completion rates of higher education students

Download the Data

Download the data that appear on the College Scorecard, as well as supporting data on student completion, debt and repayment, earnings, and more.

This data was last updated May 6, 2024 .

All Data Files Download (.zip, 436 MB)

Institution-level data files for 1996-97 through 2021-22 containing aggregate data for each institution. Includes information on institutional characteristics, enrollment, student aid, costs, and student outcomes.
Field of study-level data files for the pooled 2014-15, 2015-16 award years through the pooled 2018-19, 2019-20 award years containing data at the credential level and 4-digit CIP code combination for each institution. Includes information on cumulative debt at graduation and earnings one year after graduation.
Crosswalk files for 2000-01 through 2020-21 that link the Department’s OPEID with an IPEDS UNITID for each institution.

Most Recent Institution-Level Data

Most recent data by field of study.

Looking for help? Visit StackExchange or email the help desk at [email protected] . For inquiries by members of the press, please contact [email protected] .

The Question of College Completion Rates

Posted December 9, 2021
By Jill Anderson
College Access and Success
Higher Education Leadership

For the past few decades, college completion rates have slowly been going up. Economist Jeff Denning wanted to know why. As part of recent research, Why are College Completion Rates Increasing , Denning and co-researchers dug deep to understand this phenomenon. Denning, a Brigham Young University associate professor, says the reason isn’t what you might suspect like an increase in college enrollments or that students are studying more. Instead, his research shows that grade inflation is driving the change.

"[Whether it is] good or bad to inflate grades, I think is an open question," says Denning. "We show one side of this, which is: If you had told me before I wrote this paper, 'We have this policy lever, it's going to increase graduation [rates] and it's not going to cost any money, basically,' I'd be pretty interested in that policy lever."

In this episode, Denning provides insight into how he conducted this research and also what this might mean for higher education.

TRANSCRIPT:

Jill Anderson: I'm Jill Anderson. This is the Harvard EdCast. Economist Jeff Denning set out to better understand why college completion rates are increasing. What he didn't expect to discover is the reason why. He says it's not an increase in enrollments or even student characteristics driving this increase, but actually great inflation.

As part of his research, he looked at many different factors to attribute this change. I spoke to Jeff and wanted to know what this might mean and why he started looking at college completion rates in the first place.

Jeff Denning: There was a very famous paper, My Circles by [Don Livine Hymantan 00:00:50] that he showed from the 1970s to the 1990s college completion rates declined. That is if you went to college, how often did you complete? And that actually was declining from the seventies to the nineties.

And so that was a fact that I knew and that other researchers knew. And then I saw in more recent data, I saw that college completion rates had been increasing. Starting in about the nineties, they started increasing and suddenly over that timeframe. And so we said, "Oh, that's interesting, that's a new fact. And let's see if we can explain it."

One thing I would say is college graduation rates are increasingly show, but they're still low. By the end of our period they're almost 60%. So it's not like everyone who goes to college completes, it's just slightly more now are completing than they used to.

Jill Anderson: It's something like about a 7% increases what you saw?

Jeff Denning: That's right. So seven percentage point increase relative to a baseline of 52 percentage points. So a reasonably sized increase in graduation rates.

Jill Anderson: When you really read this and you go through it, it's shocking how much data you looked at to come to your conclusion, which was that there's some grade inflation potentially happening.

So why don't you talk to me a little bit about how you did the research came to this realization that grade inflation is part of this?

Jeff Denning: Yeah, so we started with the fact completion rates are up. Why is that? Our starting place was to use the methodology of this earlier paper and consider factors that they considered. Could those explain it statistically?

So they come away saying one of the reasons that completion rates declined is that students were going to different colleges that had lower completion rates and that there was less support for higher education as measured by student faculty ratios.

So we looked at those trends and other trends and it turns out none of those trends would predict increasing graduation rates. So students are going to the same types of colleges as they were basically, they didn't change dramatically the distribution of colleges. Colleges Weren't getting more resources per student or anything like that. It had about the same amount of resources per student, maybe a little worse, but not markedly different.

We use evidence from other papers that show students are more likely to be employed for wages, not working at school, but working to make money. So that would make you think that that's not a good thing for graduation necessarily. Prices are up, tuition is up over this time. I think that might affect it. That's going kind of the wrong way as well. Students are studying less over time.

There's another paper that shows that. We started looking at all these different trends and we're like, "Oh, these are going the wrong way. They're not going to help us explain why college completion rates are increasing." But we noticed that grades were up. So first year grades were up by a fair amount. And so we're like, "Oh, well, grades are correlated with whether or not you finish college."

And so why are grades up? That might be our story. And so then we said, "Okay, why are grades up? Well, our students more prepared for college." And so we took data from some national surveys. We took data from several colleges and we saw after we used statistical methods to see comparing people who were at a similar level of preparation over time, that is that they have the same SAT score for instance, did they have higher grades in the later years? And they did. So we couldn't explain the increase in GPA by increases in preparation.

Like I said, students were studying less, so that's not explaining it. My favorite piece of evidence comes from a single university, which we are allowed to call Public Liberal Arts College. At this Public Liberal Arts College. We have data on a lot of things, but importantly, we have your grade in the class, as well as your final exam score.

And we showed that people's grades are going up over time, even after we compare people with the same final exam score. So two people in subsequent years, they have the same exact score on the final exam, but one just comes later. The person that comes later to that school, we would predict, have a higher GPA in that class.

And even further we found for some classes, they offered the exact same tests over time. And so we could compare people who got the exact same score on the exact same test and see, did they have higher grades if they were coming at the later part of the sample?

And the answer is, yes. So over time people were having higher grades, even though they presumably learned the same amount of stuff, they had the same score on the final exam. So I think that is our strongest evidence for great inflation at that particular college, but the trends in graduation and GPA at that college mirror what we see and more nationally representative data as well.

Jill Anderson: So people are going to hear this and they're going to say, "Does this mean college has gotten easier?"

Jeff Denning: Yeah. That's a great question. Our answer would be, it seems like it a little bit. Our next question is, is that a bad thing or a good thing? We set how hard college was. I don't know when in the seventies, fifties, whatever. And the question is, is that the right level of difficulty. Our paper is one that suggests like, well, there are benefits to making college a little bit easier. More people graduate, college graduation is a good thing.

It helps people in a lot of ways. It helps them earn higher wages, et cetera. And so this seems to be a benefit. There may be costs as well, but this seems to be a benefit of making it easier for students to graduate.

Jill Anderson: What do you think that this might say in the larger discussion about the value of college? Does it change it in any way?

Jeff Denning: People often ask, this is a great next question, right? Like, "Oh, so a college degree, somehow devalued." Not our paper, but other papers have shown that the college wage premium, that is how much more you make if you have a college degree compared to someone who doesn't, it's very high and has stayed pretty constant over the last 10 or 20 years, I don't remember exactly how long, but for the last little while it's very high and it's not shrinking, it's not getting worse.

So people who go to college have higher wages, and there's a lot of good evidence to suggest that college has a causal effect on wages. It increases wages. So I think the returns to college are very strong, certainly on average. You're right to wonder, you can imagine if college changed dramatically or got way, way easier, then you might be worried about it. That doesn't seem to have played out, at least in aggregate trends that we see in the data.

Jill Anderson: What about higher ed leaders and people working in higher education? How can they take this research and use it, or think about what their goals are for students?

Jeff Denning: I think the first thing that we would say is that grades are a policy tool. That this is something that they can use to affect whether or not students graduate. It can affect other things too, like the incentive to study and that sort of thing can be affected by grades. So you want to think about grades, not as just kind of arriving like the dew or something, but something that you can actually affect and would have consequences for your students. So that's the first thing.

Is it good or bad to inflate grades, I think is an open question. We show one side of this, which is: If you had told me before I wrote this paper, "We have this policy lever, it's going to increase graduation [rates] and it's not going to cost any money, basically," I'd be pretty interested in that policy lever.

And that's basically what we found. Now, you might think there are trade offs in terms of student effort or signals to employers or that sort of thing. Our paper, I hope is one in a series of papers, that maybe other people will write, or that we'll write that say, "What's the right level here. What's the right level of grades that we should be giving out."

Jill Anderson: I don't know if there's any explanation you can give in practice. Do faculty members even realize that they're inflating the grade or is it just something that happens?

Jeff Denning: People mean different things by great inflation, but what we mean by great inflation is: Are grades higher for the same level of understanding or something like that over time, right? Could people talk about inflation across disciplines or something? We're talking about over time. So that's how we define it. Are grades higher than they were before, conditional on the same amount of knowledge or competency?

Now, why is this happening? I think is a really interesting question. When people ask me this, I usually answer it with a question is, "Why didn't it happen earlier?" Who was the constituency for tougher grades? As a professor I can tell you, it's not my students. They're not asking me to grade harsher, college administrators often are, they're interested in increasing graduation rates. And so they're I think skeptical of giving out really low grades.

So they're not really rooting for low grades often. Teachers who give out lower grades are likely to get lower student evaluations, which are important for decisions about promotion and that sort of thing.

So, there's not really an obvious constituency for lower grades within the university. So there are also other speculative reasons. I think there's some evidence to suggest adjuncts tend to give out higher grades than tenure track faculty.

There's a lot more information about professors and classes online than there used to be so you can perhaps there's incentives to find the teacher who will give higher grades or that sort of thing, or pressures to compete that way. We don't really know, but there doesn't seem to be a lot of incentives for lower grades that I can see in higher education.

Jill Anderson: Well, thank you so much Jeff.

Jeff Denning: Thanks for having me.

Jill Anderson: Jeff Denning is an associate professor at Brigham Young University. He's the lead researcher of, why have college completion rates increased, to be published in the American Economic Journal. I'm Jill Anderson. This is the Harvard EdCast produced by the Harvard Graduate School of Education. Thanks for listening.

An education podcast that keeps the focus simple: what makes a difference for learners, educators, parents, and communities

Higher Education's Resistance to Change

Visiting Professor Brian Rosenberg addresses the cultural and structural factors that impede significant transformations in higher education

A walkway and buildings on a typical college campus

College Admissions During COVID

Is the College Enrollment Decline Really a Crisis?

Access & Affordability

College completion rates highest in recent years, report finds.

A new report from the National Student Clearinghouse Research Center has found that 62.2 percent of first-time students who started college in fall 2015 graduated by spring 2021, the highest six-year completion rate in recent years. According to the Lumina Foundation , the report is perhaps the last snapshot of higher education trends before COVID-19, as most of the 2015 cohort graduated before 2020.

“What we can learn from this cohort is going to be really important, because it’s going to be the cohort we look back to, comparing back to an age of stable inequality, versus severely disrupted and impacted inequality,” Dr. Stell M. Flores, an associate professor and director at the University of Texas’s Education Research Center, explained to Diverse Issues in Higher Education .

Variation across institution types

According to Higher Ed Dive , the 2015 cohort’s completion rate is up 1.2 percentage points over that of the 2014 cohort and marks the third straight year of completion rates above 60 percent.

Compared with the year prior, completion rates rose across all institution types, including community colleges, public colleges, private nonprofit colleges, and private for-profit colleges. Students who started at community college had the largest completion rate increase, at 1.5 percentage points. However, compared with students at other types of institutions, they still had the lowest completion rate, at 42.2 percent. In contrast, four-year private nonprofit and public universities had the highest completion rate at 78.3 percent and 69 percent, respectively. The rate at four-year for-profit colleges was 46.4 percent.

Especially large increase for Black students

The report also showed racial disparities in college completion. “Reflecting other societal and educational inequities, Black, Hispanic, and Native American students are less likely to complete college degrees than white and Asian students,” Amy Feygin, principal researcher at American Institutes for Research, told Inside Higher Ed . However, Black students had the largest completion rate increase—nearly two percentage points—among all racial and ethnic groups studied. White and Latinx students’ completion rates also increased, while Asian students’ completion rates held steady.

Analyzing completion by age group, researchers found that adult learners (students who first entered college at age 24 and older) had the highest completion rate increase, although traditional college-age students (students who entered college at 20 years old or younger) continued to have higher completion rates than older students.

For Doug Shapiro, executive director of the National Student Clearinghouse Research Center, “This broad measure of performance for higher education as a nationwide system…shows long-term improvements for students and colleges alike.” The Lumina Foundation attributes the results to several factors, including “more students starting their educational journeys at bachelor’s-granting institutions, and, possibly, better academic and financial support for students, especially students of color.”

A dip on the horizon?

Feygin, however, predicts that completion rates will soon trend downward, reflecting pandemic-related enrollment loss, especially among Black and Native American male students.

The report also noted worrisome signs, including the high rate of students “stopping out” (temporarily leaving school) and withdrawing from college permanently due to economic and academic hurdles. Increasing investments in financial, academic, and basic needs support is, according to Feygin, the best way to mitigate these downward trends.

Topics in this story

‘put the money to good use’: revisiting mackenzie scott’s transformative unrestricted gifts to hbcus.

More than a year after MacKenzie Scott donated $560 million in unrestricted funds to 23 public and private historically Black colleges and universities, the gifts have added resilience to minority-serving institutions confronting existential threats.

Sign up for our weekly email

Please provide your first and last name. * First name Last name

Skip to Nav
Skip to Main
Skip to Footer

College completion rates are up for all Americans, but racial gaps persist

Please try again

completion rates of higher education students

Students’ race and ethnicity affect their chances of earning a college degree, according to several new reports on higher education released in January and February 2023. However, the picture that emerges depends on the lens you use. College degrees are increasing among all racial and ethnic groups, but white and Asian Americans are far more likely to hold a college degree or earn one than Black, Hispanic or Native Americans.

Earning a college degree involves two steps: starting college and finishing college. Before the pandemic, white, Black and Hispanic Americans were enrolling in college at about the same rates, especially when unemployment was high and jobs were hard to find. (Asian Americans enrolled in college at much higher rates.) The bigger distinction is that once a student has started college, the likelihood of making it through the coursework and tuition payments and ultimately earning a degree varies so much by race and ethnicity.

First, let’s begin with enrollment. There are two ways to look at this. One is to see how the demographic makeup of college campuses has changed over time, becoming less white and more Hispanic. The pie charts below were produced in January by the National Student Clearinghouse, a nonprofit organization that provides data reporting services to colleges. In conjunction with these services, it monitors trends in higher education by aggregating the data submitted by more than 3,600 institutions, representing 97 percent of the students at the nation’s degree-granting colleges and universities. Earlier this year, the organization launched a DEI Data Lab site to put a spotlight on how college enrollment, persistence and completion vary by race and ethnicity.

In 2011, as the pie chart on the left shows, more than 60 percent of the nation’s 20.6 million college students were white, according to an estimate by the National Student Clearinghouse. By 2020, the year represented by the pie chart on the right, the total number of college students had fallen to 17.8 million and the share of white students had dropped by almost 9 percentage points to 52 percent, still a majority. During the same period, the share of Hispanic students grew from 14 percent to 21 percent, and the share of Black students remained constant at just under 14 percent. Asian students increased from 5 to 7 percent of the college population. This represents all undergraduate college students, both younger students entering straight after high school and older nontraditional students, studying full-time and part-time, and attending both four-year universities and two-year colleges.

The 2011 figures are rough estimates because only one out of five colleges reported race and ethnicity of students to the Clearinghouse. Today, more than three out of five colleges report on the race and ethnicity of their students to the Clearinghouse. (For the original version of this pie chart, click here .)

