Brown S.D., Lent R. (Editors). Career development and counseling: putting theory and research to work

Подождите немного. Документ загружается.

526 S

PECIAL

EEDS AND

PPLICATIONS

1996; Gottfredson, 1981; Vondracek, Lerner, & Schulenberg, 1986). In contrast, con-

textual developmental and social cognitive approaches, which emphasize the recip-

rocal interaction between individuals and their contexts, propose that occupational

aspirations and choices are often facilitated or constrained by adolescents’ percep-

tions of their academic potential, gender role stereotypes, and opportunities in

their environment (Gottfredson, 1981; Lent, Brown, & Hackett, 1994; Vondracek

et al., 1986). In other words, for many individuals, vocational outcomes are influ-

enced at least as much by opportunities and constraints in the environment as they

are by personal, purposive choices. Social class appears to be an important aspect of

the opportunity structure. Research conducted during the past 30 years has consis-

tently shown that parental income and education are among the strongest predic-

tors of educational achievement and occupational attainment (Berkner & Chavez,

1997; Choy, 2002; Sewell, Haller, & Ohlendorf, 1970).

Overall, members of ethnic and racial minority groups in the United States,

particularly African Americans and Hispanics, have lower educational and in-

come levels than their White counterparts (U.S. Census Bureau, 1998). Not sur-

prisingly, and despite the fact that ethnic minority parents consistently report

high academic aspirations for their children (Delgado-Gaitán, 1990; Steinberg,

1996), ethnic minority adolescents and young adults show lower academic

achievement and attainment than their White peers. For example, the percentage

of 25 to 29 year olds with at least a bachelor’s degree in 2000 was higher for

Whites (34%) as

compared to African Americans (18%) or Hispanics (10%; Na-

tional Center for Ed

ucational Statistics, 2002).

Research findings suggest that the various ecological contexts in which chil-

dren and adolescents are embedded, including the family and the school, con-

tribute to the lower academic attainment of low-income and minority youth. In

contrast to middle-class White parents, low-income minority parents often lack

the necessary knowledge and social networks to successfully support their chil-

dren’s academic achievement and attainment (Steinberg, 1996). In addition, the

learning environment and resources available in schools that serve primarily

White and middle-class students tend to be better than in schools that primarily

serve students from low-income and ethnic minority backgrounds. Teachers and

principals in low-income schools report more problems with parent involvement

and support, teacher morale and absenteeism, students’ negative attitude, and

gang activity than in schools in more affluent areas, which tend to have stronger

teachers and offer more demanding curricula (National Center for Educational

Statistics, 2002).

Another important resource, the quality and quantity of school counselors,

differs by income level and education of parents. In most inner-city public high

schools, large student case loads (ranging from 300 to 700 students per counselor)

and substantial administrative responsibilities leave counselors with little time to

help students develop career exploration and decision-making skills, plan their

academic programs, and take the necessary steps to pursue postsecondary educa-

tion. These limited counselor resources are often dedicated to students with the

highest academic achievement, leaving a large proportion of students with col-

lege potential without much guidance (Lee & Ekstrom, 1987; Rosenbaum, Miller,

& Krei, 1996). To compensate for the lack of career counseling and guidance of-

fered in schools, middle-class parents have access to private college counselors

whose advice on effective college application strategies facilitate the access of

c22.qxd 8/5/04 9:32 AM Page 526

Promoting the Career Development and Academic Achievement of At-Risk Youth 527

high-achieving students to competitive colleges and help even mediocre students

find colleges that meet their needs (McDonough, Korn, & Yamasaki, 1997).

Many governmental and private efforts have been directed to help low-income

and minority students improve their academic achievement, graduate from high

school, and increase their odds of pursuing a college degree (Arbona, 1994; Cun-

ningham, Redmond, & Meritosis, 2003; Gándara, 2001; James, Jurich, & Estes,

2001). The purpose of this chapter is to provide an overview of college access pro-

grams and review the program evaluation literature to examine to what extent

these programs have been successful in promoting access to a college education

for at-risk students. To place the efforts of these programs in context, recent re-

search findings on the predictors of college attendance and completion among at-

risk students is presented first, together with a brief overview of how researchers

have defined the characteristics of at-risk students.

