauthorized HEA holds institutions accountable not just for the success of their overall student population, but for specific subgroups of students as well. If more students of color, low-income students, and students with disabilities graduate from college, we stand to annually add hundreds of thousands of highly-skilled individuals to the national workforce and make significant strides in addressing long-standing social and economic inequities.

A great deal of the current discussion regarding the HEA centers on the cost of college, and rightly so, given that a degree or credential is financially out of reach for many Americans. And while the push in recent years to improve transparency and provide students with more data about institutions is significant, it is crucial to understand that many students, particularly those who are low-income or older, cannot leave their geographic area to attain a degree or credential.¹⁹ This lack of real choice means that students need affordable, high-quality options in their own backyard. To ensure that all students have access to higher education that will lead to improved opportunities, the federal government must pay more attention to how colleges and universities serve students of color, low-income students, and students with disabilities.

Instead of suggesting one rigid option, this paper provides a framework for policymakers to increase accountability for historically underserved groups of students. Known as “subgroup accountability” in K-12 education policy, this same concept could be embedded within the accountability efforts currently being discussed by higher education policy stakeholders, like accreditation, risk-sharing, or a new federal-state partnership. It could also stand alone by being linked to the receipt of federal student loans or campus-based aid. It could be applied at an institutional or a programmatic level depending on where policymakers choose to focus their accountability efforts.

Which student subgroups should we use?

In K-12 education law, accountability is based on the student groups, or “subgroups,” who have been historically disenfranchised and underserved by the education system. At the postsecondary level, having key data points disaggregated by student population groups would help ensure that colleges and universities are both measuring and responding to gaps in resources and outcomes.  This type of data can be used both for transparency and accountability purposes.

Any meaningful accountability system in the HEA should rely on data disaggregated by, at a minimum:

• Major racial and ethnic groups;
• Gender;
• Low-income status;²⁰ and,
• Disability status.

Although this paper primarily focuses on improving postsecondary education for students of color, low-income students, and students with disabilities, it would also be advisable to disaggregate data on student status as a servicemember or veteran and to hold institutions accountable for these outcomes. Last year, the federal government spent over $12 billion on GI Bill educational assistance for veterans, and about half a billion more on Tuition Assistance programs for active duty members of the military. Servicemembers and veterans can face unique challenges as they transition into education, so it is important that institutions monitor these students’ outcomes and make efforts to break down barriers to success.

And while it is imperative that all students have what they need to be successful, some data disaggregation might be more appropriately used for transparency but not for accountability purposes. For instance, in ESSA, data is required to be disaggregated by homeless and foster care status; however, that information does not count towards a school’s accountability plan. There is also demonstrated bipartisan support for the HEA to include disaggregated data on enrollment status as a first-time student, attendance intensity (full- or part-time), credential-seeking status (by credential level), age, and status as a postsecondary athlete. This additional information provides valuable data to institutions, accreditors, stakeholders, and policymakers, and can help ensure resources and supports are targeted at the students who need them most.

It is also worth considering methods to support further disaggregation of Asian ethnic groups, an idea that has had bipartisan support at the K-12 level. Continuing to group all Asian ethnicities together further perpetuates a “model minority myth” and serves to mask wide variations in student outcomes. The University of Wisconsin’s Accountability Dashboard offers a good example of how data can be further disaggregated for Asian student populations. Instead of the commonly used “Asian American/Pacific Islander” subgroup, it provides information on Southeast Asian, Other Asian American, and Hawaiian/Pacific Islander subgroups.²¹

What should we measure for subgroup accountability?

At a bare minimum, the HEA should require public reporting of new disaggregated data on student population groups, but real progress towards closing equity gaps will likely only come when that same data is linked to meaningful federal accountability.

Higher education accountability should be crafted to promote equity and reflect institutional context. An accountability system should take into account institution type, mission, and the fact that, in many cases, the lowest-resourced institutions serve some of the most at-risk students. Various types of institutions could be held accountable to different metrics, as long as the metrics are not designed to let schools off the hook for outcomes based on their student populations. For instance, it might be more appropriate that four-year institutions are measured on completion rates for students finishing within 150 percent of normal time to completion, while community colleges, due to their missions and high populations of part-time students, be measured at 200 percent of normal time to completion.