How should we think about these college enrollment numbers? Do they largely mirror each racial and ethnic group’s share of the population? I was surprised to learn that the answer is yes – with a few caveats. Asian Americans are slightly overrepresented on college campuses and Hispanic Americans are slightly underrepresented.

I created this chart below, comparing the National Student Clearinghouse’s college enrollment data for 2020 with the young adult population, as reported by the U.S Census, so you can see how closely college enrollment tracks actual demographics.

Another way to look at college enrollment is to see how many young adults enroll in college. The chart below, by the U.S. Bureau of Labor Statistics, shows that the college enrollment rates of Black and Hispanic young adults improved after the 2008 recession, and approached the college going rate of white Americans. Roughly 60 percent of young Black, Hispanic and white Americans are trying for a college degree. The college going rate for Asian Americans is much higher; more than 80 percent enroll. The zigs and zags in this chart show how college going among Hispanic and Black Americans is influenced by business cycles.

The Bureau of Labor Statistics obtains enrollment data from the Current Population Survey (CPS), a monthly survey of households conducted by the Bureau of Census. (Here is the chart on the BLS site .)

When jobs are plentiful, many low-income students may join the labor force and defer their higher education. That especially reduces enrollments among Black and Hispanic young adults, among whom poverty rates are higher. When unemployment is high, more young adults enroll at college, particularly at two-year community colleges. Most recently, during the pandemic, many young Americans deferred college to help support or take care of their families. Some students chose to wait until in-person classes resumed.

Going to college is one thing; finishing it is another. This fourth chart, produced by the Lumina Foundation, shows that over time, more Americans of every race and ethnicity are earning college degrees. The Lumina Foundation is a private foundation that seeks to increase the number of adults with college degrees and other credentials, and was formed through the sale to Sallie Mae of USA Group's assets that were used to create and collect monthly payments on student loans.* It is also among the many funders of The Hechinger Report.

Share of adult population, ages 25-64, with college degrees

This chart above, originally published here on Jan. 31, is based on data from the U.S. Census Bureau’s American Community Survey. It tracks the percentage of adults 25 to 64 with two-year associate and four-year bachelor’s degrees. The share of Americans with a college degree rose from 38 percent in 2009 to nearly 46 percent in 2021 – an increase of eight percentage points.

Every race and ethnicity saw gains. The eight-percentage point gain was the same for both Black and white adults.

But racial gaps continue. In 2021, there remained an enormous 40 percentage point difference between Asian American adults, among whom 66 percent have a college degree, and Native American adults, among whom only 25 percent have a college degree. Among Black adults, 34 percent have college degrees. Among Hispanic adults, it’s 28 percent and among white adults, it’s 50 percent.

Improvements in college attainment can seem slow because graduation rates are much lower among Americans over 35. It takes years for higher college graduation rates among younger adults to raise overall college numbers. College attainment rates have jumped the fastest among young Hispanic adults under age 35, rising from below 20 percent in 2009 to above 30 percent in 2021. Courtney Brown, the chief data and research officer at Lumina, credits a variety of support programs, from tutoring to food pantries, and the convenience of online courses to explain why more young people are graduating, despite rising tuition costs. “Colleges are trying to serve students better,” said Brown. “Even the way they staff colleges, not all on getting enrollments but having more success coaches available and counselors helping students get to the finish line.”

Still, Brown acknowledges that it’s been difficult to make a dent in the stubborn gaps in college attainment between people of different races and ethnicities. “Unfortunately, everyone is increasing,” Brown said. “And so we are not seeing those gaps reduced.”

The National Student Clearinghouse’s DEI Data Lab also shows this completion problem starkly.

This chart tracks cohorts of students who began college at the same time and calculates how many of them earned any college degree within six years. Among students who started college in the fall of 2010, 62 percent of white students completed a degree by the summer of 2016, compared with only 39 percent of Black students. That’s a giant 23 percentage point gap, and a sign that a disproportionate number of Black students are dropping out of college in debt. Completion rates improved considerably for students who started college in 2015, but large gaps remain. Almost 70 percent of white students completed a degree by the summer of 2021, but only 45 percent of Black students hit this milestone. The Black-white college completion gap actually widened slightly from 23 to 24 percentage points.

The reasons for why completion rates remain much lower for Black, Hispanic and Native American students are complex. These students are more likely to attend community colleges, which have lower funding per student and fewer support services. Many students weren’t adequately prepared in high schools to handle college-level coursework, especially in math.

A poll of Black college students by Gallup-Lumina , released on Feb. 9, found that 21 percent of Black students report feeling discriminated against frequently or occasionally at the college they are attending, and that 45 percent have considered dropping out in the past six months. Black students in bachelor’s programs are far more likely to juggle family and work responsibilities alongside their studies.

“Black students are encountering so much more discrimination, and they have multiple responsibilities that no other race or ethnicity really has,” said Lumina’s Brown. “A lot of it is that Black students are more likely to have children. Working full time, having children and trying to get a bachelor’s degree at the same time is just obviously overwhelming.”

On Feb. 2, the National Student Clearinghouse Research Center released the most recent college enrollment numbers for 2022. Undergraduate enrollment for both white and Black students fell for the fifth straight year, while enrollment of Hispanic and Asian students at public two-year colleges improved. However, their numbers are below pre-pandemic levels. For example, there were roughly 975,000 Hispanic students enrolled in public two-year colleges, also called community colleges, in the fall of 2022, up from 944,000 in the fall of 2021, but considerably down from 1.14 million in 2019. (Click here and navigate to the demographics tab for these fall 2022 charts.)

And here’s a startling data point: Black student enrollment at two-year community colleges declined by a staggering 44 percent, from 1.2 million in 2010 to 670,000 in 2020, according to a Sept. 2022 report by the Joint Center for Political and Economic Studies, a think tank that studies policy issues affecting Black Americans.

Fewer students at college now certainly means fewer college-educated adults in the years ahead. And that is not a promising future.

* Correction: An earlier version incorrectly said that Lumina was founded through the sale of Sallie Mae, instead of USA Group’s sale of assets to Sallie Mae. USA Group was sold, not Sallie Mae.

This story about higher ed data 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 the Hechinger newsletter .

Improving community college completion rates by addressing structural and motivational barriers

Subscribe to the brown center on education policy newsletter, elizabeth mann levesque elizabeth mann levesque former brookings expert, student support & classroom climate consultant - university of michigan @elizkmann.

October 8, 2018

Many community college students in the U.S. do not complete a credential or degree, facing dramatically reduced earning potential. However, evidence suggests that helping students navigate the college environment and connect their coursework to their lives can help solve the community college completion puzzle. In this report, Elizabeth Mann Levesque discusses the structural and motivational barriers these students face, potential solutions, and offers policy recommendations to boost completion rates.

Table of Contents Introduction Section 1: The problem of low completion rates Section 2: The structural barrier Section 3: The motivational barrier Section 4: Policy recommendations

63 min read

Introduction

Community colleges have the potential to provide students across the United States with viable pathways into good-paying jobs. However, many students who enroll in community college in the United States do not complete a certificate or degree. Specifically, fewer than 40 percent of community college students earn a certificate or degree within six years of enrollment (Bailey et al. 2015). Individuals who do not complete any type of certificate (also referred to as a credential) beyond a high school diploma face dramatically reduced earning potential (Belfield and Bailey 2017). Low completion rates are thus an urgent problem for millions of Americans seeking a pathway into the middle class and employers reliant on a skilled and educated workforce.

To address this problem, this report discusses structural and motivational barriers to completion that community college students face, as well as strategies to overcome these barriers. The key takeaways are that making it easier for students to navigate the college environment and connect their coursework to their lives can improve student outcomes. This analysis is relevant for community colleges, policymakers seeking to provide students with a viable route to upward economic mobility, employers hoping to strengthen their workforce development partnerships, and philanthropies invested in reducing economic inequality.

Before proceeding, it is important to note that this report is not a comprehensive discussion of all the barriers to completion that community college students face. Many factors contribute to low completion rates, including inadequate academic preparation and financial constraints, despite the relative affordability of public two-year colleges (Holzer and Baum 2017; Goldrick-Rab 2016). Indeed, addressing educational and financial barriers is fundamentally important to improving community college completion. 1 The goal here is to spotlight two particular barriers, structural and motivational, that policymakers, colleges, and employers should address to improve students’ prospects.

Making it easier for students to navigate the college environment and connect their coursework to their lives can improve student outcomes.

Section 1 discusses the role that community colleges play in the higher education landscape, the well-known but intransigent problem of low completion rates, and the value of completing a postsecondary credential or degree. Sections 2 and 3 discuss the structural and motivational barriers, respectively. Section 4 discusses policy recommendations about how to address these barriers.

Section 1: The problem of low completion rates

Extensive evidence indicates that completing a credential or degree beyond a high school diploma, from an associate’s degree all the way up to a doctorate, improves employment outcomes and earnings for individuals (Oyserman 2012; Holzer and Baum 2017; Carnevale et al. 2017; Belfield and Bailey 2017). Yet, community colleges, as the institutions of higher education that serve a large share of low-income students, graduate less than 40 percent of students within six years (Bailey et al. 2015). Thus, community colleges have substantial unrealized potential to improve the earnings and employment outcomes for those at the bottom of the income distribution.

Indeed, community colleges could play a pivotal role in providing a pathway to upward economic mobility because of their ability to reach a large population of low-income and minority students. In fall 2015, community colleges served 41 percent of all U.S. undergraduates. Furthermore, community colleges serve a large share of the country’s non-white undergraduates: 56 percent of Native Americans, 52 percent of Hispanics, 43 percent of African-Americans, and 40 percent of Asian/Pacific Islanders. Public community colleges are also more affordable compared to four-year colleges. For 2017-18, the average annual tuition and fees for public, in-district community colleges was $3,570 compared to $9,970 for public, in-state four year colleges. 2 Given these differences in cost, it is unsurprising that a 2011 analysis found that “44 percent of low-income students (those with family incomes of less than $25,000 per year) attend community colleges as their first college after high school.” 3 In short, community colleges offer a pathway into higher education for a large share of low-income and minority students.

Fewer than 40 percent of community college students earn a certificate or degree within six years of enrollment.

However, many who enroll in community college do not complete a post-secondary certificate or degree. Bailey et al. (2015) summarize the problem succinctly: “Most students who enter these colleges never finish: fewer than four of every ten complete any type of degree or certificate within six years” (p. 1). For example, one analysis finds that, of the students who enroll in community college, only 26 percent earn an associate’s or bachelor’s degree after six years (Jacob 2018).

The story when we consider four-year degree completion rates for students who start at community colleges is also troubling. According to the Community College Research Center (CCRC) at Columbia University Teachers College (CCRC), 81 percent of entering community college students indicate they want to earn at least a bachelor’s degree, but only 33 percent transfer to a four-year institution within six years (Horn and Skomsvold 2011; Jenkins and Fink 2016). Of those who transfer, CCRC reports that “42 percent complete a bachelor’s degree within six years (Jenkins and Fink 2016).” The stunning implication is that only “14 percent of the entire cohort of entering community college students earns a bachelor’s degree within six years (Jenkins and Fink 2016).” 4 To be clear, the problem is not a lack of aspiration among community college students (Aelenei et al. 2017). Rather, as this report will discuss in detail, structural barriers in many community colleges often make it difficult for students to achieve their goals.

In addition, college completion rates differ systematically by household income. Comparing two age cohorts, those born in 1961-1964 and 1979-1982, Bailey and Dynarski (2011) find that college entry and completion rates rose over time for children from high- and low-income households in both cohorts. However, they note that “the increases were highly uneven, with gains largest at the top of the income distribution and smallest at the bottom” (p. 120). While the college completion rate for the 1979-1982 birth cohort is 54 percent for students from households in the top income quartile, this rate drops precipitously as income declines, to 32 percent for households in the third quartile, 21 percent in the second quartile, and only 9 percent in the lowest quartile. 5

These low completion rates are clearly problematic, particularly in light of evidence that completing a two-year degree is undeniably valuable in terms of earnings potential. Belfield and Bailey (2017) review analyses of eight states and find that the average earnings gains to completing an associate’s degree compared to enrolling in college but not completing a degree are $4,460 for men and $7,160 for women per year (p. 7). 6 In a different analysis of data from the state of Florida, Holzer and Baum (2017) similarly find large returns to higher education. 7 They find that the largest returns are to completing an advanced degree and bachelor’s degree, which are associated with, on average, 99 and 71 percent increases in earnings relative to completing only a high school diploma, respectively. While the gains to two-year degrees are smaller, they remain substantial. They find that, on average, completing an associate’s degree is associated with a 37 percent increase in earnings compared to only completing a high school diploma. 8 Further, Carnevale (2012) argues that there is important variation in potential earnings by major, describing how, in some cases, students who graduate with a two-year degree in one field stand to earn more than students with a four-year degree in a different field. He explains: “For example, an engineering technician with an associate’s degree can make more than a guidance counselor with a master’s degree” (p. 61).

A step in the right direction: Providing information on student outcomes

One approach to improving outcomes for community college students (and prospective students) is increasing awareness of the labor market outcomes associated with certain fields of study and degrees. Indeed, evidence suggests a clear need to increase awareness among community college students about outcomes associated with particular programs of study, including information on average earnings and employment opportunities by major and degree. A recent survey of community college students finds that students “overestimate salaries by 13% and underestimate the probability of finding employment by almost 25% in almost all fields” (Baker et al. 2018, p. 19). The authors conclude that providing community college students with more information on labor market outcomes and increasing the salience of this information “could serve community colleges’ goal of improving their students’ labor market prospects” (p. 19).

Evidence suggests a clear need to increase awareness among community college students about outcomes associated with particular programs of study, including information on average earnings and employment opportunities by major and degree.

In terms of providing more information, policy appears to be moving in the right direction. State governments maintain websites that provide information on the state labor market, such as the annual salary of different occupations, as well as projections about the likely growth or contraction of available jobs by occupation in the future. The ease of finding and navigating these websites varies, as does the specific type of information available. 15 Some states have partnered with research institutions to provide relevant information via easily navigable sites; the Launch My Career websites are one example of this approach. Working to bridge the information gap, the Skillful initiative is an example of a private-public partnership that seeks to provide resources for job seekers, employers, and coaches. 16 Anthony Carnevale, director of the Georgetown University Center on Education and the Workforce, has advocated for making the connection between postsecondary education and available jobs explicit by creating online matching systems that link job openings with coursework at postsecondary institutions (2012).

The federal government maintains multiple online resources to provide students with information about institutions of higher education. The National Center for Education Statistics maintains the online College Navigator. This resource allows students to explore information on colleges and universities, including information on available majors, cost, enrollment, admissions, retention, and graduation rates. The College Navigator provides information on graduation rates by major, as well as information on the rate at which students default on their student loans. The Department of Education introduced the online College Scorecard in September 2015. The scorecard provides detailed information on institutions of higher education, including student body characteristics, financial aid, debt after graduation, graduation and retention, and average salaries of former students. 17 In sum, this non-exhaustive discussion provides a glimpse into the valuable online resources that provide information to the public about the labor market outcomes associated with higher education.

