Latino Students and the Educational Pipeline

PART III
Pathways to the Bachelor?s
Degree for Latino Students
The third component of our three-part series focuses on students who attained
a bachelor?s degree and what it took to get there. We used multiple
regression analysis to determine the factors that seemed to matter on the
pathway to the BA. The appendix of this report provides methodological details
to this analysis.
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-ii
About the
Authors
Watson Scott Swail is President of the Educational Policy Institute and an internationally-
recognized researcher in the area of educational opportunity.
Dr. Swail?s work has been widely published in such education journals as
Change, Phi Delta Kappan, the Chronicle of Higher Education, and the International
Management of Higher Education (IMHE). Prior to founding EPI, Dr.
Swail served as Director of the Pell Institute in Washington, DC, Senior Policy
Analyst at SRI International, and Associate Director for Policy Analysis at the
College Board. Dr. Swail earned a Doctorate in Educational Policy from The
George Washington University, Washington, DC, a Master?s of Science from
Old Dominion University, Norfolk, Virginia, and a Bachelor?s in Education
from the University of Manitoba, Winnipeg, Manitoba.
Alberto F. Cabrera is a Senior Scholar with the Educational Policy Institute
and Professor of Education at the University of Wisconsin-Madison. Dr.
Cabrera is thoroughly familiar with the use of national databases and specializes
in research methodologies, postsecondary opportunity, and economics
of education. His research has been released in such top tier outlets as
Research in Higher Education, Review of Higher Education, Journal of
Higher Education, Economics of Education Review, New Directions for Institutional
Research, among others. He also serves on the editorial boards of
Research in Higher Education, Review of Higher Education, and Journal of
Higher Education.
Chul Lee is a graduate student at the University of Wisconsin-Madison. His
main research interests revolve around how schools contribute to college
preparation and the ways in which school effectiveness can overcome social
inequality. Trained as a quantitative methodologist through K-12 and higher
education, Mr. Lee specializes in a number of national databases, including
NELS:88, SASS, and CCD.
Adriane Williams is a Research Associate with the Educational Policy Institute.
She began her career as a Research Specialist for the Council of the
Great City Schools, an urban school advocacy organization, and continued
from there as a high school teacher. Her areas of research interest include
the middle school role in preparing the children of non-college graduate parents
for postsecondary options, high school reform, and postsecondary success
for members of underserved populations. Ms. Williams is a doctoral
candidate at the University of Wisconsin-Madison in the Educational Policy
Studies Department. She earned her Master of Education from The George
Washington University in Washington, DC and her Bachelor of Arts in Economics
and French from Wellesley College in Wellesley, MA.
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-iii
Table of
Contents
About the Authors................................................................................................ ii
Table of Contents ............................................................................................... iii
Executive Summary.............................................................................................1
Introduction .........................................................................................................4
Descriptive Overview...........................................................................................9
Research........................................................................................................... 15
Findings............................................................................................................. 15
Policy Implications............................................................................................ 21
References........................................................................................................ 26
Appendix A ? Methodology .............................................................................. 30
Appendix B ? Descriptive Statistics ................................................................ 32
Appendix C ? Delta P Values ........................................................................... 34
Appendix D ? Beta Values ............................................................................... 36
Appendix E ? Persistence as a Longitudinal Model ...................................... 38
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-iv
Preface
esearchers, policymakers, and educators as a whole often wonder
what becomes of students as they progress through the educational
system. As a former teacher, I think back to students I taught whose
names are now lost, but whose faces and personalities remain very much intact.
I often wonder what happened to them since we last met. Did they finish
high school? Go on to college? Get married and have children? Did they
meet their personal goals? Ultimately, I want to know if things worked out for
them. The memories of these students still mean a lot to me. They helped
shape me into the individual I am today, and they?well, most of them?made
my life much, much better just through the opportunity to get to know and
work with them. Unfortunately, as with most teachers, I am left mostly with
memories.
I mention this because knowing what becomes of students is a very critical
part of the development of public policy and sound educational practice. But
like teachers, only rarely do we ever get a glimpse into the lives of past students.
This report is one of a series of three reports on Latino students in the educational
pipeline, all of which are available for free download on the web at
www.educationalpolicy.org. The purpose of this series is to provide a sense
of the challenges facing Latino youth compared to White youth on the pathways
to postsecondary education and the baccalaureate. The series relies
on data from the National Educational Longitudinal Study (NELS), sponsored
by the National Center for Education Statistics in 1988 to follow 8th grade
students from middle school through to the workforce. In total, over 26,000
8th-grade students were surveyed in 1988, with followup surveys in 1990
(10th grade), 1992 (12th grade), 1994 (2 years after scheduled high school
graduation), and finally in 2000 (8 years after scheduled high school graduation).
NELS gives us the best glimpse of students in and beyond the educational
pipeline in America.
While we cannot answer questions about what happened to James, Sarah,
or Juan, we can show trends based on students as a whole and certain subsets.
We can see if these students graduated from high school, if and where
they went to postsecondary studies, and what?s happened to them since.
Because NELS is a nationally-representative and randomly-assigned database,
we have a fairly accurate portrayal of students in America. The one unfortunate
truth is that we can?t look at the state or local level. The sampling
design doesn?t allow that type of specificity.
Still, this is a magnificent research tool that provides us with a glance into
our future through a look at the past experiences of the NELS cohort. We can
wrestle with what these data mean and try to assess what educational and
social policies can make a difference. While it is true that NELS is somewhat
R
Dr. Watson Scott Swail
President
Educational Policy Institute
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-v
dated (the 1988 8th grade class?), one must remember that it is the power
of time that makes this database so unique: 12 years following one cohort of
students.
Many researchers have analyzed the information from NELS since the first
database was released in 1991. Some were commissioned directly by the
US Department of Education. Others, like us, received grants to study certain
aspects of NELS, and still others include university-based researchers and
graduate students who were simply interested in what NELS had to say. Our
purpose in this study, supported by a generous grant from Lumina Foundation
for Education, is to focus in on the Latino population as they completed
middle school, made their way through high school, and looked toward postsecondary
education and the workforce. Throughout the report, we compare
Latino students with White students. We omitted other race/ethnic groups
not because they are less important, but because discussion of more than
the two groups of specific interest tends to get overly complex.
I would also like to thank Alberto Cabrera, a senior scholar for EPI and a professor
at the University of Wisconsin, for his leadership during this series. As
well, Chul Lee provided exceptional data support and Adriane Williams
helped us with the final reporting of these findings. I also must acknowledge
Tina Gridiron Smith of Lumina Foundation for Education, who understood the
importance of this effort and provided unwavering support.
After working with these data for the past 10 years, I feel like the NELS students
are mine. While I can?t find out what happened to my middle school
students back in Winnipeg, Manitoba, and Hampton, Virginia, I have a pretty
good idea what happened to the NELS students of 1988. I think you?ll find
the discussion fascinating.
Dr. Watson Scott Swail
President, Educational Policy Institute
April 4, 2005
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-1
Executive
Summary
For many years the question of how to provide Latinos with greater access to
and through higher education has lingered. Although we have learned much
about Latinos through sponsored research, we lack a comprehensive picture
of the overall impact of various sources on educational attainment. This
study seeks to fill the gap by following students in the educational pipeline
from the moment they and their families begin to aspire to postsecondary
studies to the point of degree completion.
Less than one quarter (23.2 percent) of Latino postsecondary students
graduate with a four-year degree within 10 years of leaving high school?less
than half the rate of White students (47.3 percent). But by taking into consideration
student and family characteristics, postsecondary aspirations and
planning behaviors, secondary school activities, postsecondary activities,
and financial support factors, this study shows that the BA degree persistence
gap between Latino and White students can be dramatically reduced
by taking action in specific areas.
The analysis in this report was guided by Swail?s (2004) Integrated Model for
Student Success (IMSS) and the Pathways to College model (Cabrera and
LaNasa, 2000). Swail?s model asserts that the student experience is the result
of an interaction of cognitive, social, and institutional factors. Whereas
the cognitive and social factors impact students? abilities to persist and succeed
in education, the ability of various educational institutions and agencies
to understand these factors and make appropriate reflections about
their practice and service is also critical to students? potential success.
The sample for this study was drawn from the National Educational Longitudinal
Study of 1998 (NELS:88) and only Whites and Hispanics in the cohort
groups from 1988 base year through 2000 fourth follow-up survey are selected.
The stratified sample strategy followed an original sample of 8th
graders in 1988 to 2000; 12 years after expected high school graduation.
This analysis is based on those 8th grade students who become postsecondary
attendees between 1992 and 2000.
Research Findings
The results of this analysis fall in line with previous research on Latino students:
socioeconomic status, parental expectations, planning, course-taking
patterns, and student postsecondary behaviors have a significant impact on
postsecondary degree completion. These findings also present very clear direction
for further research and policy focuses because the factors with the
most impact?planning and postsecondary behavior?produce changes of
more than 40 percent, and some cases 60 percent, in the probability of
completion.
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-2
Family and Student Characteristics. The findings indicate that Latina students
are 20 percent more likely to complete a four-year degree than their
male counterparts. Middle-income Latino students had a 17 percent higher
probability of completing a four-year degree than low-income Latino students.
Affluence had no significant effect for Latinos.
Expectations and Aspirations. Expecting their children to attend some college
or to get a bachelor?s degree had no significant effect for Latinos, but
parental expectation of advanced degrees had a large and significant effect
for Latinos, demonstrated by an increase in the probability of completion by
46 percent. Latino students planning for some college increased the probability
of BA completion by 48 percent, and those who planned for a bachelor?s
degree increased the probability by 53 percent.
Preparation for Postsecondary Education. Receiving help in completing applications,
applying for financial aid, and writing essays produced no significant
effects for Latinos or Whites. Course-taking patterns, however, did produce
positive effects. Taking pre-calculus and calculus produced positive effects
for both Whites and Latinos with increases of 20 and 12 percent respectively.
According to this analysis, remedial math served no one. However,
remediation in English proved positive for Latinos with an increase of
26 percent.
Postsecondary Activities and Experiences. Beginning postsecondary studies
at a four-year institution increased the probability of completion by 29 and
35 percent for Latinos and Whites respectively, and maintaining continuous
enrollment increased the probability by 60 and 42 percent respectively. Additionally,
earning a GPA between 2.50 and 3.19 increased the probability of
completion by 47 percent for Latinos and 42 percent for Whites and earning
a GPA between 3.20 and 4.00 increased the probability of completion by 62
percent for Latinos and 45 percent for Whites. Choosing to delay enrollment
between high school and college reduced the probability of completion for
Latinos by 20 percent.
