Young Adults and Higher Education: Barriers and Breakthroughs to Success Young Adults and Higher Education: Barriers and Breakthroughs to Success Thomas Brock Summary Although access to higher education has increased substantially over the past forty years, stu- dent success in college—as measured by persistence and degree attainment—has not improved at all. Thomas Brock reviews systematic research findings on the effectiveness of various inter- ventions designed to help at-risk students remain in college. Brock shows how changes in federal policy and public attitudes since the mid-1960s have opened up higher education to women, minorities, and nontraditional students and also shifted the “center of gravity” in higher education away from traditional four-year colleges toward nonselective community colleges. Students at two-year colleges, however, are far less likely than those at four-year institutions to complete a degree. Brock argues that the nation’s higher education system must do much more to promote student success. Three areas, he says, are particularly ripe for reform: remedial education, student support services, and financial aid. In each of these three areas, Brock reviews programs and interventions that community col- leges have undertaken in order to raise completion rates. Some colleges have, for example, experimented with remedial programs that build social cohesion between students and fac- ulty and integrate content across courses. Other colleges have tested student support service programs that offer counseling and advising that are regular, intensive, and personalized. Still others have experimented with ways to simplify the financial aid application process and incen- tivize students to earn good grades and persist in school. Research shows that such programs and interventions can improve student outcomes, but Brock argues that more must be done to bring proven practices to scale and to test new ideas that might lead to better results. Institutions that most need help are those that provide the greatest access to nontraditional and underprepared students in community colleges and less selective universities. www.futureofchildren.org Thomas Brock is the director of the Young Adults and Postsecondary Education Policy Area at MDRC. VOL. 20 / NO. 1 / SPRING 2010 109 Thomas Brock Few decisions matter more to a private institutions have the capacity to do so young person’s future than the much good for so many. decision to attend college and earn a degree. As described by Sheldon Danziger and David Access to higher education Ratner in their article in this volume, college graduates have substantially better prospects has increased substantially, in the labor market than peers who stop their although some racial and formal education after high school. In fact, ethnic groups remain over a lifetime, an adult with a bachelor’s degree will earn about $2.1 million—roughly underrepresented. But one-third more than an adult who starts but success in college—as does not complete college and nearly twice as much as one who has only a high school measured by persistence diploma.1 College attendance and completion and degree attainment— provide other benefits as well. For example, has not improved at all. adults who have attended some college or earned a bachelor’s degree are more likely to report “excellent” or “very good” health than those who have only a high school diploma, My purpose in this article is to examine data even when they have comparable incomes.2 on college enrollment and completion in the College is often where people form their United States and to explore what might be deepest friendships and meet future spouses done to help more young people benefit from or partners. Finally, as Constance Flanagan the experience and complete college degrees. and Peter Levine discuss in their article in I begin by reviewing historical trends to show this volume, research shows that educational how the numbers and characteristics of col- attainment has positive effects on voting and lege students have changed in the past forty other measures of civic engagement.3 years. Access to higher education, it turns out, has increased substantially, although some Clearly, many of the benefits that accrue racial and ethnic groups remain underrepre- from a college education are explained by the sented. But success in college—as measured knowledge, skills, and contacts that students by persistence and degree attainment—has gain from their time on campus and in the not improved at all. I then examine some classroom. From a developmental standpoint, leading explanations for why college students colleges and universities also provide a safe do not succeed and review some research environment for young adults to explore new findings on interventions designed to help ideas and interests, interact with people who at-risk students overcome barriers. I conclude are different from themselves, and form their with some lessons and suggestions to guide identity. For all these reasons, colleges and policy makers, practitioners, and researchers. universities play an indispensable role in the transition to adulthood. At their best, they The Changing Landscape of foster both intellectual and personal growth Higher Education: 1965–2005 and prepare young people for productive Before 1965, American colleges