ERIC EJ1015252: Student Supports: Developmental Education and Other Academic Programs

2013·0.14 MB·English

by  ERIC

Student Supports: Developmental Education and Other Academic Programs
Student Supports: Developmental Education
and Other Academic Programs
Eric P. Bettinger, Angela Boatman, and Bridget Terry Long
Summary
Low rates of college completion are a major problem in the United States. Less than 60 percent
of students at four-year colleges graduate within six years, and at some colleges, the graduation
rate is less than 10 percent. Additionally, many students enter higher education ill-prepared
to comprehend college-level course material. Some estimates suggest that only one-third of
high school graduates finish ready for college work; the proportion is even lower among older
students. Colleges have responded to the poor preparation of incoming students by placing
approximately 35 to 40 percent of entering freshmen into remedial or developmental courses,
along with providing academic supports such as summer bridge programs, learning communi-
ties, academic counseling, and tutoring, as well as student supports such as financial aid and
child care. Eric Bettinger, Angela Boatman, and Bridget Terry Long describe the role, costs,
and impact of these college remediation and academic support programs.
According to a growing body of research, the effects of remedial courses are considerably
nuanced. The courses appear to help or hinder students differently by state, institution, back-
ground, and academic preparedness. The mixed findings from earlier research have raised ques-
tions ranging from whether remedial programs, on average, improve student academic outcomes
to which types of programs are most effective. Administrators, practitioners, and policy makers
are responding by redesigning developmental courses and searching for ways to implement
effective remediation programs more broadly. In addition, recent research suggests that colleges
may be placing too many students into remedial courses unnecessarily, suggesting the need for
further examining the placement processes used to assign students to remedial courses.
The authors expand the scope of remediation research by discussing other promising areas of
academic support commonly offered by colleges, including advising, tutoring, and mentoring
programs, as well as supports that target the competing responsibilities of students, namely car-
ing for dependents and balancing employment with schoolwork. They conclude that the limited
resources of institutions and equally limited funds of students make it imperative for postsec-
ondary institutions to improve student academic supports and other services.
www.futureofchildren.org
Eric P. Bettinger is an associate professor of economics and education at Stanford University; Angela Boatman is a postdoctoral
scholar at Stanford University; and Bridget Terry Long is the Xander Professor of Education and Economics at Harvard University.
VOL. 23 / NO. 1 / SPRING 2013 93
Eric P. Bettinger, Angela Boatman, and Bridget Terry Long
Students often arrive at college fac- To help them succeed, many postsecondary
ing multiple challenges, including institutions offer a range of academic and
inadequate academic preparation, cocurricular supports.
competing obligations to work and
family, and limited experience Remedial courses, which fall under the
navigating the complexities of collegiate sys- broad term of remediation, are the support
tems and requirements. Although all these most widely used by colleges to address the
challenges pose problems for college comple- academic needs of underprepared students.6
tion, the primary obstacle is poor prepara- These courses (commonly referred to as
tion for college-level coursework. Data from developmental courses by practitioners)
the National Center for Education Statistics target underprepared students with the
indicate that in 2004 only 26.8 percent of purpose of improving their abilities to handle
high school seniors had completed “high- college-level material and succeed in college.
level” academic coursework, defined as four Research suggests that more than one-third
years of English, three years of mathemat- of all first-year students in college today
ics (including at least one year of a course are taking some form of remedial course-
higher than algebra II), three years of sci- work in either English or mathematics; the
ence, three years of social studies, and two share can climb to six out of ten students at
years of a single non-English language.1 A some postsecondary institutions.7 The bulk
separate study found that only 32 percent of of remediation is provided by nonselective
students leave high school at least minimally public institutions, the point of entry for 80
prepared for college academically.2 A lack of percent of four-year students and virtually all
alignment between the K–12 and postsec- two-year students.8
ondary education systems compounds the
problem, frequently resulting in confusing
messages to students and their parents about
what students should do to enter and suc- The bulk of remediation is
ceed in college.3
provided by nonselective
Although all students face challenges in public institutions, the point
higher education, underprepared students
of entry for 80 percent of
confront more urgent problems, both aca-
demically and more broadly. While adjusting four-year students and
to a new environment, they must simul-
virtually all two-year
taneously acquire college-level academic
skills. Difficulties in the classroom can be students.