Financial Aid. There was no difference in the effect of receiving grants, loans,
or participating in work study programs.
Policy Implications
Postsecondary planning, academic preparation, and taking the right steps in
college matter for Latino persistence. These are the areas where we found
the most significant effects. It is important to note that the nature of the
analysis means that the size of the effect is dependent on the presence of
all of the other factors as well. We present policymakers and practitioners
with a set of options at several important levels: middle school, high school,
and postsecondary institution.
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-3
Middle School. The middle school level is critical according to our findings
because Latinos who enter high school with a plan for any type of postsecondary
study?with a sense of the purpose of their high school work?are far
more likely to graduate college than those who have no plan. In fact, Latino
students who had a plan to attend college improved the chances of graduation
by 48 percent compared to other students. Of course, having a plan to
attain a BA degree improves the chances by an additional 5 percent (53 percent).
Thus, developing college knowledge among students and families can
have a major impact on future educational opportunities.
High School. Academic preparation for college must begin immediately with
the first math courses that students take. Latinos need to be enrolled in and
master Algebra I no later than the ninth grade in order to reap the benefits of
high mathematics achievement on postsecondary persistence. Latinos who
take more than three years of mathematics beginning with Algebra I have a
higher probability of graduating from college than those who take fewer than
three courses. In addition, remediation in English is also important for Latinos.
It is during high school that solid advising must take place about postsecondary
education and the type of institutions students and families should
consider. Latinos seeking a four-year degree are somewhat disserved by beginning
at a two-year institution.
Postsecondary Institution. Helping Latinos maintain continuous enrollment
and providing academic support while they are enrolled is the primary role of
the postsecondary institution in the effort to improve the completion rates of
Latinos. The provision of academic support is evinced as important by the
large effect of grade point average on the probability of completion. Of
course, being academically prepared when they enter is important for all
students, but being able to take advantage of academic support services
undoubtedly helps students to maintain high performance in an academic
culture that differs significantly from high school.
Conclusion
Latino students who are supported by their families in the pursuit of a postsecondary
education, create a plan by the eighth grade, take three years of
mathematics or more, start at a four-year institution, maintain continuous
enrollment and a GPA of 2.50 or above can close the gap between Latinos
and Whites in the completion of four-year degrees. The findings in this study
do not suggest that the work to make these things a reality is easy, but they
do suggest where to begin.
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-4
Introduction
he research literature is full of papers discussing the plight of Latino
students. Most focus on the barriers that these students face as
compared to others, most notably White students, but also Asian,
Black, and Native Americans. Some studies focus on particular school districts
or college campuses. Others use broader databases, while still others,
unfortunately, use little data and even less analysis.
Lumina Foundation for Education was generous enough to provide the Educational
Policy Institute with a grant to study Latino students in the educational
pipeline using the most powerful longitudinal database available: the
National Educational Longitudinal Study (NELS). Started in 1988, the NELS
study randomly sampled 26,000 8th-grade students, and followed them up
four times over the course of the next 12 years. The final followup, in 2000,
provides us with a unique glimpse into the lives of this student cohort 8
years after scheduled graduation. This long-range view allows us to see what
happened to these students in high school, postsecondary education, and
into the job market.
This report series is divided into three sections to answer three questions
regarding Latino progress through the educational pipeline:
Question One. What happened to 1988 NELS 8th-grade Latino students
in the 12 years that followed? How did their progress compare
with White students throughout the various stages of the educational
and occupational pipeline? (Part I)
Question Two. What are the primary differences between Latino and
White students for those who completed a BA and other levels of
education? (Part II)
Question Three. What factors seem to have the most impact on Latino
students? ability to navigate the educational system and
achieve higher levels of learning? (Part III)
This report, Part III, focuses on the latter question. As with the other two
parts of our series, we used the NELS database to gain a perspective on
what really matters to those students who earn a bachelor?s degree.
For many years, the question of how to provide Latinos with greater access
to and through higher education has lingered. The importance of this question
is heightened by the fact that in the coming years Latinos in the US will
be the largest minority by a significant margin. Although we have learned
much about Latinos through sponsored research, we lack a comprehensive
picture of the overall impact of various factors on educational attainment.
In this report we want to inform readers about what factors relate to the persistence
of Latino students to BA degree completion. Latino postsecondary
T
e
l
t
r
m
h
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-5
students have a 23.2 percent probability of graduating with a four-year degree
within 10 years of leaving high school?less than half the rate of White
students (47.3 percent). Our analysis suggests that it is possible to reduce
this gap significantly by adjusting public policy in several critical areas.
By taking into consideration student and family characteristics, postsecondary
aspirations and planning behaviors, secondary school activities, postsecondary
activities, and financial support factors, this study shows that Latino
persistence to degree completion can be directly affected by taking action
in specific areas. Informing the educational expectations that parents
have for their children, helping them to develop high aspirations in their children,
and supporting students in making postsecondary plans by the 8th
grade are vital steps to take toward the goal of postsecondary persistence.
As well, ensuring that students who aspire to a four-year degree begin at
four-year institution immediately after graduation, maintain continuous enrollment,
and perform well while in school are critical actions at the postsecondary
level.
Significant Challenges
The educational experiences of Latinos from the earliest grades through
postsecondary completion are colored by significant challenges. An analysis
of data from the Early Childhood Longitudinal Study found that as early as
kindergarten, Latinos score on average 5.5 points lower than Whites (45.5)
on mathematics assessments. The average of 40 depends heavily on high
scoring Cubans (46.1) and South Americans (42.4). All other Latino groups
score much lower with the highest score at 39.7 by Salvadorans (Center
2004; Pew Hispanic Center 2004; PHC 2004).
Once Latinos arrive in elementary school this gap continues. An analysis of
NAEP data shows that Latinos and Whites have starkly different performances
on math assessments. Thirty-four percent of Hispanics score in the
first (lowest) quartile with only 11 percent in the fourth (highest) quartile.
Conversely, 16 percent of Whites are in the first quartile with 32 percent in
the fourth quartile (PHC 2004).
In secondary school, where academic performance becomes important for
college admission and success, Latinos are less likely than White and Asian
students to take a rigorous set of mathematics courses or demonstrate solid
mathematical skills on college entrance tests. A prior analysis of NELS data
shows that Latino completion of three years of math or more falls below the
national average by 14 percentage points at 42 percent (Swail, Cabrera et
al. 2004). This was accompanied by the high rate of remediation in math at
16 percent compared with 8 percent for Whites. Latinos score below the national
average on the ACT and the SAT by 2 and 75 points respectively. White
In secondary school,
where academic
performance
becomes important
for college admission
and success, Latinos
are less likely than
White and Asian
students to take a
rigorous set of
mathematics
courses.
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-6
students regularly score above the national average (CGCS 2001a; CGCS
2001b).
Possibly the result of the academic challenges confronted to this stage in the
educational pipeline, Latinos are less likely to graduate from high school or
receive a GED than any other group (86.4 percent vs. 92.3 percent; (Swail,
Cabrera et al. 2004). Two-thirds of Latinos do enroll in postsecondary institutions,
but the majority (40 percent) enrolls in two-year institutions compared
with 30 percent of Whites. Twenty-two percent of Latinos enroll in four-year
institutions whereas 41 percent of Whites do. The end of postsecondary careers
for Latinos in the NELS:88 database is largely disappointing. As previously
stated, 23.2 percent received four-year degrees within eight years of
graduating from high school compared to 47.3 percent of Whites students.
This study seeks to understand the set of circumstances that contribute to
this inequitable outcome. More specifically, we ask: For Latinos who completed
four-year degrees, what factors affected their probability of degree
completion and to what extent?
Integrated Model of
Student Success
This analysis was guided by a model based on the Geometric Model of Persistence
and Achievement (Swail 1995; Swail, Redd et al. 2003). The model
asserts that the student experience is the result of an interaction of cognitive,
social, and institutional factors, which are represented in Figure 1 as
the three sides of a triangle. Whereas the cognitive and social factors impact
students? abilities to persist and succeed in education, the ability of various
educational institutions and agencies to understand these factors and make
appropriate reflections about their practice and service is also critical to a
students? potential success. Thus, dynamic changes on the systemic side of
the model allow for the inclusion of K-16 coordination and put forth the idea
that there are multiple institutions at play producing a larger system in which
students and families interact and function. In addition to the geometric
model, we also rely on the pathways to college model (Cabrera and LaNasa
2000). This model argues that success in college begins in middle school
when parents and children being to aspire and plan for college.
The foundation of the triangle consists of systemic factors, K-16 coordination,
outreach programs, financial aid, climate and diversity, and facilities
and services. The first two elements can contribute to the development of
cognitive and social factors conducive to postsecondary success early in the
academic experiences of students.
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-7
Exhibit III-1. Swail?s Integrated Model of Student Success
The cognitive factors consist of postsecondary planning, academic preparation,
college knowledge, and academic integration. Although the first three
elements happen prior to entering a postsecondary institution, they are crucial
to postsecondary success and can be the result of early interaction with
postsecondary institutions through K-16 coordination and outreach programs.
Academic integration is taken from the larger research on persistence,
specifically Tinto?s (1975) student integration model.
The social factors consist of family encouragement, student aspirations, social
integration, and development. Again, the family encouragement and
student aspiration elements must begin far in advance of the postsecondary
institution. In fact, they are crucial to ensuring postsecondary access. It is
these elements that influence the pre-collegiate cognitive factors. The social
integration and development elements to be considered in the context of
student development theories are supported in the research literature by
Tinto (1975) and Pascarella and Terenzini (1991; 2005).
In addition to the elements found on the different sides of the triangle is the
geometric character of the model itself. Just as Swail?s geometric model
suggests, when one side of the triangle is lacking, it changes the character
of the entire interaction and potentially weakens the structure. In other
words, all of the elements working in concert produce the ideal outcome. In
less than ideal circumstances, it is possible for elements from one side to
contribute to the strengthening of elements from the other. For instance, in
the event that families are not encouraging and students have low aspirations,
one can expect to see a lack of academic planning and preparation. At
this point, systemic elements can contribute to the college knowledge of
families and students potentially promoting family encouragement and helping
both the families and students to develop higher aspirations (Auerbach
2001).