While important, providing students with information as they make choices about which program of study to embark upon is only part of the solution to the completion problem. Students who need help making informed decisions may not be the ones who are most likely to use information provided online. Noting variation in how students responded to the initial release of the College Scorecard in September 2015–with private school students demonstrating more responsiveness to new information on reported earnings (Hurwitz and Smith 2018)–Holzer and Baum (2017) argue that “[j]ust making general information available is unlikely to significantly improve the college decisions of students from less-privileged backgrounds” (p. 123). Indeed, as discussed above, Baker et al. (2018) recommend not only making information on earnings available, but also making it more salient in students’ decisionmaking.

As the rest of this report discusses in detail, structural and motivational barriers may derail students even when they are equipped with information about labor market outcomes. To preview the nature of these barriers, consider that students who select a major with high potential earnings may find themselves confused about how to complete their chosen program of study. Or, during their coursework, students may not see the connection between their courses and their selected career path. Further, recall that while 81 percent of students enter community college intending to earn a four-year degree, only 14 percent of these students actually do so within six years (Horn and Skomsvold 2011; Jenkins and Fink 2016).

The intent to graduate with a postsecondary degree is clearly present among a majority of community college students. In addition to inaccurate or incomplete information about the potential value of different degrees, other barriers prevent students from reaching their goals. In the subsequent sections, this report discusses structural and motivational barriers to college completion that we should consider alongside strategies designed to provide students with information on labor market outcomes.

Section 2: The structural barrier

The traditional structure of community colleges hampers students’ progress toward completing a credential or degree. This is the core critique by researchers who have recently made a strong and convincing case for rethinking the structure of community colleges (Bailey et al. 2015; Holzer and Baum 2017). Bailey et al. (2015) characterize this traditional structure as a “cafeteria model” (p. 13). In this section, I explain the barriers students face in cafeteria-style colleges, as well as promising strategies for addressing these problems. To be clear, the problems in cafeteria-style colleges are attributable to the structure and design features of colleges, not to faculty, staff, and other individuals who work hard on the behalf of students.

The traditional structure of community colleges hampers students’ progress toward completing a credential or degree.

To appreciate the nature of this problem, consider the community college environment. Completing a credential or degree requires students to sort through an overwhelming amount of information to make complicated decisions, such as what to major in, what courses to take to satisfy program requirements, whether and how to get involved in a job training program, whether and how to transfer to a four-year program, and what kind of job to pursue after graduation, just to name a few (Scott-Clayton 2011; Bailey et al. 2015).

In the cafeteria model, it is often difficult for students to identify a clear pathway from enrollment to obtaining a credential, earning a two-year degree, or transferring to a four-year program. Program requirements and options are often unclear, and students have insufficient guidance and support to make informed choices. Bailey et al. (2015) describe how in a cafeteria-style college, “Students are left to navigate often complex and ill-defined pathways mostly on their own” (p. 13). The authors link this structure, and students’ difficulties navigating it, to dropping out: “We find that the typical student is overwhelmed by the many choices available, resulting in poor program or course selection decisions, which in turn cost time and money, and likely lead many students to drop out in frustration” (p. 22).

Cafeteria-style colleges often do not have enough advising resources to support students as they try to navigate this complex environment. The ratio of students to academic advisers is staggering. Estimates of the typical adviser’s caseload vary, but in all cases, it is clear that advisers are stretched incredibly thin. Bailey et al. (2015) cite estimates of approximately one adviser per every 800-1,200 students (p. 58-59). The National Academic Advising Association reports a median of 441 students per adviser in community colleges (Scrivener et al. 2015 p. 26, citing Robbins 2013).

During initial advising sessions, there is typically not enough time for long-term planning, goal-setting, or a thorough orientation to college life and how to navigate from enrollment to completion (Bailey et al. 2015). Bailey et al. (2015) report that on the Community College Survey of Student Engagement, “Only 38 percent of students reported that an advisor helped them to set academic goals and create a plan for achieving those goals” (p .55). Through the orientation and registration process, students often do not learn about the college’s program offerings or career or transfer pathways. Without a clear understanding of available opportunities, “They embark on a slate of courses whose relevance to their own nascent goals and interests may not be entirely clear” (Bailey et al. 2015, p. 56). While cafeteria-style colleges may offer support services to help students navigate this terrain, Bailey et al. (2015) conclude that navigating even the support services can be “complicated and confusing” (p. 58-59).

Holzer and Baum (2017) draw a clear connection between the structural barriers in cafeteria-style colleges and low completion rates that is worth quoting at length:

The cafeteria approach reduces success rates among students for a variety of reasons. Students have too many options, too little information about the alternatives, and too little insight into their own skills and preferences when choosing courses and majors…Instead of gathering sufficient information to make sensible choices, students defer difficult choices like choosing a major and fall back on default options, taking the path of least resistance. They struggle to complete the degree or credential they seek, with too little guidance about what courses to take and in what sequence. Students have particular difficulty transferring into other programs and to institutions because the credits they have earned are frequently not accepted (p. 160).

In short, students entering community college often face many decisions about how to define and achieve their goals. Confronted with an overwhelming number of options, many students do not know how to progress from enrollment to program completion, lacking sufficient support in charting a course and adhering to their plans. As a result, students invest time and money in classes that may not feel related to their own goals and may not adequately satisfy specific program requirements. In this environment, students may not see the value in remaining enrolled.

Addressing the structural barrier

Addressing the structural barrier requires rethinking the organization of community colleges to reduce the confusion and frustration that contributes to low completion rates. This type of approach is referred to as the “guided pathways” model. In contrast to the cafeteria model, guided pathways models feature clearly structured programs and extensive advising support. Bailey et al. (2015) describe the guided pathways approach as a comprehensive strategy to address the problems found in the cafeteria model, focusing on program structure, intake and student supports, instruction, and developmental education.

In contrast to the traditional “cafeteria-style model” of community colleges, the guided pathways approach features clearly structured programs and extensive advising support.

Bailey et al. (2015) explain the guided pathways framework as follows. 18 The steps for moving toward completing a credential or degree are much clearer, and the decisionmaking process is simplified. Academic programs are clearly defined, with coherent trajectories. Students who enter without a clear direction are assisted in choosing an area of study with a default curriculum that the student can modify to fit their interests. The intake and student support services provide students with consistent, structured advising. Students are not left on their own to make important decisions about career pathways or to ask for help when they are struggling. Rather, students receive career counseling at the outset, and advisers track students’ progress through their programs. In terms of instruction, faculty help to clarify how students will achieve the learning outcomes associated with each program of study. Developmental education courses, which can often be a stumbling block for students, are more deliberately incorporated into programs of study.

Evidence suggests that implementing reforms aligned with this guided pathways approach can improve student outcomes. Perhaps the most high-profile success story of a guided pathways approach is the City University of New York’s (CUNY) Accelerated Study in Associate Programs (ASAP). The ASAP program is “designed to help more students graduate and help them graduate more quickly” (Scrivener et al. 2015, p. iii). The program requires participants to attend classes full time, and students are encouraged to take developmental courses early and to graduate within three years. Students receive advising services from advisers with small caseloads, and they receive enhanced career services and tutoring. Students can take blocked courses in the first year, and they can also take a seminar during their first few semesters that emphasizes skills like goal-setting. ASAP participants receive a tuition waiver that covers all expenses including tuition and fees, free use of textbooks, and free passes for public transportation provided that students participate in certain program services (Scrivener et al. 2015, p. iii).

MDRC, a nonprofit, nonpartisan education and social policy research organization, conducted a rigorous evaluation of the ASAP program using a random assignment study at three CUNY community colleges (Scrivener et al. 2015). The study compared low-income students who needed developmental coursework and were randomly assigned to participate in ASAP with similar students who were randomly assigned to a control group; the control group students had access to the standard CUNY services for students. The results were overwhelmingly positive. The evaluation found that over three years, ASAP students earned nine more credits than control group students. Graduation rates almost doubled among ASAP students—40 percent of ASAP students had received a degree within three years, compared to 22 percent of the control group. Further, transfer rates to four-year programs were higher among ASAP students (25 percent of the group) compared to control group students (17 percent). The program was also cost effective; because the graduation rate was so much higher among ASAP students, the cost of services per graduate was actually lower.

The authors of the MDRC report underscore the importance of these results: “ASAP’s effects are the largest MDRC has found in any of its evaluations of community college reforms. The model offers a highly promising strategy to markedly accelerate credit accumulation and increase graduation rates among educationally and economically disadvantaged populations” (p. iii). Without a doubt, the ASAP program represents an effective strategy for improving college completion rates among disadvantaged populations.

Several aspects of the ASAP program target the structural barriers found in cafeteria models. In particular, the ASAP program provides comprehensive advisement services, described as such by the authors of the MDRC report:

During the course of the study, ASAP students were required to meet with their assigned adviser in person twice per month throughout each semester; advisers also sometimes communicated with students by phone, e-mail, or text message. Advising appointments were tracked, and attendance was linked to students’ receipt of monthly Metro-Cards. ASAP students were assigned an adviser during their first semester and usually continued to see the same person throughout their college careers (Scrivener et al. 2015, p. 26).

As part of the evaluation, ASAP and control group students were surveyed during their first year of college. These survey responses show that students in the ASAP program group reported seeing their advisers much more frequently compared to students in the control group (21 times on average in the first semester compared to four, and 17 times in the second semester compared to two). ASAP students were more likely to rate their advising experiences as high quality, and on average, they reported spending more time with their advisers during each session (64 percent of ASAP survey respondents reported advising sessions of 16 to 30 minutes on average compared to 47 percent of control group survey respondents). 19 Students in the ASAP program also report discussing a wider range of topics with their advisers. For example, 95 percent of ASAP program participants reported discussing academic progress with their adviser compared to 62 percent of control group respondents (p. 29).

Over a three-year evaluation, graduation rates almost doubled among ASAP students—40 percent of ASAP students had received a degree within three years, compared to 22 percent of the control group. Further, transfer rates to four-year programs were higher among ASAP students compared to control group students.

ASAP students and control group students worked with different advisers. The ASAP advisers had much smaller caseloads, serving an average of 60 to 80 students per semester. At the three colleges in the ASAP study, non-ASAP advisers had caseloads that ranged from 600 to 1,500 students per adviser (p. 26). The report notes that “[while] dedicated, the colleges’ non-ASAP advisers managed very large caseloads that typically did not permit the more personalized touch that students experienced in ASAP advising” (p. 28).

ASAP students were also required to visit an ASAP career and employment specialist (CES) once a semester. These specialists discussed topics such as finding part-time jobs, networking, writing resumes, and aligning majors with career paths. Students in the control group were not required to visit a specialist, and survey results indicate that ASAP students used these services more frequently: 80 percent of ASAP survey respondents indicated that they spoke with CES staff in their first year, compared to 29 percent of control group students (p. 31, Table 3.4).

As mentioned above, the program mandated specific requirements of students and provided additional resources beyond advisement and career services, such as financial support, free use of textbooks, and free public transportation. Indeed, the authors of the MDRC report observe that ASAP program is an “uncommonly comprehensive and long-term program” (p. iii). Furthermore, the MDRC analysis discussed here was not designed to evaluate the impact that individual components of the program had on students’ credit accumulation and graduation rates. Rather, the evaluation estimates how the “full package” of ASAP services affected students’ academic outcomes (p. 85). Policymakers hoping to learn from and adapt this approach must be aware that it is not necessarily the case that implementing some, but not all, of the ASAP approach will produce the same positive outcomes for students.

Students face substantial structural barriers to college completion when colleges are organized in the traditional “cafeteria-style” model. As the ASAP program convincingly demonstrates, addressing these design flaws can dramatically improve completion rates among students. Admittedly, enacting this entire suite of structural changes in a community college may be a daunting and costly prospect. 20 Nonetheless, the evidence is clear that embracing the guided pathways approach is an essential step toward improving community college completion rates. Section 4 below offers recommendations about steps aligned with the guided pathways approach that community colleges can take to improve their students’ experiences and outcomes.

Section 3: The motivational barrier

Community college environments fundamentally shape students’ experiences, as the previous section illustrated. In cafeteria-style environments, it can be very difficult for students to identify connections between their coursework and their own lives and future goals. This disconnect is what I refer to as a motivational barrier. Unfortunately, but perhaps unsurprisingly, the structural problems in cafeteria-style colleges contribute to this barrier. For this reason, while this section primarily focuses on the motivational barrier, it also builds on the prior discussion of structural challenges and acknowledges the relationship between structural and motivational barriers.

This section discusses the importance of motivation in educational settings and reflects on how adopting a guided pathways approach and incorporating specific instructional strategies can help address this motivational barrier. It is important to emphasize at the outset that community college students do not have inherent shortcomings that create a motivational barrier. As discussed above, the vast majority of community college students enter college intending to complete a four-year degree. Furthermore, low-income students who attend four-year colleges and universities have “very similar earnings outcomes” compared to high-income students who attend the same school (Chetty et al. 2017, p. 2). However, in cafeteria-style colleges, students face structural barriers that make the road to completion unclear and difficult. Compounding these structural challenges, community college students are more likely than their four-year university peers to experience life circumstances that make navigating academic settings more difficult (Aelenei et al. 2017).

Defining the motivational barrier

The subsequent discussion focuses on a particular aspect of motivation that can have powerful, positive effects on student outcomes, including long-term interest in a subject and academic performance: students’ perceptions that their coursework is meaningful in relation to their lives. Maintaining this type of motivation can unfortunately be quite difficult in cafeteria-style college settings. Bailey et al. (2015) note that “without a specific goal to work toward, it can be difficult for students to maintain their motivation to persist in school” (p. 57). They describe how this lack of motivation may be a barrier for students, citing survey results in which current and former community college students indicate that having a goal helps them stay on track in college. Karp (2011) similarly describes how students need to perceive a connection between their coursework and their lives to stay sufficiently motivated: “Students who do not see the value in their coursework often behave in counterproductive ways, for example, by failing to complete assignments or dropping required courses” (p. 12).

This perceived disconnect between coursework and students’ own lives may be exacerbated within the cafeteria-style college. Without sufficient knowledge of the programs of study available or how to choose courses related to a particular academic or career path, students may enroll in courses that have little relevance to their career goals. Even if students are enrolled in courses that are related to their long-term interests, it may be difficult to see how a specific course helps them come closer to earning a credential or degree if program pathways are unclear. Indeed, these are some of the main shortcomings of the cafeteria model as discussed by Holzer and Baum (2017) and Bailey et al. (2015). Perceiving little to no “real-world” application of their coursework, students may decide that remaining enrolled in coursework is a bad investment, opting instead to drop out of college to join the workforce immediately. In other words, for students to stay motivated to persist on their academic trajectory, they need to both see the destination (the careers and earnings an education will provide) and the pathway to get there (the connections between what they are doing in school and what they would like to achieve) (Oyserman and Lewis 2017). The motivational barrier is thus an obstacle to degree completion, with negative consequences for students and the workforce more broadly.

For students to stay motivated to persist on their academic trajectory, they need to both see the destination (the careers and earnings an education will provide) and the pathway to get there (the connections between what they are doing in school and what they would like to achieve).

