University of Southern California Dissertations and Theses

Acceptance, belonging, and capital: the impact of socioeconomic status at a highly selective, private, university

(USC Thesis Other)

Acceptance, belonging, and capital: the impact of socioeconomic status at a highly selective, private, university

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Running Head: IMPACT OF SOCIOECONOMIC STATUS 1

Acceptance, Belonging, and Capital: The Impact of Socioeconomic Status at a Highly Selective,
Private, University

By

Christopher Erik Mattson

A Dissertation Presented to the
FACULTY OF THE ROSSIER SCHOOL OF EDUCATION
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the Degree
DOCTOR OF EDUCATION

December 2014

Dissertation Committee Members:
Dr. Tracy Tambascia (Chairperson)
Dr. Patricia Tobey
Dr. Janice Schafrik

Copyright 2014 Christopher Erik Mattson
IMPACT OF SOCIOECONOMIC STATUS 2
Dedication
My own post-secondary journey has spanned three different decades and four different
states. At each stop along the way I have felt increasingly comfortable with my sense of self and
direction. This was not the case when I first arrived at Linfield College in the fall of 1994. I was
lost in many ways and lacked the capital necessary to afford my mistakes. Eventually these
feelings fueled my desire to help college students and my pursuit of work within the field of
student affairs. That is why this study is dedicated to students who have questioned whether or
not they belonged. Furthermore, this study is dedicated to improving the sense of belonging of
future students.

IMPACT OF SOCIOECONOMIC STATUS 3
Acknowledgments
I could not have reached this point without good timing and support. My wife, Denitsa,
has regularly inspired and challenged me to become a better person. Our daughters, Mila and
Niki, have provided me with endless energy, love, laughter, and hope.
Professionally, when I first decided to pursue student affairs, Michele Rosenthal gave me
insight and direction. The team at Susman, Duffy, and Segaloff gave me a generous send off as
Denitsa and I took a big risk moving across the country to Los Angeles. Felicia Hunt found the
misplaced documents of my application, allowing me to start a life-changing master’s program.
Michele Dunbar and others gave me opportunities to apply my learning within a division led by
the great Michael Jackson. My first higher education class was taught by Shaun Harper who
helped me feel like I belonged. Pat Tobey repeatedly led me to multiple opportunities. Eddie
Roth gave me the flexibility to grow as a professional. And, most importantly, the many students
I was lucky to work with helped me feel valued.
When it came to pursuing a doctoral program, I was enrolled and ready to begin the
Ph.D. program in Leadership Studies at the University of San Diego. After a change of heart, I
was thankful for the flexibility and opportunity that the Rossier School of Education provided. It
is rather fitting that Dean Gallagher is a fellow northwest soul and graduate of Western
Washington University.
Lastly, this dissertation would not be possible without the people who helped shape it.
My chair, Tracy Tambascia gave me honest and necessary criticism, especially when my
interests were overwhelming. Pat Tobey, as always, kept me motivated and hopeful. Janice
Schafrik was my quantitative savior. Additionally, the data could not have been collected
without the aid of Paul Dieken and David Glasgow. To all of these people, I am forever grateful.
IMPACT OF SOCIOECONOMIC STATUS 4
Table of Contents
List of Tables 6
List of Charts 8
Abstract 9
Chapter 1: Overview 10
Background of the Problem 10
Statement of the Problem 16
Purpose 18
Research Questions 20
Significance of the Study 21
Limitations 22
Delimitations 23
Definitions 25
Conclusion 27
Chapter 2: Literature Review 28
The Need for Quality Assessment 29
Socioeconomic Status 38
Theoretical Framework 60
Methodology 70
Significance 71
Conclusion 71
Chapter 3: Methodology 72
Research Methodology 73
Research Questions 76
Methodology of Other Studies 77
Reasons for Support Program Methodology 77
Reasons for SES Analysis Methodology 79
Population 80
Data Collection Procedures 82
Measurement 83
Data Analysis 84
Limitations 88
Ethical Considerations 90
Conclusion 91
Chapter 4: Data and Findings 92
Comparing Cohort Years 92
Exclusions 93
Research Question One 96
Research Question Two 119
Chapter 5: Conclusions and Implications 131
Summary 131
Limitations and Delimitations 138
Discussion 140
Conclusions 143
Implications 148
IMPACT OF SOCIOECONOMIC STATUS 5
Future Research 156
Closing 157
References 158
Tables 171
Charts 223

IMPACT OF SOCIOECONOMIC STATUS 6
List of Tables
Table 1: Prominent Studies and Their Connection to SES
Table 2: Ranges of SES as Calculated by Institutional Expected Family Contribution
Table 3: Academic Preparedness Composite Score Calculation
Table 4: Academic Preparedness Score Examples
Table 5: Degree Program Categories
Table 6: All Students by SES
Table 7: Cohort Details
Table 8: Students Examined for Study
Table 9: Initial Data on Students in Study
Table 10: Additional Data on Students in Study
Table 11: Outcomes for Students in Study
Table 12: Support Program Totals for SES, Preparedness, and First Generation Status
Table 13: Support Program Results for Degree Program Categories
Table 14: Correlations for Independent and Dependent Variables
Table 15: Additional Correlations for Independent and Dependent Variables
Table 16: Comparison of Support Program Participant and Non-Participant Outcomes
Table 17: Comparison of Similar Preparedness Groups (AP .80 < .90)
Table 18: Comparison Totals for SES, Gender, and First Generation Status
Table 19: Comparison Results for Race/Ethnicity and Degree Program Categories
Table 20: Composition for Participants by SES for Additional Variables (AP .80-.90)
Table 21: Composition for Non-Participants by SES for Additional Variables (AP .80-.90)
Table 22: Composition for Non-Participants by SES for Additional Variables (AP.80-.868)
Table 23: Lower 50
th
Percentile Preparedness Comparison (AP less than .83139)
Table 24: Comparison totals for SES, Gender, and First Generation Status (AP < .83139)
Table 25: Comparison Results for Race/Ethnicity and Degree Program Categories (AP < .83139)
Table 26: Composition for Participants by SES for Additional Variables (AP <.83139)
Table 27: Composition for Non-Participants by SES for Additional Variables (AP <.83139)
Table 28: Composition for Non-Participants by SES for Additional Variables (AP < .803)
Table 29: 25
th
to 75
th
Percentile Preparedness Comparison (AP .81039 to <.8664)
Table 30: Comparison Totals for SES, Gender, and First Generation Status (AP .80139-.8664)
Table 31: Comparison Results for Race/Ethnicity and Degree Program (AP .80139-.8664)
Table 32: Composition for Participants by SES for Additional Variables (AP .80139-.8664)
Table 33: Composition for Non-Participants by SES for Additional Variables (AP .80139-.8664)
Table 34: Composition for Non-Participants by SES for Additional Variables (AP .80139-.854)
Table 35: 50
th
to 100
th
Percentile Preparedness Comparison (AP .83139-.936)
Table 36: Comparison Totals for SES, Gender, and First Generation Status (AP .83139-.936)
Table 37: Comparison Results for Race/Ethnicity and Degree Program (AP .83139-.936)
Table 38: Composition for Participants by SES for Additional Variables (AP .83139-.936)
Table 39: Composition for Non-Participants by SES for Additional Variables (AP .83139-.936)
Table 40: Composition for Non-Participants by SES for Additional Variables (AP .83139-.892)
Table 41: Comparison of Reasonably Similar Groups
Table 42: Research Question Two Composition Totals by SES
Table 43: Outcome Totals by SES
Table 44: Outcomes for AP ranges by SES
IMPACT OF SOCIOECONOMIC STATUS 7
Table 45: Composition Totals by SES for AP .80 < .90
Table 46: Composition Totals by SES for AP .90 < 1.0
Table 47: Composition Totals by SES for AP 1.0+
Table 48: Outcomes by Race/Ethnicity
Table 49: Outcomes by Race/Ethnicity Part One (AP .85 < .95)
Table 50: Outcomes by Race/Ethnicity Part Two (AP .85 < .95)
Table 51: Outcomes for All Students by Gender
Table 52: Outcomes for All Students by First Generation Status