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-8
Interaction is the key to this model. The condition of each element on each
side of the triangle has an impact on the student outcome. There are four
possible outcomes for students: secondary school dropout, high school
graduate, postsecondary dropout (immediately or after stop-out) and postsecondary
success (directly or after stop-out). The elements in this model are
central to understanding the causes of student outcomes. The integrated
model of student success guided the variable selection and choice of analytical
methods.
A key understanding of the model is that educational institutions?whether
elementary, secondary, or postsecondary?must measure and understand
the cognitive and social realities of the student. If these are not systematically
measured and considered, the institution has little ability to make the
type of systemic and programmatic alterations in practice that can make a
difference. Student unit-record information is crucial to changing educational
practice and expanding educational opportunities for students.
Research Methodology
The target population for this study involves 1988 eight-grade students who
went on to some form of postsecondary education between 1992 and 2000.
Drawn from the National Educational Longitudinal Study of 1998 (NELS:88),
our cohort follows White and Hispanic students from the 1988 base year
through the fourth follow-up survey in 2000. This sample represents 72 percent
of the entire 1988 eight-grade cohort.
Created under the auspices of NCES, two databases were used in this study
(NELS:88 CD 2003-348 and NELS:88 CD 2003-402). The databases contain
extensive socioeconomic status, parental expectation, planning for college,
high school-based support, curriculum, high school characteristics, diverse
activities in postsecondary education, financial support, and types of PSE
completion. NCES followed a stratified sample strategy in creating the
NELS:88 whereby the original sample of 8th graders was adjusted to represent
about 3 million population. Subsequent weights reflect the number of
individuals attending postsecondary institutions. In this study, we used the
panel weight of fourth follow-up survey (F4BYPNWT), which adjusts the
NELS:88 data to reflect the number of 1988 middle school juniors who coherently
succeeded first through fourth follow-up surveys. As noted by Adelman
(1999), standardized statistical packages such as SPSS significantly
underestimate the sampling error when handling stratified samples. We
used AM to correct for this problem during the regression analysis.
A key understanding of
the model is that
educational
institutions?whether
elementary,
secondary, or
postsecondary?must
measure and
understand the
cognitive and social
realities of the student.
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-9
Descriptive
Overview
The following section provides more detail about the sample group studied,
including descriptive statistics about their background characteristics, preparation
behaviors, access variables, and persistence behaviors. This analysis
paints a detailed picture of the educational path of Latinos and sets the
stage for our regression analysis to follow. It should be noted that the results
of this section will differ from those of Part I of this report series due to a difference
in the cohort group. Part I studied all NELS 8th grade students, while
Part III focuses exclusively on those students who entered some form of
postsecondary education by 2000. However, we strongly encourage readers
to review both Part I and Part II reports, as they provide indepth descriptive
and inferential analysis of the NELS cohort.
Family & Student
Characteristics
Background characteristics do not provide specific educational policy implications
regarding the educational path of Latino students as compared to
White students. Data on these characteristics do, however, point to policy
areas that have an impact on educational outcomes.
Educational Legacy. The educational attainment of parents and older siblings
is an indication of the kinds of educational goals and expectations for
the child (Sewell and Shah 1968; Auerbach 2001; Choy 2001). Latino students
were much less likely to have a parent with an earned educational
credential?at any level?than White students. In fact, 42 percent of Latino
students had parents whose highest level of education was less than a high
school diploma compared to 18.2 percent of for White students. At the other
end of the spectrum, 81.8 percent of White students had parents with some
college or more, while 58 percent of Latinos had the same educational legacy.
Family Income. The socioeconomic background of students weighs heavily
on their academic achievement at the K-12 level and is a significant barrier
in preparing for postsecondary education and gaining access to institutions
(Alexander, Pallas et al. 1987; Baker and Velez 1996; Cabrera and La Nasa
2001; ACFSA 2002). Latino 8th-grade students were much more likely to
hail from low-income backgrounds than White students. Approximately 46
percent of Latino students came from a family with family income below
$25,000 (1988 dollars) and only 9.2 percent were from high income families
(above $75,000). Comparatively, less than one quarter (17 percent) of
White students were low-income and 21.7 percent were from high income
families.
Approximately 46
percent of Latino
students came from
a family with family
income below
$25,000 (1988
dollars) and only 9.2
percent were from
high income families
(above $75,000).
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-10
Gender. The pool of eighth-grade Latino students was slightly more female
than the White cohort. About 55 percent of Latino students were female
compared to 52.7 percent of White students.
Expectations
& Aspirations
The expectations and aspirations of parents, peers, family members, and the
students themselves have a direct impact on the educational planning and
eventual educational opportunities of students. These issues have policy
implications with regard to our ability to create programs to educate these
individuals about the importance of postsecondary education and career
planning.
Parental Expectations. Parental encouragement is pivotal in the development
of a predisposition for postsecondary study and eventual occupational
attainment (Sewell and Shah 1968; Sewell and Hauser 1975; St. John and
Noell 1989; Stage and Hossler 1989; Flint 1993; Hossler and Vesper 1993;
Berkner and Chavez 1997; Flint 1997; Horn 1997; Hossler, Schmit et al.
1999; Cabrera and LaNasa 2000; Perna 2000; Terenzini, Cabrera et al.
2001). Expectations signal the level of encouragement that parents will provide.
Based on our analysis, Latino parents are less likely to have high postsecondary
expectations for their children. Forty-two percent of Latino parents
have as their highest expectation a high school diploma or a general equivalency
diploma, while 18.2 percent of White parents have those expectations.
Conversely, 81.8 percent of White parents expect their children to have
some college or above by the end of their academic careers, while 58 percent
of Latino parents have that expectation.
Postsecondary Aspirations. What students aspire to attain, in educational
terms, has an effect on their academic behaviors and outcomes (Sewell and
Shah 1968; Chapman 1981; Cabrera and LaNasa 2000; Cabrera and La
Nasa 2001; Cooper 2002). White students were much more likely than Latino
students to aspire to a postsecondary degree while in the 8th grade, especially
a BA or higher (see Exhibit III-2 on page 11). In total, 79 percent of
White students expected to earn at least a BA; 28.4 percent planned on an
advanced degree. By comparison, 62.9 percent of Latino students planned
on earning at least a BA, with 23.3 percent looking forward to an advanced
credential.
White students were
much more likely
than Latino students
to aspire to a
postsecondary
degree while in the
8th grade, especially
a BA or higher.
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-11
Exhibit III-2. Postsecondary Aspirations of 1988 8th Grade Latino and White
Students
51
10 11
28
18 19
40
23
0
20
40
60
Less Than College Some College Bachelor's Advanced Degree
Percentage
White Latino
*** ** *** **
* = p<0.1 ** = p<0.05 *** = p<0.01
Preparation for
Postsecondary
Education
How students prepare for postsecondary education is a critical area for concern
among educators and policy makers, and no less important to the students
themselves. Academic preparation impacts the attention one spends
to the prospect of postsecondary education and has a direct impact on the
type of school a student applies and is ultimately admitted. Arguably the
most important characteristic of students who succeed in postsecondary institutions
is that they are academically prepared for postsecondary studies
(Tinto 1993; Adelman 1999). A rigorous course load is used as a signal of
postsecondary fitness and is highly correlated with high performance on college
entrance-type examinations like the ACT and the SAT (CGCS 2001a;
CGCS 2001b).
Remedial Course-Taking Patterns during High School. Latino students were
more likely than White students to take mathematics and English remedial/
developmental courses in during high school. Latino students were also
more likely to be multiple remedial-course takers. In remedial mathematics,
15.8 percent of Latino students took one or more remedial courses compared
to 8 percent of White students. In English, 12.6 percent of Latino students
took one or more remedial courses, compared with 5.6 percent of
White students.
Course-Taking Patterns. It is widely acknowledged that students require at
least three years of mathematics consisting of courses in Algebra I, Geometry,
and Algebra II to enroll into a four-year institution without being placed
A rigorous course
load is used as a
signal of
postsecondary
fitness and is highly
correlated with high
performance on
college entrance-type
examinations like the
ACT and the SAT.
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-12
on academic probation or being placed in remedial postsecondary courses.
Taking advanced mathematics courses makes a student far more competitive
and is more likely to be part of an academically-focused curriculum
(Adelman, 1999).
Exhibit III-3. High School Mathematics Course-Taking Patterns of 1988 8th Grade
Latino and White Students
28
46
45
38
15
27
0%
25%
50%
75%
100%
White Latino
Percentage
Pre-Calculus
Calculus***
Algebra II
Trigonometry**
Algebra I
Geometry
Other Math***
* = p<0.1 ** = p<0.05 *** = p<0.01
However, our analysis shows that almost half (46.2 percent) of Latino students
are concentrated in courses below the Algebra II level while White students
are concentrated at the Algebra II and Trigonometry level (45.4 percent).
At the highest mathematics levels, 26.5 percent of White students
take Pre-Calculus and Calculus courses, compared to only 15.4 percent of
Latino students. Latinos are also less likely to be enrolled in an academically-
oriented high school program than White students (65.6 percent vs.
75.5 percent).
Access to
Postsecondary
Education
The type of postsecondary institution a student attends is an important factor
in their development and future opportunity. Although many students
who earn a BA start at a two-year institution, credits earned at two-year institutions
are not always transferable or do not count toward a degree
(Cabrera, Burkum et al. Forthcoming).
Postsecondary Enrollment. By the year 2000, 8 years after high school
graduation for the cohort, 33.7 percent of Latinos in the sample were enrolled
in a four-year institution and 66.3 percent were enrolled at the twoyear
level. Conversely, 54.8 percent of White students were enrolled in fouryear
programs with 45.2 percent in two-year programs.
Our analysis shows
that almost half of
Latino students are
concentrated in
courses below the
Algebra II level while
White students are
concentrated at the
Algebra II and
Trigonometry level.
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-13
Exhibit III-4. First Postsecondary Institution Attended, Two-Year versus Four-Year
45
55
66
34
0
25
50
75
Two-Year Institution or Less Four-Year Institution
Percentage
White Latino
* = p<0.1 ** = p<0.05 *** = p<0.01
*** ***
Postsecondary
Attendance,
Persistence,
& Completion
How students attend postsecondary education has an important impact on
their ability to persist and complete education. This section looks at the attendance
patterns, delayed entry into postsecondary education, academic
achievement, and degree completion.