discouraging and can complicate the aca-
demic, social, and financial adjustments to
college.4 Ultimately, academic struggles may
lead to lower self-esteem, greater frustration, To better meet the needs of underprepared
and higher drop-out rates.5 Asked to make students, some colleges have implemented
complex choices about their field of study and interventions such as summer bridge pro-
future plans, students may find it increasingly grams, learning communities, academic
difficult to respond to the demands of college. counseling, and tutoring. Others have tried to
94 THE FUTURE OF CHILDREN
Student Supports: Developmental Education and Other Academic Programs
address nonacademic student needs that may language.10 Nonnative English speakers in
affect academic performance, such as devel- need of remediation most commonly enroll in
oping programs to help older, nontraditional developmental English or English as a Second
students with nonacademic barriers such as Language (ESL) courses.11 Remedial and
child care and transportation. developmental courses allow colleges to offer
access to students whose life circumstances or
In this article we examine remedial educa- earlier academic experiences might otherwise
tion and other kinds of student supports. have been a barrier to college entry.
We first consider remediation, including the
students who need it, how it is organized, Students in remediation may have earned
how much it costs, and what researchers have a high school diploma, but may still not be
learned about its effects on student out- academically prepared for success in college.
comes. We then discuss additional academic The need for remediation in college is closely
supports, including advising, tutoring, and tied to a student’s high school curriculum.12
mentoring programs. Next, we focus on A 2002 study by the Ohio Board of Regents
supports that target the competing respon- found that students who had completed an
sibilities of students, namely caring for academic core curriculum in high school were
dependents and balancing employment with half as likely to need remediation in college
academic obligations. Overall, we consider as students who had not.13 Completing a high
whether and how these supports help stu- school core curriculum, however, does not
dents to be successful and how colleges and necessarily ensure that a student will avoid
universities might improve their efforts to remediation in college. Many students who
support students. complete upper-level math courses in high
school still require math remediation courses
Helping the Underprepared or need to repeat subjects in college.14 That
Student: College Remediation students who are “academically prepared” still
Most students in remediation are recent high need to be in remediation suggests that the
school graduates who have exited secondary problem is larger than just poor high school
school without grade-level competency or the course selection or the lack of a college-
proper preparation for college-level material, preparatory curriculum at some schools.15
but more than 25 percent are over the age of
thirty.9 Recent structural shifts in the labor The Organization and Delivery
market have displaced many adult workers, of Remedial Education
who subsequently enroll in college to acquire Postsecondary institutions across the nation
the skills necessary for re-employment; many offer remedial courses structured in a variety
need to refresh their math, reading, and writ- of ways. Traditional remedial courses gen-
ing skills. Another group of students in reme- erally take a fifteen-week, semester-long
diation includes those who were not born in format. Courses are typically, but not univer-
the United States or who grew up speaking sally, offered for credit and count toward a
languages other than English, or both. Of the student’s overall grade point average but not
2.6 million students attending community toward graduation requirements. The vast
colleges in California, for example, an esti- majority of colleges offer multiple levels of
mated 25 percent speak English as a second remedial and developmental courses within
VOL. 23 / NO. 1 / SPRING 2013 95
Eric P. Bettinger, Angela Boatman, and Bridget Terry Long
a subject area (for example, English language Sung-Woo Cho, factors that extend the time
arts or mathematics) to meet the needs of it takes students to complete degrees are
students from a wide range of academic also associated with a lower probability of
backgrounds. For example, many institutions degree completion.19
offer up to three developmental mathemat-
ics courses below college-level mathemat- Students are usually assigned to remedial
ics: Developmental Algebra II for students courses based on an exam or assessment
just below college-level mathematics and taken when they arrive on campus. About
in need of algebraic computational skills, 92 percent of institutions use some kind
Developmental Algebra I for those needing of standardized placement exam to assign
to learn functions, quadratic equations, and students to remedial or developmental
inequalities, and Remedial Arithmetic for courses.20 The most widely used place-
those in need of computational arithmetic
ment exams are the Computerized Adaptive
skills.16 The course sequence in remedial
Placement Assessment and Support Systems
English courses is generally similar.