I draw on the expectancy-value (EV) framework to define this type of motivational barrier more precisely. 21 Hulleman et al. (2016) explain that since the 1980s, psychologists have applied the EV framework to explain student behavior in educational contexts. According to the EV framework, there are “two critical aspects of motivation that are necessary for students to be optimally engaged. First, students need to believe that they can succeed (i.e. positive expectancies). Second, students need to perceive an important reason to engage in the behavior (i.e. need to have positive values)” (Hulleman et al. 2016, p. 242). Hulleman and colleagues add a third dimension to this framework: cost. “Even if students believe they can do a task and have a reason to do a task, they still might not be motivated if they experience significant cost preventing them from engaging in that task” (Hulleman et al. 2016, p. 258). Thus, in terms of the EV framework, a motivation barrier may occur if students have low expectancy (do not believe they can succeed), low value (do not perceive value in a particular behavior), or high costs (insufficient time or resources to engage in the behavior).

In the context of community college completion, students may be particularly vulnerable to low expectancy, low value, and high costs. In terms of expectancy, students who enter community college and who have previously struggled in academic settings may not believe that they will be successful in their coursework. In terms of value, particularly in cafeteria-style settings, students may not see a connection between completing current courses and progressing toward academic and/or career goals. Given the demands on many community college students’ time, including work and family commitments (Johnson and Rochkind 2009), perceiving value in their coursework may be critically important to sustaining the necessary motivation to stay enrolled. Finally, in terms of cost, community college students may lack sufficient financial resources to stay enrolled (Goldrick-Rab 2016). 22

Expectancy value interventions

Research on the relationship between students’ perceptions of value and their academic outcomes provides insight into the role of motivation in student success. Hulleman et al. (2010) argue that “it is plausible that students who are disengaged from school due to a history of poor performance or low expectations may benefit the most from a utility value intervention” (p. 882). Indeed, a large body of rigorous, experiment-based research on EV interventions in educational settings illustrates the importance of helping students to build motivation. Further, this work suggests that increasing students’ perception of the value of their coursework can have powerful, positive effects, particularly for students who initially have low expectancies.

A large body of rigorous, experiment-based research on expectancy-value interventions in educational settings illustrates the importance of helping students to build motivation.

These EV interventions vary, but in general, they prompt students to connect their coursework to their lives–in the words of Hullemen and colleagues, to “generate their own connections and discover for themselves the relevance of course materials to their lives” (Hullemen et al. 2010, p. 881). Several interventions implemented in different educational contexts have demonstrated that increasing students’ value in a particular task can positively impact student outcomes, including performance and interest in a course (Hulleman and Harackiewicz 2009; Hulleman et al. 2010; Harackiewicz et al. 2015; Rozek et al. 2015). This type of intervention appears particularly effective for students with low expectations at the beginning of a course (Hulleman et al. 2010). These experiments have been conducted in high school (Hulleman and Harackiewicz 2009; Rozek et al. 2015) and college settings (Hulleman et al. 2010; Harackiewicz et al. 2015; Hullemen et al. 2017).

A brief discussion of a college-level EV intervention helps exemplify what this type of intervention entails and how it can affect student outcomes. Hullemen et al. (2010) conducted two EV interventions via two experiments, one in a laboratory setting and one in an introductory psychology course. This discussion focuses on the latter. Students were randomly assigned to treatment and control groups, and each group received a different writing assignment (the assignments were part of the syllabus and were completed for course credit). These assignments required students to choose a topic covered in the course and write a one to two page essay on the topic; students were required to complete this assignment twice.

In the treatment group, students were either assigned to write a letter to a significant person in their lives “describing the relevance of their topic to this person” or “find a media report … that pertained to their topic and write an essay that discussed the relevance of the media report to information they were learning in class” (p. 888). Both assignments “asked students to connect the course material to their lives through their social connections or the popular media” (p. 888). In the control group, students were not asked to make this personal connection. Instead, they either summarized a topic from class or searched an online database for abstracts related to the topic and discussed how the abstracts “expanded upon the material they were learning in class” (p. 888). Neither control group prompt encouraged students to identify the relevance of their coursework for their own lives.

The researchers measured students’ interest in psychology and their perceptions of the value of the course (utility value) based on their responses to a number of items. For example, one item used to measure perceptions of utility value was: “I think what we are studying in Introductory Psychology is useful for me to know” (p. 888). The researchers also measured students’ grades in the course. The study finds that the intervention increased students’ perceptions of utility value, and in turn, students were more interested in majoring in psychology. Further, students’ grades improved as their utility values increased. These effects were strongest for students with lower prior performance in the course. In short, the intervention increased students’ perception of value in the course material, which in turn increased interest in the field of psychology and improved academic performance in the course. 23

Keep in mind that this intervention only required students to write two short essays reflecting on the relevance of their coursework for their lives. Furthermore, this intervention worked particularly well for students with low expectancies. The authors summarize the practical application of these results:

Practically speaking, the relevance intervention is easy and inexpensive to implement, produces effects in as few as one or two trials, can be flexibly implemented during class or on the students’ own time, and is applicable to a diverse array of topics or activities. Although the participants in our research only wrote about two topics (mental math and psychology), there is no reason to assume that similar results cannot be obtained in other domains, such as history, English, or chemistry. In fact, some recent research indicates that these results also obtain in high school science and college statistics classes (Hulleman, Hendricks, & Harackiewicz, 2007; Hulleman & Harackiewicz, 2009). Importantly, the students who most often concern teachers—those who perform poorly and have low performance expectations— benefited the most from our intervention, and those with high performance expectations were not harmed by it (p. 892).

In short, EV interventions are promising strategies to help improve student motivation. This research illustrates the positive academic outcomes associated with helping students identify the applicability of their coursework to their lives outside the classroom.

Students benefit from perceiving a connection between their coursework and their lives. The implications for improving community college completion rates are two-fold. First, this research on motivation underscores the urgency of addressing structural barriers that make it difficult for students to identify and complete a course of study. When students, particularly those with low expectations, identify the relevance of their coursework to their lives, they perform better and express more interest in the subject they are studying. Yet, this is precisely the type of connection that can be difficult to make in cafeteria-style settings, where paths of study are unclear and resources are confusing to navigate. Structural reforms can help students identify and progress along clear pathways from enrollment to graduation. In this environment, it may be easier for students to identify a clear connection between their coursework and their own lives, which may have positive outcomes for their performance and interest in their program of study.

Second, EV interventions may be useful tools for instructors and perhaps advisers in a community college context. These interventions can be tailored to specific courses, and if implemented properly, EV interventions may help students perceive value in their coursework, increase their interest in the course, and improve their performance. These positive outcomes may be building blocks that help students complete the necessary courses to earn a postsecondary credential or degree. As Section 4 discusses at length, this is not to say that including EV interventions within courses can singlehandedly solve the completion problem; far from it. However, these interventions may be valuable components of a broader strategy to create a college environment conducive to student success.

Section 4: Policy recommendations

Community colleges can play a pivotal role in providing individuals with viable pathways into the American middle class, maintaining a strong workforce, and building a competitive 21 st -century economy. Delivering on this promise requires innovative solutions to increase the number of community college students who complete a postsecondary credential or degree.

Delivering on the promise of community college requires innovative policy solutions to increase the number of students who complete a postsecondary credential or degree.

The final section of this report offers several policy recommendations about how to address structural and motivational barriers to college completion. These recommendations are designed to help college leaders, employers, researchers, and policymakers identify steps to improve community college completion rates.

Among employers, these lessons may be particularly relevant for those who partner with community colleges to create career pathways (and for those who hope to do so in the future). Challenges that students face in navigating the college environment may extend to identifying and completing education and training programs developed in partnership with employers. In developing these programs with college partners, employers can benefit from the insights discussed in this report and the subsequent recommendations.

Policy recommendation 1: Address both structural and motivational barriers to community college completion.

The structural problems discussed above–such as unclear pathways from enrollment to graduation, enormous student-to-adviser ratios that make it challenging to provide one-on-one support, and a lack of easily navigable support services–are fundamental barriers to college completion. Addressing these issues should be a core component of colleges’ strategies to increase completion rates.

In providing clearer pathways from enrollment to graduation, addressing the structural barriers may help to alleviate some of the motivational barriers that students face. On a more granular level, colleges can also address motivational barriers by supporting instructors, employer partners, and advisers in implementing strategies like the EV interventions described in Section 3 that help students draw connections between their coursework and their lives. These interventions may be feasible, relatively low-cost ways to help students find value in their coursework before, during, and after a college’s transition from a cafeteria-style model to a guided pathways model. However, without addressing the structural problems inherent in a cafeteria-style model, fundamental barriers to completion remain.

The importance of implementing structural changes bears underscoring. Encouragingly, research suggests that EV interventions that are well tailored to specific courses can produce positive outcomes for students in those courses (Hullemen et al. 2010). At the same time, it is likely that addressing structural barriers is essential to realizing the full potential of EV interventions in a community college context. Take, for example, the proposition of including EV interventions in student orientation. Orientation may be optional, and students do not necessarily take advantage of advising services offered during this process (Bailey et al. 2015, p. 55). Incorporating EV interventions into orientation, while useful in theory, may have limited impact if a small share of incoming students participate. Similarly, in cafeteria-style colleges, this type of intervention may be difficult to implement effectively through advising sessions, as students may not meet regularly enough or long enough with advisers for these interventions to work. As part of a guided pathways model where students meet regularly with advisers for longer sessions, however, implementing EV interventions through advisement services may be feasible. Advisers could assign a writing prompt or incorporate into their conversations questions designed to help students think about the relevance of their coursework to their career and education goals. Overall, structural improvements could facilitate effective incorporation of EV interventions into multiple aspects of students’ community college experiences.

Policy recommendation 2: Apply lessons from proven and promising models.

There is a strong and convincing case that the cafeteria-style model impedes students from completing credentials and degrees (Scott-Clayton 2011; Holzer and Baum 2017; Bailey et al. 2015). Fortunately, evidence from the City University of New York’s (CUNY) Accelerated Study in Associate Programs (ASAP), discussed in Section 2, illustrates that interventions aligned with the guided pathways approach have resulted in dramatically higher completion rates among program participants. Community colleges interested in adopting a guided pathways model can learn from CUNY’s ASAP program. Resources include the evaluation by MDRC discussed in Section 2 of the full report (Scrivener et al. 2015) and a resource guide to ASAP produced by CUNY (Boykin and Prince 2015). Community colleges can also learn from applications of the ASAP approach in other contexts , including elsewhere in New York, California, and Ohio. Results from these evaluations can provide insight into whether and how to tailor the original ASAP program to community colleges outside the CUNY system. Drawing on this growing research base, colleges can adapt tested models of the guided pathways approach to meet their needs.

In addition, Bailey et al. (2015) offer a comprehensive review of the guided pathways approach, along with examples of ways in which different colleges have pursued these types of reforms. These recommendations can inform pilot programs designed to test new approaches that address the problems inherent in the cafeteria-style model.

Policy recommendation 3: Leverage emerging technologies to reduce structural barriers, particularly in the context of student support services.

One dimension of the guided pathways approach is improving advisement and student support services. The ASAP program, for example, reduced the student-to-adviser ratio. However, similar changes may be impractical for many community college systems given their potential cost. Fortunately, innovative research that leverages emerging technology has identified scalable, cost-effective mechanisms to provide students with targeted support.

Innovative research that leverages emerging technology, like artificial intelligence, has identified scalable, cost-effective mechanisms to provide students with targeted support.

Researchers Benjamin Castleman and Lindsay Page, in their book “Summer Melt: Supporting Low Income Students through the Transition to College,” describe the potential to use multiple strategies, including text messaging outreach, to reduce “summer melt,” wherein students who were accepted to and intend to enroll in college do not matriculate in the fall (p. 2). They offer detailed guidelines for how to adapt these lessons, which can serve as a starting point for colleges seeking strategies to strengthen their student support services.

More broadly, this work provides a foundation for leveraging emerging technologies to reach many more students than was possible before. Lindsay Page and Hunter Gehlbach recently partnered with Georgia State University and AdmitHub to reduce summer melt using Pounce, an artificial intelligence chatbot that uses individually tailored text message outreach (Page and Gehlbach 2017). They describe the positive results of the program as follows: “Students planning to go to GSU who received Pounce outreach completed their required pre-matriculation tasks and enrolled on-time at significantly higher rates than those who received GSU’s standard outreach. Pounce reduced GSU’s summer melt by 21%. These impacts mirror previous summer melt interventions but with far fewer staff.”

Thus, one viable avenue for improving student support and advisement services is through tailored outreach and support using technological innovations. While these examples focus on reducing summer melt, there is reason to be optimistic that this type of outreach may be valuable in additional areas of student support services. Indeed, Bailey et al. (2015) recommend leveraging technology to improve case management systems and provide feedback to students on their progress. To effectively incorporate “e-advising” innovations, they argue that “colleges must set in place some foundational program structures, encourage end-users to use the tools in their daily lives, and integrate human points of contact into the system” (p. 71).

Policy recommendation 4: Rigorously pilot and evaluate expectancy value interventions.

EV interventions that help students see the relevance of their coursework for their lives can have powerful effects on student outcomes, including increased interest in a subject and improved academic performance. Particularly when paired with a guided pathways approach that helps students navigate the college environment, this type of intervention may help students perceive the relevance between their coursework and their career and education goals. Community colleges, employer partners, and research partners should design evidence-based pilot programs and rigorously evaluate their outcomes to identify the most effective strategies to incorporate EV interventions in students’ community college experiences.

Correct implementation of EV interventions is important to obtain the desired results. Recent research on a different type of intervention delivered in an education setting suggests that the level of fidelity matters for student outcomes (Horowitz et al. 2018). Thus, pilot programs should include adequate training for those who implement the intervention. Ideally, these pilot programs will use a randomized control trial approach to assess the effect of these interventions on relevant student outcomes (see Scrivener and Coghlan 2011 for a brief discussion of the benefits of randomization), such as academic performance, interest in the course and the relevant field of study, persistence, and graduation rates.

In addition, colleges, employer partners, and research partners should consider the following questions when developing pilot programs:

How frequently should instructors include EV assignments in their coursework?
How long should EV writing assignments be such that they are effective without being overly burdensome in terms of students’ and instructors’ time?
How can instructors tailor these assignments to the particular courses they are responsible for?
What kind of resources, initial training, and/or ongoing professional development is necessary to support instructors in effectively implementing these interventions?

With appropriate attention to program design, professional development, evaluation, and improvement, community colleges could incorporate these interventions into different aspects of students’ experiences, including the following:

Regular coursework
Employer content delivered as part of public-private partnerships/career pathways
Developmental coursework
Student success courses
Initial orientation and intake sessions
Advising sessions

Policy recommendation 5: Evaluate innovative efforts and disseminate information widely.

In addition to knowing how well a program works , college leaders, employer partners, and policymakers need answers to practical questions about how to implement effective programs . Embedding the following types of questions into evaluations of guided pathways reforms would provide valuable insight into how to implement these changes:

What motivated college leaders to implement these changes?
How did college leaders convince stakeholders to invest in a guided pathways approach?
How were programs tailored to local contexts without sacrificing the key components?
What resources were necessary to design and implement these changes?

Furthermore, employers that partner with colleges to provide training and education programs related to career pathways may be interested in drawing on the guided pathways approach in their work with community colleges. Questions that may serve as a starting point for exploring these possibilities include:

What challenges do existing programs face in terms of retention and completion?
How can employers and colleges incorporate a guided pathways approach in their career pathways programs to address these issues? What adjustments to existing programs are necessary?
What steps can the college and employers take to jointly navigate these changes productively?