and universi- lives at work and in society. Few public or ties were rarefied places populated mostly by 110 THE FUTURE OF CHILDREN Young Adults and Higher Education: Barriers and Breakthroughs to Success white males from middle- or upper-income the federal government launched a major families. In part, the lack of diversity program for facilities construction, target- reflected the fact that for much of the ing “developing institutions” like community nation’s history, a college education was not colleges and historically black colleges and needed to make a decent living. Indeed, after universities.8 Federal spending on higher World War II, the difference between the education increased exponentially, from $655 average wages of high school and college million in 1956 to $3.5 billion in 1966.9 graduates was small and shrinking. After 1950, however, the trend moved in the During the same period, the civil rights opposite direction and accelerated as the movement influenced higher education by demand for highly skilled labor increased.4 In challenging public laws and practices that 1975, year-round workers with a bachelor’s excluded blacks and other minority groups degree earned 1.5 times the annual pay of from attending some colleges and universi- workers with only a high school diploma; by ties, particularly in the South. Early battles 1999, that ratio had risen to 1.8.5 focused on winning admittance for individual students. In 1964, Congress passed the Civil Prevailing social norms and a limited federal Rights Act, which outlawed discrimination role in higher education also served to keep based on race in schools, public places, and higher education an exclusive domain before employment and mandated equal opportu- the 1960s. In many parts of the country, nity for women. By the late 1960s, civil rights discriminatory laws and attitudes kept many activists broadened their perspective to blacks and other racial or ethnic minorities encompass poverty and income inequality from pursuing a college degree. Prevailing and helped launch dozens of Great Society attitudes about the role of women limited programs that funded education and job their college-going as well. Finally, before training programs targeted to low-income 1965, financial aid was not generally available Americans.10 for college students. The federal G.I. Bill had covered college costs for tens of thousands of Demographic trends, combined with the veterans after World War II, but it, too, had social activism of the 1960s, also created “masculinized” campus life and had aided pressure for change. As the baby boom whites far more than African Americans.6 generation reached maturity, young adults poured onto college campuses in record The mid-to-late 1960s marked a major turn- numbers. Colleges and universities became ing point. Changes in federal policy, coupled centers of protest, most famously against the with big changes in public attitudes and Vietnam War, but also against all manner of expectations, opened up higher education as social convention and custom.11 Rules gov- never before. From a policy perspective, the erning higher education were not above the passage of the Higher Education Act of 1965 fray. Questions of who should have access— was arguably the most important change, as and what role colleges and universities should it extended need-based financial assistance play in confronting and reducing inequities to the general population for the first time.7 in the larger society—were hotly debated. The federal role expanded in other ways, The “open admissions” movement gained too, fueling growth on college and university currency during this era, most famously with campuses. Starting in 1963, for example, the 1970 decision by the City University of VOL. 20 / NO. 1 / SPRING 2010 111 Thomas Brock Figure 1. Fall Enrollment in Two- and Four-Year Degree-Granting Institutions, 1963–2005, by Millions 2.0 1.8 1.6 1.4 1.2 nt e m 1.0 oll nr .8 E .6 .4 .2 0 1963 1967 1971 1975 1979 1983 1987 1991 1995 1999 2003 2005 Source: T. D. Snyder, S. A. Dillow, and C. M. Hoffman, Digest of Education Statistics 2007 (NCES 2008-022) (Washington: National Center for Education Statistics, Institute of Education Sciences, U.S. Department of Education, 2008), table 180. New York to allow all high school graduates Along with increased enrollments, the to pursue college degrees regardless of aca- demographics of students attending colleges demic preparation. Other institutions across and universities changed. Figure 2 compares the country, notably community colleges, the characteristics of students by gender, race adopted similar policies.12 or ethnicity, and age starting in the 1970s, when the federal government first began Trends in Student Enrollment reporting on student demographics. The top panel shows how the gender balance reversed and Demographics between 1970 and 2005, from mostly male to The effects of changing laws and attitudes mostly female. The second panel depicts the are evident in the dramatic rise in college increase in the percentage of students from enrollments depicted in figure 1. Total fall racial or ethnic minority groups, which more enrollment increased from just over 5.9 than doubled from 1976 to 2005. By far the million students in 1965 to about 17.5 mil- largest percentage increases were among lion students in 2005—a