(COMPASS) and the Assessment of Skills
for Successful Entry and Transfer (ASSET),
Because students most commonly have to
each published by ACT, Inc., as well as the
succeed in their assigned developmental
ACCUPLACER published by the College
course before moving on to the next course
Board. Some schools also use state standard-
in the sequence, remedial courses are often
ized test scores and high school transcripts to
the gateway to college-level courses. More
help make assignments. Typically, administra-
than four-fifths of campuses nationally
tors base course assignments on “hard” cut-
restrict enrollment in some college-level
offs—students scoring below a single given
classes until remediation is complete, and
threshold are assigned to a remedial course.
most require students placed into remedia-
Students are placed into mathematics reme-
tion to enroll in the course recommended
diation more often than into English lan-
by the institution.17 Students in need of
guage arts (that is, reading or writing or both)
multiple remedial courses in the same
remediation,21 but English remediation may
subject could thus take courses for more
be even more critical to a student’s academic
than a year before fulfilling their remedial
success because reading and writing skills are
requirements. Not surprisingly, students
fundamental to most other subjects.
assigned first to the higher developmental
mathematics courses complete their devel-
The Costs of Remediation
opmental course sequence and move on to
a college-level mathematics course at higher A study by the Alliance for Excellent
rates (45 percent) than those assigned to the Education concluded that the total cost of
lowest-level courses (17 percent). Rates are delivering remediation nationwide during
even lower for men, older students, African the 2007–08 school year was $3.6 billion
American students, part-time students, and in the form of direct costs both to students
students in vocational programs.18 Because (for example, tuition) and to institutions (for
remedial courses rarely count toward a stu- example, instructional costs). The study also
dent’s graduation requirements, remediation estimated additional costs beyond these
may decrease rates of degree completion. As direct costs in the form of lost earning poten-
shown in a 2012 study by Davis Jenkins and tial for those remedial students who may be
96 THE FUTURE OF CHILDREN
Student Supports: Developmental Education and Other Academic Programs
more likely to drop out of college without instructional spending on the students most
completing a degree.22 In 2008, a report by in need of high-quality instruction.27 Although
the Strong American Schools project used the costs of remediation are generally high,
higher education expenditure data col- the social costs of not offering remediation
lected by the U.S. Department of Education may be higher still. Unskilled individuals are
to estimate that the total taxpayer cost of more likely not only to collect unemployment
remediation per student ranged from $1,607 and welfare benefits but also to commit crime
to $2,008 in two-year colleges and between and be incarcerated. Moreover, the changing
$2,020 and $2,531 in four-year colleges.23 demands of the twenty-first-century economy
require efficient retraining. A 2005 study con-
Although remedial and developmental ducted for the Texas Public Policy Foundation
courses often do not count toward graduation
estimated that when students leave high
requirements, students must nevertheless pay
school without acquiring basic reading,
tuition for these courses and bear the oppor-
writing, and math skills, the state loses more
tunity cost of forgone earnings. In 2003–04,
than $13.6 billion annually in terms of lower
Florida community college students who
earnings potential, poor worker productiv-
required remediation took an average of nine
ity, and increased spending on social pro-
credit hours of remedial coursework and paid
grams.28 Proponents argue that remediation
an additional $504 for college prep course-
programs help postsecondary institutions
work during their first year of college.24 Given
fulfill their obligation to assist students who
that many remedial students also receive
may have attended poor-quality K-12 schools.
federal financial aid, taxpayers shoulder a
Remediation efforts can provide such stu-
portion of the cost of remediation as well.
dents a second chance to learn the basic skills
necessary for future labor market success.
Although remediation is expensive for col-
leges to provide, it may be less costly than
The Effects of Remediation on
other college courses, as remedial courses
Student Outcomes
often have comparatively larger class sizes
A growing body of research is emerging on
and a higher prevalence of lower-paid adjunct
both the scope and effectiveness of college
instructors.25 The Ohio Board of Regents
remediation. Many earlier descriptive studies
found that although 38 percent of students
merely compare samples of remedial students
in the state’s public two-year colleges were
to their peers, without noting that students
enrolled in developmental courses, such
courses accounted for only 3.6 percent of in need of remediation may differ from their
the total budget for instructional spending.26 more academically prepared peers in both
Because institutions are able to lower the cost their observed and unobserved background
of offering remedial courses through a variety characteristics. Comparing the outcomes of
of administrative and instructional deci- these two very different types of students
sions, while students are unable to lower the without taking into account these unob-
amount they pay in tuition, the cost of reme- served differences, such as student ability
diation can be unevenly distributed. In cases and motivation, can lead to biased estimates
like this, remedial and developmental courses of the impact of remediation on subsequent
have the ability to generate revenue, which academic outcomes.29 Short of randomly
gives perverse incentives to schools to reduce assigning students on the margins of needing
VOL. 23 / NO. 1 / SPRING 2013 97
Eric P. Bettinger, Angela Boatman, and Bridget Terry Long
remediation to either remedial or college- receive an associate’s degree or certificate.34
level courses, it can be difficult to ascertain The mixed results from these studies suggest
whether differences in student outcomes are that the causal effect of remedial courses on
caused by students’ enrollment in remedial student outcomes for students at the mar-
classes, or are instead explained by their gin of passing out of remediation is not yet
lower levels of academic preparation—the fully understood. Because many educational
very thing that required them to be remedi- interventions have had varying effects on
ated in the first place. students of different genders, races, and other
demographic characteristics, however, it is
Given the importance of remediation and the plausible that remedial courses may also have
limitations posed by bias in past research,
varying effects on different types of students.