Research-practice partnerships may be particularly valuable in developing, implementing, and evaluating guided pathways reforms. Programs designed without practitioners’ input are unlikely to meet the needs of a college’s student population, and researchers can help to ensure that the program design draws on available evidence and allows for rigorous evaluation and policy learning. Ultimately, partnerships between researchers, colleges, and employers can lead to programs based on a deep understanding of the student population, the college system, and the local labor market. Working together, these partners can develop evidence-based innovations, continuously improve their programs, and share their results with the wider research, policy, and education communities. This work can help create a shared road map to reducing barriers to community college completion.

Improving community college completion rates should be a top priority for policymakers at all levels of government, employers, community colleges, and the philanthropic community. Earning a postsecondary credential or degree provides a gateway to higher average earnings and opens up career pathways for graduates, while higher completion rates help strengthen the American workforce. Yet, far too many students who enroll in community college do not complete a degree. While not the subject of this analysis, academic and financial barriers loom large for many students. Ongoing efforts to improve the quality of public K-12 education and to make college more affordable are essential to improving completion rates.

Improving community college completion rates should be a top priority for policymakers at all levels of government, employers, community colleges, and the philanthropic community.

As discussed at length in this report, improving completion rates also requires addressing structural and motivational barriers that students face. Transitioning from the cafeteria-style model to a guided pathways approach is a promising strategy to reduce structural barriers. These structural changes can pave the way for helping students overcome motivational barriers (Oyserman and Lewis 2017). In addition, through EV interventions, instructors can help students perceive the relevance of their coursework for their lives.

Community colleges hold enormous potential for students across the United States. Realizing this full potential is vital for students to obtain the education and training they need to pursue their career goals, obtain good-paying jobs, and contribute to a vibrant American economy. Moving forward, we must invest in innovative, evidence-based solutions, enable students to complete postsecondary credentials and degrees, and ultimately help them achieve their academic and career goals.

Acknowledgements

The author is deeply grateful to Elizabeth Bell and Neil Lewis for their invaluable feedback on this project and to Hana Dai, Bethany Kirkpatrick, Francesca Royal, and Kim Truong for their excellent research assistance. This project was made possible through the generous support of the Arconic Foundation.

Brookings recognizes that the value it provides is in its absolute commitment to quality, independence, and impact. Activities supported by its donors reflect this commitment.

Aelenei, Cristina, Neil A. Lewis Jr, and Daphna Oyserman. “No pain no gain? Social demographic correlates and identity consequences of interpreting experienced difficulty as importance.” Contemporary Educational Psychology 48 (2017): 43-55. https://doi.org/10.1016/j.cedpsych.2016.08.004 .

“American Association of Community Colleges Fast Facts 2018.” American Association of Community Colleges . Last modified March 2018. https://www.aacc.nche.edu/wp-content/uploads/2018/04/2018-Fast-Facts.pdf .

American Institutes for Research. “AIR Research Fuels ‘Launch My Career’, a New Website to Help Students Envision Return on Investment for Higher Education Choices.” AIR News Releases, Last modified: 9 June 2016. https://www.air.org/news/press-release/air-research-fuels-launch-my-career-new-website-help-students-envision-return.

Bahr, Peter R. “The Earnings of Community College Graduates in California.” Center for Analysis of Postsecondary Education and Employment, Teachers College, Columbia University. New York, NY: (December 2016). https://capseecenter.org/wp-content/uploads/2016/12/the-earnings-of-community-college-graduates-in-california.pdf.

Bailey, Martha J., and Susan M. Dynarski. “Inequality in Postsecondary Education” in Whither opportunity?: Rising inequality, schools, and children’s life edited by Greg Duncan and Richard Murnane, 117-131. New York: Russell Sage Foundation, 2011.

Bailey, Thomas, Shanna Smith Jaggars, and Davis Jenkins. Redesigning Americas Community Colleges: A Clearer Path to Student Success . Cambridge, MA: Harvard University Press, 2015.

Baker, Rachel, Eric Bettinger, Brian Jacob, and Ioana Marinescu. “The Effect of Labor Market Information on Community College Students’ Major Choice.” Economics of Education Review 65 (June 5, 2018): 18-30. doi:10.1016/j.econedurev.2018.05.005.

Belfield, Clive, and Thomas Bailey. “The Labor Market Returns to Sub-Baccalaureate College: A Review.” Center for Analysis of Postsecondary Education and Employment, Teachers College, Columbia University. New York, NY: (2017).

Billings, Meredith. “Understanding the design of college promise programs, and where to go from here.” Brown Center Chalkboard, Brookings Institution, September 17, 2018, https://www.brookings.edu/blog/brown-center-chalkboard/2018/09/18/understanding-the-design-of-college-promise-programs-and-where-to-go-from-here/ .

Boykin, Daniela, and Amy Prince. “Inside ASAP: A Resource Guide on Program Structure, Components, and Management.” Office of Academic Affairs, City University of New York, 2015. http://www1.cuny.edu/sites/asap/wp-content/uploads/sites/8/2015/05/Resource-Guide.pdf.

Bradley, David H. “The Workforce Innovation and Opportunity Act and the One-Stop Delivery System.” Congressional Research Service. Washington, D.C.: 2015. https://fas.org/sgp/crs/misc/R44252.pdf .

Bureau of Labor Statistics. “Occupational Outlook Handbook: Occupation Finder.” Last modified April 2018. https://www.bls.gov/ooh/occupation-finder.htm.

Bureau of Labor Statistics. “Occupational Employment Statistics: May 2017 State Occupational Employment and Wage Estimates.” Last modified March 2018. https://www.bls.gov/oes/current/oessrcst.htm.

Carnevale, Anthony P. “Two Key Actions to Align Postsecondary Education with the Labor Market.” In Governing to Win: Enhancing National Competitiveness Through New Policy and Operating Approaches, edited by Charles L. Prow, 59-64. Washington, D.C.: Rowman & Littlefield Publishers, 2012.

Carnevale, Anthony P., Jeff Strohl, Ban Cheah, and Neil Ridley. “Good Jobs That Pay without a BA.” Georgetown University Center on Education and the Workforce, 2017. http://hdl.handle.net/10822/1047863 .

Castleman, Benjamin L., and Lindsay C. Page. Summer Melt: Supporting Low-Income Students through the Transition to College . Cambridge, MA: Harvard Education Press, 2014.

Cellini, Stephanie R., and Nicholas Turner. “Gainfully employed? Assessing the employment and earnings of for-profit college students using administrative data.” National Bureau of Economic Research (NBER) no. w. 22287, Revised January 2018. DOI: 10.3368/jhr.54.2.1016.8302R1 .

Chetty, Raj, John N. Friedman, Emmanuel Saez, Nicholas Turner, and Danny Yagan. “Mobility Report Cards: The Role of Colleges in Intergenerational Mobility.” National Bureau of Economic Research (NBER) no. w. 23618, 2017.

The City University of New York. “ASAP: Replication.” http://www1.cuny.edu/sites/asap/replication/.

Cohen, Patricia. “A rising call to promote STEM education and cut liberal arts funding.” The New York Times . February 21, 2016.

Community College Research Center at Columbia University Teachers College. “Community College FAQs.” 2018. https://ccrc.tc.columbia.edu/Community-College-FAQs.html.

Deil-Amen, Regina, and James E. Rosenbaum. “The Social Prerequisites of Success: Can College Structure Reduce the Need for Social Know-How?” The ANNALS of the American Academy of Political and Social Science 586, no. 1 (March 1, 2003): 120-43. doi:10.1177/0002716202250216.

Deming, David and Susan Dynarski. “Into College, out of Poverty? Policies to Increase the Postsecondary Attainment of the Poor.” National Bureau of Education Research (NBER) no. w. 15387, 2009.

Duncan, Arne and John Bridgeland. “Free college for all will power our 21st-century economy and empower our democracy.” Brown Center Chalkboard, Brookings Institution, September 17, 2018.

Dynarski, Susan M. “Does aid matter? Measuring the effect of student aid on college attendance and completion.” American Economic Review 93, no. 1 (2003): 279-288. DOI: 10.1257/000282803321455287.

Goldrick-Rab, Sara. Paying the Price: College Costs, Financial Aid, and the Betrayal of the American Dream . Chicago: University of Chicago Press, 2016.

Harackiewicz, Judith M., Elizabeth A. Canning, Yoi Tibbetts, Stacy J. Priniski, and Janet S. Hyde. “Closing Achievement Gaps with a Utility-value Intervention: Disentangling Race and Social Class.” Journal of Personality and Social Psychology 111, no. 5 (November 2, 2015): 745-65. doi:10.1037/pspp0000075.

Holzer, Harry J. and Sandy Baum. Making College Work: Pathways to Success for Disadvantaged Students . Washington, D.C.: Brookings Institution Press, 2017.

Horn, Laura, and Paul Skomsvold. “Community College Student Outcomes: 1994–2009.” NCES 2012253. Washington, D.C.: National Center for Education Statistics, 2011.

Horowitz, Eric, Nicholas Sorensen, Nicholas Yoder, and Daphna Oyserman. “Teachers can do it: scalable identity-based motivation intervention in the classroom.” Contemporary Educational Psychology 54 (2018): 12-28. https://doi.org/10.1016/j.cedpsych.2018.04.004 .

Hulleman, Chris S., Kenneth E. Barron, Jeff J. Kosovich, and Rory A. Lazowski. “Student Motivation: Current Theories, Constructs, and Interventions Within an Expectancy-Value Framework.” In Psychosocial Skills and School Systems in the 21st Century: Theory, Research, and Practice , 241-78. Switzerland: Springer International Publishing, 2016.

Hulleman, Chris S., Jeff J. Kosovich, Kenneth E. Barron, and David B. Daniel. “Making connections: Replicating and extending the utility value intervention in the classroom.” Journal of Educational Psychology 109, no. 3 (2017): 387.

Hulleman, C. S., and J. M. Harackiewicz. “Promoting Interest and Performance in High School Science Classes.” Science 326, no. 5958 (July 03, 2009): 1410-412. doi:10.1126/science.1177067.

Hulleman, Chris S., Olga Godes, Bryan L. Hendricks, and Judith M. Harackiewicz. “Enhancing Interest and Performance With a Utility Value Intervention.” Journal of Educational Psychology 102, no. 4 (2010): 880-95. doi:10.1037/a0019506.

Hurwitz, Michael, and Jonathan Smith. “Student Responsiveness To Earnings Data In The College Scorecard.” Economic Inquiry 56, no. 2 (April 2018): 1220-243. doi:10.1111/ecin.12530.

Jacob, Brian A. “Building Knowledge to Improve Degree Completion in Community Colleges.” Washington, D.C.: Brookings Institution, 2018.

Jepsen, Christopher, Kenneth Troske, and Paul Coomes. “The labor market returns to community college degrees, diplomas and certificates.” Journal of Labor Economics 32, no. 1 (2014), 95–121. Doi:10.1086/671809.

Jenkins, Davis, and John Fink. “Tracking Transfer: New Measures of Institutional and State Effectiveness in Helping Community College Students Attain Bachelor’s Degrees.” Community College Research Center, Teachers College, Columbia University. New York, NY: 2016.

Johnson, Jean, and Jon Rochkind. “With Their Whole Lives Ahead of Them: Myths and Realities about Why So Many Students Fail to Finish College.” New York, NY: Public Agenda, 2009.

Jones, Tiffany and Katie Berger. “A Promise Fulfilled: A Framework for Equitable Free College Programs.” The Education Trust, 2018.

Karp, Melinda M. “Toward a New Understanding of Non-Academic Student Support: Four Mechanisms Encouraging Positive Student Outcomes in the Community College.” Community College Research Center, Teachers College, Columbia University. New York, NY: 2011.

Lazowski, Rory A., and Chris S. Hulleman. “Motivation Interventions in Education.” Review of Educational Research 86, no. 2 (June 1, 2016): 602-40. doi:10.3102/0034654315617832.

Liu, Vivian Y.T., Clive Belfield, Madeline J. Trimble. “The medium-term labor market returns to community college awards: Evidence from North Carolina.” Economics of Education Review 44 (2014): 42-55. DOI: 10.1016/j.econedurev.2014.10.009.

National Center for Education Statistics. “College Navigator.” Accessed September, 2018. https://nces.ed.gov/collegenavigator/ .

The National Center for Public Policy and Higher Education. “Affordability and Transfer: Critical to Increasing Baccalaureate Degree Completion.” Last modified June 2011. http://www.highereducation.org/reports/pa_at/index.shtml .

Oyserman, Daphna. “Not Just Any Path: Implications of Identity-based Motivation for Disparities in School Outcomes.” Economics of Education Review 33(C) (September 2012): 179-90. doi:10.1016/j.econedurev.2012.09.002.

Oyserman, Daphna and Neil A. Lewis. “Seeing the destination AND the path: Using identity‐based motivation to understand and reduce racial disparities in academic achievement.” Social Issues and Policy Review 11, no. 1 (2017): 159-194. doi: 10.1111/sipr.12030.

Page, Lindsay C. and Hunter Gehlbach. “How an artificially intelligent virtual assistant helps students navigate the road to college.” AERA Open 3, no. 4 (2017): doi:10.1177/2332858417749220 .

Page, Linsey C. and Hunter Gehlbach. “How Georgia State Used an Algorithm to Help Students Navigate the Road to College.” Harvard Business Review, January 2018. https://hbr.org/2018/01/how-georgia-state-university-used-an-algorithm-to-help-students-navigate-the-road-to-college.

Poutré, Alain and Mamie Voight. “Do Tennessee Promise and New York’s Excelsior Scholarship Help Students with Limited Means Afford College?” The Institute for Higher Education Policy, 2018.

Robbins, Rich. Implications of Advisor Load . Monograph No. 25. National Academic Advising Association, Kansas State University. Manhattan, KS: NACADA, 2013.

Rozek, Christopher S., Janet S. Hyde, Ryan C. Svoboda, Chris S. Hulleman, and Judith M. Harackiewicz. “Gender Differences in the Effects of a Utility-value Intervention to Help Parents Motivate Adolescents in Mathematics and Science.” Journal of Educational Psychology 107, no. 1 (June 2, 2014): 195-206. doi:10.1037/a0036981.

Scott-Clayton, Judith. The Shapeless River: Does a Lack of Structure Inhibit Students’ Progress at Community Colleges? Community College Research Center, Teachers College, Columbia University. New York, NY: 2011.

Scrivener, Susan, Michael J. Weiss, Alyssa Ratledge, Timothy Rudd, Colleen Sommo, and Hannah Fresques. “Doubling Graduation Rates: Three-Year Effects of CUNY’s Accelerated Study in Associate Programs (ASAP) for Developmental Education Students.” MDRC. New York, NY: 2015.

Scrivener, Susan and Erin Coghlan. “Opening Doors to Student Success: A Synthesis of Findings from an Evaluation at Six Community Colleges.” MDRC. New York, NY: 2011.

Skillful: A Markle Initiative. 2018. http://skillful.com/about .

United States Department of Education. “College Scorecard.” Accessed September, 2018. https://collegescorecard.ed.gov/.

West, Darrell M. The Future of Work: Robots, AI, and Automation . Washington, D.C.: Brookings Institution Press, 2018.