nearly 300 percent Hispanics and Asian and Pacific Islanders, increase.13 The rise was steepest through though all minority groups experienced 1975 and was far greater than could be growth in college enrollment while the share accounted for by population growth alone. of whites declined. Finally, the third panel To put the enrollment figures into perspec- shows an increase in the percentage of tive, in 1965 the number of young adults in students aged twenty-five and older and a the prime college-going years of eighteen to proportionate decline in those aged twenty- twenty-four was approximately 20.3 million; four and under. The U.S. Department of by 2005 that number had increased 44 per- Education projects that the trend toward cent, to about 29.2 million.14 older students will continue in coming years.15 112 THE FUTURE OF CHILDREN Young Adults and Higher Education: Barriers and Breakthroughs to Success The shift in demographic characteristics hints Figure 2. Characteristics of Students Attending College, by Gender, Race and Ethnicity, and at another significant development in the Age, 1970 and 2005 student population. The so-called traditional undergraduate—the high school graduate Gender who enrolls full-time immediately after fin- Female ishing high school, relies on parents for finan- Male cial support, and either does not work during the school year or works only part-time—is now the exception rather than the rule. Only 27 percent of undergraduates met these 1970 2005 criteria in 1999–2000. By comparison, in the same year, 28 percent of undergraduates met Race and ethnicity the Department of Education’s definition of White “highly nontraditional”: they were likely in Black their twenties or older, working while going Hispanic to school, and raising children (possibly as Asian/Pacific Islander single parents), among other criteria. Some American Indian/ highly nontraditional students did not have a 1976 2005 Alaskan Native high school diploma.16 Non-resident alien Patterns of Institutional Attendance Age Government statistics show that a large 24 + up majority of undergraduates enroll in four- 24 + under year colleges and universities. At the same time, the “center of gravity” in higher educa- tion has gradually shifted, with community colleges playing a much more prominent role 1970 2005 today than in the past. In 1969 (when the government adopted its current methodology Source: Same as figure 1, tables 187, 217, and 181. for categorizing two- and four-year schools), 26 percent of all college students attended two-year institutions. By 2005, that figure had education again. Indeed, two of the five risen to 37 percent.17 largest higher education institutions in 2005 rely principally on online instruction: the The vast majority of students enroll in University of Phoenix, with an enrollment publicly funded colleges and universities. of more than 117,000 students, and Western In 2005, private institutions accounted for International University, with an enrollment about one-fourth of all undergraduates—a of nearly 51,000 students.19 figure that has increased only slightly in the past decade. Nearly all of these students are Although the demographic composition enrolled in four-year institutions, though a small percentage of students is enrolled of colleges and universities has become in private two-year colleges.18 The advent more diverse over the past forty years, the of online courses may be changing higher increased diversity is largely accounted for by VOL. 20 / NO. 1 / SPRING 2010 113 Thomas Brock nonselective institutions. Specifically, female, began collecting data during the 1970s. A black, and Hispanic students are dispropor- recent study suggests that there has been a tionately enrolled in community colleges.20 slight uptick in the persistence rate at public Nontraditional students are also much more four-year colleges.24 Although that increase likely to be enrolled in community colleges may seem contradictory, it likely reflects the and to participate in distance education via longer time it now takes students in four-year the Internet.21 colleges, particularly at less selective public institutions, to earn degrees.25 Historical data Trends in Persistence and Completion on students attending community college go Government statistics indicate that student back only to 1990, but show no significant outcomes differ markedly by type of institu- change in persistence or completion.26 tion attended. Specifically, undergraduates who begin at four-year colleges and universi- Persistence and completion rates differ ties are about twice as likely to complete a significantly by race and ethnicity and by gen- postsecondary degree as undergraduates who der. At public two- and four-year institutions, begin at two-year institutions. The five-year Asian and Pacific Islanders have the highest completion rate for students who began at a persistence and completion rates of any racial four-year college or university—taking into or ethnic group, followed by non-Hispanic account certificates, associate’s degrees, or whites, Hispanics, and non-Hispanic blacks. bachelor’s degrees—was 60 percent in the (The longitudinal studies commissioned by 1995–96 academic year. For students who the government lack sufficient numbers of began at a community college, the rate of American Indians and Alaska