several recent studies have tried to estab-
lish the causal effects of remediation using
Do the Effects of Remediation Differ
quasi-experimental research designs. Much
by Student Characteristics and Ability?
of the existing research, however, focuses
Additional work by Long and Calcagno focus-
on students at the margin of passing out of
ing on Florida found that the effects of reme-
remediation and compares students who
diation differed by student background and
score just above and below the cutoff on the
demographics.35 Women, for example, expe-
remediation placement exam.30 Remediation
rienced more positive effects from placement
was found to increase the probability of col-
into remediation than did men. This finding
lege persistence at a large state university
could relate to other differences documented
in the Northeast and in four-year colleges
by gender—such as learning styles, levels of
in Ohio.31 In a study of more than 100,000
engagement, or amount of study time—and
community college students in Florida, Juan
may give clues about why remediation works
Carlos Calcagno and Bridget Terry Long
for some students but not others.
found that assignment to developmental
courses increased both persistence to the
Older students placed into remediation also
second year and the total number of credits
had more positive outcomes than did younger
completed, although not degree completion.32
peers. One explanation could be that older
A study of Texas students concluded that
students are more focused or ready to take
placement into remedial courses had little
advantage of “refresher” courses or the oppor-
effect on the number of credits attempted,
tunity to “catch up.” It could also be that
receipt of a college degree, or future labor-
market earnings among students scoring older students have a greater need for devel-
around the test-score cutoff.33 Using data opmental courses because they have been
from the National Education Longitudinal out of high school longer. If so, then older
Study of 1988, Paul Attewell and colleagues students who score just high enough not to
used a propensity score matching technique be placed in remediation might benefit from
to create observationally similar groups of taking the courses regardless of placement
students, half of whom had taken remedial status.
courses and half of whom had not, and con-
cluded it was less probable that students in Remediation’s effectiveness also appears to
remedial courses would receive a bachelor’s vary by income. Low-income students (that
degree but no less probable that they would is, students receiving Pell Grants) had more
98 THE FUTURE OF CHILDREN
Student Supports: Developmental Education and Other Academic Programs
negative outcomes in remediation than Dadger on students in Virginia’s twenty-three
higher-income peers in terms of persistence, community colleges, however, found that
associate’s degree completion, transfer rates, being assigned to three rather than two levels
and credits earned. Because income is often of remediation reduced the likelihood of
highly correlated with high school quality earning a community college credential by
and the availability of “high-level” academic 9 to 15 percentage points.38 Dadger concluded
coursework,36 the underlying cause of these that students assigned to the lowest level of
differences may be academic preparation. remedial math would have benefited if they
Because Pell Grants do not usually cover had been able to skip that remedial course.
the full costs of education, it may also be Recent research from Judith Scott-Clayton
that affordability interacts with performance
found that remedial assignment may be a
in remediation and afterwards. Further
significant discouragement to students whose
investigation of the interaction of financial
test scores underrepresent their ability.39
need and remedial experiences may clarify
these relationships.
The effects of remediation, then, are con-
siderably nuanced: remedial courses appear
More recently, research has explored whether
to help or hinder students differently by
the mixed results of earlier studies may be
state, institution, background, and academic
explained by differences in students’ aca-
preparedness. The mixed findings in earlier
demic preparation. Using data from the
research present an interesting puzzle about
public colleges in Tennessee, Angela Boatman
why remedial and developmental courses
and Bridget Terry Long examined how
have such different effects. Only by first
remedial and developmental courses affect
identifying the subgroups of students whom
the academic outcomes of students with
remedial programs appear to be helping or
varying levels of academic preparation.37 The
hindering and the delivery methods associ-
state’s system of assigning students to one of
ated with the largest effects can administra-
four levels of math and one of three levels of
tors, practitioners, and policy makers design
reading or writing enabled them to explore
and implement effective remediation pro-
the effects of more and less remediation, from
grams more broadly. Further experimentation
students who need only one developmental
with different types of instructional models
course to those who need multiple courses.