The Brookings Institution is a nonprofit organization devoted to independent research and policy solutions. Its mission is to conduct high-quality, independent research and, based on that research, to provide innovative, practical recommendations for policymakers and the public. The conclusions and recommendations of any Brookings publication are solely those of its author(s), and do not reflect the views of the Institution, its management, or its other scholars.

For example, financial aid plays an important role in college enrollment and persistence (Dynarski 2003; Deming and Dynarski 2009). Recently, momentum for “free college” seems to be growing, from Sen. Bernie Sanders’s rallying cry during his campaign for the presidency in 2016 to more recent, bipartisan advocacy . While these programs hold promise for increasing access to college among low-income students, recent reports caution that in order to serve the students with the most financial need, promise programs and free college programs should be designed using an equity framework (Jones and Berger 2018; Poutré and Voight 2018). With respect to free college programs, to produce the desired outcomes, policymakers should pay close attention to program design.
Information in this paragraph is from the American Association of Community College Fast Facts.
National Center for Public Policy and Higher Education
From the Community College Research Center at Columbia University Teachers College “Community College FAQs.”
See Figure 6.3, p. 121.
These estimates are based on analyses of separate data sets compiled for eight states. The data sets are “large-scale, longitudinal student transcripts merged with individual quarterly earnings records” that allow for the inclusion of individual fixed effects when estimating returns to education (Belfield and Bailey 2017, p. 4).
This data set merges administrative data from the state of Florida with quarterly earnings data (p. 73).
See Figure 3-3a, p. 77.
See Figure 3-3b, p. 76.
Some point to these types of differences in returns across disciplines as justification for shifting public funding for higher education away from the humanities and toward STEM fields. Careful consideration of such proposals is important but is outside the scope of this report. It is worth noting, however, Patricia Cohen’s astute observation with respect to these policy proposals: “Informing students better is one thing. Penalizing certain majors in the form of reduced funding is another.”
See Figure 3-3a, p. 76.
Compared to only earning a high school diploma, higher average earnings are associated with completing some course credits without finishing a degree. However, these gains are relatively small according to both Belfield and Bailey (2017) and Holzer and Baum (2017). Belfield and Bailey (2017) find the following: “For North Carolina, Liu et al. (2015) estimate that each credit accumulated by male (female) community college students is associated with gains of $17 ($29) per quarter (0.4 [0.7] percent higher earnings). For Kentucky, Jepsen et al. (2014, Table 7) estimate returns per credit of $9 ($18). However, for California, Bahr (2016, Table 3, Model 3) identifies mixed returns per credit and finds that these returns to credits vary by field of study” (p. 10). In their analysis of Florida, Holzer and Baum (2017) similarly find small average returns for noncompleters. They calculate estimated earnings by multiplying the estimated return per credit hour by the average number of credits completed by those who do not complete a credential or degree, and they find that “average community college noncompleters earn just a 4 percent return over what they would earn with a high school diploma” (p. 81).
For 29 percent of these (14 jobs), the growth rate is predicted to be as fast as average. For another 40 percent of these (19 jobs), the growth rate is supposed to be faster than average.
30 percent of these (14 jobs) are predicted to grow as fast as average, and another 48 percent (22 jobs) are predicted to grow faster than average.
Under the Workforce Innovation and Opportunity Act (WIOA), states receive funding from the federal government to operate “One-Stop” centers that provide a “central point of service for those seeking employment, training, and related services” (Bradley 2015, p. 6). There are approximately 3,000 of these centers nationwide, and they are one avenue through which prospective or current students may learn about job opportunities and training pathways (Bradley 2015). Unlike the websites discussed in the text, these centers do not only provide information via online publications, but offer more comprehensive services as well.
For a description of Launch My Career websites, see: https://www.air.org/news/press-release/air-research-fuels-launch-my-career-new-website-help-students-envision-return . See more on the Skillful initiative here: http://skillful.com/about.
Additional federal resources include the Bureau of Labor Statistics (BLS) in the Department of Labor, which publishes occupational employment statistics for each state. Another resource, the Occupational Outlook Handbook, includes an occupation finder that provides information by occupation on entry-level education, on-the-job training, projected number of new jobs, projected growth rate, and median pay.
For a full description of the guided pathways model, please see “Redesigning America’s Community Colleges” by Thomas Bailey, Shanna Jaggars, and Davis Jenkins.
See Table 3.3 and discussion on pages 27-28 of MDRC report.
Scrivener et al. (2015) report that over three years, “On an annual basis, [the ASAP program] increased the investment in the average program group member by roughly $5,428” (p. 71). However, they also find that the ASAP program lowers the “cost per degree earned within three years” (p. 71). Students in the ASAP program were 83.9 percent more likely to earn a degree, which means that “[e]ven though ASAP spent more money overall, this estimated effect actually lowered the cost per degree earned for ASAP students by 11.4 percent compared with students who receive the usual college services” (p. 71).
Notably, other types of motivation theories are also highly relevant to community college students’ experiences, such as identity-based motivation (Aelenei et al. 2017). These theories and related interventions that have the potential to improve student outcomes should be considered along with the interventions discussed here.
See discussion p. 78-80 and Figure 9, p. 80 in “Paying the Price . ”
See p. 890-891 in Hullemen et al. (2010) for a discussion of these results.

Higher Education

Governance Studies

Brown Center on Education Policy

Dr. Neil A. Lewis, Jr.

May 14, 2024

Katharine Meyer

May 7, 2024

Jamie Klinenberg, Jon Valant, Nicolas Zerbino

Selected Higher Education Statistics – 2022 Student data
Selected Higher Education Statistics – 2021 Student data
Selected Higher Education Statistics – 2020 Student data
Selected Higher Education Statistics – 2019 Student data
Selected Higher Education Statistics – 2018 Student data
Selected Higher Education Statistics – 2017 Student data
Selected Higher Education Statistics – 2016 Student data
Selected Higher Education Statistics – 2015 Student data
Selected Higher Education Statistics – 2014 Student Data
Selected Higher Education Statistics – 2013 Student Data
Selected Higher Education Statistics – 2012 Student Data
Selected Higher Education Statistics – 2011 Student Data
Selected Higher Education Statistics – 2010 Student Data
Selected Higher Education Statistics – 2009 Student Data
Selected Higher Education Statistics – 2008 Student Data
Selected Higher Education Statistics – 2007 Student Data
Selected Higher Education Statistics – 2006 Student Data
Selected Higher Education Statistics – 2005 Student Data
Selected Higher Education Statistics – 2004 Student Data
Selected Higher Education Statistics – Time Series Data
Selected Higher Education Statistics – 2023 Staff data
Selected Higher Education Statistics – 2022 Staff data
Selected Higher Education Statistics – 2021 Staff data
Selected Higher Education Statistics – 2020 Staff data
Selected Higher Education Statistics – 2019 Staff data
Selected Higher Education Statistics – 2018 Staff data
Selected Higher Education Statistics – 2017 Staff data
Selected Higher Education Statistics – 2016 Staff data
Selected Higher Education Statistics – 2015 Staff data
Selected Higher Education Statistics - 2014 Staff data
Selected Higher Education Statistics – 2013 Staff Data
Selected Higher Education Statistics – 2012 Staff Data
Selected Higher Education Statistics – 2011 Staff Data
Selected Higher Education Statistics – 2010 Staff Data
Selected Higher Education Statistics – 2009 Staff Data
Selected Higher Education Statistics – 2008 Staff Data
Selected Higher Education Statistics – 1997 to 2007 Staff Data
Data Requests, Data Protocols and Data Privacy and Visual Analytics Guide
uCube – Higher Education Data Cube
Undergraduate Applications Offers and Acceptances Publications
Upholding Quality - Quality Indicators for Learning and Teaching

Completion Rates - Cohort Analyses

Cohort analysis reports present the results of tracking commencing higher education student outcomes over time. That is, after nominated periods, how many commencing students have completed their studies, how many are still studying, and how many have not completed an award course. Cohort analysis publications also present outcomes of students by selected characteristics.

Summaries and Tables

Completion Rates of Higher Education Students – Cohort Analysis, 2005-2022
Completion Rates of Higher Education Students - Cohort Analysis, 2005-2021
Completion Rates of Higher Education Students – Cohort Analysis, 2005-2020
Completion Rates of Higher Education Students - Cohort Analysis, 2005-2019
Completion Rates of Higher Education Students - Cohort Analysis, 2005-2018
Completion Rates of Higher Education Students - Cohort Analysis, 2005-2017
Completion Rates of Higher Degree Research Students – Cohort Analysis, 2007-2017
Key findings from the 'Completions Rates of Higher Education Students - Cohort Analysis, 2005-2015' report
Completion Rates of Higher Education Students - Cohort Analysis, 2005-2015
Completion rates of higher education students: Cohort analysis, 2005-2014
Completion rates of domestic bachelor students 2005 – 2013: A cohort analysis
Completion rates of domestic bachelor students 2005 – 2012: A cohort analysis

Today’s Students Caret Left
Today’s Institutions Caret Left
Today’s Credentials Caret Left
Ending Racial Injustice Caret Left
Today's Student at a Glance
Students of Color
Financially Stressed Students
First-Generation Students
Working Adults
Community Colleges
Public Universities
Minority-Serving Institutions
College Affordability
Bachelor’s Degrees
Associate Degrees
Workforce Training
Quality in Higher Learning
Disparities in Higher Learning
Our Journey Toward Racial Justice
How to Talk about Education and Race
Education Levels Nationally Caret Left
Public Policy
Lumina Data & Research
Lumina Impact Ventures Caret Left
A Stronger Nation Data Report
Investment Inquiry
Lumina News and Blog
Focus Magazine
Photojournalism
Daily Headlines
State Strategy Labs Caret Left
Student Supports Caret Left
Local Talent Handbook
Apply for Technical Assistance
Technical Assistance FAQ
About Strategy Labs
Beyond Financial Aid
History of Federal Aid
Borrowers of Color
Tuition Reimbursement
I Want That Job!
Our People Caret Left
Strategic Plan
Grants Caret Left
Contact Us Caret Left
Civic Leadership
Lumina Disclosures
Board of Directors
President and CEO
Our Stories
Grantee FAQ
Grants Database
Your Experience with Lumina’s Grant Process
Media Outreach
Careers at Lumina
Creative Commons
Brand Guidelines

The State of Higher Education 2024

A valuable, but obstructed path to great jobs and lives.

The 2024 State of Higher Education study includes responses from over 14,000 U.S. adults aged 18 to 59 without a college degree

Since 2020, Lumina Foundation and Gallup have partnered to produce the State of Higher Education study, an annual survey of thousands of U.S. adults without a college degree. Each year, Lumina and Gallup measure their attitudes toward education beyond high school, interest in pursuing a degree or credential, the barriers they face to enrolling or completion, and the experiences currently enrolled students report having in their programs.

Critical findings from the 2024 State of Higher Education study include:

Nearly all adults without a college degree say at least one type of credential is “extremely” or “very” valuable.
Career outcomes — such as earning a raise, promotion or a more fulfilling role — are the primary motivators for pursuing higher education.
Meanwhile, the primary barriers to enrollment are cost and a lack of financial aid.
More than one in three currently enrolled students have considered stopping out of their degree or credential program within the last six months.
About one in six currently enrolled students say they have at least occasionally felt disrespected, discriminated against or unsafe in their program.

Related Resources

Insights to Support Adult Learners at Community Colleges
Fewer Young Men Are in College, Especially at Four-Year Schools
Ending Unfunded Mandates in Higher Education

Higher Education Today
American Council On Education
Carnegie Classifications
Race and Ethnicity In Higher Education

Mental Health Supports Are Key in in Community College Student Success

May 13, 2024

Title: Challenges and Opportunity: An Examination of Barriers to Postsecondary Academic Success

Authors: Sade Bonilla and Veronica Minaya

Source: Annenberg Institute at Brown University

Community colleges are a critical component of the U.S. higher education system, providing access to students from traditionally underserved communities. While they are a vital entry point to postsecondary education, community colleges face lower levels of completion compared to four-year institutions, stemming from educational, economic, and social inequities that persist among low-income and first-generation students.

The Annenberg Institute launched a descriptive study to examine the relationship between academic success and students’ time utilization, campus engagement, financial and mental well-being. The study focuses on community college students and finds that a nationwide increase in adverse mental health among college students intersects with high rates of food and housing insecurity.

The report emphasizes the significance of mental health factors as a barrier to academic attainment and suggests that facilitating access to mental health supports is a prominent avenue for supporting student engagement and success.

The study investigates how academic support can tackle adverse student mental health and increase engagement. It argues that students and institutions are equally responsible for students’ persistence despite hardships. This theory rests on the idea that the level of learning and effort linked to academic success is directly tied to a sense of community and social integration within a campus. Accordingly, community colleges with more student activities often find higher levels of student success.

Additionally, the study examines how financial resources influence student success. Food, housing, and financial insecurity remain pervasive issues that affect first-generation, low-income students’ ability to complete college. Nearly 40 percent of community college students were low-income in 2020, and 31 percent of these students were living at or below the federal poverty line.

Not only does students’ inability to access basic needs harm their physical health and well-being, but it also has negative implications for their academic performance. The need to prioritize basic needs leads to a limited bandwidth for students to focus on academics. The study argues that integrating basic needs resources for students is part of combating student mental health.

To read the full report and access more information, please click here .

—Eliza Gonzalez

If you have any questions or comments about this blog post, please contact us .

Keep Reading

Discovering Hidden Barriers to Community College Enrollment and Success

When students apply to a community college, they expect enrollment to be a clear, orderly process. Unfortunately, for far too many students, it feels more like a maze, full of unknown steps, unfamiliar terminology and unexpected delays. David Bevevino of the Education Advisory Board looks at how community college student onboarding is like a game of Chutes and Ladders—and what institutions can do to change it.

It Begins with Difficult Conversations: How Community College Leaders Can Support Faculty-led Student Success Efforts

If community colleges want to make a difference in student persistence and attainment—and to make the improvement sustainable over time—these efforts must be led by those on campus who have the most frequent contact with students: the faculty. Carrie B. Kisker, a director of the Center for the Study of Community Colleges, looks at where to begin.

The Rise of Mental Health on College Campuses: Protecting the Emotional Health of Our Nation’s College Students

Nance Roy, chief clinical officer of the Jed Foundation, provides action steps to support emotional well-being and reduce suicide and substance abuse for college students.

Amid a crisis of public college enrollment, tuition money sees biggest plunge in 40 years

Campuses emptied during the pandemic—and students have returned in far lesser numbers.

Tuition at public universities fell sharply last year amid higher education’s ongoing enrollment crisis—but, in a silver lining, support from states and federal stimulus funds has so far kept institutions afloat.

Net tuition and fee revenue—the amount institutions take in from students, net of financial aid—fell 3.3% from 2022, to an average of $7,353 per full-time student, according to a report from the State Higher Education Executive Officers’ Association.

Although tuition has been falling in recent years, “this is the largest decline going back to 1980,” said SHEEO policy analyst Kelsey Kunkle.

The post-pandemic era has marked a sharp reversal of a decades-long trend in which public colleges became ever-more reliant on student tuition as states pulled back on their support. In 1980, when SHEEO started tracking this data, student tuition and fees, on average, made up 20% of a public university’s revenue, Kunkle told Fortune. By 2013, public colleges relied on tuition and fees for nearly they made up nearly half — 47% — of their operating costs,o and byin 2018, average tuition rose to a high of $8,117, according to SHEEO.