Attendance Patterns. There are two important indicators of attendance patterns
for students. The first is whether students attend full-time or part-time,
the other an observation of the consistency of their attendance. Although attending
part-time and in a consistent pattern may be a necessity and choice
for students, it is known that both options have negative consequences on
the ability of students to persist and complete a degree program (Adelman,
1999).
Over half of Latino postsecondary students attended in a part-time status
(51.8 percent), compared to 37.1 percent of White postsecondary students.
With regard to continuous enrollment, 39.6 percent of Latino students attended
their postsecondary studies continuously compared to two-thirds
(64.2 percent) of White postsecondary students.
Delay of Entry to PSE. Latino students were more likely to delay entry into
postsecondary education following successful graduation from high school. A
gap of 5.6 points between Latinos (76.7 percent) and White students (82.3
percent) exists with regard to entering postsecondary education within 7
months of high school graduation. Conversely, 23.3 percent of Latino youth
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-14
delay entry to postsecondary education at least 8 months past high school
graduation compared to 17.7 percent of White youth.
Academic Achievement. How well a student performs is also an important
factor to consider in postsecondary persistence. Performance as measured
by grade point average (GPA) differs significantly between Latinos and
Whites. Fifty-two percent of Latinos earned GPAs of 2.49 and below, and
68.1 percent of White students earned GPAs of 2.50 and above. Additionally,
18.6 percent of Latino students scored above 3.20, 10 percent lower
than White students (28.6 percent).
Degree Attainment. Low-income students and students of color are generally
less likely than middle/high SES and White students to complete a degree
program. The gaps are wide and widening in this regard (Cabrera and
LaNasa 2000; Harvey 2003; Walpole 2003). More than three-quarters of Latino
students who enrolled in a four-year postsecondary institution, or 76.8
percent, did not earn a degree by the year 2000. Comparatively, 52.7 percent
of White students did not complete.
Financial Aid
Financial concerns tend to loom large in persistence to degree completion.
The actual availability of resources and how students perceive the availability
of resources and their capacity to acquire them also has an impact on
persistence (Bean 1986; Metzner and Bean 1987; Cabrera, Castaneda et al.
1992; Cabrera, Nora et al. 1992). Types of aid also matter inasmuch as type
of aid and combination and cost are considered, especially for students of
color (Pantages and Creedon 1978; Porter 1989; Murdock 1990; St. John
1991; St. John and Starkey 1995; Mumper 1996; St. John, Paulsen et al.
1996; Perna 1998; Fenske, Porter et al. 2000).
The only significant differences between Whites and Latinos are in grant receipts,
work study programs, and holding on-campus jobs. Latinos received
grants more often than Whites (51.4 percent versus 43.8 percent), and
Whites received work-study aid at a higher rate than Latinos (10.1 percent
versus 6.9 percent). As well, White students were more likely to hold oncampus
jobs at a rate of 23.6 percent compared to 16.7 percent for Latinos.
Although the differences were not statistically significant, more White students
received loans (30.2 percent) than Latinos (26.1), and a higher percentage
of White parents borrowed money for college (14.6 percent) than
Latino parents (12.3 percent).
Low-income students
and students of color
are generally less
likely than
middle/high SES,
and White students
to complete a degree
program. The gaps
are wide and
widening in this
regard.
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-15
Research
Findings
This section of the report focuses on the regression analysis of the
NELS:88/00 database. The results of our analysis fall in line with other research:
socioeconomic status, parental expectations, planning, coursetaking
patterns, and student postsecondary behaviors have a significant impact
on postsecondary degree completion. These findings also present very
clear direction for further research and policy focuses because the factors
with the most impact?planning and postsecondary behavior?produce
changes of more than 40 percent, and some cases 60 percent, in the probability
of completion. Detailed tables of findings can be found in Appendix C.
What is extraordinary about the analysis is that the 24.1 percentage point
gap in BA probability between Whites and Latinos is substantially reduced
when all factors are held constant, which is to say this model suggests that
attending to the following factors would significantly increase the chances
that Latinos will graduate from a four-year institution at rates comparable if
not equal to those of Whites.
Family & Student
Characteristics
Gender. While it is already established that Latinos attend four-year institutions
at much lower rates than Whites, this analysis shows that being Latina
increased the probability of graduation by 20 percent1 (Baseline p=.232). It
is unclear from this analysis what extra barriers presented themselves to
young men, but the difference between males and females is statistically
significant (p<0.01).
Income. Middle-income Latinos had a 17 percent higher probability of earning
a BA compared to low-income Latinos. Being upper income had no significant
effect for Latinos. Socioeconomic status worked differently for
Whites. Being middle class only increased the probability of completion by
10 percent (Baseline p=.473), but high-income Whites had a 24.0 greater
probability of completion than low-income students. No significant effects of
having college educated parents were found for either group.
Educational Legacy. This last finding can be seen as a challenge to the
standing research on the value of having college-educated parents. There
are several ways to consider this finding, which would suggest that it is complimentary
and not contradictory. Having a college education places individuals
in a higher-income bracket (Baum and Payea 2004), meaning that
the family income variable does contain some of the effect of parental educational
attainment. Further, subsequent findings concerning expectations
1 All findings are reported in terms of Delta-p unless otherwise noted. For a discussion of the interpretations
see Appendix A.
Middle-income
Latinos had a 17
percent higher
probability of earning
a BA compared to
low-income Latinos.
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-16
and planning reflect a degree of knowledge on the part of parents who have
college degrees or some college experience.
Expectations
& Aspirations
Parental Expectations. Parent expectations and the degree of planning that
took place in the 8th grade were also analyzed. Expecting their children to attend
some college or to get a bachelor?s degree had no statisticallysignificant
effect for Latinos (see Exhibit III-5). However, parental expectation
of advanced degrees had a large and significant effect for Latinos increasing
the probability of completion by 46 percent. Parents who have high expectations
for their children clearly behaved in ways that had strong positive effects
on their children?s behaviors. This finding reflects the research suggesting
that parental engagement and involvement are crucial to student success
(Conklin and Dailey 1981; Hoover-Dempsey and Sandler 1995; Cabrera
and LaNasa 2000; Cabrera and La Nasa 2001).
Exhibit III-5. Percent Change in the Probability of Completing a Four-Year Degree
Due to Expectations and Aspirations
-27
4
12
8
4
20
28
46
48
53
41
2
-30 -15 0 15 30 45
Parental Expectations--
Some College
Parental Expectations--
Bachelor's
Parental Expectations--
Advanced Degree
Student Planning Some
College
Student Planning
Bachelor's
Student Planning
Advanced Degree
Marginal Probablity (Delta-p)
Latino (23.2%)
White (47.3%)
Baseline ps
***
***
**
*
*
***
**
* = p<0.1
** = p<0.05
*** = p<0.01
Planning for College. Planning for college at the 8th-grade level also had large
and significant effects for the probability of completion. Latino students
planning for some college increased the probability of BA completion by 48
percent, and those who planned for a bachelor?s degree increased the probability
by 53 percent. This is one of the most robust findings of the study. It
suggests that students who entered high school with a plan and sense of
purpose had a much greater probability of taking all the necessary steps to
succeed in high school, apply to and enroll in a four-year institution, and persist
to degree completion. Research suggesting that students without a purpose
are more likely to drop out is supported by these findings (Connell
Planning for college
at the 8th-grade level
also had large and
significant effects for
the probability of
completion. Latino
students planning for
some college
increased the
probability of BA
completion by 48
percent.
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-17
1982; Rosenbaum 2001). Furthermore, policymakers interested in promoting
persistence to degree completion would be well served by focusing their
efforts on helping families and students develop and work toward clearly
formed educational goals. The goals should be clearly formed and articulated
because although it is beneficial to plan for some college, it is far more
beneficial to plan for a bachelor?s degree if that is the desired goal.2
Preparation for
Postsecondary
Education
Course Taking Patterns. Measures of high school based support?such as receiving
help in completing applications, applying for financial aid, and writing
essays?produced no significant effects for Whites or Latinos. Course-taking
patterns, however, did produce positive effects. For White students taking
Algebra II or Trigonometry versus taking courses below Algebra I produced
positive effects. Taking pre-calculus and calculus produced positive effects
for both Latino and White students, with increased probabilities of 12 and
20 percent respectively. It is reasonable to assume that these findings indicate
the value of academic preparation over simple college preparatory assistance
programs. No amount of assistance in going through the motions of
applying to college will go as far as being academically prepared to succeed
once a student enrolls.
Exhibit III-6. Percent Change in the Probability of Completing a Four-Year Degree
Due to Academic Preparation
9
20
-26
22
6
12
26
2
-30 -20 -10 0 10 20 30
Algebra II or Trigonometry
Pre-Calculus and Calculus
Remedial English Courses
Academically Oriented H.S.
Program
Marginal Probability (Delta-p)
Latino (23.2%)
White (47.3%)
Baseline ps
***
**
***
**
**
**
***
* = p<0.1
** = p<0.05
*** = p<0.01
2 For a description of a longitudinal model that considers the importance of the temporal nature
of planning see Appendix E.
No amount of
assistance in going
through the motions
of applying to college
will go as far as being
academically
prepared to succeed
once a student
enrolls.
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-18
The level of remediation also mattered. According to our analysis, remedial
math served no one particularly well. Remediation in English, however,
proved positive for Latinos with an increase of 26 percent, but proved negative
for Whites at a decrease of 26 percent. This finding very likely reflects a
significant number of English language learners and non-native English
speakers among Latinos as opposed to Whites. The effect of remediation in
English for native speakers is similar to remediation in any other subject: it
detracts from the time available to study and master rigorous academic
course material. Conversely, remediation in English for non-native speakers
can provide a necessary foundation. There are larger policy issues to which
this analysis cannot speak, but this finding suggests that more research
should be done to determine which remedial English courses provide benefits
and which do not.
Public, private, or vocational orientation in high school had no significant effect
on student postsecondary outcomes. However, schools with academic
orientations produce small-but-significant positive effects of 2 percent for
Latino students and 22 percent for White students.
Postsecondary
Activities & Experiences
What students do during their postsecondary careers has a significant effect
on students. Latino and White students who began their postsecondary studies
at a four-year institution increased their probability of completion by 29
and 35 percent respectively. This finding is supported by existing research
which asserts that beginning at a two-year institution is not necessarily the
best step to take if one?s ultimate goal is a four-year degree, even though
many students use the two-year institution as a step stone due to financial
concerns (Astin 1975; Breneman and Nelson 1981; Cabrera, Burkum et al.