One lesson learned from K-12 education policy is that it is essential to measure “growth,” or improvement on a metric, in addition to “proficiency,” or whether the bar for that metric is met.²³ NCLB relied too heavily on a single bar for success, which led many schools to focus their supports and resources on so-called “bubble kids” who could meet the proficiency score with a bit of extra help. Because of this, higher-achieving students and students who were further behind were left with fewer supports and attention. To avoid this trap at the postsecondary level, accountability should measure institutions on both growth and overall success when it comes to improving outcomes for subgroups of students.

To provide a comprehensive perspective on how institutions are performing and reduce “gaming,” metrics for subgroup accountability should be based on data that span students’ educational journeys. Basing accountability on only one metric would not give an adequate picture of institutional success and could have unintended negative consequences. For example, research shows that completion alone as a metric can encourage institutions to act as “diploma mills,” providing students with certificates and degrees that hold little labor market value.²⁴ Taking a more holistic perspective will ensure that institutions focus on enrolling diverse student populations, providing them with a supportive and high-quality learning experience while they are in school, and preparing them for the world of work after they complete.

Subgroup accountability in the HEA should include data on students:

• prior to enrollment (access);
• in school (progression);
• finishing school (completion); and,
• post-school (earnings and repayment rates).

Access

Information related to access to higher education is critical because it helps institutions and policymakers understand the population of students who aspire to attend college and those who actually do attend.²⁵ It also identifies trends related to student enrollment, including institutions’ recruitment and acceptance practices. Currently, major equity gaps exist in access. For instance, research shows that students of color often “undermatch” when they choose institutions of higher education, attending lower-tier schools than what they are qualified to attend.²⁶ Furthermore, students of color are often concentrated in institutions with low graduation rates and poor loan-repayment outcomes.²⁷

Access metrics used for accountability purposes could include information on the students who applied to, were admitted to, and enrolled at an institution, disaggregated by student subgroup. With increased attention to access metrics, institutions could seek to ensure that they are recruiting a diverse student population and proactively identify and remove barriers to access for student subgroups. In the context of increasing accountability, measuring access is also critical because it can help mitigate the chance that institutions might choose to enroll fewer students perceived as “risky.”

Progression

Higher education is currently a “leaky pipeline,” as only half of all students who enter postsecondary education earn a degree.²⁸ A focus on student completion must include an examination of how students progress through school. Persistence and credit accumulation data is crucial because it provides institutions and stakeholders with meaningful information related to student success over time and insight into how practices might be changed to help more vulnerable students stay in school through to completion. Nearly a quarter of students who drop out default on their loans, which is a financial disaster for the individual and a loss for the federal government.²⁹ A focus on college completion and transparency of progression data is crucial both for the benefit of students and, in the context of the billions of taxpayer dollars spent on aid, to ensure a solid return on investment. Progression data used in accountability can include metrics such as first- to second-year retention rates, credit accumulation, and gateway course completion, disaggregated by student subgroup.

Success in developmental education (DE), previously known as remedial education, is another progression metric that can provide information to help close equity gaps. While estimates vary, some sources cite that up to 68 percent of community college students, and 40 percent of four-year university students, enroll in at least one DE course.³⁰ Black, Latinx, and low-income students also enroll in DE at higher rates than their White and higher-income peers.³¹ And over $7 billion is spent on DE annually, with mixed success.³² Enrollment in DE correlates with low course and degree completion rates, as students enrolled in DE are not earning credit towards completion for those courses, while at the same time they are expending limited fiscal resources.³³

Laudably, many states and institutions are seeking to improve and reform DE.³⁴ In the past, a metric measuring the rate at which institutions moved students from non-credit-bearing to credit-bearing courses could be used, but many of the improvements to DE mean students are now gaining credit throughout their time in DE. A disaggregated subgroup accountability metric that could account for this factor would be the percent of students who enter an institution as academically unprepared (as defined by the institution) who then go on to complete a degree or certificate.