IMPACT OF SOCIOECONOMIC STATUS 8
List of Charts
Chart 1: One Way Analysis of Variance for Means between Cohorts
Chart 2: Additional Comparison of Means between Cohorts
Chart 3: Relationship between SES and Academic Preparedness Scores
Chart 4: Relationship between SES and Four Year Graduation Rates
Chart 5: Distribution of Academic Preparedness Scores for Non-Participants
Chart 6: Distribution of Academic Preparedness Scores for Participants
Chart 7: RQ1: Comparison of Means: Lowest SES (AP .80 < .90)
Chart 8: RQ1: Comparison of Means: Lowest SES (AP .83139 < .936)
Chart 9: RQ1: Comparison of Means: Low SES (AP .80 < .90)
Chart 10: RQ1: Comparison of Means: Did Not Apply for Aid (AP .80 < .90)
Chart 11: RQ1: Comparison of Means: Did Not Apply for Aid (AP < .80139)
Chart 12: RQ1: Comparison of Means: Did Not Apply for Aid (AP.80139 < .8664)
Chart 13: Four Year Graduation Means Plot for Non-Participants (n= 6,557)
Chart 14: ANOVA by SES: All Non-Participants
Chart 15: ANOVA by SES: Non-Participants (AP .80 < .90)
Chart 16: ANOVA by SES: Non-Participants (AP .90 < 1.0)
Chart 17: Asian American Distribution of Academic Preparedness Scores
Chart 18: Black/African American Distribution of Academic Preparedness Scores
Chart 19: Hispanic/Latino Distribution of Academic Preparedness Scores
Chart 20: White Distribution of Academic Preparedness Scores
Chart 21: Distribution of Students by SES Category for Racial/Ethnic Groups (.85 < .95)
Chart 22: Four Year Graduation Rates by SES and Race/Ethnicity (AP .85 < .95)
Chart 23: Data for Graduation Rates by SES and Race/Ethnicity (AP .85 < .95)
Chart 24: Differences in Four Year Graduation Rates by Major and SES
Chart 25: Differences in Graduation Rates by Participation and SES (AP .80 < .90)
Chart 26: Impact of SES and Academic Preparedness on Graduation
Chart 27: Data for Impact of SES and Academic Preparedness on Graduation

IMPACT OF SOCIOECONOMIC STATUS 9
Abstract
There will be challenges resulting from the goals of the completion agenda (Lee & Rawls, 2010),
underrepresentation of low socioeconomic status (SES) students at highly selective colleges
(Carnevale & Rose, 2003), and relationship between institution type and social mobility
(Haveman & Smeeding, 2006). If rates of access and success for low SES students are not
improved then the economic intentions behind the completion agenda may be compromised.
This study measured the impact of SES and academic preparedness on academic outcomes at a
highly selective, private, research university. Academic outcome data consisted of grade point
average (GPA) and completed units after the first and fourth year, persistence to the second year,
and graduation after the fourth year for the 2007, 2008, and 2009 freshmen cohorts. A
composite score of high school GPA and test scores was used to determine the academic
preparedness of students and the variable was statistically significantly for all academic
outcomes measured. The comparison of students of similar academic preparedness revealed SES
was statistically significant for GPA after the first and fourth year, first year units completed, and
four year graduation. When further examining the effectiveness of a student support program,
the results were inconclusive. Although the potentially at-risk students required to participate in
the program achieved similar outcomes when compared to non-participants of similar SES and
academic preparedness, the support program did not minimize the effects of SES. The findings
of this study further advance previous research pointing to the challenges faced by low SES
students in the areas of acceptance, belonging, and capital in higher education. The
identification of potential best practices to respond to this will require future research examining
the impact of SES at other universities, especially when academic preparedness is factored.