Forthcoming).
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-19
Exhibit III-7. Percent Change in the Probability of Completing a Four-Year Degree
Due to Postsecondary Activities
35
42
42
45
29
60
47
62
-25
-19
-20
-12
-25 0 25 50 75
Started at a 4-year
institution
Maintained continuous
postsecondary enrollment
Delayed enrollment in
postsecondary education
Enrolled part time
College GPA of 2.50 - 3.19
College GPA of 3.20 - 4.00
Marginal Probability (Delta-p)
Latino (23.2%)
White (47.3%)
Baseline ps
* = p<0.1
** = p<0.05
*** = p<0.01
**
***
***
***
**
***
**
***
**
**
**
Maintaining continuous enrollment in postsecondary education increased
the probability of earning a BA by 60 and 42 percent for Latino and White
students respectively. This finding, in combination with the literature on financial
aid concerning the type of work that students do and the number of
hours worked per week (Pascarella 1994; Horn and Chen 1998), paints a
clear picture that shows the importance of continuous enrollment for success.
Stopping out often leads to dropping out (Tinto 1975; Swail, Cabrera et
al. 2004).
Having a GPA between 2.50 and 3.19 increased the probability of completion
by 47 percent for Latinos and 42 percent for Whites, while a GPA between
3.20 and 4.00 increased the probability of completion by 62 percent
and 45 percent respectively. Again, academic preparation remains critical to
postsecondary success. Students who are better prepared academically are
more likely to perform at higher levels at college.
There were postsecondary behaviors that had negative effects on students
as well. Choosing to delay enrollment between high school and college reduced
the probability of completion for Latinos by 20 percent and 25 percent
for White students. Additionally, attending in a part-time capacity also
had a significant negative effect for Whites with a 19 percent reduction in
probability of earning a BA, but the negative effect for Latinos (12 percent)
was not statistically significant.
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-20
Financial Aid
Although research supports the importance of financial factors in persistence,
we found no significant effects of the five financial aid variables in
this analysis. For instance, there was no significant difference in the effect of
receiving grants, loans, or participating in work study programs (0.09), nor
did holding a campus-based (0.02) or parent borrowing (-0.23). These findings
support the research cited earlier showing that it is types of funding in
certain combinations along with the costs that have an effect on persistence
(Pantages and Creedon 1978; Porter 1989; Murdock 1990; St. John 1991;
Blanchette 1994; St. John and Starkey 1995; Mumper 1996; St. John,
Paulsen et al. 1996; Perna 1998; Fenske, Porter et al. 2000).
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-21
Policy
Implications
Postsecondary planning, academic preparation, and taking the right steps in
college matter for Latino persistence. These are the areas where we found
the most significant effects. It is important to note that the nature of this
analysis means that the size of the effect is dependent on the presence of
all of the other factors as well.
At the outset of this report, we presented a framework for consideration of
how student outcomes are the result of the interplay of student proficiencies
in the cognitive and social areas with the strategies and conditions of a system
or organization. Our analysis found that these items matter to Latino
students and that the ability of schools and communities to provide the resources
to develop these skills and knowledge bases is critical to the educational
improvement of Latino students.
For purposes of clarity, we have partitioned our policy discussion into three
pieces: middle school, high school, and postsecondary institution.
Middle School
Because Latino students are more likely to be from low-income families and
less likely to have college-educated parents than other students, it is less
likely that their families will have the knowledge and skills necessary for the
development of aspirations for postsecondary study. From a policy perspective,
while poverty and educational legacy are outside the control of the educational
institutions serving Latino students, it is a viable option to work with
families to develop college knowledge?the knowledge and skills guiding
preparation and success for postsecondary education. Two federal programs,
GEAR UP and Upward Bound, help low-income students learn more
about postsecondary education, therefore providing them with the foundation
to make further educational strides. Other, non-federal programs, such
as AVID, MESA, Puente, College Summit, and numerous other communitybased
programs are examples of efforts to engage students early. Many of
these programs begin at the middle school level and follow students into
high school. The guidance counselor, however, remains a critical piece in the
pre-planning of students in middle school before high school courses have
been selected. Career-orientation programs help students understand the
possibilities before they become tracked outside the academic pipeline.
According to our findings, middle school is important because Latinos who
enter high school with a plan for any type of postsecondary study?with a
sense of the purpose of their high school work?are far more likely to graduate
from college than those who have no plan. In fact, having a plan to attend
college with no real degree aspirations improves the chances of
It is a viable option to
work with families to
develop college
knowledge?the
knowledge and skills
guiding preparation
for and success in
postsecondary
education.
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-22
graduation by 48 percent for Latino youth. Of course, having a plan to attain
a BA degree improves the chances by an additional 5 percent (53 percent).
Planning for college and taking appropriate math courses are intertwined
(Cabrera & La Nasa, 200). Those two factors themselves are in turn dependent
upon the extend parents themselves maintain high degree aspirations
for their children. It stands to reason that policies addressing these
three factors simultaneously would enable Latino students and their families
to be better equipped to undertake a successful postsecondary education
journey.
Recommendations:| Make certain that middle school children and their parents are aware of
the economic and social benefits of professional and technical occupations
while spelling out the curricular choices needed to eventually attain
such occupations;| Clearly communicate to parents and their middle high school children
the practical applications of math and its connection to access to professional
and technical occupations;| Actively involve Latino parents in school activities while providing them
with useful information about curriculum planning, financing options,
and applying for college;| Increase the amount of information on career and postsecondary options
for students, starting in the 6th grade, if not earlier. Included is the
realization of the importance of the importance of course-taking patterns
during both middle and high school;| Engage influential Latino organizations to the point they themselves can
disseminate information on career and postsecondary options for students,
starting in the 6th grade or earlier;| Fully utilize the expertise of guidance counselors who are trained in career
and academic development to appropriately support the needs of
students and families;| Encourage the development of supplementary pre-college programs and
community partnerships that support career and academic development
of students and parents;| Encourage increased parental involvement in the career and academic
development of the child;| Develop policy to encourage the selection of Algebra I at the 8th grade in
order to open up further academic options for students in high school.
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-23
High School
The popular idea that high school is where planning for college begins is disproved
by this analysis. Planning must begin earlier. Still, academic preparation
during high school sets the stage with the first math courses that students
take. Latinos need to be enrolled in and master Algebra I no later than
the ninth grade in order to reap the benefits of high mathematics achievement
on postsecondary persistence. Latinos who take more than three years
of mathematics beginning with Algebra I have a higher probability of graduating
from college than those who take fewer than three courses. Thus, those
students who start on the mathematics track in their freshman year of high
school have a much higher propensity to complete those courses before high
school graduation.
Remediation in English is also important for Latinos. More research must to
be done to determine if the finding is reflective of the needs of non-native
speakers and/or first generation English speakers. Additionally, more research
is needed to determine if different types of courses have different effects.
While some courses may be considered remedial, they may be linguistically
appropriate thus providing benefits to students who need it and producing
negative effects for those who do not.
It is also at the high school level that solid advising needs to take place
about the type of institutions students and families should consider. Latinos
seeking a four-year degree are somewhat disserved by beginning at a twoyear
institution. Our analysis found that BA-directed Latino students who
started at a four-year institution had a 29 percent increase in probability of
graduating than those who started at a community college. Thus, high school
counselors armed with appropriate knowledge and materials are in a position
to make sure that Latino families and students make prudent decisions
related to education and career goals.
The most important information that families and students need to have in
high school is that delaying entry and taking breaks is detrimental to academic
careers. If staying in school continuously increases the probability of
graduating by 60 percent, the second largest and most significant effect in
the analysis, then students and families need to be armed with that information
before they begin.
Recommendations:| Develop policy to encourage and allow for the completion of Algebra II by
9th grade and Geometry by 10th grade;| Ensure that math competencies are articulated throughout K-12 and
that Latino students master them at grade level;| Provide remedial English programs for Latino students;
The popular idea that
high school is where
planning for college
begins is disproved
by this analysis.
Planning must begin
earlier.
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-24| Provide adequate advising and counseling support at every grade level
during high school, and ensure a responsible student:counseling ratio;| Provide dropout prevention programs for Latino students that help them
to understand the purpose of high school and to develop a plan for
postsecondary life.
Postsecondary
Institution
Helping Latinos to maintain continuous enrollment and providing academic
support while they are enrolled should be the primary role of the postsecondary
institution in the effort to improve the completion rates of Latinos.
The provision of academic support is evinced as important by the large effect
of grade point average on the probability of completion. Of course, being
academically prepared when they matriculate is important for all students,
but being able to take advantage of academic support services undoubtedly
helps students maintain high performance in an academic culture that differs
significantly from high school.
This study does not sufficiently address the impact of financial aid due to
statistically-insignificant findings for Latinos. Regardless, as mentioned
throughout this paper, the literature on financial aid is somewhat consistent
in this area. Although the analysis includes variables on different types of financial
aid, it could not measure the combination of aid packages with
costs, which is what affects student persistence.
Recommendations:| Make certain that classroom and out-of-classroom experiences are
geared to enhance learning and acquisition and use of competencies;| Provide appropriate levels of academic and social ?safety nets? for students;| Track student progress throughout the postsecondary experience, with
special emphasis on the freshman year;| Stress financial aid policies and programs that enable Latino students to
maintain continuous enrollment while bringing about engagement with
faculty and academic staff;| Adopt a long-term strategy whereby year-to-year persistence strategies
are articulated with the objective of securing a four-year degree;| Provide invasive counseling for students to encourage continuous enrollment
and prudent course selection;
Helping Latinos to
maintain continuous
enrollment and
providing academic
support while they
are enrolled should
be the primary role of
the postsecondary
institution.
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-25| Link the financial aid office with academic and social services to ensure
that students are provided with a coordinated level of support that encourages
continued enrollment and progress toward student goals.
In Conclusion
Encouraging Latino students to explore and develop their career and academic
interests is important to keeping them in the academic and career
pipeline. This analysis found that Latino students who are supported by their
families in the pursuit of a postsecondary education, create a plan by the
eighth grade, take three years of mathematics or more, start at a four-year
institution, maintain continuous enrollment and a GPA of 2.50 or above can
significantly improve their chances of postsecondary and career success.
The findings in this study do not suggest that the work to make these things
a reality is easy, but they do suggest where to begin.
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-26
References
ACFSA (2002). Empty Promises: The Myth of College Access in America.