Completion Rates

Completion of a credential or degree is generally accepted to be a key indicator of institutional success. Measuring the rate at which students complete their degree or certificate program is of obvious importance to gauging the success of institutions of higher education. The number one reason students go to college is to get a job so they can have a stable and secure life, and the completion of a degree or credential is key to unlocking economic opportunities.³⁵ In the past, completion rates were only available for first-time, full-time students enrolling in the fall, which was particularly problematic when trying to gauge institutional success with student population groups that don’t meet a “traditional student” definition. However, the US Department of Education has recently worked to expand the federal measure of completion rates to part-time and transfer students, as well as measuring outcomes for students enrolling year-round beginning in 2019.³⁶ While this new data is a big step forward, completion data used for accountability will need to be disaggregated by student subgroups as well.

Policymakers will need to determine an appropriate window of time in which to measure completion. Completion metric windows can help encourage institutions to move more students towards on-time graduation, which is beneficial both for students and federal spending. A completion window of 150 percent of normal time to completion (six years for a four-year degree, three years for a two-year degree) could be chosen as a more aggressive option to incent on-time completion. Two-hundred percent of normal time to completion (eight years for a four-year degree, four years for a two-year degree) would be a more generous option, and it may better account for demographics at institutions that enroll a high percentage of part-time students. This would still allow for the design of an accountability system intended to flag higher education’s worst actors.

Earnings and Repayment Rates

Even the completion of a degree is no protection against economic insecurity for students of color, as a White dropout is less likely to default on their student loans than a Black graduate.³⁷ Measuring student outcomes post-completion conveys information regarding how well an institution prepares students for the workplace. These post-college metrics can illustrate the real-world value of a degree and provide crucial data to combat inequity in higher education outcomes. Additionally, post-college metrics are valuable because they can help deter “diploma mill” institutions that confer degrees that hold little value in the labor market.

The currently-used cohort default rate (CDR), which measures how many students in a given institution default on their student loans, is not an ideal metric for gauging the success of vulnerable populations of students.³⁸ It can encourage “gaming” behaviors, such as pushing students into forbearance, even when it is not in a student’s best interest.³⁹ Additionally, with increased use of income-based repayment (IBR), CDR is becoming a less meaningful metric than it once was. While IBR is a beneficial tool for many students, it is possible that those who would otherwise be in default can enroll in IBR but not pay down debt or even keep up with accumulating interest. While alternative metrics are suggested below, CDR could be maintained as a complementary metric—it is simply not enough alone.

One option for measuring student outcomes after completion is to use disaggregated loan repayment rates. The College Scorecard’s repayment rate measures the share of borrowers who have reduced their principal amount by at least one dollar over a period of one, three, five, and seven years. While paying such a small amount on a loan balance is admittedly a low bar, if the goal is to create an accountability system that is ensuring students are protected from very low-performing institutions, it is a fair target to establish. Repayment rates also produce a more accurate measure of who is struggling to pay down their educational debts before the worst case scenario of reaching default.

Additionally, a disaggregated earnings metric could be used to gauge student outcomes after completion and the real-world value of a degree or credential. Research shows that the top reasons students enter postsecondary education are to get a good job and to make more money.⁴⁰ One way to measure earnings would be by comparing the percentage of graduates earning more than the average regional high school graduate salary.⁴¹ If a student expends money and time to attain a degree or credential, in almost all cases, they should earn more than someone without a higher education.

How can we identify and compare institutions for subgroup accountability?

Compared to the K-12 education system, the higher education system is an even more varied patchwork of types of institutions and governing bodies, each attracting different student populations according to geography, selectivity, program offerings, mission, and culture. Some of these institutions may have few equity gaps, some may have equity gaps but strong or moderate outcomes overall, and some may have low outcomes for almost all students. While it is crucial to address this latter category in the next HEA reauthorization, that is not the focus of this paper as it is a topic that has already been widely discussed.

One way to measure equity gaps is by looking within an institution at outcomes for low-income students, students of color, and students with disabilities as compared to higher-income students, White students, and students without disabilities. Policymakers should both measure the severity of equity gaps at an institution and give institutions credit for any growth they make in closing gaps. 