IMPACT OF SOCIOECONOMIC STATUS 10
Chapter 1: Overview
Background of the Problem
Each fall, at private colleges and universities throughout the United States, there are life-
changing moments experienced by lower socioeconomic status students. These situations
happen before the academic and social integration of Tinto (1975; 2006), before the student
interactions of Pascarella and Terenzini (1983), and long before the student involvement of Astin
(1993). According to Aries and Seider (2005), these students are awed by how pristine and
manicured the campus is. They see expensive cars, people wearing suits and fancy clothes, large
televisions, computers, gadgets, pricy bookstores and restaurants, and they see other students
who are acting as if everything is normal. And the student from a lower socioeconomic status
wonders and questions. How did I get here? How am I going to do this? How am I going to pay
for this? I guess I'm not normal. I'm poor... Do I even belong here?
The Completion Agenda
The United States has identified a goal to improve graduation rates nationally (Lee &
Rawls, 2010). In 2007, 55.8% of the Canadian population between the ages of 25 to 34 had an
associate degree or higher. That rate led the world and the United States, with a rate of 40.4%,
trailed by a considerable margin. The completion agenda of President Obama calls for the
United States to increase its percentage to 55% by the year 2025 (Lee & Rawls, 2010). These
goals are widely accepted because the general view is that an educated workforce is necessary
for the economic health and vitality of the country (Crellin, Kelly, & Prince, 2012).
As this pursuit continues, it will likely be guided by research. There is substantial
literature on the predictive quality of pre-college input variables on graduation rates (Astin,
1997), on the issue of equity within higher education (Bensimon, 2005), and the challenges of
IMPACT OF SOCIOECONOMIC STATUS 11
assessing institutional efficiency (Kuh & Pascarella, 2004; Webber & Ehrenberg, 2010). The
economically disadvantaged have also been shown to be the least likely to pursue a college
degree and also the ones that could benefit the most from receiving it (Brand & Xie, 2010).
There has been relatively limited research, however, when it comes to the impact of
socioeconomic status on college outcomes such as retention and graduation (Reason, 2009;
Langhout, Drake, & Roselli, 2009; Walpole, 2003).
Highly Selective Colleges and Universities
Individuals who earn a college degree are more likely to increase their earnings potential
and social class (Haveman & Smeeding, 2006). Considering graduation rates are highest at
highly selective universities (Carnevale & Rose, 2003), degree seekers at these universities are
more likely to reach their goal. Studies have also found that students graduating from highly
selective colleges and universities have a higher earnings potential when compared to graduates
of less selective institutions (Brewer, Eide, & Ehrenberg; 1999; Fox, 1993; Thomas & Zhang,
2005). Highly selective universities are generally defined as those that admit the lowest
percentage of entering freshmen. According to the U.S. News & World Report (2011) guide to
colleges, the institutions with the lowest acceptance rates in 2011 ranged from 3.2% to 33.1% at
the 100 most selective public and private schools. In other words, 66.9% to 96.8% of freshman
applicants at these schools were denied admission.
In addition to finding that students attending highly selective colleges and universities are
also more likely to graduate, Carnevale and Rose (2003) also found that as institutional
selectivity increased so too did graduation rates. This finding also applied for students from
every socioeconomic status (SES) level. Students from higher SES backgrounds still graduated
at higher rates than their lower SES peers, but students of similar SES backgrounds were
IMPACT OF SOCIOECONOMIC STATUS 12
graduating at higher rates based on the selectivity of the college or university attended
(Carnevale & Rose, 2003). In other words, the two greatest predictors of graduation were the
selectivity of the school and the SES of the student. According to Haveman and Smeeding
(2006), however, the number of low SES students enrolled in highly selective institutions was far
fewer than the total of low SES students that were academically qualified.
Rising Costs of Tuition
The rising costs of attending college are a national concern (Driscoll, Comm, &
Mathaisel, 2013). These increasing costs are particularly meaningful to students of low SES.
Financial aid is viewed as a means of making college more affordable, but low SES students are
responsible for paying a disproportionately higher rate, when comparing the net cost of attending
to family income at highly selective, private, institutions (Hill, Winston, & Boyd, 2005). The
full cost of attending after accounting for aid that does not need to be repaid is referred to as the
net cost of attending and presents a more accurate depiction of what students and families are
responsible for paying immediately or through loans (Hill, et al., 2005). The costs of attending
highly selective, public, institutions are also of concern because state subsidies and policies
regularly lower the cost of attending for all students even though the higher SES students are
fully capable of paying full price (Haveman & Smeeding, 2006). No matter what type of
institution a low SES student attends, on average, the cost of attending they are responsible for,
in relation to their family income, will be inequitably higher when compared to that of their
higher SES peers.
Students from low SES backgrounds experience a multitude of challenges. The rate of
low SES students attending highly selective institutions is significantly low (Carnevale & Rose,
2003; Goldrick-Rab, 2006; Walpole, 2005). Socioeconomic status can be linked to standardized
IMPACT OF SOCIOECONOMIC STATUS 13
test scores and many low SES students may not have the necessary test scores to gain admission
into highly selective schools (Hoffman & Lowitzki, 2005). Low SES students who do attend
must then adjust to the challenging upper-middle class norms of their campus (Langhout, et al.,
2009) and the wealth of their peers (Aries & Seider, 2005).
Sense of Belonging
A student’s sense of belonging on a college campus has been connected with academic
outcomes (Freeman, Anderman, & Jensen, 2007; Hurtado & Carter, 1997; Locks, Hurtado,
Bowman, & Oseguera, 2008; Pittman & Richmond, 2008). Researchers have also found that
social class background influences the sense of belonging of a student (Ostrove & Long, 2007).
A study by Aries and Seider (2005) revealed that the prominence of wealth at prestigious private
colleges results in feelings of exclusion and powerlessness for lower income students.
Social and Cultural Capital
Social capital generally refers to the potential or actual resources available through
exchanges between persons within a social network (Avery & Daly, 2010; Bourdieu, 2008;
Coleman, 1988; Stanton-Salazar, 1997). Within education, the relationships students potentially
establish with teachers and counselors, according to Stanton-Salazar (1997), are extremely
valuable because they provide students with the opportunity to gain institutional knowledge.
Certain aspects of the campus-specific knowledge described by Stanton-Salazar (1997) can be
viewed as similar to the cultural capital of Lareau and Weininger (2003). In their review, they
posed that cultural capital equates to "the direct or indirect "imposition" of evaluative norms
favoring the children or families of a particular social milieu" (pp. 597-598).
Low SES students are at a disadvantage when it comes to the economic, social, and
cultural capital demands of attending college (Aries & Seider, 2005; Langhout, et al., 2009;
IMPACT OF SOCIOECONOMIC STATUS 14
Ostrove & Long, 2007; Walpole, 2003). These three concepts of capital come from the writings
of Bourdieu (2008) and are relevant to the challenges that low SES students face when
attempting to earn a college degree to move up the social class ladder. The economic capital that
low SES students lack is evident in their financial disadvantages. The social and cultural capital
available to students of varying SES also impacts academic outcomes (Aries & Seider, 2005;
Avery & Daly, 2010; Langhout, et al., 2009; Ostrove & Long, 2007; Walpole, 2003).
Based on the literature (Aries & Seider, 2005; Langhout, et al., 2009), low SES students
could be expected to have challenges adjusting to the upper middle class norms of a highly
selective, private, university because they lack the cultural capital relevant to the practices and
nuance within the campus environment. These challenges are evident in the findings of
Langhout et al. (2009) that students from low SES backgrounds are more likely to experience
concerns with school belonging, negative psychosocial outcomes, and greater intentions of
leaving school. Similarly, Johnson, Richeson, and Finkel (2011) found that low SES students at
an elite university felt socially stigmatized and were more likely to question their academic fit
and struggle with self-regulation when balancing their social concerns and school work.
Walpole (2003) found that low SES students work more, study less, are less involved,
and report a lower grade point average than high SES students at four-year schools. Low SES
students simply do not have the economic and cultural capital that an upper-middle class college
campus desires. Social capital is also more challenging to acquire for low SES students because
they need to work more and typically have less to exchange or trade when establishing a social
connection.
Within the literature on low SES students there have been recommendations for
improving access (Carnevale & Rose, 2003; Park, Denson, & Bowman, 2012) and sense of
IMPACT OF SOCIOECONOMIC STATUS 15
belonging (Aries & Seider, 2005; Langhout, et al., 2009; Ostrove & Long, 2007). Upon
analyzing whether low-income students would benefit from affirmative action policies at the top
universities in the United States, Carnevale and Rose (2003) found that low SES students from
the lowest quartile only accounted for three percent of the college attending population and only
nine percent of the student population came from the lowest 50% based on SES. They examined
simulations for five alternative methods of admission at the 146 most selective colleges and
recommended that racial affirmative action be maintained as much as possible and that economic
affirmative action be expanded. Similarly, Park et al. (2012) challenged highly selective colleges
to increase efforts toward the recruitment of academically prepared students from low-SES