Washington, DC, Advisory Committee on Student Financial Assistance.
Adelman, C. (1999). Answers in the tool box: Academic intensity, attendance
patterns, and bachelor's degree attainment. Washington, DC, U.S. Department
of Education, Office of Educational Research and Improvement.
Alexander, K. L., A. M. Pallas, et al. (1987). "Consistency and change in educational
stratification: Recent trends regarding social background and college
access." Research in Social Stratification and Mobility(6): 161-185.
Astin, A. W. (1975). Preventing students from dropping out. San Francicso,
Jossey-Bass.
Auerbach, S. (2001). Under Co-construction: Parent Roles in Promoting College
Access for Students of Color. Los Angeles, University of California: 472.
Baker, T. L. and W. Velez (1996). "Access to and opportunity in postsecondary
education in the United States: A review." Sociology of Education Extra
Issue: 82-101.
Baum, S. and K. Payea (2004). Education Pays 2004. New York, NY, The College
Board.
Bean, J. P. (1986). Assessing and reducing attrition. Managing college enrollment.
D. Hossler. San Francisco, Jossey-Bass: 47-61.
Berkner, L. and L. Chavez (1997). Access to postsecondary education for the
1992 high school graduates. Washington, DC, U.S. Department of Education,
Office of Educational Research and Improvement, National Center for Education
Statistics.
Blanchette, C. M. (1994). Higher education grants effective at increasing minorities?
chances of
graduating. Washington, DC, Health, Education, and Human Services Division,
General Accounting Office.
Breneman, D. W. and S. C. Nelson (1981). "The future of community colleges."
Change 13(5): 17-25.
Cabrera, A. F., K. R. Burkum, et al. (Forthcoming). Pathways to a Four-Year
Degree: Determinants of Transfer and Degree Completion Among Socioeconomically
Disadvantaged Students. College Retention: A Formula for
Success. A. Seidman.
Cabrera, A. F., M. B. Castaneda, et al. (1992). "The convergence between
two theories of college persistence." Journal of Higher Education 63(2): 143-
164.
Cabrera, A. F. and S. M. La Nasa (2001). "On the path to college: Three critical
tasks facing America's disadvantaged." Research in Higher Education
42(2): 119-149.
Cabrera, A. F. and S. M. LaNasa, Eds. (2000). Understanding the college
choice of disadvantaged students. New Directions for Institutional Research,
No.107. San Francisco, Jossey-Bass.
Cabrera, A. F., A. Nora, et al. (1992). "The role of finances in the persistence
process: A structural model." Research in Higher Education 33(5): 571-593.
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-27
Center, P. H. (2004). Hispanic School Achievement: Catching Up Requires
Running Faster Than White Youth (Fact Sheet). Washington, DC, Pew Hispanic
Center.
CGCS (2001a). An analysis of ACT achievemnt in Urban Schools. Washington,
DC, Council of the Great City Schools.
CGCS (2001b). An analysis of SAT I achievement in urban schools. Washington,
DC, Council of the Great City Schools.
Chapman, D. W. (1981). "A model of student college choices." Journal of
Higher Education 52(5): 490-505.
Choy, S. (2001). Students whose parents did not go to college: Postsecondary
access, persistence, and attainment. Washington, DC, U.S. Department
of Education, National Center for Education Statistics, Office of Educational
Research and Improvement.
Conklin, M. E. and A. R. Dailey (1981). "Does consistency of parental educational
encouragement matter for secondary students?" Sociology of Education
54(4): 254-262.
Connell, R. W. (1982). Making the difference: schools, families and social division.
Sydney ; Boston, G. Allen & Unwin.
Cooper, C. R. (2002). "Five Bridges Along Students' Pathways to College: A
Developmental Blueprint of Families, Teachers, Counselors, Mentors, and
Peers in the Puente Project." Educational Policy 16(4): 607-622.
Fenske, R. H., J. D. Porter, et al. (2000). "Tracking .nancial aid and persistence
of women, minority, and needy students in science, engineering and
mathematics." Research in Higher Education 41(1): 67-94.
Flint, T. A. (1993). "Early awareness of college financial aid: Does it expand
choice?" The Review of Higher Education 16(3): 309-327.
Flint, T. A. (1997). "Intergenerational effects of paying for college." Research
in Higher Education 38(3): 313-344.
Harvey, W. B. (2003). Minorities in Higher Education: Annual Status Report.
Washington, DC, American Council on Education: 98.
Hoover-Dempsey, K. V. and H. M. Sandler (1995). "Parental Involvement in
Children's Education: Why Does it Make a Difference?" Teachers College Record
95: 310-331.
Horn, L. J. (1997). Confronting the odds: Students at risk and the pipeline to
higher education. Washington, DC, U.S. Department of Education, National
Center for Education Statistics, Office of Educational Research and Improvement.
Horn, L. J. and X. Chen (1998). Toward Resiliency: At-Risk Students Who
Make It to College. Washington, DC, U.S. Department of Education, Office of
Educational Research and Improvement, National Center of Educational Statistics.
Hossler, D., J. Schmit, et al. (1999). Going to college: How social, economic,
and educational factors influence the decisions students make. Baltimore,
MD, The John Hopkins University Press.
Hossler, D. and N. Vesper (1993). "An exploratory study of the factors associated
with parental saving for postsecondary education." Journal of Higher
Education 64(2): 140-165.
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-28
Metzner, B. S. and J. P. Bean (1987). "The estimation of a conceptual model
of nontraditional undergraduate student attrition." Research in Higher Education
27(1): 15-38.
Mumper, M. (1996). "Beyond .nancial aid: Alternative approaches to improving
college participation." Review of Educational Research 22(83-97).
Murdock, T. (1990). "Financial aid and persistence: An integrative review of
the literature." NASPA Journal 27(3): 213-221.
Pantages, T. J. and C. F. Creedon (1978). "Studies of college attrition, 1950?
1975." Review of Educational Research 48(1): 49-101.
Pascarella, E. T. (1994). "Impacts of on-campus and off-campus work on
first-year cognitive outcomes." Journal of College Student Development
35(5): 364-370.
Pascarella, E. T. and P. T. Terenzini (1991). How College Affects Students.
San Francisco, Jossey-Bass.
Pascarella, E. T. and P. T. Terenzini (2005). How College Affects Students.
San Francisco, CA, Jossey-Bass.
Perna, L. W. (1998). "The contribution of financial aid to student persistence."
Journal of Student Financial Aid 28(3): 25-40.
Perna, L. W. (2000). "Differences in the decision to attend college among African
Americans, Hispanics, and Whites." The Journal of Higher Education
71(2): 117-141.
Pew Hispanic Center (2004). Hispanic School Achievement: Catching Up Requires
Running Faster Than White Youth (Fact Sheet). Washington, DC, Pew
Hispanic Center.
PHC (2004). Hispanic School Achievement: Catching Up Requires Running
Faster Than White Youth (Fact Sheet). Washington, DC, Pew Hispanic Center.
Porter, O. F. (1989). Undergraduate completion and persistence at four-year
colleges and universities: Completers, persisters, stopouts, and dropouts.
Washington, DC, National Institute of Independent Colleges and Universities.
Rosenbaum, J. (2001). Beyond College for All. New York, Russell Sage Foundation.
Sewell, W. H. and R. M. Hauser (1975). Education, occupation, and earning:
Achievement in the early career. New York, Academic Press.
Sewell, W. H. and V. P. Shah (1968). "Social class, parental encouragement,
and educational aspirations." American Journal of Sociology 73(5): 559-572.
St. John, E. P. (1991). "What really influences minority attendance?: Sequential
analyses of the high school and beyond sophomore cohort." Research in
Higher Education 32(2): 141-158.
St. John, E. P. and J. Noell (1989). "The effects of student financial aid on
access to higher education: An analysis of progress with special consideration
of minority enrollment." Research in Higher Education 30(6): 563-581.
St. John, E. P., M. B. Paulsen, et al. (1996). "The nexus between college
choice and persistence." Research in Higher Education 37(2): 175-220.
St. John, E. P. and J. B. Starkey (1995). "An alternative to net price: Assessing
the influence of prices and subsidies on within-year persistence." Journal
of Higher Education 66(2): 156-186.
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-29
Stage, F. K. and D. Hossler (1989). "Differences in family influences on college
attendance plans for male and female ninth graders." Research in
Higher Education 30(3): 301-315.
Swail, W. S. (1995). A conceptual framework for student retention in science,
engineer-ing, and mathematics. Graduate School of Education and
Human Development (GSEHD). Washington, DC, The George Washington
University.
Swail, W. S., A. F. Cabrera, et al. (2004). Latino Youth and the Pathway to
College. Washington, DC, Educational Policy Institute.
Swail, W. S., K. E. Redd, et al. (2003). Retaining Minority Students in Higher
Education: A Framework for Success. San Francisco, Jossey-Bass.
Terenzini, P. T., A. F. Cabrera, et al. (2001). Swimming against the tide: The
poor in American higher education. New York, College Entrance Examination
Board.
Tinto, V. (1975). "Dropout from higher education: A theoretical synthesis of
recent research." Review of Higher Education 45(1): 89-125.
Tinto, V. (1993). Leaving college: Rethinking the cause and cures of student
attrition. Chicago, The University of Chicago Press.
Walpole, M. (2003). "Socioeconomic Status and College: How SES Affects
College Experiences and Outcomes." The Review of Higher Education 27(1):
45-71.
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-30
Appendix A ? Methodology
Database. The sample for this study was drawn from the National Education Longitudinal Study of 1998 (NELS:88) and
only Whites and Latinos in the cohort groups from 1988 base year through 2000 fourth follow-up survey were selected.
Created under the auspices of NCES, the two databases (NELS:88 CD 2003-348 and NELS:88 CD 2003-402) are used
in this study, and key variables demographic and school related variables were retrieved. Those relate to socioeconomic
status, parental expectation, planning for college, high school based support, curriculum, high school characteristics,
diverse activities in postsecondary education, financial support, and types of PSE completion.
Weight Employed in the Analyses. NCES followed a stratified sample strategy in creating the NELS:88 whereby the
original sample of 8th graders was adjusted to represent about 3 million population. Subsequent weights reflect the
number of individuals attending postsecondary institutions. In this study, we used the panel weight of fourth follow-up
survey (F4BYPNWT), which adjusts the NESL:88 data to reflect the number of 1988 middle school juniors who coherently
succeeded first through fourth follow-up surveys. As noted by Adelman (1999), standardized statistical packages
such as SPSS significantly underestimate the sampling error when handling stratified samples. To correct for this problem
we used AM to run all logistic regression models and estimate the correct standard errors for all analyses reported
through this report. The AM statistical software, developed by the American Institutes for Research (2002), is distributed
is endorsed and distributed by the Association of Institutional Research.