It is also important to gain more perspective on institutional performance by comparing gaps in equity to other institutions. Setting up a benchmarking system is challenging because postsecondary institutions vary so widely in mission and students served, but a number of methodologies have been suggested by stakeholders. Four options to compare institutional equity gaps discussed in this paper are:

• Setting a reasonable bar;
• Identifying the bottom five percent;
• Establishing peer groups; and,
• Using input-adjusted metrics.

Setting a Reasonable Bar

A straightforward way to compare institutions would be to set an expected outcome for each metric and measure whether institutions meet that goal, or if not, how they are progressing towards it.⁴³ These bars would need to be reasonable, so missing them would flag that something is truly going wrong at an institution. For example, institutions could be measured by the percentage of their graduates earning more than the region’s average high school graduate salary, disaggregated by student subgroup to identify equity gaps. While this is not a lofty goal, it does set a floor for what’s generally considered an acceptable outcome to ensure an institution or program of higher education is providing value to its students.

This type of system would be relatively simple and would serve to identify the country’s poorest-performing institutions. Measuring all institutions according to the same yardstick would clearly delineate where equity gaps exist. However, depending on where the bar is set, it would not be likely to drive widespread change, since most colleges and universities would not fall under the established bar.

Identifying the Bottom Five Percent

A second option others have suggested is drawn from K-12 education, where the poorest-performing five percent of schools in each state are identified for intensive interventions and supports.⁴⁴ At the higher education level, the bottom five percent of institutions across the country could be determined, either through flagging the schools that demonstrate the poorest equity outcomes for each metric or through the use of a rubric that aggregates all subgroup accountability metrics. This approach is driven by the idea that the initially-identified institutions would improve over time, and a new bottom five percent would be identified periodically.

Identifying a bottom five percent would be a fairly direct approach to hold accountable the institutions with the worst outcomes for historically underserved student groups. This method could have real merit in identifying and addressing the worst actors in the system, but it is also unlikely to drive widespread change, as only a limited number of institutions would be subject to accountability.

Establishing Peer Groups

Some stakeholders have also suggested setting up “peer groups” of institutions to use for comparison of outcomes.⁴⁵ Peer groups could be groupings of schools that:

• are in the same sector (public or private);
• are of similar size;
• share the same predominant degree;⁴⁶
• serve a similar student population;⁴⁷
• have students with a similar level of academic preparedness;
• are similarly-resourced or with similar per-student expenditures;
• have a similar level of selectivity;⁴⁸ and/or,
• share specific missions.⁴⁹

Benchmarking using peer grouping could allow for a fair comparison of schools’ success with subgroups of students, since the schools in a group would share comparable characteristics. However, setting up a peer group system is complex, and care would have to be taken to ensure the parameters for grouping are developed in a way that would lead to accurate comparisons. Another drawback of peer grouping is that it doesn’t necessarily set a high bar for all institutions. An institution that has fairly low outcomes could look better simply because its peer institutions are extremely low performing.

Using Input-Adjusted Metrics

Input-adjusted metrics, or risk-adjusted metrics, have also been a widely discussed option for comparing institutions.⁵⁰ Similar to the “value-added” concept in K-12 education, input-adjusted metrics seek to measure the success of an institution while taking into account the various inputs that differ from school to school. While these inputs do include resources and other characteristics, the concept of input-adjusted metrics generally focuses on taking into account an institution’s student population.

Caution must be urged when considering the use of input-adjusted metrics, which can essentially codify low expectations for schools serving historically underserved student populations, and for the students themselves—sending the message that it’s simply not possible for certain kinds of students to succeed. Additionally, complex input-adjusted metrics algorithms could make data too difficult for students and families to easily interpret.

What should happen when an institution has low equity outcomes?

After data has been disaggregated and reported, and low-performing institutions have been identified, sanctions should be put in place for schools that over a period of years consistently fail to help underrepresented students succeed. At the same time, supports should be offered to help struggling institutions do the hard work of improving student subgroup outcomes. Additionally, rewards could be provided to institutions that are outperforming their peers when it comes to serving at-risk student population groups. In order not to set up an NCLB-like system with very small carrots and very large sticks, policymakers will also have to ensure that supports and rewards are proportional to sanctions.