backgrounds. Carnevale and Rose (2003) point out, however, that admission for low SES
applicants is only a single hurdle to surpass. The next, and more challenging, barrier is figuring
out how to pay the costs at a top college or university.
The sense of belonging of low-SES students attending elite universities also remains a
concern. Aries and Seider (2005) recommended that faculty and administrators at prestigious
colleges and universities consider how low SES students can better be supported and involved.
Langhout et al. (2009) insisted that schools, in addition to admitting students from low SES
backgrounds, must also address issues within the campus culture that might prevent the low SES
students from feeling like they belong. Ostrove and Long (2007) encouraged institutions to be
more aware of how welcoming and inclusive they are for students from low SES backgrounds.
Johnson et al. (2011) recommended that future studies attempt to measure which indicators
related to SES; such as family income, parental education, parental occupation, and financial aid;
have a greater impact on educational outcomes.
IMPACT OF SOCIOECONOMIC STATUS 16
Unfortunately, there are challenges when it comes to the recommendations found within
the research and additional complications in regards to the assessment of these
recommendations. How can highly selective institutions convince greater pools of academically
qualified students from low SES backgrounds to apply? What might happen to make highly
selective colleges and universities more affordable to low SES students, and how might potential
applicants learn about this information? What does a more welcoming campus look like? What
additional support is beneficial to low SES students and how would that support be provided?
The recommendations lead to multiple questions that have yet to be answered.
Considering the goal of increasing the graduation rate of 25 to 34 year olds in the United
States, the lack of attention given to SES is surprising. Students from low SES backgrounds are
underrepresented in higher education (Carnevale & Rose, 2003) and this is especially true at
highly selective universities. Researchers have shown that low SES students struggle more with
finding a sense of belonging (Aries & Seider, 2005; Langhout et al., 2009; Ostrove & Long,
2007) and also are limited when it comes to the many forms of capital required within higher
education (Avery & Daly, 2010; Walpole, 2003). Although support programs exist (Dennis,
Phinney, & Chuateco, 2005; Grant-Vallone, Reid, Umali, & Pohlert, 2003; Padgett & Reid,
2002), a plan to increase graduation rates nationally will need to include a more informed
strategy for aiding students from low SES backgrounds.
Statement of the Problem
There have been studies that examined access to higher education (Walpole, 2003) and
the net cost of attending for low SES students (Hill, Winston, & Boyd, 2005). Researchers have
not yet fully examined the resulting persistence, grade point averages, units earned, and
graduation rates of low SES students. The results and potential gaps are unknown. The research
IMPACT OF SOCIOECONOMIC STATUS 17
to date has also relied heavily on student reported data pulled from surveys to calculate SES
(Astin & Oseguera, 2004; Carnevale & Rose, 2003; Goldrick-Rab, 2006; Walpole, 2003), as
opposed to the actual data provided by families to financial aid offices. Whether college students
generally have the perspective and knowledge necessary to accurately answer survey questions
about the finances and job occupations of their parents is a question worth asking.
Considering the popular belief and evidence that receiving a college degree increases the
opportunity to achieve greater future earnings, it is surprising that the impact of current family
earnings on academic outcomes has not received more attention. If low SES students are not
receiving the same access or achieving the same success then it would appear that highly
selective institutions have become a filter that further divides and reciprocates the division by
socioeconomic status within this country. This concern is especially magnified considering the
increasingly capitalistic practices of colleges and universities (Brewer, Gates, & Goldman, 2002;
Giroux, 2002; Rhoades, 2006). This trend will likely further push the campus culture at highly
selective institutions from upper-middle class toward upper class and elite class norms and
beliefs. Continuation of these practices could further confuse, alienate, and push away current
and potential low-SES students during a time when the country is trying to increase graduation
rates and more equitable academic and economic opportunities.
The limited research on the impact of student support programs on low SES students also
requires further attention (Rhoades, 2006). The importance of sense of belonging has been
related to SES (Aries & Seider, 2005; Ostrove & Long, 2007), as has social (Avery & Daly,
2010; Stanton-Salazar, 1997) and cultural capital (Walpole, 2003). Recommendations for
improving the success of low SES students have emphasized the importance of campus culture
(Langhout et al., 2009; Ostrove & Long, 2007). The importance of connecting minorities with
IMPACT OF SOCIOECONOMIC STATUS 18
institutional agents has also been detailed (Stanton-Salazar, 1997). The variables that contribute
to the SES of a student and how they individually correlate to academic outcomes have also been
recommended for future research (Johnson et al., 2011). So it would reason that programs
offering support, similar to what has been recommended within the literature, might already have
established best practices that could be generalized across different institutions.
A full examination of the literature reveals the presence of gaps when it comes to access
and outcomes for low SES students, in addition to gaps within the research itself. These gaps are
problems that require attention. Table one demonstrates how a more complete study of SES is
needed. Given the overlapping areas of study, there is a need for additional research. More
research and information gathering must occur in order to better understand how the persistence,
grade point averages, units earned, and graduation of low SES students can be supported and
improved, particularly at highly selective, private, colleges and universities.
Table 1 reveals a need for research that examines a more complete array of the key issues
and components related to SES in higher education. As example, a study could not be located
that examined the actual pre-college SAT scores and HSGPA, financial aid information, and
college outcomes data for individual students. As such, there are questions left unanswered by
the existing research. For instance, how are low SES students performing when compared to
other students with similar SAT scores and HSGPA? Are low SES students of a different
race/ethnicity achieving similar outcomes? Do student support programs improve outcomes?
These questions and many others remain unanswered.
Purpose
The purpose of this study was to assess how entering freshmen of various socioeconomic
backgrounds performed academically at a highly selective, private, research university.
IMPACT OF SOCIOECONOMIC STATUS 19
Researchers have examined equity, institutional efficiency, pre-college variables, financial aid,
sense of belonging, social capital, cultural capital, and student support programs and all signs
indicate that low SES students are struggling to gain admission and earn degrees at the highly
selective colleges and universities that may be beneficial to their future. This study further
examined this situation and extends the research further.
The study attempted to account for demographics, pre-college variables, and other inputs
such as choice of major and financial aid. Examining the descriptive statistics based on these
variables made it possible to answer multiple questions. For instance, did the lower SES
population have a different composition of race/ethnicity and/or pre-college high school grade
point average and/or test scores? How also did SES impact the major pursued by students?
Research that answers all of these basic questions has not been found in the literature.
A student support program was also examined to determine whether students within the
program perform differently when compared to similar students outside of the program. By
comparing the academic outcomes of participants to non-participants, the study aimed to learn
whether the additional support impacted academic outcomes differently based on the SES of the
student. For instance, did the academic outcomes of low SES participants compare well to the
outcomes of low SES non-participants?
Existing data from three consecutive freshmen cohorts were assessed. The data included
students from the entering freshmen cohort of 2007 through to the cohort of 2009. Persistence
rates, cumulative grade point average (GPA), units earned, and graduation rates were examined.
Demographics such as gender, race/ethnicity, and first-generation college student status were
also included. Pre-college variables consisted of best SAT scores, high school GPA (HSGPA),
and adjusted HSGPA. These pre-college variables were used to form a composite range of
IMPACT OF SOCIOECONOMIC STATUS 20
academic preparedness. For the research questions the key differences between groups were
based on SES, academic preparedness, or participation in the support program.
Research Questions
There are two primary research questions for this study.
1. For first-time freshmen, what differences exist when the persistence, grade point average,
units earned, and degree completion are compared for students required to participate in a
support program and non-participants of similar socioeconomic status and academic
preparedness?
a. What other differences between the participants in the support program and non-
participants can be found when additionally examining first-generation status,
gender, ethnicity, and pursued major?
2. For the general population of non-participants outside of the support program, how did
socioeconomic status and academic preparedness impact the persistence, grade point
average, units earned, and degree completion of first-time freshmen?
a. When also examining first-generation status, gender, ethnicity, and major, were
differences found when comparing students of similar SES and academic
preparedness?
b. If there were differences in persistence, grade point average, units earned, and
degree completion for students within particular SES and academic preparedness
ranges, were these differences similar to those found for participants of the
support program population?