PREDICTORS OF COLLEGE ATTENDANCE

AMONG AT-RISK YOUTH

Researchers in the social sciences and education have extensively examined the

predictors of school success and college attendance among socioeconomically

vantaged and disadvantaged children and youth (see Arbona, 1994; Baker &

Velez, 1996, for reviews of this literature). More recently, researchers have exam-

ined the high school and college trajectories of American youth using informa-

tion gathered as part of the National Education Longitudinal Study started in

1988 (Choy, 2002). Data for this study, known as the NELS: 88, were first collected

from a nationally representative sample of eighth graders (26,432 students across

the United States) in the spring of 1988, and follow-up surveys were conducted in

1990, 1992, 1994, and 2000. The data gathered included interviews with students,

teachers, and parents as well as students’ high school and postsecondary school

transcripts (Ingels et al., 2002). This section provides an overview of the major

findings on the predictors of college attendance and graduation among at-risk

students obtained primarily from longitudinal studies conducted with data from

the NELS:88.

EFINITION OF

-R

ISK

TUDENTS IN THE

NELS:88.

The factors most commonly used to identify children at risk academically have

been low socioeconomic background and membership in racial or ethnic minor-

ity groups. Based on these demographic factors, entire groups of children have

been labeled at risk, even though many students from low-income households and

racial and ethnic minority groups do well academically (Borman & Rachuba,

2001; National Center for Educational Statistics, 2002). Other researchers have,

therefore, attempted to refine the at-risk category by examining family and

school factors that, independent of family income and racial/ethnic group mem-

bership, distinguish students who do and do not achieve well academically.

Using data from the NELS:88, Chen and Kaufman (1997) identified five factors

that increased the odds of students dropping out of high school when socio-

eco

nomic status and membership in racial or ethnic minority groups were con-

trolled for. These factors included being from a single-parent family, having an

older sibling who dropped out of high school, changing schools two or more times

c22.qxd 8/5/04 9:32 AM Page 527

528 S

PECIAL

EEDS AND

PPLICATIONS

from first to eighth grade, low grades (C or lower) from sixth to eighth grade,

and repeating a grade between first and eighth grade. Students were considered

at low risk academically if they showed only one factor, at moderate risk if they

showed two factors, and at high risk if they showed three or more factors.

Follow-up studies with the NELS:88 data have shown that students considered at

moderate and high risk in eighth grade were less likely to complete high school,

attend college, and obtain a college degree as compared to students at low or no

risk (Ingels et al., 2002). These risk factors were associated with dropping out of

school regardless of the student’s family income or racial-ethnic group member-

ship. However, students at moderate- or high-risk status were more likely than

students at lower risk to be African American or Hispanic and to have parents

with lower levels of education (Chen & Kaufman, 1997; Horn & Chen, 1998).

Close to 58% of the 1992 high school graduates in the NELS:88 cohort had at least

one risk factor (33% had one factor, 16% had two factors, and 9% had three or

more risk factors).

Another group of students considered at risk for educational achievement is

first-generation college students, those whose parents did not continue their edu-

cation past high school. About 27% of the 1992 high school graduates in the

NELS:88 cohort were first-generation college students. Similar to at-risk students

as defined by Chen and Kaufman (1997), first-generation college students were

more likely than students who had at least one parent with a college degree to

come from low-income families and to be Hispanic or African American (Horn,

Nuñez, & Bobbitt, 2000). Even though first-generation college students are less

likely to obtain a college degree than students with college-degreed parents

(Choy, 2002; Horn et al., 2000), the occupational outcomes for both groups tend to

be similar. A study using data from the 1989/1990 Beginning Postsecondary Lon-

gitudinal Study (BPS:90/94) revealed that in 1994, five years after enrolling in

college, first-generation students who had attained a bachelor’s or an associate’s

degree were as likely to be employed, earn comparable salaries, and work in sim-

ilar occupations as their non-first-generation peers with similar degrees (Nuñez

& Caroll, 1998). These findings suggest that obtaining a postsecondary degree

serves to equalize the playing field for students from different socioeconomic and

educational backgrounds.