Institutions should be provided a reasonable window of time to improve before sanctions are put in place. For a four-year institution, one planning year plus four years of implementation time should be enough to boost metrics of completion, although access metrics could improve in a shorter period. Institutions that primarily provide associate’s degrees and certificates could be given a shorter time window, or for simplicity’s sake, the same time window could be applied to all institutions given that many students pursuing associate’s degrees or credentials are attending part-time.

Sanctions

Currently, the vast majority of postsecondary institutions sail through the existing nominal federal accountability mechanisms, with only a handful losing Title IV eligibility or closing due to egregious practices. For instance, in one recent school year, out of over 7,000 US colleges and universities, only 55 failed the cohort default rate test, while students continued to enroll in countless more schools with abysmal student outcomes.⁵⁴ It has become evident that this “all or nothing” approach is not an effective method to push institutions to improve.

Instead, policymakers should put in place a stairstep of sanctions for poor-performing institutions who fail to improve over time. Sanctions that increase in severity, instead of the current binary system, might also deter institutions from pulling out of the federal system altogether and restricting their students’ access to federal aid.

Some ideas for graduated sanctions could include:

• public disclosures or website and College Scorecard warning labels;
• a mandatory campus climate survey;⁵⁵
• financial set-aside requirements for a school to dedicate a percentage of institutional funding to improve student outcomes;
• financial penalties, with the money paid by the institution going into a federally-funded program to support access and completion; and,
• in extreme cases, loss of beneficial tax credit statuses including tax-exempt bonds to nonprofit organizations and the charitable interest deduction. 

Supports for Improvement

When considering new subgroup accountability mechanisms in the HEA, policymakers will have to determine what level of resourcing is appropriate to provide institutions identified for improvement. At a minimum, robust technical assistance and support for institutional leaders will be vital to assist schools that need to address equity gaps.

However, since many schools serving predominantly low-income students and students of color are underresourced compared to institutions serving higher income and mostly White student populations, it would be ideal to offer financial support to identified institutions.⁵⁶ This funding could be temporary and would provide institutions increased capacity to analyze their data and outcomes and put in place sustainable practices to improve student achievement.

Another option to consider, either alone or alongside funding for identified institutions, would be to provide fiscal support to not-for-profit institutions enrolling high percentages or numbers of low-income students, students of color, and students with disabilities. This idea is akin to ESSA Title I funding. Policymakers could consider appropriating this funding beginning several years before subgroup accountability is fully deployed, in order to help institutions prepare for the new policies.

If financial support is offered, funding should be used, as much as possible, for evidence-based practices. ESSA created tiers of evidence that were to be applied when states, districts, and schools utilized the bill’s funding. An analogous approach could be included in the HEA to ensure that federal funds are being spent as wisely as possible. For-profit institutions should not be eligible for any new federal support, as these institutions seek to make money off students and should be encouraged to use their existing resources to improve student outcomes.

Rewards for Enrollment and Outcomes

While accountability is implemented to target the worst actors in postsecondary education, it would also be beneficial to recognize that there are institutions who are outperforming their peers when it comes to equity. High-performing institutions could be provided with non-financial incentives and rewards. Rewards might include things like decreased data reporting, reduced accreditation burdens, extra points or preference in Department of Education grant competitions, and increasing funding for schools to offset the costs of compliance with federal laws and regulations.⁵⁷ The Department of Education could also create a new category of distinction for schools in each peer group or type of institution that have high levels of outcomes for historically underserved student populations. This category could be utilized on an institution’s website, on the College Scorecard, and in marketing.

How can we mitigate unintended consequences?

Unlike in K-12 schools, many institutions of higher education can choose who is admitted. Some higher education stakeholders argue that increasing accountability will lead to unintended consequences for the very populations of students policymakers would be intending to support through these policies. Because of this possibility, it is important to ensure policy is designed in a way to minimize incentives to bar enrollment of students an institution would view as “risky” in some way. This is key when designing sanctions, supports, and rewards for institutions to ensure they bear out as few unintended consequences as possible.