IMPACT OF SOCIOECONOMIC STATUS 21
Significance of the Study
This study is significant because the literature related to SES and academic outcomes
research is limited (Reason, 2009; Walpole, 2003). The knowledge generated from this study is
beneficial to university administrators, researchers, and student affairs practitioners. Bensimon
(2005, 2007) has noted that practitioners can benefit from participating in the collection and
analysis of research data. Harper and Kuh (2007) provide argument that faculty within higher
education should use more qualitative methods in their research. The opposite, as Bensimon
proposes for student support practitioners, is also likely true in that student affairs practitioners
should be more familiar with the quantitative data regarding the students they are serving.
Practitioners already working closely with students could benefit from gaining more
specifics regarding the performance of low SES students. The increased knowledge could help
provide understanding and motivation that aid feelings of belonging, scheduling of more
appropriate event programming, and productive discourse on how SES relates to campus climate.
Possible adjustments to the costs of certain activities could also further promote a campus
climate that is more understanding of socioeconomic differences.
Admission and financial aid officers are capable of examining the data and background
information within this study to decide how current practices match the mission and execution of
goals at the university. In theory, admission and financial aid offices do not work in a vacuum.
Communication and collaboration with the rest of the university is vital, especially when it
comes to the educational outcomes of students. If a certain population is struggling, then
perhaps they could be better directed to support services upon admission.
Senior level administrators could use the knowledge generated in this study to think
about how fees and mandatory deposits might influence whether students feel they can afford to
IMPACT OF SOCIOECONOMIC STATUS 22
attend and persist. The concept of students following different paths within higher education
based on their SES should be tangible to senior level administrators. In the increasingly
capitalistic world of higher education (Giroux, 2002), where students are viewed as consumers
with discretionary money to spend, how much do schools value a low SES student who arrives
with a smaller wallet? What impact would the completion agenda efforts of Obama have if low
SES students were not valued by the emerging capitalistic system? This study is significant
because it points out how the goals and direction of higher education may not be aligned.
Limitations
There were multiple limitations within this study. Educational outcomes were only
examined for entering freshmen at one particular university. Applicants who applied but were
denied admission, as well as those who were admitted but chose not to attend, were not included
in the study. An analysis of freshmen applicants who did not attend the university would
certainly provide for greater learning, but that was not possible in this study. Carnevale and
Rose (2003) even recommend that researchers look at the academic outcomes of students with
similar characteristics who attend colleges and universities of different quality. For this study
though, that data was not accessible. There also were a small number of students with missing or
incomplete data. Some students may have also had incorrect data. For instance, a transfer
student may have been coded as a freshman, or vice versa.
The fall entering first-time freshman cohort years of 2007, 2008, and 2009 were analyzed
because that data and the graduation status for those students were available. The peak of the
financial crisis of 2007 and 2008 likely had a significant impact on the students within this study,
as well as on the students that decided not to attend the university during that time period. There
likely were students that entered prior to the economic crisis that needed to withdraw later on for
IMPACT OF SOCIOECONOMIC STATUS 23
financial reasons. The entire admission to matriculation process after 2007 for students, families,
and the university likely also changed along with the economic conditions of the time.
Delimitations
Delimitations within this study can be found with the calculation of SES. Whereas
previous studies have relied on survey data completed by students (Aries & Seider, 2005; Astin
& Oseguera, 2004; Carnevale & Rose, 2003; Goldrick-Rab, 2006; Walpole, 2003), the
calculation of SES in this study will be reflective of the internally calculated expected family
contribution totals used by the university for the first academic year of attendance. Although
parental work occupation data were accessible, it was not used. The applicant provided
occupation of parents, found within the admission applications, can be too inconsistent and often
times vague. It is also uncertain how much communication occurs between applicants and their
parents when answering questions about parental occupation. Attempting to measure social class
from this data would be troublesome. Figuring out how to weigh this information to determine
SES would also be complicated. For these reasons, the parental occupations reported by students
within their college application were not used and the internally calculated expected family
contribution for the first academic year was relied upon instead. A known delimitation related to
this calculation in the expected family contribution of a student may have changed from year to
year due to changes in family income and assets. These changes could have impacted whether a
student was able to persist at the university.
There are other decisions that were made for this study that are also delimitations.
Academic preparedness consisted of the composite score for high school grade point average
(HSGPA), internally adjusted HSGPA, and standardized test scores. Even though research has
demonstrated that high school grade point average has been a better predictor of college success
IMPACT OF SOCIOECONOMIC STATUS 24
(Geiser, 2007; Kobrin, Patterson, Shaw, Mattern, & Barbuti, 2008; Zwick & Sklar, 2005), these
same authors have also recommended that HSGPA be used together with standardized test scores
when it comes to predicting the academic preparedness of a student and making decisions
regarding admission.
The degree programs pursued by students were also combined into categories. Pursued
degree programs were placed into a category, with the options being architecture, arts, business,
communication, engineering, humanities, natural sciences, social sciences, and undecided. For
instance, fine arts majors and theatre majors were both placed into the arts category. Biology
and mathematics majors were categorized within natural sciences. The pursued degree program
for students at the end of their first semester of enrollment was used in the analysis. There most
certainly were students that changed their degree program later on, even those that double-
majored within different categories. This delimitation accepted the reality of this challenge
within the research methodology.
This study also focused solely on domestic students. International students have an
enormous impact on colleges and universities, particularly when it comes to impacting the
campus environment and culture. The international population also heavily influences the
economic capabilities and strategies of an institution. International students are not eligible for
financial aid, however, and present a complication when it comes to analyzing SES. This study
did not fully examine the data for international students. Descriptive data was provided on the
international student population to provide a better context of the campus environment, but
further analysis was not performed.
Additional students were also excluded from the full analysis. Student athletes were not
included due to their wide variance in aid packages, academic preparedness, time committed to
IMPACT OF SOCIOECONOMIC STATUS 25
extracurricular activities, and support available to them. Students pursuing an intensive five year
architecture degree program were also excluded. The remaining exclusions were the result of
insufficient or missing data. For instance, students with missing high school grade point
averages or standardized test scores were removed from the full analysis.
Definitions
Academic Outcomes – Grade point average, units earned, persistence, and degree completion.