Natives on completion was 32 percent.22 which to report.) Asian and Pacific Islanders who entered public four-year institutions in Many students take longer than five years to 1995–96 were nearly twice as likely to earn earn a degree: some are enrolled part-time, a degree or still be in school after six years some change their majors, some need to as non-Hispanic blacks who entered the drop out temporarily, and some have other same year. The story by gender is a bit more reasons for the delay. Measures of persistence complicated. At public four-year institutions, take into account those who have earned a women have slightly higher persistence and certificate or degree as well as those who completion rates than men (a difference are still enrolled in college. Eighty percent of about 5 percentage points); at public of students who began at a four-year col- two-year institutions, the gender difference lege or university in 1995–96 persisted after is reversed.27 It is important to recall that five years. Among students who began at a because more women than men enroll in community college, the persistence rate was college, many more associate’s and bachelor’s 52 percent. The data also show that students degrees are awarded to women—a pattern attending private institutions (both four-year that has held true at both two- and four-year and two-year) persist at higher rates than institutions since at least the late 1980s.28 their counterparts at public institutions.23 Despite these patterns, neither race and Viewed historically, rates of comple- ethnicity nor gender is a good predictor of tion at four-year institutions have been who will earn a college degree, owing to large unchanged since the federal government variation within these demographic groups. 114 THE FUTURE OF CHILDREN Young Adults and Higher Education: Barriers and Breakthroughs to Success Research by Clifford Adelman for the compared with 32 percent of blacks and 25 Department of Education shows that the two percent of Hispanics.31 Rates of college best predictors are entering college immedi- attendance for black and Hispanic males are ately after finishing high school and taking a particularly low. A recent national survey of high school curriculum that stresses reading college-qualified students who did not enroll at grade level and math beyond basic algebra. in college underscores that college costs, Higher socioeconomic status is also a predic- availability of aid, and uncertainty about the tor, though only moderately so.29 Consistent steps needed to enroll in college remain with these findings, being classified as a significant deterrents.32 Inadequate prepara- traditional student is another strong predictor tion for college is another factor, though with of college completion. Conversely, all of the the rise of nonselective institutions, it is less a characteristics used to define nontraditional barrier to access than to success once students status—delayed entry into college from high have enrolled in college. school, working full-time, single parenthood, and so on—are considered “risk factors” From a public policy standpoint, it makes because they are negatively correlated with little sense to promote greater college access persistence.30 As noted, community colleges if students are failing once they get there. account for a disproportionate share of Figuring out how to boost college completion nontraditional students; they are also the is the challenge. The United States has seen institutions that raise the most concern about no progress on this measure since the advent persistence and completion. of statistics on it and is losing ground to other nations in the share of the adult population Summary of Trends and Key Issues with college degrees.33 The costs of such Access to higher education has been greatly failure—to students especially, but also to expanded since the mid-1960s. More students colleges, governments, and society at large— are attending college—both in real terms and are extremely high. I next examine why some as a percentage of the population—and they students don’t succeed and what might be are demographically more diverse. Actions done in response. taken by the federal government clearly played a major part in these trends, though Improving Academic Outcomes for larger economic, demographic, and social Students in Higher Education forces were also at play. Finally, the growth of The search for solutions to the college persis- nonselective institutions like community tence and completion problem begins with an colleges and, more recently, online courses understanding of its underlying causes. In his and programs has made it easier for people to seminal book Leaving College, Vincent Tinto attend college even if they lack good prepara- examines why students depart prematurely tion or are working while going to school. from both two-year and four-year institutions. He acknowledges that students come to col- Despite these gains, college access remains lege with different skills and abilities, vary- problematic, and gaps in enrollment between ing personal motivation and objectives, and certain racial and ethnic minority groups are diverse external commitments that will influ- substantial. In 2006, for example, 44 percent ence their ability to succeed, but he argues of whites between the ages of eighteen and