In addition to carefully choosing metrics that are less likely to be gamed by institutions, policymakers could utilize a very familiar policy approach and include a “hold harmless” for student populations. For instance, a hold harmless could require institutions to prove that their numbers or percentages of low-income students, students of color, and students with disabilities did not decrease in the period after which subgroup accountability was implemented. To account for normal fluctuations in student populations, the hold harmless could be based on an average of three years prior to and after the new provisions go into effect.

If financial penalties are leveraged, it would be prudent to require the penalized institutions to prove that they are not increasing tuition and fees or decreasing student supports to fund the penalty. Additionally, sanctions could be prorated based on the student population a school serves so that penalties are proportional to the rates at which they enroll historically underserved student groups. For example, a school that enrolls high percentages of low-income students, students of color, and students with disabilities could pay a lower financial penalty than a school that enrolls fewer of those students, even if the schools have the same outcomes for the underserved student groups. 

How can we improve student outcomes?

The goal of implementing subgroup accountability at the higher education level is to push institutions to move away from a “sink or swim” mentality and be more intentional about helping all students succeed in a postsecondary environment. Similar to K-12 education, it is important to encourage the use of data to identify what each institution needs to do in order to improve student outcomes and for federal policy not to be overly prescriptive of those improvement actions.⁵⁸

A variety of resources are already available for institutions concerned about equity and the outcomes of their historically underserved students. The Association of American Colleges and Universities has an extensive set of resources designed to “help colleges and universities integrate diversity, equity, and inclusion into their missions and institutional operations.”⁵⁹ Complete College America provides Scaling Standards and resources to increase completion, with a focus on closing equity gaps.⁶⁰ And the Education Trust offers a Practice Guide that draws from the experiences of institutions that have been successful in closing gaps in college completion for low-income students and students of color.⁶¹

Another promising practice, universal design for learning (UDL), has been commonly-used at the K-12 level for many years. More recently, some institutions of higher education, including Boston College, the California State University System, and the University of Tennessee, have begun to implement UDL to improve their student outcomes.⁶² UDL provides a research-based “blueprint for creating instructional goals, methods, materials, and assessments that work for everyone - not a single, one-size-fits-all solution but rather flexible approaches that can be customized and adjusted for individual needs.”⁶³ Although UDL is often associated with the teaching of students with disabilities, it has wide-ranging benefits for students at all levels because it is based on the belief that what helps students at the ends of a bell curve will help all students. The incorporation of UDL at institutions of higher education can lead to coursework, physical environments, and student services that better support students of color, low-income students, and students with disabilities.

Quality of teaching is also an important factor in student success. At the K-12 level, the federal government mandates that teachers be trained, and encourages and funds ongoing professional learning. To close equity gaps, colleges and universities could focus more on training faculty, improving the quality of teaching, and measuring student learning as a formative process to inform teaching practice. For instance, institutions struggling to graduate particular student population groups can offer professional learning opportunities for administrators to utilize data in cycles of inquiry to bring about change that leads to increased equity, as well as implement training for faculty to learn how to provide culturally-responsive curricula and pedagogy.

For many students, the services they receive outside of the classroom are paramount to their success, and institutions should be intentional about understanding the needs of their student populations and responding appropriately. Research shows that student support services such peer tutoring and workshops are associated with improved academic outcomes, as are intrusive advising, mentoring, coaching, and first-year experience and summer bridge programs. ⁶⁴ Addressing students’ basic needs is also increasingly becoming a focus of institutional leaders who realize that hungry or homeless students will not be able to succeed in school.⁶⁵

Conclusion

Over the past several years, there has been clear bipartisan support for the idea of increasing accountability for colleges and universities in order to ensure that students have high-quality postsecondary options. Federal higher education policy must also provide increased transparency of data and hold schools accountable for the outcomes of their historically underserved student populations. Incorporating this new accountability will require working through many of the thorny questions discussed in this paper. However, given the current postsecondary outcomes for historically underserved students, the impact these outcomes have on our nation’s economic success in the 21^st century, and the billions of dollars in federal support institutions receive each year, policymakers cannot afford to ignore this issue. Federal policy should be leveraged to ensure that colleges and universities are no longer “sink or swim” environments for students of color, low-income students, and students with disabilities.