would be useful in helping to identify best
The study found that the effects of the
practices. Randomly assigning students to
courses differ by the level of student prepara-
the same remedial courses taught in different
tion, with students on the margin of needing
ways could help to identify more specifically
remediation having large negative effects and
those practices most effective in improving
students most needing remediation having
smaller negative effects and sometimes even student outcomes.
positive effects. Students in the lowest levels
of remedial writing, for example, persisted Reforming the Delivery of Remediation
through college and completed degrees at The mixed results of research have illumi-
higher rates than their peers in the next level nated critical questions regarding not only
up, thus indicating that remediation could be whether remedial programs, on average,
beneficial for students with weaker prepara- improve student academic outcomes, but also
tion. Similar research conducted by Mina which types of programs are most effective.40
VOL. 23 / NO. 1 / SPRING 2013 99
Eric P. Bettinger, Angela Boatman, and Bridget Terry Long
Educators are beginning to address those learning labs, online-learning models, and
questions as institutions start to experiment high-tech classrooms.45 The aim is to shorten
with redesigning their developmental courses. the time students spend in developmental
In their work describing recent innovations in courses, enabling them to move more quickly
developmental education, Elizabeth Zachry into their college-level courses, while also
and Emily Schneider distill the multitude of ideally creating efficiencies in the delivery
these redesign efforts into four broad types: of developmental education. Although few
reforms that shorten the time students spend researchers have rigorously evaluated the
in remedial courses, programs that com- effectiveness of technology in remedial educa-
bine basic skill attainment with college-level tion, the article by Bradford Bell and Jessica
coursework, supplemental programs such as Federman in this issue provides a review of
tutoring or advising, and interventions tar- the research to date.46
geted to students before they enter college.41
Both individual institutions and state systems
Research findings on such redesigned courses have shown increasing interest in redesigning
are enlightening. A recent quasi-experimental their developmental education curriculum to
study of an accelerated developmental English address student needs. In the fall of 2007, the
course at Chabot College found that students Tennessee Board of Regents received a three-
whose developmental English course was year grant through the U.S. Department of
combined with their college-level English Education to implement the Developmental
courses into one singular course were sig- Studies Redesign Project.47 The goals were to
nificantly more likely than students who took enable postsecondary institutions to improve
a developmental course alone to transfer to the effectiveness of their remedial math,
a four-year college, earn more college-level reading, and writing courses and to serve
credits, and earn a certificate or degree, more students better and at less cost.48 Six
although these findings were mixed for ESL colleges were selected to receive funding,
students.42 Similarly, a descriptive study of but only four were able to fully implement
the FastStart program at the Community changes in the first semester of the project.
College of Denver, which combined four Although the separate course-redesign efforts
developmental math courses into two, found differed in details, all involved a shift to using
that the program helped students to success- learning technology, both in and out of the
fully complete their developmental sequence classroom, to enable students to focus on the
at higher rates.43 In Washington state, the specific skills in which they were deficient.49
Integrated Basic Education Skills Training For example, one of the three institutions
(I-Best) program combines instruction in that reformed their developmental math
basic skills with college-level material. Results curriculum, Austin Peay State University,
from a multivariate analysis of this alterna- eliminated its developmental math courses
tive model suggest higher rates of credit and created enhanced sections of the two
accumulation for participants over time, as core college-level courses, Fundamentals
well as persistence to the second year, with of Mathematics and Elements of Statistics,
the largest gains found for adult basic educa- for students whose ACT exam scores placed
tion students and English language learners.44 them in developmental math. The college-
Remediation redesign efforts are also using level courses were linked to Structured
technological strategies such as self-directed Learning Assistance workshops, which
100 THE FUTURE OF CHILDREN
Student Supports: Developmental Education and Other Academic Programs
provided students additional tutoring and County (CCBC) also “mainstreams” students
assistance for any course material with which placed into upper-level developmental writing
they were struggling. courses directly into college-level English
courses and offers a companion ALP course
taught by the same instructor. A recent quasi-
Future research on the effects experimental study of the four-year effects
of the program concluded that ALP students
of these redesigned courses
were 29 percentage points more likely to com-
will show the extent to which plete college-level English within one year,
and 6 percentage points more likely to persist
the particular instruction
to the next year, than students who enrolled in
and delivery methods of traditional developmental English.51 The study
design, however, cannot rule out the possibil-
remedial courses affect
ity of self-selection bias—that is, that students
subsequent student academic who would elect to take the ALP companion
course may, for example, be more motivated
outcomes, thus informing
than those who did not and would therefore
administrators and policy skew the findings. But the large positive
effects at CCBC—together with the findings
makers how best to help
from Austin Peay—suggest that mainstream-
underprepared students. ing approaches merit further investigation.