that what happens to them after they arrive twenty-four were enrolled in college, on campus is at least as important as what VOL. 20 / NO. 1 / SPRING 2010 115 Thomas Brock happened before. Specifically, he focuses on years to shift the primary teaching responsi- how well students are integrated into the bility for remedial education to community classrooms and laboratories where instruc- colleges. The most recent data from the tion takes place and into the informal meet- Department of Education indicate that 42 ing places such as dormitories, cafeterias, percent of freshmen at community colleges and hallways. In all of these situations, he enroll in at least one remedial reading, writ- notes, the quality and frequency of interac- ing, or mathematics course. At private and tions between students, faculty, and staff will public four-year institutions, the figure ranges shape students’ experiences and determine from 12 to 24 percent.37 how well they “fit” at a particular institution.34 Other theorists have placed greater weight Research and anecdotal evidence suggest that on how cultural norms and organizational many students who are assigned to remedial structures and processes may affect student education drop out of the classes (and often success, but generally support Tinto’s dual out of college) and that those who remain emphasis on student attributes and institu- make slow progress. An analysis of data from tional practices as the keys to understanding the Department of Education’s National college persistence and completion.35 Education Longitudinal Study shows that only 28 percent of remedial students in Some observers have argued that the prob- two-year colleges attain a degree or certifi- lem of low persistence and completion would cate within eight and one-half years of entry go away if primary and secondary schools did (compared with 43 percent of nonremedial a better job of preparing students for col- students), and that 52 percent of remedial lege. There is certainly truth to this claim, students in four-year colleges finish bach- and another article, if not a book, could be elor’s degrees within this period (compared written on educational reforms and college with 78 percent of students without remedial preparatory programs that target younger course work). The analysis also shows that students.36 But these reforms and programs remedial education delays time-to-degree for will always fail to reach some students, and students in two-year colleges. Though seldom blaming the nation’s schools for poor college acknowledged, remedial education acts as a completion rates lets colleges and universi- gatekeeper and quality control mechanism ties off the hook too easily. America’s higher in most institutions.38 It allows underpre- education system has many strengths, but it is pared students access to campus facilities far from perfect, and policy makers and insti- and resources, yet clearly divides them from tutions can do much more to promote greater students considered to be “college ready.” student success. Three areas seem particu- larly ripe for reform: remedial education, Is this the best that colleges and universities student support services, and financial aid. can do, or is it possible to remake remedial education so that greater numbers of stu- Rethinking Remedial Education dents acquire basic skills and go on to earn That many students arrive on college cam- college degrees? Many educators believe that puses unprepared to do college-level work is change is possible, noting the propensity of well known. Both two- and four-year institu- remedial education classes to use outmoded tions face this deficit, though the nation has teaching methods—including repetition made a conscious policy choice in recent and memorization of material that does not 116 THE FUTURE OF CHILDREN Young Adults and Higher Education: Barriers and Breakthroughs to Success connect to students’ interests—and to isolate partnership with the MacArthur Network on and marginalize students from the rest of Transitions to Adulthood. I discuss the other the college community.39 Many reforms have Opening Doors interventions later. been suggested or tried, but relatively few have been evaluated in a way that establishes Kingsborough’s Learning Communities pro- a causal relationship between the reforms gram targeted incoming freshmen, the great and educational attainment.40 majority of whom required remedial English. Students were placed into groups of fifteen to twenty-five to take three of their first- Many reforms have been semester courses together: an English course (based on level of proficiency, but usually suggested or tried, but remedial); a regular college course like intro- relatively few have been ductory psychology or health; and a student evaluated in a way that success course, taught by a college counselor, that covered time management, effective establishes a causal study habits, and other skills considered relationship between the necessary to succeed in college. Faculty who reforms and educational taught in the Learning Communities were expected to coordinate assignments and meet attainment. periodically to review student progress. The idea was to build social cohesion between students and faculty and to make the sub- A notable exception is an evaluation of a ject matter more meaningful to