One of the most often cited explanations for the underrepresentation of minor-

ity and low-income students in higher education is the high incidence of school

dropouts among this population. For example, in 2000 the school dropout rates

among 16- to 24-years-olds was much higher for African Americans (13%) and

Hispanics (28%) than for their White (7%) counterparts (National Center for Ed-

ucational Statistics, 2002). Researchers have identified a long and consistent list

of social (e.g., race/ethnicity, family income, parental education, family struc-

ture) and academic factors (e.g., test scores, grades, held back in school) associ-

ated with dropping out of school (Arbona, 1994; Chen & Kaufman, 1997; Dryfoos,

1990). Studies based on the NELS:88 have confirmed these findings. Compared to

their peers who in eighth grade showed none of the risk factors (single-parent

household, a high school dropout sibling, frequent school changes, low grades,

and repeating a grade) identified by Chen and Kaufman, four years later students

at low risk (only one risk factor) were four times more likely to have dropped out

of school. Meanwhile, students at moderate (two risk factors) and high risk (three

c22.qxd 8/5/04 9:32 AM Page 528

Promoting the Career Development and Academic Achievement of At-Risk Youth 529

or more factors) were, respectively, 13 and 30 times more likely to have left school

without a degree than their no-risk peers.

Nevertheless, some at-risk students beat the odds and succeed academically.

Approximately 60% of students in the NELS:88 who showed multiple risk factors

as eighth graders graduated from high school in 1992. Chen and Kaufman (1997)

found that several family, individual, and peer factors differentiated the resilient

high-risk students from their peers who did not obtain a high school diploma. A

positive school attitude, a cohesive family, parental support for schooling, and

having friends engaged in school played a protective role in reducing the impact

of risk on the resilient students. Previous studies also have found that school en-

gagement and parental involvement are associated with academic persistence

among at-risk high school students (Dryfoos, 1990; Finn, 1993). Data from the

NELS:88 have allowed researchers to examine longitudinally factors that predict

college attendance and completion among different types of at-risk students who

beat the odds and graduate from high school (e.g., Berkner & Chavez, 1997; Horn,

1997; Horn et al., 2000).

REDICTORS OF

OLLEGE

TTENDANCE AMONG

-R

ISK

TUDENTS

The pathway to college includes five sequential steps starting with students’ as-

pirations to continue their education beyond high school and culminating with

matriculation in a four-year college. Obtaining academic skills, completing col-

lege entrance exams, and submitting college applications are the necessary inter-

mediate steps to go from aspirations to college enrollment (Horn, 1997). Results

of logistic regression analyses show that when controlling for parental education,

family income, and completion of the five steps in the college pipeline, similar

proportions of White, Asian, Black, and Hispanic high school graduates in the

NELS:88 cohort enrolled in four-year colleges. Small but statistically significant

differences in college enrollment related to parental income and education per-

sisted even when controlling for students’ academic qualifications and prepara-

tion for college (Berkner & Chavez, 1997).

Findings from the NELS:88 cohort indicated that at-risk students face difficul-

ties in completing most of the steps in the college pipeline (Berkner & Chavez,

1997). College aspirations, the first step in the pipeline, were less prevalent

among low-income eighth-grade students (58%), compared to their middle in-

come (76%) and high income (92%) peers. A large proportion of at-risk high

school graduates also lacked the academic preparation necessary to gain admis-

sion to four-year colleges. Berkner and Chavez developed a composite index of

academic qualifications for college that considered students’ high school GPA,

senior class rank, scores in aptitude standardized tests, and rigor of the high

school curriculum. According to this index, close to 65% of the 1992 high school

graduates from the NELS:88 cohort were at least minimally qualified to attend

college. However, most of these academically qualified students came from high-

income families (86%) and had college-educated parents (82%). Similarly, among

academically qualified students, those whose parents were college graduates

were more likely than their peers with nondegreed parents (90% versus 60%) to

complete steps two and three in the college pathway: taking entrance exams and

submitting college applications (Berkner & Chavez, 1997).

c22.qxd 8/5/04 9:32 AM Page 529

530 S

PECIAL

EEDS AND

PPLICATIONS

Researchers (Horn, 1997; Horn & Chen, 1998) have identified several parent,

peer, and school engagement factors that increased the odds of attending college

among academically qualified high school graduates in the NELS:88 cohort who,

as eighth graders, had exhibited two or more of the at-risk characteristics identi-

fied by Chen and Kaufman (1997). The two factors that most strongly predicted

attendance at a four-year college among at-risk students, when controlling for

standardized academic achievement in eighth grade and highest level of math

taken in high school, were:

1. Having parents who reported speaking frequently with their children

(when they were in the tenth grade) about school-related issues.