Academic Preparedness – The composite score of high school grade point average, adjusted high
school grade point average, and best standardized test score.

Classism – Prejudice, discrimination, and/or the unfair treatment of a person or group of people
based on their social and/or economic class

Federal Expected Family Contribution (EFC) – According to the Free Application for Federal
Student Aid (FAFSA) website the Expected Family Contribution (EFC) is a measure of your
family's financial strength based on a formula that includes taxed and untaxed income, assets,
benefits, family size, and number of family members attending college during a particular year
(http://www.fafsa.ed.gov/help/fftoc01g.htm).

Institutional Expected Family Contribution (IEFC) – This is similar to the Federal EFC, but the
internal calculation by the school also accounts for additional family assets, such as home equity,
when figuring what each family will be expected to contribute financially.

IMPACT OF SOCIOECONOMIC STATUS 26
First Generation College Student – Generally defined as the first generation within a family to
attend a college or university.

Highly Selective Colleges or Universities – Institutions that generally admit the lowest
percentage of freshman applicants. These schools are thus highly selective in regards to
admitting students.

High School Grade Point Average (HSGPA) – For this study and site, a 4.0 HSGPA refers to the
maximum score of an “A” in every high school course.

Adjusted HSGPA – For this study and site, the adjusted HSGPA accounts for rigor of the school
and coursework, including AP and IB coursework, and the maximum score is equal to a 4.7.

Net Cost of Attending – This is the yearly price that students are responsible for paying.
Scholarships and grants are deducted from the full cost of tuition, room, board, books, and other
expected expenses. Loans and work-study amounts do not impact the net cost of attending
because this is money the student and/or family will either have to pay off or earn.

Social Mobility – Typically related to an increase in social class and family income, as opposed
to a decrease.

Socioeconomic Status (SES) – Although the calculation of SES varies, it generally refers to a
combination of family income and social class.
IMPACT OF SOCIOECONOMIC STATUS 27
Conclusion
This introductory chapter has presented the problem. Evidence has been shown that the
needs and opportunities of low SES students require further understanding and support. If social
mobility is the primary goal of achieving a college degree, and attending a highly selective
college or university offers the greatest opportunity for upward mobility in this country, then
what purpose will the completion agenda of Obama solve if qualified low SES students continue
to be underrepresented and continue to underachieve at highly selective institutions? Will highly
selective institutions become an increasingly stringent filter that further divides and reciprocates
the division of wealth? This study seeks to further measure the validity of this concern while
also exploring the assistance that support programs can offer to improve the situation. Chapter
Two reviews and elaborates on the literature related to the problem and frames it within a
theoretical perspective.