Other states are also experimenting with
course redesign. Since the spring of 2012,
In her evaluation, Boatman used a regression
Virginia community colleges have taught
discontinuity research design to conclude
developmental math as a series of nine one-
that students taking redesigned courses at the
credit modules, with students taking only the
three institutions that redesigned their reme-
modules that are required for their field of
dial math courses had more positive outcomes
study and in which the diagnostic placement
than similar students both from institutions test indicates a need for improvement.52 The
that did not participate in the redesign and goal is to reduce the time required to com-
from previous cohorts at the same institu- plete remediation. Future research on the
tions.50 Austin Peay saw the largest positive effects of these redesigned courses will show
effects on persistence, suggesting that the cut- the extent to which the particular instruction
off used to assign students to developmental and delivery methods of remedial courses
math may be too high and that some students affect subsequent student academic outcomes,
who are now placed into developmental math thus informing administrators and policy mak-
courses would have better outcomes if they ers how best to help underprepared students.
were placed directly into college-level math
courses that offer additional support. Accelerating Remediation with Summer
Bridge Programs
The Accelerated Learning Program (ALP) One broad aim of several of the large-scale
at the Community College of Baltimore redesign efforts is to move students through
VOL. 23 / NO. 1 / SPRING 2013 101
Eric P. Bettinger, Angela Boatman, and Bridget Terry Long
their remedial courses more quickly. Summer and college-level courses—a remedial
bridge programs offer students a similar writing course, for example, linked with
opportunity by enabling them to take sum- an entry-level psychology course. In 2002,
mer courses at their college for several weeks the National Survey of First-Year Academic
before they begin their freshman year. The Practices found that 62 percent of responding
bridge programs are generally voluntary and colleges used the learning community model,
differ by type and length. Most, however, although at most only a small portion of the
share a common target population: first- student body participated in those communi-
generation, low-income, and minority stu- ties.57 Although a recent report suggests that
dents in need of remedial coursework. The the learning community model is difficult
goal is to help students make the academic to scale up,58 it is nevertheless a popular
and social transition to college.53 approach to remediation.
Proponents of the learning communities
To date, research on summer bridge pro-
model offer several reasons why it may
grams is limited, as few studies have adequate
be more effective than traditional models
control groups for comparison purposes.
for teaching students with low basic skills.
Descriptive studies examining persistence
Linking a course like remedial English with
and grade point average indicate that students
a course of special interest to a student may
in summer support programs tend to perform
make the material more engaging and moti-
better in their courses than students who do
vate the student to work harder. Learning
not receive such support.54 In 2009, research-
communities also offer students the opportu-
ers at the National Center for Postsecondary
nity to form deeper ties with their peers and
Research reported results from a random-
with faculty, thereby strengthening their sup-
ized experiment focusing on summer bridge
port networks and institutional attachment.59
programs in eight institutions in Texas. They
found that the program did not affect stu-
Recently, as part of its Opening Doors
dents’ persistence from the first to the second
Demonstration, the social policy research
semester, but also found some evidence that
organization MDRC conducted a random-
summer bridge students were more likely to
assignment evaluation of a learning commu-
pass college-level courses in math and writing
nities program at Kingsborough Community
in their first semester.55 This initial modest
College in Brooklyn. Analysts found that
boost in achievement, however, did not last.
students in the learning community moved
After two years, the treatment group students
more quickly through their developmental
and the control group showed no statistically
English requirements, enrolled in and passed
significant difference in the number of semes-
more courses, and earned more credits in
ters in which they had enrolled.56
their first semester overall than their peers
who were not selected to participate in a
Remediation in the Context of Learning learning community. The researchers also
Communities saw a positive effect of participating in a
Learning communities offer another model learning community on graduation after six
for helping students through their remedial years. It is worth noting that the Opening
courses. Learning communities organize stu- Doors program at Kingsborough also served
dents into cohorts that take paired remedial students who did not need developmental
102 THE FUTURE OF CHILDREN