2. Students’ reports that their friends planned to attend college (Horn, 1997).

Friends seemed to have the strongest influence; at-risk students who had re-

ported during their sophomore high school year that most or all of their friends

planned to attend college were four times more likely to have enrolled in college

two years after high school graduation than their at-risk peers with similar aca-

demic qualifications but who had reported that none of their friends planned to

attend college (Horn & Chen, 1998). Participating in high school outreach pro-

grams, receiving help with college applications from school staff or counselors,

and preparing for entrance exams also increased the odds of at-risk high school

graduates’ attending four-year colleges (Horn, 1997; Horn & Chen, 1998). Similar

factors predicted college attendance among first-generation college students

(Horn et al., 2000).

The parent, peer, and school factors that had a positive influence on the

chances of at-risk students attending a four-year college did not seem to have

much effect on increasing the odds of their persisting in college once they en-

rolled (Horn, 1997). Access to college is important because it is the first step to-

ward obtaining a college degree. However, it is obtaining the degree, and not just

college attendance, that may open doors to career choices and opportunities.

Adelman (1999) examined variables that predicted college completion among a

nationally representative group of high school students (High School and Beyond

sophomore cohort of 1,980 tenth graders), who, at some point in the 11 years after

high school graduation, had attended a four-year college. He found that the

strongest predictors of bachelor’s degree attainment were continuous enrollment

in college and the academic resources students brought from high school, mea-

sured in terms of quality and intensity of high school curriculum, standardized

test scores, and class rank. Of these three academic indexes, the intensity and

quality of the high school curriculum were the academic factors most strongly re-

lated to college completion among ethnic minority students (Adelman, 1999). Un-

fortunately, not all secondary schools provide students access to a rigorous

curriculum that includes advanced mathematics beyond Algebra 2, the three

basic laboratory sciences, foreign languages, and advanced placement courses

(Spade, Columba, & Vanfossen, 1997). Furthermore, low-income and ethnic mi-

nority students generally have less access to advanced curricula, even in schools

where such opportunities exist (Monk & Haller, 1993).

Results of longitudinal studies with the NELS:88 data confirm the generalized

perception that a large proportion of high school graduates from low-income

backgrounds lack the academic skills necessary to gain admissions to four-year

colleges. Furthermore, many at-risk students with adequate academic credentials

c22.qxd 8/5/04 9:32 AM Page 530

Promoting the Career Development and Academic Achievement of At-Risk Youth 531

fail to complete two important steps for college admissions: taking college en-

trance exams and submitting college applications. In sum, research findings con-

sistently point to several barriers or impediments to college access faced by at-risk

students, including lack of aspirations, inadequate academic preparation, peers

who are not engaged in school, parents who are not involved in students’ schooling

or knowledgeable about college planning, and lack of financial resources. Fortu-

nately, interventions provided by school counselors and outreach programs seem

to increase the chances of at-risk students attending a four-year college. Only a

small proportion of at-risk high school graduates (5%) in the NELS:88 study sur-

vey indicated that they had participated in college outreach programs; however,

these students were twice as likely to attend a four-year college than their peers

who had not participated in these programs (Horn, 1997). These findings under-

score the need for outreach programs to facilitate the academic development and

college access of low-income and at-risk children and youth.

YPICAL

EATURES OF

OLLEGE

CCESS

ROGRAMS

Since the 1980s, a large number of programs have been implemented to help low-

income and ethnic minority students achieve academically, graduate from high

school, and attend college. College access programs vary along several dimen-

sions, including sources of funds, type of agency that implements the program,

characteristics of program participants, services offered, and strategies used to

evaluate program outcomes. In this section, findings from several reviews that

have identified the typical features of these programs are first summarized

(Ar

bona, 1994; Cunningham et al., 2003; Gándara, 2001; Swail, 2000). Then, eight

programs that exemplify specific program features are described.