The precise breakdown of universities’ public-to-tuition funding varies dramatically from state to state, but over the past four decades, public colleges have grown increasingly reliant on student funds. States have gradually reduced the amount of funding they give universities, pulling back especially after the Great Recession even as the costs of attendance rose—and, with each incoming class bigger than the last and students willing to take out loans, institutions were under little pressure to cater to the cost-conscious.

That trend might have kept going if not for the pandemic, which shuttered campuses nationwide and sent millions of students to learn remotely. Unlike typical economic downturns, the COVID-19 recession didn’t boost college enrollment; instead, between 2020 and 2021, enrollment fell by 3.3%—the largest drop in students since 1980. It has kept falling since. The number of students enrolled in public colleges and universities now stands at 10.2 million in 2023, down 12% from its 2011 peak.

So far, universities, wary of their falling enrollment numbers and increased scrutiny over college affordability, have been loath to hike tuition.“There’s pressure to make tuition consistent, and many public institutions are unable to raise tuition or increase enrollment,” Kunkle said.

Many have also been restrained by state officials trying to keep education costs down. Colorado’s recently passed budget limits tuition increases to 3% for in-state students , while colleges in Connecticut, Maine , and Texas have recently expanded free-tuition programs for residents.

“States, systems, and individual institutions are noticing and wanting to make efforts to address college affordability, and the ways they’re doing that is trying to keep tuition flat or keep it to minimum growth, below inflation,” Kunkle said.

For now, falling tuition has been offset by increased state funding, some of which has been boosted by federal pandemic aid. When state funding and tuition is combined, revenue per full-time student reached a record $18,301 last year, according to SHEEO.

“We hope states prioritize higher education in the way they have been,” Kunkle said, “but with upcoming budget strains, they’re going to face really hard choices.”

Latest in Personal Finance

Nearly one-fifth of Americans have ‘maxed out’ their credit cards as inflation and high interest rates push delinquencies to 3-year high

Stocks continue to trade around the world outside of U.S. market hours.

24/7 trading is already in full swing. Now Wall Street is considering ditching the opening and closing bells altogether

Gen Z worker looking worried at he computer

Gen Z really do have it worse: Those in their early 20s are earning less and have more debt than millennials did at their age

people moving in and out of a tube station

U.K. labor market cools but business confidence returns. Are high interest rates paying off?

Portrait of senior couple in front of suburban home

A record 4.1 million Americans may retire this year: Financial planners say they should take these 5 steps

Cancer survivor says embryos with an ex-husband are her last chance to have a baby. Ex-husband wants to avoid procreating against his will

The collapsed Baltimore bridge will be demolished soon, and the crew of the ship that’s trapped underneath will be onboard when the explosives go off

The housing crisis in the U.S. is flipped upside down in Japan, where each home that’s occupied could be next to an empty one by 2033

Meet the boomers who’d rather spend $100k to renovate their homes than risk the frozen housing market: ‘It would be too hard to purchase anything else’

Hedge fund billionaire Ken Griffin says college protests are the result of a ‘cultural revolution’ and Harvard should ’embrace our Western values’

TV chef Gordon Ramsay spends an extra $7.6 million on staff as U.K. restaurant empire losses triple

Consumers were deprived of rare bourbons, including Pappy Van Winkle’s 23-year-old whiskey, by alcohol overseers

Using interpretable machine learning approaches to predict and provide explanations for student completion of remedial mathematics

Published: 10 May 2024

Cite this article

Thomas Mgonja ORCID: orcid.org/0000-0002-3099-4048 1

29 Accesses

Explore all metrics

The successful completion of remedial mathematics is widely recognized as a crucial factor for college success. However, there is considerable concern and ongoing debate surrounding the low completion rates observed in remedial mathematics courses across various parts of the world. This study applies explainable artificial intelligence (XAI) tools to interpret predictions on whether students will complete mathematics remediation. Various machine learning models are compared, with random forest emerging as superior in predicting non-completion. Global interpretations using correlation analysis, logistic regression, feature importance, permutation importance, and SHapley Additive exPlanations (SHAP) summary plots identify significant predictors such as college grade point average (G.P.A), high school G.P.A, starting point in the remedial sequence, number of failed remedial courses, delay in remediation, Rate My Professor scores, and age. Additionally, local interpretations using Local Interpretable Model-Agnostic Explanations (LIME) and Diverse Counterfactual Explanations (DiCE) analyses were utilized to garner tailored advise for at-risk students. It was observed that instructor attributes cannot be overlooked, especially, when exploring local interpretations. Future research should consider other features such as a students’ socio-economic status (SES), employment status, and placement exam scores. Future studies could also involve data from multiple institutions and examine user experience in implementing these models.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price includes VAT (Russian Federation)

Instant access to the full article PDF.

Rent this article via DeepDyve

Institutional subscriptions

Data availability

The dataset used in this study is available from the corresponding author upon request and provided Institutional Review Board (IRB) approves it.

Alyahyan, E., & Düştegör, D. (2020). Predicting academic success in higher education: Literature review and best practices. International Journal of Educational Technology in Higher Education, 17 , 1–21.

Article Google Scholar

Bailey, T. (2009). Challenge and opportunity: Rethinking the role and function of developmental education in community college. New Directions for Community Colleges, 2009 (145), 11–30.

Bailey, T., Jeong, D. W., & Cho, S. W. (2010). Referral, enrollment, and completion in developmental education sequences in community colleges. Economics of Education Review, 29 (2), 255–270.

Baranyi, M., Nagy, M., & Molontay, R. (2020). Interpretable deep learning for university dropout prediction. Proceedings of the 21st Annual Conference on Information Technology Education, 13–19.

Benner, A. D., & Graham, S. (2011). Latino adolescents’ experiences of discrimination across the first 2 years of high school: Correlates and influences on educational outcomes. Child Development, 82 (2), 508–519.

Bettinger, E. P., & Long, B. T. (2010). Does cheaper mean better? The impact of using adjunct instructors on student outcomes. The Review of Economics and Statistics, 92 (3), 598–613.

Bristol, T. J., & Martin-Fernandez, J. (2019). The added value of Latinx and Black teachers for Latinx and Black students: Implications for policy. Policy Insights from the Behavioral and Brain Sciences, 6 (2), 147–153.

Brock, C. (2011). Education and conflict: A fundamental relationship. In J. Paulson (Ed.), Education, conflict and development (pp. 17–32). Symposium Books.

Google Scholar

Cahill, A., Henning, J., Nelson, R., Diaz, C., & Henning, P. (2022). Using Rate My Professor as a Student Evaluation Tool. College Student Journal, 56 (2), 109–118.

Calcagno, J. C., Crosta, P., Bailey, T., & Jenkins, D. (2007). Stepping stones to a degree: The impact of enrollment pathways and milestones on community college student outcomes. Research in Higher Education, 48 (7), 775–801.

Celik, I., Dindar, M., Muukkonen, H., & Järvelä, S. (2022). The promises and challenges of artificial intelligence for teachers: A systematic review of research. TechTrends, 66 (4), 616–630.

Chapman, D. D., & Joines, J. A. (2017). Strategies for increasing response rates for online end-of-course evaluations. International Journal of Teaching and Learning in Higher Education, 29 (1), 47–60.

Chapman, P., Clinton, J., Kerber, R., Khabaza, T., Reinartz, T., Shearer, C., & Wirth, R. (2000). CRISP-DM 1.0: Step-by-step data mining guide. SPSS Inc, 9 (13), 1–73.

Chawla, N. V., Bowyer, K. W., Hall, L. O., & Kegelmeyer, W. P. (2002). SMOTE: Synthetic minority over-sampling technique. Journal of Artificial Intelligence Research, 16 , 321–357.

Chen, X. (2016). Remedial coursetaking at US public 2-and 4-year institutions: Scope, experiences, and outcomes. Statistical analysis report. NCES 2016-405 . National Center for Education Statistics.

Chung, J. Y., & Lee, S. (2019). Dropout Early Warning Systems for High School Students Using Machine Learning. Children and Youth Services Review, 96 , 346–353.

Coussement, K., Phan, M., De Caigny, A., Benoit, D. F., & Raes, A. (2020). Predicting student dropout in subscription-based online learning environments: The beneficial impact of the logit leaf model. Decision Support Systems, 135 , 113325.

Delen, D., Davazdahemami, B., & Rasouli, D. E. (2023). Predicting and mitigating freshmen student attrition: A local-explainable machine learning framework. Information Systems Frontiers , 1–22. https://doi.org/10.1007/s10796-023-10397-3

Delen, D., Topuz, K., & Eryarsoy, E. (2020). Development of a bayesian Belief Network-based DSS for predicting and understanding freshmen student attrition. European Journal of Operational Research, 281 (3), 575–587. https://doi.org/10.1016/j.ejor.2019.03.037

Fan, Y., Shepherd, L., & d. Slavich, d. Waters, M. Stone, R. Abel, and E. Johnston. (2019). Gender and Cultural Bias in Student Evaluations: Why Representation Matters. Plus ONE, 14 (2), e0209749. https://doi.org/10.1371/journal.pone.0209749

Fay, M. P. (2020). Learning to fail ? Student experiences in remedial mathematics in community colleges (Doctoral dissertation). https://academicworks.cuny.edu/gc_etds/3759/

Fernández-García, A. J., Preciado, J. C., Melchor, F., Rodriguez-Echeverria, R., Conejero, J. M., & Sánchez-Figueroa, F. (2021). A real-life machine learning experience for predicting university dropout at different stages using academic data. Ieee Access : Practical Innovations, Open Solutions, 9 , 133076–133090.

Figlio, D., Cassandra, H., Krzysztof, K. (2020). “Effects of Scaling Up Private School Choice Programs on Public School Students.” NBER Working Paper w26758 .

George, M., Khazanov, L., & McCarthy, C. (2015). Peer coaching of at-risk students in a remedial mathematics course. Community College Journal of Research and Practice, 39 (11), 1022–1038.

Gottfried, M., Kirksey, J. J., & Fletcher, T. L. (2022). Do high school students with a same-race teacher attend class more often? Educational Evaluation and Policy Analysis, 44 (1), 149–169.

Greene, R. A. (2018). "The effect of modality on student achievement and course completion in a developmental mathematics course" (2018). UNF Graduate Theses and Dissertations , 815. https://digitalcommons.unf.edu/etd/815

Greenwell, B. M., Boehmke, B. C., & McCarthy, A. J. (2018). A simple and effective model-based variable importance measure . arXiv preprint arXiv:1805.04755.

Grinsztajn, L., Oyallon, E., & Varoquaux, G. (2022). Why do tree-based models still outperform deep learning on typical tabular data? Advances in Neural Information Processing Systems, 35 , 507–520.

He, J., Bailey, J., Rubinstein, B., & Zhang, R. (2015). Identifying at-risk students in massive open online courses. Paper presented at the International Conference on Artificial Intelligence .

Helal, S., Li, J., Liu, L., Ebrahimie, E., Dawson, S., & Murray, D. J. (2019). Identifying key factors of student academic performance by subgroup discovery. International Journal of Data Science and Analytics, 7 , 227–245.

Hornstein, H. A. (2017). Student evaluations of teaching are an inadequate assessment tool for evaluating faculty performance. Cogent Education, 4 (1), 1304016.

Howell, E., & Walkington, C. (2020). Factors associated with completion: Pathways through developmental mathematics. Journal of College Student Retention: Research, Theory & Practice . https://doi.org/10.1177/1521025119900985

Jimenez, L., Sargrad, S., Morales, J., & Thompson, M. (2016). Remedial education: The cost of catching up . Washington, DC: Center for American Progress.

Karlos, S., Kostopoulos, G., & Kotsiantis, S. (2020). Predicting and interpreting students’ grades in distance higher education through a semi-regression method. Applied Sciences, 10 (23), 8413.

Kraft, M. A. (2019). Teacher effects on complex cognitive skills and social-emotional competencies. Journal of Human Resources, 54 (1), 1–36.

Kumar, V. S., & Boulanger, D. (2021). Automated Essay Scoring and the Deep Learning Black Box: How are rubric scores determined? International Journal of Artificial Intelligence in Education, 31 , 538–584.

Lundberg, S. M., & Lee, S. (2017). A unified approach to interpreting model predictions. In I. Guyon, U. V. Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, & R. Garnett (Eds.), Advances in Neural Information Processing Systems 30 (pp. 4765–4774). Curran Associates Inc.

Marquez-Vera, C., Cano, A., Romero, C., Noaman, A. Y. M., Mousa Fardoun, H., & Ventura, S. (2016). Early Dropout Prediction Using Data Mining: A Case Study with High School Students. Expert Systems, 33 (1), 107–124.

Mgonja, T., & Robles, F. (2022). Identifying Critical Factors When Predicting Remedial Mathematics Completion Rates. Journal of College Student Retention: Research, Theory & Practice, 2 , 1–23.

Mingyu, Z., Sutong, W., Yanzhang, W., & Dujuan, W. (2021). An Interpretable Prediction Method for University Student Academic Crisis Warning. Complex & Intelligent Systems, 8 , 1–14.

Molnar, C. (2019). Interpretable Machine Learning . Available online: https://christophm.github.io/interpretable-ml-book/ (accessed on 23 May 2023).

Monks, J. (2009). Who are the part-time faculty? Academe, 95 (4), 33–37.

Mothilal, R. K., Sharma, A., & Tan, C. (2020). Explaining machine learning classifiers through diverse counterfactual explanations. In: FAT*’20: Conference on Fairness, Accountability, and Transparency, Barcelona , Spain, January 27–30, 2020, ACM (pp. 607–617).

Nagrecha, S., Dillon, J., & Chawla, N. (2017). MOOC dropout prediction: Lessons learned from making pipelines interpretable . In Proceedings of the 26th International Conference, World Wide Web Companion (pp. 351–359). International World Wide Web Conferences Steering Committee.

Nagy, M., & Molontay, R. (2023). Interpretable Dropout Prediction: Towards XAI-Based Personalized Intervention. International Journal of Artificial Intelligence in Education, 1 , 1–27.

Pascarella, E., & Terenzini, P. (2005). How college affects students : A third decade of research . San Francisco: JosseyBass. (vol. 2).

Ran, F. X., & Xu, D. (2019). Does contractual form matter? The impact of different types of non-tenure-track faculty on college students’ academic outcomes. Journal of Human Resources, 54 (4), 1081–1120.

Rasheed, D. S., Brown, J. L., Doyle, S. L., & Jennings, P. A. (2020). The effect of teacher–child race/ethnicity matching and classroom diversity on children's socioemotional and academic skills. Child Development , 91 (3), e597–e618.

Ribeiro, M. T., Singh, S., & Guestrin, C. (2016). Why should I trust you?: Explaining the predictions of any classifier. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 34–60). ACM.

Rosen, A. S. (2018). Correlations, trends and potential biases among publicly accessible web-based student evaluations of teaching: a large-scale study of RateMyProfessors.com data. Assessment & Evaluation in Higher Education, 43 (1), 31–44.

Rovira, S., Puertas, E., & Igual, L. (2017). Data-driven system to predict academic grades and dropout. PLoS ONE, 12 (2), e0171207.