students by Learning Communities program at integrating the content and helping students Kingsborough Community College in apply the concepts and lessons across the Brooklyn, New York, one of four sites in a courses.42 national demonstration project called Opening Doors.41 The goal of Opening Doors More than 1,500 students participated in was to test interventions designed to increase the Learning Communities evaluation and persistence and raise academic achievement were, as noted, randomly assigned to either a among low-income community college program group that participated in Learning students. At all of the Opening Doors sites, Communities or a control group that took students were randomly assigned either to a regular, unlinked courses. The students program group that received an enhanced set were young (mostly seventeen to twenty of services or to a control group that experi- years old), low-income, and highly diverse enced “business as usual” at the college. in terms of race and ethnicity. Researchers Random assignment ensures that the students tracked program and control group members in the program and control groups are similar for two years and found that students in the at baseline; subsequent differences in educa- Learning Communities were more likely to tional attainment or other outcomes can feel integrated at school and to be engaged therefore be attributed to the intervention, in their courses and with fellow students and rather than to differences in student motiva- instructors. They also passed more courses tion or characteristics. The Opening Doors and earned more credits during their first research was conducted by MDRC, in semester, moved more quickly through VOL. 20 / NO. 1 / SPRING 2010 117 Thomas Brock remedial English requirements, and were A recent evaluation comparing academic more likely to take and pass an English skills outcomes for 900 I-BEST participants with assessment test that was required for gradua- those of more than 31,000 students in regular tion. It is important to note that these effects, remedial courses in Washington State com- while statistically significant, were generally munity colleges found that I-BEST students modest. For example, after four semesters, had higher persistence rates, earned more students in the program group earned an credits toward a college credential, earned average of 33.2 college credits, compared more occupational certificates, and showed with an average of 30.8 credits for the control greater increases on remedial education tests. The evaluation did not use a random- group (a difference of 2.4 credits, or less assignment design and therefore cannot than one standard college course). Moreover, eliminate the possibility that students who contrary to expectations, the Learning were selected into I-BEST were more Communities did not have an immediate motivated or had other characteristics that effect on persistence. Kingsborough is only may have distinguished them from students in one test, however, and a new set of random- regular remedial courses, although the authors ized trials is under way to determine whether controlled for observed differences in student similarly structured programs at six com- backgrounds and enrollment patterns. The munity colleges around the country will help results suggest that the model holds promise students complete remedial English or math and should be subject to more rigorous requirements and persist.43 experimental evaluation in the future.44 As noted, one of the underlying concepts of Another approach to reforming remedial Learning Communities is to make the course education is to accelerate the pace at which material more meaningful to students by students move through it. For example, linking the information covered in one class students who test just below college level to the discussions and assignments in another may be assigned to a short-term review class. Another way to make remedial educa- class rather than a full-semester course. tion more meaningful is exemplified by Alternatively, colleges may set up basic skills Washington State’s Integrated Basic “immersion” courses that are shorter in Education and Skills Training (I-BEST) duration but require more hours of atten- program. I-BEST offers “contextualized dance each week, to help students master instruction,” which integrates basic English the material more quickly. Such courses can and math skills into college-level career or be offered during the regular semester or technical training in fields such as nursing, during breaks between semesters.45 Incoming early childhood education, automobile repair, students may also be targeted for intensive and data entry. For example, nursing students remediation before they start classes to in the I-BEST program attend English help them do better on basic skills assess- classes that emphasize medical terminology ment tests and possibly avoid remediation and writing used in health care settings. If altogether. A summer “bridge” program in students have trouble making sense of the Texas, for example, offers remedial English words because of limited English proficiency, and math instruction along with general they receive additional help in learning basic study skills to students who just finished language skills. high school and are about to enroll in a 118 THE FUTURE OF CHILDREN