Results of an extensive survey of all types of precollege outreach programs in

the United Stares and its territories conducted by the College Board in 1999/2000

(Swail, 2000) indicated that more than half (54%) of the 1,100 programs that re-

sponded to the survey were funded by the federal government. Other sources of

funds included state governments (15%), universities (9%), and corporations and

foundations (20%). These outreach programs were sponsored and implemented

by diverse institutions including colleges and universities (57%), schools (16%),

community-based organizations (13%), and other private nonprofit entities such

as foundations or corporations (14%).

For the most part, precollege outreach programs recruit low-income, minority,

and first-generation students offering services to individual students, entire

classrooms, or whole schools (Arbona, 1994; Gándara, 2001; Swail, 2000). About

10% of the programs in the College Board’s national survey (Swail, 2000) served

students younger than middle school, and about half of the programs targeted

only high school students. Interventions provided by college outreach programs

primarily focus on college and career awareness, social support, and academic

enrichment. The most common components included in these programs are coun-

seling, academic enrichment, parental interventions, social and cultural enrich-

ment, mentoring, college scholarships, and evaluation of program’s activities

(Arbona, 1994; Cunningham et al., 2003; Gándara, 2001; Swail, 2000).

Counseling In most outreach programs, counseling services, offered individually

and in small groups, emphasize the dissemination of college information and ad-

vising. Specific services include career planning and decision making, academic

c22.qxd 8/5/04 9:32 AM Page 531

532 S

PECIAL

EEDS AND

PPLICATIONS

advising, social skills development, study skills training, college fairs and career

days, and assistance with the completion of college and financial aid applications.

In some programs, counselors also provide personal emotional counseling.

Academic Enrichment Academic enrichment activities are provided to increase

students’ school retention, graduation rates, and preparedness for college. Aca-

demic interventions include tutoring, SAT test preparation workshops, and the

opportunity to take college-level courses while in high school. Some programs

offer extensive courses to complement the school’s curriculum in areas such as

reading, writing, and analytical skills, whereas other programs attempt compre-

hensive changes in the content and delivery of the school curriculum, with the

tent of offering most students access to rigorous college preparatory course-

work. Academic enrichment activities are offered during regular school hours,

after school, and during the summer.

Parental Intervention Research findings as well as field experiences indicate that

parental involvement is a critical factor in students’ academic achievement. Swail

(2000) reported that about two-thirds of college access programs offered a parental

component and, in about one-fifth of the programs, parents were required to ac-

tively participate in program activities. Services specifically provided for parents

include orientation sessions about the program, information sessions on college re-

quirements, college and financial aid applications workshops, and psychoeduca-

tional interventions. These interventions emphasize adolescent development and

how to help students succeed in school, including communication, discipline, and

monitoring homework. Some programs also recruit parents as volunteers and mem-

bers of school teams and advisory committees.

Social and Cultural Enrichment Social and cultural enrichment activities are

designed to help students become more knowledgeable of the world around

them and to develop a network of peers that promotes and supports a culture of

“col lege

-going.” In some cases, graduates of the programs are recruited to serve

as tutors and counselors for the younger cohorts. Social and cultural enrichment

activities organized by programs include leadership development, goal-setting,

and confidence-building workshops, community service activities, cultural ac-

tivities connected to the arts, visits to college campuses, and field trips to recre-

ational and cultural sites. It is expected that these experiences will help students

to develop a positive sense of self and motivate them to pursue their academic

and career goals.

Mentoring Formal mentoring programs pair participants with volunteers who

provide one-to-one friendship, guidance, and support to students. Mentors usu-

ally are older peers, university or school staff and faculty, volunteers from the

community, corporate personnel, and professionals. It is expected that the men-

toring relationship will improve students’ academic performance indirectly by

improving their attitudes about school and raising their personal goals and moti-

vation. Mentors also support youth during times of personal or social stress and

help in decision making. In career mentoring programs, students are paired with

adults employed in the student’s field of interest who are willing to help students

c22.qxd 8/5/04 9:32 AM Page 532

Promoting the Career Development and Academic Achievement of At-Risk Youth 533

become familiar with the world of work in general along with the requirements of

their specific occupational field.