Dependent Variables
Four-year degree completion. Students who secured at least a bachelor?s degree by 2000 were considered degree
completers for this study. Degree completion is coded as 1 and 0 for those who did not complete a four-year degree.
This variable was ascertained from PETS (NESL:88 2003-402) and the original code for this variable is HDEG.
Independent Variables
Background. Gender (F4SEX) is coded as 0 for male and 1 for female. Ethnicity (RACE4) only includes White and Hispanic
(White=0, Hispanic=1), and other ethnic groups are excluded from the logistic regression analysis.
Family Income. The original code of this variable is FAMINC 92 in NELS:88 CD 2003-402. The 6 original income categories
were collapsed into three: low (less than $25,000), middle ($25,000 to $74,999) and high ($75,000 or more).
Highest Parental Expectation. This variable was attained from EDEXP92 in NELS:88 2003-402. This variable reports
how far in school 8th graders believe their parents think they should go. The basic level is high school graduate or less
and it is compared to some college, bachelor?s or advanced degree.
Planned for college at 8th grade. Created by (Berkner & Chavez, 1997), BYS45 (NELS:88 CD 2003-348) identifies the
highest degree planned to obtain when the subject was in the 8th grade. This nominal variable is also coded high school
graduate or less, some college, bachelor?s, and advanced degree.
High school based support. Three variables in regard to school based support have been used in this study. The first
one received help PSE applications (F2S57A), the second is received help filling out financial aid form (F2S57B), and
the third to receive assistance in writing PSE application (F2S57C). All these variables are retrieved from NELS:88 CD
2003-348.
Highest mathematics in high school. Adelman and associates developed this variable, labeled HIGHMATH in CD#
2003-402, based on high school and postsecondary transcripts. They originally categorized highest mathematics
course students took into seven categories, but in this study is it re-categorized into three. Base category includes algeLatino
Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-31
bra I, geometry, and other math. Next category includes algebra 2 and trigonometry. The top category includes precalculus
and calculus.
Remediation. Two measures were employed signifying the number of remedial courses in math (REMMTH) and reading
(REMREAD) the community college student took. These variables were derived from college transcripts CD 2003-402
(see Adelman, 1999). It is modified to binary variable, in which 0 indicates no remedial course and 1 means 1 or more
remedial course.
High school type. This variable came from NELS:88 CD 2003-402. The public school is used as base group.
High school program. High school program (F2RTRPRG) is determined from high school credits and the three groups
are undecided or general, mostly vocational, and academic or rigorous academic program. The undecided is used as
base group. F2TRTPRG is in NELS:88 CD 2003-348.
PSE activity. Five indicators were employed in the logistic regression analysis--if students started at a four-year institution
or not (REFINST by INSTCOMB), if they continued college enrollment (CONTIN), if they delayed PSE entry (DELAY),
and if they attended as part time (F4EPARTT). Whether attending part time was based on self-reported information contained
in NELS 2003-348. The other variables were derived from college transcript databases (NELS 2003-402).
Undergraduate Grade Point Average. PETS database (NELS:88 CD 2003-402) has GPA. It is derived from PETS
(NELS:88 2003-402). We categorize GPA score from 0 to 4 as following: GPA less than 2.5, GAP 2.5 through 3.19, and
GPA 3.2 through 4.0 The base groups is less GPA less than 2.5.
Financial support. Five important variables are related to financial support in postsecondary education. They are all
derived from NELS:88 2003-348, and they signify whether the student received grants to attend PSE (GRANTS), loans
(LOANS), work study programs (WORKSTDY), or held a campus-based job (CAMPJOB), or whether his or her parent borrowed
(PARNBORW).
Interpretation of Logistic Regression Results
Baseline p - observed probability of the dependent variable. For instance, the observed probability that 1988 cohort
would eventually secure a four-year degree by 2000 is 0.4468, meaning that 45% of them graduated. Observed probabilities
are also referred as "unadjusted probabilities." Baseline p serves as a benchmark to assist in assessing how
much each independent variable contributes to the probability of the dependent variable.
Beta weights. In contrast to OLS, interpretation of logistic parameter estimates is not straightforward. Unlike OLS, the
metric of individual coefficients is expressed in terms of logits rather than in terms of the original scale of measurement.
This problem is particularly accentuated for categorical variables; the corresponding beta weights represent contrasts
among categories summarized in terms of differences of logits. For instance, the SES effect of 1.07 for all indicates
that Highest-SES originated students, on the average, are 1.07 logit units more likely to obtain a bachelor?s degree
than are Lowest-SES students. To overcome this problem, logistic regression results are usually presented in
terms of changes in probabilities and adjusted probabilities.
Delta-p. Developed by Peterson (1985), delta-p reflects the incremental change in the dependent variable (e.g., completing
a 4-year degree) due to a unit change in the independent variable (e.g. college academic performance). For instance,
the delta-p value of .47 associated with higher undergraduate GPA means that if a student gets a higher GPA
compared to the lowest one, the probability of degree completion increases by 47% percent. When the independent
variable is dichotomous (e.g. gender), delta-ps are interpreted as differences between the two categories.
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-32
3Appendix B ? Descriptive Statistics
Descriptive Statistics for the Variables Employed in the Logistic Regression Models
Comparison between
White and Latino
Factors Variables
All (%) White (%) Latino (%)
% difference S.E.
Completion of 4-year degree Four-year Degree or higher 44.7 47.3 23.2 24.1*** 0.0266
Whites 88.6 Ethnicity
Latinos 11.4
Male 47.1 47.3 45.4 1.9 0.0376 Gender
Female 52.9 52.7 54.6 -1.9 0.0376
Low (Less than $25,000) 19.9 17.0 46.2 -29.2*** 0.0333
Middle ($25,000-74,999) 59.6 61.3 44.6 16.7*** 0.0318
Family Income
High ($75,000+) 20.5 21.7 9.2 12.5*** 0.0293
No HS, or HS diploma, or GED 20.6 18.2 42.0 -23.8*** 0.0293 Parental Education
Some college or above 79.4 81.8 58.0 23.8*** 0.0293
Parent expected No PSE or don't know 6.6 6.3 9.3 -3.0** 0.0176
Parent expected Some college 12.2 11.9 14.5 -2.6 0.0198
Parent expected Bachelor's 43.6 44.6 35.2 9.4*** 0.0251
Parental Expectations for child
Parent expected Advanced degree 37.6 37.2 41.0 -3.8 0.0275
Less than college 11.6 10.7 18.3 -7.6*** 0.0252
Some college 11.3 10.3 18.8 -8.5** 0.0401
Bachelor's 49.3 50.6 39.6 11*** 0.0325
Planned for College at 8th grade
Advanced degree 27.8 28.4 23.3 5.1** 0.0253
Received help with college application 47.2 46.9 49.2 -2.3 0.0265
Received in applying for financial aid 37.4 35.9 50.0 -14.1*** 0.0269
Received help with college essays 30.1 29.4 35.3 -5.9** 0.0267
Algebra I, Geometry, and Other Math 30.0 28.1 46.2 -18.1*** 0.0346 High School Math Taken
Algebra II and Trigonometry 44.7 45.4 38.4 7** 0.0346
3 The 2000 panel weight F4BYPNWT was used to estimate the number of 8th graders in the population that participated in both the base year and the fourth followed
up that took place 12 years later. Due to the complex stratification procedures used in selecting the cases, the AM statistical software (American Institutes
for Research, 2002) was used in estimating the correct standard errors for the t-test comparisons reported in the table.
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-33
Comparison between
White and Latino
Factors Variables
All (%) White (%) Latino (%)
% difference S.E.
Pre-Calculus and Calculus 25.3 26.5 15.4 11.1*** 0.0206
None 91.2 92.0 84.2 7.8*** 0.0218 Remedial courses in MATH
1 or more remedial courses 8.8 8.0 15.8 -7.8*** 0.0218
None 93.7 94.4 87.4 7.0*** 0.0204 Remedial courses in English
1 or more remedial courses 6.3 5.6 12.6 -7.0*** 0.0204
Public 88.3 88.1 90.4 -2.3 0.0247 High school type
Private 11.7 11.9 9.6 2.3 0.0247
Undecided 15.9 14.9 24.2 -9.3*** 0.0288
Vocational oriented 9.6 9.6 10.1 -0.5 0.0176
High school program
Academically oriented 74.5 75.5 65.6 9.9*** 0.0314
First Postsecondary institution attended less than 2-year or 2-year institution 47.6 45.2 66.3 -21.1*** 0.0310
4-year 52.4 54.8 33.7 21.1*** 0.0310
Maintained continuous PSE enrollment 61.4 64.2 39.6 24.6*** 0.0316
Entered within 7 months 81.6 82.3 76.7 5.6* 0.0291 Delayed enrollment in PSE
Delayed 8~20 or more than 20 months 18.4 17.7 23.3 -5.6* 0.0291
Enrolled part time 38.8 37.1 51.8 -14.7*** 0.0348
0.00 ~ 2.49 34.0 31.8 52.0 -20.2*** 0.0381
2.50 ~ 3.19 38.4 39.5 29.4 10.1*** 0.028
College-GPA
3.20 ~ 4.00 27.6 28.6 18.6 10.0*** 0.0277
Received grants Yes 44.6 43.8 51.4 -7.6* 0.0426
Received loans Yes 29.7 30.2 26.1 4.1 0.0295
Received work study programs Yes 9.8 10.1 6.9 3.2** 0.0144
Parents borrowed Yes 14.3 14.6 12.3 2.3 0.0221
Held a campus-based job Yes 22.9 23.6 16.7 6.9*** 0.0247
1. The 2000 panel weight F4BYPNWT was used to estimate the number of 8th graders in the population that participated in both the base year and the fourth followed
up that took place 12 years later. Due to the complex stratification procedures used in selecting the cases, the AM statistical software (American Institutes
for Research, 2002) was used in estimating the correct standard errors for the t-test comparisons reported in the table.
* p < 0.1 ** p < 0.05 *** p < 0.01
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-34
Appendix C ? Delta P Values
Change in the probability of completing a four-year degree among members of the 1988 8th cohort that became postsecondary education attendees due
to background, socioeconomic status, parental expectations, preparation for college, high school factors and postsecondary experiences.