Sargsyan, A., Karapetyan, A., Woon, W. L., & Alshamsi, A. (2020). Explainable AI as a Social Microscope: A Case Study on Academic Performance. International Conference on Machine Learning, Optimization, and Data Science, 1 , 257–268.

Schneider, S., Krieglstein, F., Beege, M., & Rey, G. D. (2022). The impact of video lecturers’ nonverbal communication on learning–An experiment on gestures and facial expressions of pedagogical agents. Computers & Education, 176 , 104–350.

Smith, B. I., Chimedza, C., & Bührmann, J. H. (2021). Individualized help for at-risk students using model-agnostic and counterfactual explanations. Education and Information Technologies, 27 , 1539–1558.

Spooren, P., & Christiaens, W. (2017). I liked your course because I believe in (the power of) student evaluations of teaching (SET). Students’ perceptions of a teaching evaluation process and their relationships with SET scores. Studies in Educational Evaluation, 54 , 43–49.

Swiecki, Z., Ruis, A. R., Gautam, D., Rus, V., & Williamson Shaffer, D. (2019). Understanding when students are active-in-thinking through modeling-in-context. British Journal of Educational Technology, 50 (5), 2346–2364. https://doi.org/10.1111/bjet.12869

Thammasiri, D., Delen, D., Meesad, P., & Kasap, N. (2014). A critical assessment of imbalanced class distribution problem: The case of predicting freshmen student attrition. Expert Systems with Applications, 41 (2), 321–330.

Tindall, N. T., & Waters, R. D. (2017). Does gender and professional experience influence students’ perceptions of professors? Journalism & Mass Communication Educator, 72 (1), 52–67.

United Nations Development Programme. (2019). Human Development Indicators 2019. Retrieved from https://hdr.undp.org/content/human-development-report-2019

Van Looveren, A., & Klaise, J. (2019). Interpretable counterfactual explanations guided by prototypes. Preprint retrieved from https://arxiv.org/abs/1907.02584 . (Preprint version)

Varga, E. B., & S.t.n,. (2021). Detecting at-risk students on computer science bachelor programs based on pre-enrollment characteristics. Hungarian Educational Research Journal, 3 (11), 297–310.

Vultureanu-Albişi, A., & Bădică, C. (2021). Recommender systems: An explainable AI perspective. In 2021 International Conference on Innovations in Intelligent Systems and Applications (INISTA) (pp. 1–6). IEEE.

Yu, B., & Wright, E. (2017). Academic Adaptation Amid Internationalisation: The challenges for local, mainland Chinese, and international students at Hong Kong’s universities. Tertiary Education and Management, 23 , 347–360.

Zawacki-Richter, O., Marín, V. I., Bond, M., & Gouverneur, F. (2019). Systematic review of research on artificial intelligence applications in higher education–where are the educators? International Journal of Educational Technology in Higher Education, 16 (1), 1–27.

Zhang, S., Yao, L., Sun, A., & Tay, Y. (2019). Deep learning based recommender system: A survey and new perspectives. ACM Computing Surveys (CSUR) , 52 (1), 1–38.

Zhu, M. (2021). Limited contracts, limited quality? effects of adjunct instructors on student outcomes☆. Economics of Education Review, 85 , 102–177.

Zientek, L. R., Lane, F. C., Sechelski, A., & Shupp, S. (2022). Effects of delaying college-level mathematics course enrollment for remedial students. Journal of College Student Retention: Research, Theory & Practice, 24 (2), 474–498.

Download references

Not Applicable.

Author information

Authors and affiliations.

Department of Mathematical and Quantitative Reasoning, Utah Valley University, Mail Stop 176, 800 W University Parkway, Orem, UT, 84058, USA

Thomas Mgonja

You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Thomas Mgonja .

Ethics declarations

Conflict of interest.

I have no conflict of interest to disclose.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Mgonja, T. Using interpretable machine learning approaches to predict and provide explanations for student completion of remedial mathematics. Educ Inf Technol (2024). https://doi.org/10.1007/s10639-024-12647-6

Download citation

Received : 23 August 2023

Accepted : 21 March 2024

Published : 10 May 2024

DOI : https://doi.org/10.1007/s10639-024-12647-6

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

Interpretable machine learning
Mathematics remediation
Completion rates
Explainable artificial intelligence
Global interpretation
Local interpretation
Find a journal
Publish with us
Track your research

The Complex Factors Affecting English-Learner Graduation Rates

The likelihood of an English learner graduating from high school within four years may depend on the student’s race and ethnicity, gender, and socioeconomic status, according to a new study from New York University and University of Houston researchers.

Federal data from 2019-20 show that while the national English-learner graduation rate has increased to about 71 percent, those students still lag behind the overall rate of 86 percent.

For their study published in the Educational Researcher journal this month, researchers reviewed graduation data of more than 127,900 New York City students who began 9th grade in 2013 and 2014. While students overall who received federally mandated language services—English learners—were about 4 percent less likely to graduate on time, the narrative quickly changed when data for race, gender, and socioeconomic status were broken down.

For instance, girls—who at any point in their K-12 education were identified as English learners—were more likely to graduate than their male counterparts. Black English learners had higher graduation rates than Black peers who never had specialized language services. Asian and Pacific Islander English learners graduated at higher rates than all Latino and Black peers, while Latino English learners were the group least likely to graduate out of all . Median household incomes of where English learners lived also influenced graduation outcomes.

“What our work suggests is that we have to think about the intersectional identities that students hold,” said Michael Kieffer, an associate professor of literacy education at the NYU Steinhardt School of Culture, Education, and Human Development and a co-author of the report.

What happens when you disaggregate English-learner data

While qualitative research in the past has looked into what happens when you disaggregate English-learner data, the new study was intended to identify any trends across a large data set, Kieffer said.

Another motivation for the study stemmed from the field of raciolinguistics, where racialized students are perceived differently because of their multilingualism, based on how their race and ethnicity are viewed. For instance, multilingual white students may have their linguistic assets praised more than multilingual Black and Latino students, said Ben Le, a co-author of the study and a doctoral student at NYU Steinhardt.

This intersectional lens was crucial to the study.

For instance, for the cohort of students identified as English learners, no distinction was made between those who were reclassified out of English-learner status in their elementary years and those who received language services for a longer period of time. Yet, disparities among race, gender, and socioeconomic status lines remained, Le said.

For instance, all English learners in low-income neighborhoods, where the median household income was below $40,000, were equally likely to graduate compared with their non-English-learner peers in similar neighborhoods. But in middle- and high-income neighborhoods, English learners in general were less likely to graduate than non-English learners in similar neighborhoods.

It’s why Le recommends that when educators disaggregate their local data on English-learner graduation rates, they should take stock of what resources were available to the students both in and out of school and what perceptions of a student’s assets are at play.

What educators should keep in mind

The study overall not only highlights the value of educators and researchers alike breaking down English-learner data, but it also brings into question how the English-learner label works, said Lucrecia Santibañez, a professor at the University of California, Los Angeles School of Education & Information Studies.

When the NYU and University of Houston study pointed out the disparities at play when English learners were in middle- and high-income neighborhoods, Santibañez said the label of English learner might lead to tracking students out of advanced coursework as other national research studies have found.

The label itself is needed to denote who legally qualifies for language-instruction support, yet educators need to be careful not to let deficit mindsets tied to that label get in the way of students accessing high-quality education, she added.

Although the study focuses on New York City data, past research suggests such disparities are not unique to this data set but play out in different contexts, Le said.

“Thinking about this approach in different places, this kind of intersectional approach is still very valuable,” Kieffer added.

Sign Up for EdWeek Update

Edweek top school jobs.

Sign Up & Sign In

COMMENTS

Measuring Student Success In IPEDS: Graduation Rates (GR), Graduation
Measuring student success (i.e., college completion) can serve critical purposes for the higher education community. Students and their families may search for college choices that demonstrate high student success rates, and policymakers may make decisions informed by institutional performance on such measures.
The State of American Higher Education Outcomes in 2023
This report unpacks critical outcomes like completion, post-enrollment earnings, and student loan repayment at over 5,000 institutions of higher education and breaks down the data points by predominant degree awarded and sector, allowing for side-by-side examination of trends in multiple student success metrics across certificate programs and two- and four-year degrees at public, private non ...
PDF Completing College
About This Report. As the tenth in the Completing College series, this report updates the six-year college completion rates nationally bystate, tracking theenrollment completion outcomes for the fall 2015 cohort of beginning college students through June 2021. The national completion rate measures the performance of the entire higher education ...
College completion rate rises nationally after last year
The national six-year completion rate for students who started college in 2015 reached 62.2 percent, ... The report tracks enrollment and completion outcomes for all students who entered higher education for the first time in fall 2015, enrolling full-time or part-time at more than 3,600 two-year and four-year institutions, through June 2021. ...
PDF Completing College
The national completion rate measures the of the entire higher education system. It counts all students who enter postsecondary education for the first time each fall, enrolling full-time or part-time at two-year or four-year institutions, and completing at any degree-granting institution. It includes those who complete after transfer, not just
Completing College
Progress in the national college completion rate has stalled. The six-year completion rate for the fall 2017 cohort was 62.2 percent, essentially unchanged since 2015. Six-year completion rates increased in over half of states, with nine states increasing 1 percentage point (pp) or more. This is up from the previous year when only five states ...
The College Completion Landscape: Trends, Challenges, and Why it
The differences in completion rates by institutional type are partly driven by the types of students each part of higher education tends to serve. As shown in Table 4, students in different demographic groups have varying levels of completion, and this can help explain why some sectors of higher education have lower rates than others.
Data Home
The U.S. Department of Education's College Scorecard has the most reliable data on college costs,graduation, and post-college earnings. ... as well as supporting data on student completion, debt and repayment, earnings, and more. This data was last updated May ... enrollment, student aid, costs, and student outcomes. Field of study-level data ...
The Question of College Completion Rates
For the past few decades, college completion rates have slowly been going up. Economist Jeff Denning wanted to know why. As part of recent research, Why are College Completion Rates Increasing, Denning and co-researchers dug deep to understand this phenomenon.Denning, a Brigham Young University associate professor, says the reason isn't what you might suspect like an increase in college ...
College completion rates highest in recent years, report finds
A new report from the National Student Clearinghouse Research Center has found that 62.2 percent of first-time students who started college in fall 2015 graduated by spring 2021, the highest six-year completion rate in recent years. According to the Lumina Foundation, the report is perhaps the last snapshot of higher education trends before COVID-19, as most of the 2015 cohort graduated before ...
College completion rates are up for all Americans, but racial gaps
Students' race and ethnicity affect their chances of earning a college degree, according to several new reports on higher education released in January and February 2023. However, the picture that emerges depends on the lens you use. ... Completion rates improved considerably for students who started college in 2015, but large gaps remain ...
Report: Progress on College Completion Rates Stalls
A new report from the National Student Clearinghouse Research Center shows that college completion rates have stagnated, with 62.3 percent of students who enrolled in 2016 completing a degree by June 2022—virtually unchanged from last year's six-year completion rate of 62.2 percent.. The report examines all students who enter postsecondary education for the first time, whether they enroll ...
Census Bureau Releases New Educational Attainment Data
The high school completion rate in the United States for people age 25 and older increased from 87.6% in 2011 to 91.1% in 2021. The percentage of the population age 25 and older with associate degrees rose from 9.5% to 10.5% between 2011 and 2021.
Education at a Glance 2022
True cohort completion rates correspond to the share of students from a specific entry cohort who graduate within a particular timeframe. This is the preferred methodology for analysing completion rates, but only countries with longitudinal surveys or registers are able to provide such information.
New Report Shows Recent Increase In College Completion
The female rate increased 2.7 percentage points to 48.2% and the male rate increased 2.5 percentage points to 50.7%. However, the overall completion rate for male students is far lower than for ...
Education at a Glance 2023: Putting U.S. Data in a Global Context
High School Completion Rate. The United States has a higher upper secondary (high school) completion rate than most other OECD countries. In 2021, some 87 percent of U.S. students completed their high school program in the expected timeframe, compared with the OECD average of 72 percent ().Early Childhood Education
Improving community college completion rates by addressing structural
Yet, community colleges, as the institutions of higher education that serve a large share of low-income students, graduate less than 40 percent of students within six years (Bailey et al. 2015).
A statistical overview of higher education in England
The progression rate for full-time undergraduate students gaining a qualification in higher education in 2019-20 (first degree) is 71.9 per cent (a 1.9 percentage point rise from the previous year). For their part-time counterparts it is higher, at 86.3 per cent (a 3.9 percentage point increase from the previous year).
Completion Rates of Higher Education Students
Completion Rates of Higher Education Students - Cohort Analysis, 2005-2019. Download Completion Rates of Higher Education Students - Cohort Analysis, 2005-2019 as a XLSX (1.08mb) Creation Date. 8 September 2020. FOI Reference. D20/978562. Stream. Higher education. Creator ...
NY Financial Aid Expansion to Benefit About 93,000 Students
The idea was first introduced by lawmakers on the assembly's higher education committee late last year, and came to fruition through enactment of the fiscal year 2025 budget.. A new report, titled the New York State TAP Difference and released in tandem with the expansion announcement, demonstrates how the program broadens college access, boosts completion rates and fosters educational ...
Completion Rates
Completion Rates - Cohort Analyses. Cohort analysis reports present the results of tracking commencing higher education student outcomes over time. That is, after nominated periods, how many commencing students have completed their studies, how many are still studying, and how many have not completed an award course.
The State of Higher Education 2024
The 2024 State of Higher Education study includes responses from over 14,000 U.S. adults aged 18 to 59 without a college degree. Since 2020, Lumina Foundation and Gallup have partnered to produce the State of Higher Education study, an annual survey of thousands of U.S. adults without a college degree. Each year, Lumina and Gallup measure their ...
Mental Health Supports Are Key in in Community College Student Success
The Annenberg Institute launched a descriptive study to examine the relationship between academic success and students' time utilization, campus engagement, financial and mental well-being. The study focuses on community college students and finds that a nationwide increase in adverse mental health among college students intersects with high rates of food and housing insecurity.
What's keeping adults interested in college from enrolling?
Among unenrolled adults, 51% said they were likely to pursue a postsecondary credential in the next five years, the survey found. That's the highest rate of interest recorded by Lumina and Gallup since they started the survey in 2020. Researchers polled a sample of 14,000-plus adults between the ages 18 and 59 who do not have a college degree.
Public colleges see largest tuition drop in 40 years amid enrollment
Unlike typical economic downturns, the COVID-19 recession didn't boost college enrollment; instead, between 2020 and 2021, enrollment fell by 3.3%—the largest drop in students since 1980. It ...
PDF December 1, 2021
completion, regardless of which postsecondary institution that students attend (e.g., out of state, private, public, or for-profit institutions) and regardless of transfers. Immediate college enrollment rates of high school graduates at the national level are available through the Condition of Education published by NCES annually.1 Thousands of ...
Using interpretable machine learning approaches to predict ...
That is, remediation is associated with negative implications for college completion rates and costs, adding to the student debt crisis, with an estimated cost of $1.3 billion for remediation in English and mathematics (Jimenez et al., 2016). As a result, remedial education is often regarded as higher education's "bridge to nowhere ...
The Complex Factors Affecting English-Learner Graduation Rates
Federal data from 2019-20 show that while the national English-learner graduation rate has increased to about 71 percent, those students still lag behind the overall rate of 86 percent. For their ...