IMPACT OF SOCIOECONOMIC STATUS 28
Chapter 2: Literature Review
The previous chapter introduced the need for colleges and universities to better
understand and assess the impact that socioeconomic status (SES) has on academic outcomes
such as persistence and graduation. This has become necessary given the need to improve the
college graduation rates of students. The United States has fallen behind other countries when it
comes to citizens aged 25 to 34 earning an associate degree or higher. According to the 2010
Progress Report by the College Board (Lee & Rawls, 2010), Canada led the world in 2007 with a
55.8% of its population completing a postsecondary degree and the United States trailed
considerably with only a 40.4% of its citizens aged 25 to 34 with an associate degree or higher.
In response, President Obama has increased national expectations in an environment already
riled by longstanding competition between institutions for respect, prestige, quality of students,
and money (Brewer, Gates, & Goldman, 2002). The goal for the United States is to increase the
percentage of 25 to 34 year olds with an associate degree or higher to 55 percent by 2025 (Lee &
Rawls, 2010). The future demands have already led to concerns over how these national goals
will be assessed (Humphreys, 2012; Schneider, 2012) and whether students will be negatively
impacted by the shifts in strategies likely to occur (Evenbeck & Johnson, 2012; Kelderman,
2011; Walters, 2012). In a review of persistence research, Reason (2009) finds that students
from low SES backgrounds are underrepresented in the current research and the impact of SES
must be further explored in future studies.
Driscoll, Comm, and Mathaisel (2013) reported that the American higher education
system needs to become more accessible and sustainable due to rising costs, tuition, class sizes,
and the continued exclusion of poor and minority students. The authors, in their review,
identified availability, dependability, capability, affordability, and marketability as the five areas
IMPACT OF SOCIOECONOMIC STATUS 29
requiring attention in order to improve the sustainability of higher education. Availability brings
people to a university and provides access to technology, materials, facilities, tools, and
relationships with others. Dependability impacts whether students, faculty, and staff persist.
Capability insists that proper assessment measures are taken. Affordability requires that schools
examine their financial assets and consider how costs can be lowered without extensively
sacrificing quality. In marketability, an institution needs to understand its advantages and
properly communicate them as part of a marketing plan. Collectively, these five areas (Driscoll
et al., 2013) capture the core business components of higher education. These abilities are also
vital when it comes to understanding SES at a highly selective, private, research university.
This chapter will examine how student success is typically measured at four year
universities. The research related to the impact of SES on academic outcomes will then be
synthesized, particularly at highly selective, private, institutions. Theoretical frameworks will
then be connected to the literature findings to offer additional perspectives and guidance. The
aim will be to present the current condition of the problem and the necessity for additional
research.
The Need for Quality Assessment
Much of the research on retention and graduation can be separated into three areas. Each
of these three types of studies offers a distinct viewpoint for understanding the impact of SES.
There is research that looks to better predict graduation rates based on input variables (Astin,
1997; Reason, 2003; Reason, 2009), there are those that explore the equity within the
populations and outcomes (Alon & Tienda, 2005; Bensimon, 2005; Hoffman & Lowitzki, 2005),
and there are examinations of overall institutional efficiency (Goenner & Snaith, 2003; Jones-
White, Radcliffe, Huesman, & Kellogg, 2010; Kokkelenberg, Sinha, Porder, & Blose, 2008; Kuh
IMPACT OF SOCIOECONOMIC STATUS 30
& Pascarella, 2004; Ryan, 2004; Sanford & Hunter, 2011; Webber & Ehrenberg, 2010;
Wellman, Johnson, & Steele, 2012).
Predictive
Astin (1997) called for the evaluation of student inputs to better predict retention in the
study titled "How 'good' is your institution's retention rate?" The student inputs described
consisted of high school grades, admission test scores, race, and gender. The question posed is
whether a university's performance data, such as average retention and graduation rates, tell the
full story of what actually happened. For instance, does a high or low retention and graduation
rate reflect the quality of a school or the quality of the students attending? As such, the author
obtained data from a national sample of 365 baccalaureate-granting institutions through a 1985
Cooperative Institutional Research Program (CIRP) survey and a follow up with the registrar’s
office of each school in the summer of 1989. A model was developed using high school grades,
SAT scores, gender, and race to calculate an expected retention rate. Institutions considered to
be successful would have a higher actual retention rate than that which the prediction model
would expect. Even though Astin (1997) did not include SES in this study, the need to examine
additional variables was demonstrated.
Astin (1993) previously identified the importance of SES. Student segregation by SES
within peer groups was prevalent at the undergraduate level and Astin claimed that segregation
by SES on college campuses was influenced more by the structures and policies of the institution
than that of the choices of an individual student. High SES students were more likely to attend
institutions with large numbers of their high SES peers. Meanwhile, low SES students primarily
attended schools with a wider mix of students by SES.
IMPACT OF SOCIOECONOMIC STATUS 31
Research examining the predictive quality of high school grade point average (HSGPA)
and standardized test scores, such as the SAT, have been relatively uniform in declaring that
HSGPA is a better predictor of academic success in college when compared to standardized test
scores such as the SAT (Geiser, 2007; Kobrin et al, 2009; Zwick & Sklar, 2005), but that both
should be used together in combination. Acknowledging that differences between high schools
make it challenging for college admission offices to compare HSGPA between schools, Geiser
(2007) identified HSGPA, particularly in college-preparatory class-work, as being superior to
standardized test scores when it comes to the prediction of college grades. Kobrin, et al (2009)
found that the best combination for predicting first-year college GPA utilized both HSGPA and
SAT scores. Similarly, Zwick and Sklar (2005) examined how HSGPA and SAT scores
predicted college grades and graduation for students of difference ethnic backgrounds and first
language. They found that looking at HSGPA and test scores together allowed them to account
for a larger variance in first year GPA when compared to looking at the variables separately.
For many of the studies examining HSGPA, it is not clear whether a pure HSGPA on a
4.0 scale was used or an adjusted HSGPA that accounted for advanced placement and honors
coursework. For instance, Sadler and Tai (2007) have examined how honors and advanced
placement (AP) courses are regularly examined by college admission offices when weighing
HSGPA. They noted how high schools do not follow any standard weighting system, so there is
pressure placed on college admission offices to figure out how to compare HSGPA for different
students from the same high school and also for students from various different high schools.
Sadler and Tai (2007) found that students who earned a B in an AP course at the end of high
school do not do better in the same college subject when compared to students that received an A
in the regular high school course. Additionally, a student that received a grade of C in an AP
IMPACT OF SOCIOECONOMIC STATUS 32
course performed significantly worse when compared to students that received an A in the same
regular high school class. Overall though, Sadler and Tai (2007) claimed that additional weight
added to HSGPA is reasonable for AP coursework, but they found that honors coursework was
not as consistent or impactful.
Reason (2003; 2009) has thoroughly examined research on persistence. Reason (2009)
acknowledged that research on student persistence occurs at multiple institutions but that efforts
to improve persistence must still be institutionally specific. In a review of persistence studies
published between 1990 and 2002, Reason (2003) found that the changing demographics within
higher education made it important to further examine how the increasing diversity would impact
the predictive capability of certain demographic variables. For instance, the literature reviewed
by Reason revealed that predictor variables such as SES, race/ethnicity, and gender should
continue to be used. Over time though, significant changes in the percentage of students on a
campus of a particular SES, race/ethnicity, or gender, as examples, would not only impact the
campus climate, but also potentially influence the relationship the variable has on persistence. It
is uncertain though whether these continued changes to the campus climate will increase or
decrease persistence and graduation rates. What is clear, however, is that the variables
previously found to have an impact on retention must constantly be examined, especially as the
diversity of a campus climate changes.
Equity Driven
Studies focused on equity acknowledge the complications from using a purely predictive
approach because some variables are not addressed. Furthermore, it is not just a matter of which
variables are more predictive of success and which are not. Equity research seeks to understand
why the results are what they are and to gain perspective on what could and should be happening
IMPACT OF SOCIOECONOMIC STATUS 33
instead. Bensimon (2005) addressed disparate outcomes for underrepresented minority students,
calling for the disaggregation of student outcome data by race, ethnicity, and gender. Taking a
qualitative approach, Bensimon examined the attitudes, values, and actions of faculty members,
administrators, and other employees. Bensimon learned that practitioners within higher
education would need to see the disaggregated data, and even participate in the collecting, before
they could accept the reality of the achievement gap and become more equity minded. The need
for organizational learning by those closest to the problems was identified as an overall finding
(Bensimon, 2005). Only by participating in the search for a solution did professionals shift from
viewing underrepresented students from a deficit model, blaming them for their own problems,
to that of an equity perspective that sought fairness and a deeper understanding of the problem.
Concerns over equity have also led to research that challenges variables traditionally
considered predictors of success. Hoffman and Lowitzki (2005) found support that HSGPA was
a better predictor of academic achievement than SAT scores at a predominantly White Lutheran