College Scholarships Scholarship programs usually promise middle and high

school students money to cover college tuition costs as a way to motivate them to

pursue the necessary steps to attend college. Last-dollar scholarships provide funds

to cover the difference between all other scholarships and aid a student receives

and the actual college costs. Statewide scholarship programs generally provide

full tuition waivers for public institutions to students who meet academic perfor-

mance eligibility requirements (Southern Regional Education Board, 2000). Other

programs offer college money only for students pursuing specific fields or careers.

Program Evaluation Descriptions of educational programs for at-risk youth

abound in the literature; however, rigorous evaluations of such programs are

scarce. Most rigorous evaluations use an experimental design where applicants

who meet eligibility criteria for the program are randomly assigned to either an

experimental group, who is given the opportunity to participate in the program,

or to a control group of nonparticipants. Because it is expected that randomiza-

tion distributes unmeasured characteristics of students equally between the two

groups, outcomes observed in the control group are a good estimate of what

would have happened without the program’s intervention. Less rigorous empiri-

cal program evaluation designs use for comparison purposes information ob-

tained from national and local data or from groups of students in the same or

other schools who share participants’ sociodemographic characteristics.

As part of their evaluation efforts, most programs keep track of the percentage

of participants who complete the program, graduate from high school, and enroll

in college (Swail, 2000). Few programs use pre-post designs to compare students’

behaviors before and after receiving the program’s intervention or compare out-

comes observed in participants to a control group of similar students not exposed

to the program (Arbona, 1994; Cunningham et al., 2003; Gándara, 2001). Never-

theless, studies that carefully examine programs’ outcomes, even if they do not

rigorously assess programs’ impacts, provide valuable information to policymak-

ers and program managers.

DESCRIPTION OF COLLEGE ACCESS PROGRAMS

FOR AT-RISK STUDENTS

In this section, eight college access programs that exemplify different ap-

proaches to service delivery, selection of participants, and evaluation of program

outcomes are briefly described. These eight programs are meant to serve as ex-

amples of types of programs and not as a representative sample of all the types

of programs that have been documented in the literature. In recent years, the

American Youth Policy Forum has published several comprehensive compendi-

ums of school and youth programs that have documented outcomes across a

wide range of academic achievement indicators (James, 1997, 1999; James et al.,

2001; Jurich & Estes, 2000). Other publications have provided detailed descrip-

tions of a wide variety of college access programs (e.g., Cunningham et al., 2003;

Gándara, 2001; Southern Regional Education Board, 2000). Interested readers

may consult these resources.

c22.qxd 8/5/04 9:32 AM Page 533

534 S

PECIAL

EEDS AND

PPLICATIONS

The eight programs described here have been implemented in more than one

site and have reported a behavioral outcome evaluation by either comparing stu-

dents’ behaviors before and after the program’s intervention or by comparing

program participants to a similar group of nonparticipants. Programs differed in

the population of students they served. Most programs work with high school

students by offering services to individual students or to specific cohorts of stu-

dents within a high school. Other programs target students before they reach

high school. In some cases, programs have attempted to institute school-wide

changes at the middle school level. College sponsorship programs have attempted

to motivate entire classrooms of middle school students by guaranteeing college

scholarships to students who persevere and graduate from high school.

IGH

CHOOL

ROGRAMS

HAT

ERVE

NDIVIDUAL

TUDENTS

Two programs, Upward Bound and Career Beginnings, served high school stu-

dents individually and reported the most rigorous outcome evaluations.

Upward Bound Upward Bound is a federally funded program established in 1965

to help low-income, first-generation college students enroll and succeed in col-

lege. Currently, there are 563 regular Upward Bound projects across the country

serving approximately 44,000 students. Most of these projects are sponsored by

two- and four-year colleges, although some projects are hosted by high schools

and community-based organizations. Projects provide a wide variety of services

to high school students (freshmen to seniors) during the school year and the

sum

mer, including formal instruction in high school subjects (e.g., English,

math, sci

ence), tutoring, mentoring, counseling, career planning, cultural and

recreational field trips, visits to college campuses, and college planning activities

(e.g., workshops on completing financial aid and college applications). In most

programs, the summer session includes an intensive instructional program and a

residential component on a college campus. Services in Upward Bound are pri-

marily directed at individual students who tend to be educationally motivated,

have few behavioral or disciplinary problems, and earn mostly Bs or Cs in school

(Moore, 1997; Upward Bound Program, n.d.).