Marginal Probability Delta-P Factors
Variables and Values All White Latino
Background
Latino v White -0.03
Female v Male 0.00 -0.01 0.20***
Socioeconomic Status
Middle income v low income 0.11** 0.10* 0.17**
High income v low income 0.26** 0.24** 0.20
College educated parents 0.08 0.10 -0.01
Parental expectations
Some college v none or undefined -0.24*** -0.27** 0.20
Bachelor's v none or undefined 0.05 0.04 0.28
Advanced degree v none or undefined 0.14** 0.12* 0.46*
Planned for college at 8th grade
Some college v none or undefined 0.09 0.08 0.48**
Bachelor's v none or undefined 0.06 0.04 0.53***
Advanced degree v none or undefined 0.03 0.02 0.41***
High school based support
Received help in writing college application v none 0.01 0.01 0.01
Received help in applying for financial aid v none -0.05** -0.04 -0.11
Received help with school application essays v none 0.02 0.03 0.00
High school math course
Algebra II or Trigonometry v less than Algebra II 0.09* 0.09*** 0.06
Pre-Calculus and Calculus v less than Algebra II 0.19** 0.20** 0.12**
High school remediation course
Took remedial courses in math -0.01 0.02 -0.11
Took remedial courses in English -0.23* -0.26** 0.26***
High school characteristics
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-35
Marginal Probability Delta-P Factors
Variables and Values All White Latino
Private v Public 0.17* 0.15 0.22
Vocational oriented program v undefined or general 0.16** 0.16* -0.03
Academic oriented program v undefined or general 0.21*** 0.22** 0.02***
Postsecondary activity
Started at a 4-year institution 0.35** 0.35** 0.29**
Maintained continuous postsecondary enrollment 0.44*** 0.42** 0.60***
Delayed enrollment in postsecondary education -0.24** -0.25** -0.20***
Enrolled part time -0.19** -0.19** -0.12
College GPA of 2.50 ~ 3.19 v less than 2.50 0.43*** 0.42*** 0.47***
College GPA of 3.20 ~ 4.00 v less than 2.50 0.47*** 0.45*** 0.62**
Financial support
Received grants 0.03 0.02 0.09
Received loans -0.07 -0.08 0.09
Received work study programs 0.08 0.08* 0.09
Held a campus-based job 0.02 0.02 0.02
Parents borrowed 0.03 0.02 0.19
Baseline P : Probability of Securing a 4-year degree 0.447 0.473 0.232
Model Chi-square, df 653092***, 32 590689***, 31 54220***, 31
Percent of Correctly Predicted Cases 89.4 89.8 84.6
* p<0.1
** p<0.05
*** p<0.01
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-36
Appendix D ? Beta Values
Effects of background, socioeconomic status, parental expectations, preparation for college, high school factors and postsecondary experiences on the
probability of completing a four-year degree among members of the 1988 8th cohort that became postsecondary education attendees
Betas Factors Variables and Values
All S.E. White S.E. Latino S.E.
Background
Latino v White -0.14 0.19
Female v Male 0.01 0.04 -0.05 0.10 0.94*** 0.04
Socioeconomic Status
Middle income v low income 0.43** 0.07 0.40* 0.09 0.79** 0.10
High income v low income 1.07** 0.22 1.04** 0.20 0.91 0.36
College educated parents 0.33 0.18 0.39 0.16 -0.08 0.14
Parental expectations for child
Some college v none or undefined -1.15*** 0.10 -1.25** 0.19 0.93 0.47
Bachelor's v none or undefined 0.22 0.19 0.16 0.20 1.23 0.76
Advanced degree v none or undefined 0.56** 0.11 0.48* 0.16 2.03* 0.66
Planned for college at 8th grade
Some college v none or undefined 0.35 0.54 0.33 0.48 2.09** 0.45
Bachelor's v none or undefined 0.24 0.38 0.18 0.33 2.38*** 0.12
Advanced degree v none or undefined 0.11 0.54 0.08 0.49 1.80*** 0.39
High school based support
Received help in writing college application v none 0.05 0.11 0.05 0.11 0.08 0.21
Received help in applying for financial aid v none -0.22** 0.03 -0.17 0.07 -0.81 0.56
Received help with school application essays v none 0.07 0.10 0.11 0.10 -0.02 0.30
High school math
Algebra II or Trigonometry v less than Algebra II 0.37* 0.09 0.36*** 0.03 0.32 0.33
Pre-Calculus and Calculus v less than Algebra II 0.79** 0.13 0.82** 0.15 0.60** 0.11
High school remediation
Took remedial courses in math -0.03 0.20 0.09 0.32 -0.74 0.37
Took remedial courses in English -1.08* 0.29 -1.20** 0.27 1.16*** 0.09
High school characteristics
Private v Public 0.68* 0.20 0.61 0.19 1.02 0.52
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-37
Betas Factors Variables and Values
All S.E. White S.E. Latino S.E.
Vocational oriented program v undefined or general 0.63** 0.09 0.67* 0.19 -0.19 0.44
Academic oriented program v undefined or general 0.87*** 0.05 0.91** 0.14 0.13*** 0.01
Postsecondary activity
Started at a 4-year institution 1.58** 0.21 1.63** 0.20 1.30** 0.19
Maintained continuous postsecondary enrollment 2.25*** 0.18 2.24** 0.24 2.81*** 0.23
Delayed enrollment in postsecondary education -1.14** 0.21 -1.12** 0.23 -2.19*** 0.20
Enrolled part time -0.84** 0.12 -0.83** 0.14 -0.84 0.37
College GPA of 2.50 ~ 3.19 v less than 2.50 2.14*** 0.03 2.18*** 0.09 2.07*** 0.10
College GPA of 3.20 ~ 4.00 v less than 2.50 2.63*** 0.12 2.65*** 0.08 2.93** 0.53
Financial support
Received grants 0.12 0.20 0.08 0.18 0.45 0.36
Received loans -0.28 0.24 -0.34 0.21 0.45 0.52
Received work study programs 0.33 0.12 0.33* 0.11 0.45 0.51
Held a campus-based job 0.08 0.08 0.08 0.12 0.10 0.06
Parents borrowed 0.13 0.23 0.10 0.26 0.86 0.89
Constant -6.13 0.15 -6.08 0.29 -9.58 1.09
Baseline P : Probability of Securing a 4-year degree 0.447 0.473 0.232
Model Chi-square, df 653092, 32 590689, 31 54220, 31
Percent of Correctly Predicted Cases 89.4% 89.8% 84.6%
* p<0.1
** p<0.05
*** p<0.01
Latino Students and the Educational Pipeline Part III: Pathways to the Bachelor?s Degree
Educational Policy Institute Part III-38
Appendix E ?
Persistence as a Longitudinal Model
Persistence is a process that begins no later than middle school. It is during the middle school years that the aspirations
of families for their children concerning college are made explicit. The academics of middle school lay the foundation
for high math achievement in high school because students must be ready to start a college preparatory course of
study in math in the ninth grade. It is also in middle school where the aspirations of parents can be bolstered by clear
information about what it takes to prepare for, apply to, and enroll in college. This college knowledge allows families
and students to be more active participants in process of planning for college.
The foundation of family encouragement and involvement, making clear plans based on family aspirations, academic
preparation, and a solid body of college knowledge not only helps students to get prepared for and into college, but it
also helps them make it through. Once students are in college a number of other factors come into play. The students
themselves will be much better served if they become integral parts of the academic and social communities of the
institution. This integration can be helped or hindered by institutional characteristics such as racial/ethnic and economic
diversity and the academic and social climate. The facilities of the institution and the degree that services are
accessible also have an impact on how well students integrate themselves.
How institutional characteristics interact with the foundation of students when they arrive and affect student integration
efforts can lead to positive and negative pre-baccalaureate outcomes such as transfer, stop-out, and dropout. An additional
factor that contributes to these outcomes is financial aid. How well a family plans for the financial realities of
postsecondary education, how much they save, and how savvy they are about financial aid, impacts every part of the
process from deciding to prepare in middle and high school to choosing which institution to attend. Whether to remain
in college is often affected by financial considerations as well.
All of these things, from the elements of the foundation laid in middle school to the entire experience of the postsecondary
institution, affect whether a student will complete a four-year degree. Having a four-year degree means having
more opportunities. The path that is begun in middle school can result in further academic studies or better employment
opportunities than the average citizen. Although the analysis here stops with the completion of the degree, further
studies can consider the extension of this longitudinal model which includes several post-baccalaureate outcomes including
higher incomes, better job performance, more job satisfaction, higher rates of loan repayment and higher degrees
of satisfaction with the postsecondary institution and a commitment to support it.
Recent Publications from the
Educational Policy Institute epi
www.educationalpolicy.org
Washington, DC ?Improving Educational Policy & Practice Through Research? Toronto, ON
The Affordability of
University Education
This report looks at the
relative affordability of public
university education in the
United States and Canada,
comparing all 50 US states
and 10 Canadian provinces
on postsecondary access,
student financial aid,
tuition and fee charges,
and overall net cost of
attendance for the years
1999-01.
Latino Youth and the
Pathway to College
Authors Swail, Cabrera, and
Lee use data from the U.S.
Department of Education?s
National Educational
Longitudinal Study (NELS)
to show how Latino
students fair in the educational
pipeline.
Value Added: The Costs and
Benefits of College Preparatory
Programs
This report considers issues
related to the complex
proposition that the cost of
early intervention program
delivery is directly and
positively tied to the ability
of programs to successfully
enable students to get into
college.
Changes in Tuition Policies:
Natural Policy Experiments
in Five Countries
This international study
reviews tuition and fee policy
changes and strategies in 5
countries and 9 jurisdictions.
Funded by the Canada
Millennium Scholarship
Foundation, this review
provides insight into the
impact of tuition policies
on enrolment.
EPICenter/EPICentre
EPIcenter is a quarterly
report by EPI which provides
information on recent
research conducted by EPI.
A US/international version
is distributed out of the DC
office, while a Canadian
EPICentre is distributed
out of the Toronto office.
A New Measuring Stick
This report is the first to
attempt to quantify how well
different jurisdictions fare in
terms of ensuring equitable
access to university to
students from different
socio-economic backgrounds,
through use of
the Educational Equity
Index (EEI).
www.educationalpolicy.org