Upward Bound is undergoing a six-year longitudinal evaluation commissioned

by the Planning and Evaluation Service of the U.S. Department of Education

(Myers & Schirm, 1999). Of students who were eligible applicants to a nationally

representative sample of 67 Upward Bound programs from 1992 to 1994, the eval-

uators randomly assigned 1,524 to Upward Bound and 1,320 to a control group of

nonparticipants. Myers and Schirm (1999) reported that retention in Upward

Bound tends to be low; typically, participants remained in the program for about

19 months. Upward Bound students had slightly higher educational expectations

and had earned a few more high school credits in math and social studies than

did students in the control group. The two groups did not differ in other high

school outcomes associated with college-going, including participation in school

activities, talking with parents about school, or parental involvement at school.

Furthermore, a similar number of students in the treatment and control groups

attended two- and four-year colleges. However, students who entered Upward

Bound as 10th graders and remained in the program for 13 to 24 months were

more likely than their counterparts in the control group to attend a four-year

c22.qxd 8/5/04 9:32 AM Page 534

Promoting the Career Development and Academic Achievement of At-Risk Youth 535

col

lege, indicating that longer exposure to the program led to stronger effects.

Once in college, Upward Bound students earned more nonremedial credits from

four-year colleges, were more likely to receive financial aid (loans, grants, and

work study appointments), and reported more informal contacts with advisors

and faculty and greater participation with peers in study groups and school

clubs. The outcomes related to postsecondary education must be interpreted with

caution because the last published report of the evaluation included data gath-

ered when most students were in their last years of high school, and only about

one-fourth of them had entered college (Myers & Schirm, 1999).

Even though Upward Bound seems to have limited impact on students’ aca-

demic behaviors during high school, it shows stronger effects once students en-

roll in college. Upward Bound also seemed to have a larger impact on more

disadvantaged students. Among students with lower educational expectations

(less than a college degree) and lower academic performance, students who par-

ticipated in Upper Bound completed more high school credits, were less likely to

drop out of school, and were more likely to attend college than their counterparts

in the control group. Similarly, Upward Bound had stronger impact on high

school retention and course taking among low-income students.

Career Beginnings Career Beginnings brings together high schools, colleges,

and businesses in working partnerships to facilitate college attendance among

low-income high school juniors and seniors of average academic achievement.

Students participate in structured workshops on career awareness and college

readiness topics, remedial classes in basic academic skills, entry-level summer

jobs, and mentoring relationships with adults from the business or professional

community. A rigorous evaluation of the program implemented in seven sites

(of 24) across the country was conducted by the Manpower Demonstration Re-

search Corporation (MDRC; Cave & Quint, 1990). Eligible applicants to the pro-

gram in each of the seven sites were randomly assigned in equal numbers either

to the program or to a nonparticipant control group. The evaluation of the pro-

gram’s impact is based on the experiences of 1,233 youth who responded to two

follow-up interviews (78% response rate) conducted one and two years after

random assignment.

In terms of college attendance, Career Beginnings did not lead to an overall in-

crease in access to postsecondary education. However, a larger proportion of stu-

dents who had participated in the program (53%) attended a four-year college as

compared to students in the control group (48%). In contrast to the results re-

ported for Upward Bound, students with greater academic or economic barriers

to college attendance were not helped more by Career Beginnings. Cave and

Quint (1990) concluded that receiving more services was, however, associated

with college-going and that mentoring seemed to serve as a catalyst that in-

creased the effectiveness of other services. Attending three types of services—

college entrance exam preparation classes, college fairs and college information

sessions, and job-readiness workshops—was related to college enrollment,

though participating in tutoring, remedial courses, and workshops in study-skill

and test-taking strategies were not. Because the characteristics of students who

participated in the different types of activities were not controlled for, these find-

ings must be interpreted with caution. It is possible that students with better aca-

demic skills were more likely to attend college and that, because they did not

c22.qxd 8/5/04 9:32 AM Page 535