university, particularly for students of color. Using data from 522 full-time degree seeking
students who completed coursework during the fall semester of 2000, the authors found that
HSGPA was a stronger predictor of academic achievement than SAT scores for all types of
students and that HSGPA was also a significant predictor of retention for Latino students.
Hoffman and Lowitzki (2005) used correlations and a priori path analysis model to find their
results. Niu and Tienda (2011) found that class rank is a better predictor of college performance
than test scores at five Texas universities. Niu and Tienda accounted for how the economic
composition of high schools could impact test scores and discussed the implications of
residential segregation along class and racial lines.
IMPACT OF SOCIOECONOMIC STATUS 34
A section on equity cannot be complete without addressing affirmative action. Some of
the research related to SES stems from and has been in direct response to the court battles
addressing affirmative action in the last decade (Alon & Tienda, 2005; Carnevale & Rose, 2003;
Hoffman & Lowitzki, 2005; Park, Denson, & Bowman, 2012). Although affirmative action can
be closely linked with SES (Carnevale & Rose, 2003) this paper will aim to keep the focus on
SES and the factors related to it.
Alon and Tienda (2005) studied the mismatch hypothesis that believed minority students
who attended selective colleges and universities would have lower graduation rates than those
that went to less selective institutions. The authors looked at same-group student comparisons
across institutions of varying selectivity. Their findings suggest that the mismatch hypothesis
should be rejected because the likelihood of graduation for minority students increases when the
selectivity of the school is higher. As part of the findings, the authors acknowledge the overall
graduation rates for underrepresented populations remain lower and that more must be done to
help disadvantaged students at all types of colleges and universities.
Park, Denson, and Bowman (2012) tried to identify whether socioeconomic diversity on a
campus made a difference. Their aim was to understand if admission preferences based on SES
as opposed to race could offer some of the same benefits that race-conscious admission policies
provided. Using data from the Cooperative Institutional Research Program (CIRP) at the Higher
Education Research Institute at the University of California, Los Angeles, the authors examined
the 2003 Student Information Form (SIF) and the 2007 College Student Survey (CSS). These
methods made it possible for them to gather pre and post college entrance data on 14,894
students from 88 different institutions. The authors found that engagement with racial diversity
is influenced by socioeconomic diversity. They also found that students who engaged in high
IMPACT OF SOCIOECONOMIC STATUS 35
levels of cross-class interactions were more likely to experience co-curricular diversity and
cross-racial interactions. With these findings, the authors suggest that selective and highly
selective institutions could improve racial diversity and cross-racial interactions by more
aggressively attracting academically talented low income students of all races. They do,
however, acknowledge that admission preferences made by SES instead of race may possibly
reduce the racial diversity that is provided through race-conscious admission practices. For this
reason, they argue that universities should understand there are benefits to having both racial and
socioeconomic diversity.
Institutional Efficiency
The bulk of research examining institutional efficiency is similar to the predictive studies,
but is typically focused on expenditures in relation to persistence and/or graduation rates at the
initial institution (Ryan, 2004; Sanford & Hunter, 2011; Webber & Ehrenberg, 2010). Whether
students transfer to and graduate from another institution (Jones-White et al., 2009) and the
complication with evaluating the quality of a college without accounting for selectivity (Kuh &
Pascarella, 2004) are also included in this section. As opposed to predictive studies, research
addressing institutional efficiency pays more attention to how schools are directing resources and
reaching expected goals.
Goenner and Snaith (2003) examined institutional factors along with student
characteristics in connection with graduation rates. They focused their study on 258 doctoral
universities and found that positive correlates to graduation included high school class rankings,
standardized-test scores, and the percentage of out-of-state students. The institutional factors
easily linked with improved graduation rates were student-faculty ratios, percentage of full-time
faculty, total expenditures, and tuition and fees helped predict student outcomes. In sum,
IMPACT OF SOCIOECONOMIC STATUS 36
however, the authors essentially provided evidence that an institution could increase its
graduation rates by charging more, spending more, and changing admission policies so that
enrollment is more highly composed of students that are better prepared.
According to Kuh and Pascarella (2004), selective institutions have high graduation rates
and are positively connected to better earnings after college. They point out that university
ranking publications tend to demonstrate which schools are most selective rather than most
effective. When it comes to gauging the quality of an institution, the authors suggest that more
should be done to show what happens to students during college. This supports the notion that
an institution should account for the selectivity of its student population when assessing
institutional efficiency.
Hoping to fill a gap in the literature, Ryan (2004) examined the connection of
institutional expenditures to six year graduation rates at 363 Carnegie-classified Baccalaureate I
and II institutions. Ryan aimed to address five questions.
1) Is there a relationship between expenditures and persistence to degree attainment?
2) Does support for student services, academic support and instruction help to explain
variations in persistence to degree?
3) Do the findings clarify contradicting claims about expenditure effects?
4) Do the findings warrant further investigation?
5) What are the potential implications for the development of theories of student
persistence, institutional decision making and public policy? (pp. 8-9)
Utilizing the IPEDS Peer Analysis System, the researcher obtained data from 363
institutions (Ryan, 2004). Using a non-experimental, applied research design, the expenditures
per full-time equivalent student at each institution was examined for the areas of instruction,
IMPACT OF SOCIOECONOMIC STATUS 37
academic support, student services, and institutional support. The findings by Ryan suggest that
expenditures do impact graduation, particularly instructional and academic support expenditures.
Spending on student services, despite the literature connecting this expenditure to student
development, did not appear to have an effect on degree attainment. A complication with the
study by Ryan though, is that colleges and universities are organized differently. The offices and
programs that are classified as student services or academic support oriented might vary
depending on how each institution is organized. Thus, the attempt of Ryan to find results that
can be generalized for all colleges and universities is somewhat negated because it looks at the
forest and assumes that each tree has the same branches serving the same purposes.
Webber and Ehrenberg (2010) sought to find out whether expenditures other than those
on instruction affect graduation and persistence rates in American higher education. The authors
looked to see if increasing tuition, especially with lagging graduation rates of young Americans,
is a reflection of inefficiency or if the expenditures outside of instruction do have an impact on
graduation. The four expenditure categories that received focus were instructional, academic
support, student services, and research. The findings within this study, in contrast to that of
Ryan (2004), provide evidence that student services expenditures influence graduation and first-
year persistence rates. The impact is more prominent at schools with lower entrance test scores
and/or a larger number of Pell Grant dollars per undergraduate student. Pell Grant eligibility is
commonly used as a measure of financial need for undergraduate students. A campus with a
high rate of Pell eligible students reflects a generally low SES. It should be noted though that the
specifics within these expenditures and how that is different between institutions is unknown.
Increasing levels of instructional and/or research expenditures in this study appeared to have a
negative effect on graduation rates. So there appears to be a tipping point where too much
IMPACT OF SOCIOECONOMIC STATUS 38
money spent on instruction and/or research at a college could negatively impact the balance.
There also is the concern that an institution might spend money specifically with the hope of
increasing selectivity and ranking, but in ways that run counter-intuitive to the needs of current
students at the university.
Jones-White et al. (2009) obtained the student records of three cohorts of entering
freshmen at the University of Minnesota-Twin Cities through the National Student
Clearinghouse and looked for students who either received a baccalaureate degree from their
home institution, from another institution, or earned an associate degree/certificate from another
school. The authors point out that a college or university should not be judged solely on its
retention and graduation rates, but that it should also be examined for whether students that
departed eventually graduated from another college or university.
A limitation of the study by Jones-White et al. (2009) is how it can be generalized
because it focused on a single institution. The exclusion of socio-psychological factors, the
potential for students graduating outside of the National Student Clearinghouse net, and the focus
solely on student outcomes are also limitations. The findings allow for a more complete
measurement of graduation and suggest that institutions should strive for having a high rate of
students make successful progress toward a degree, even if it is received elsewhere (Jones-White
et al., 2009).
Socioeconomic Status
In comparison to research on racial/ethnic disparities within higher education, the
research that focuses on the relationship between socioeconomic status (SES) and educational
attainment is limited. Goldrick-Rab (2006) looked at 12
th
graders in the United States who
graduated in 1992 and found that only five percent of students from the lowest SES started
IMPACT OF SOCIOECONOMIC STATUS 39
college at a four-year institution of any tier. Carnevale and Rose (2003) examined 1988 data and
found that students from the lowest SES only accounted for three percent of the total student
population within the highest rated tier of colleges and universities, whereas 74% of students at
the same institutions consisted of individuals from the highest SES. These findings, and others
found within the literature, point to the importance of understanding variations in how SES is
calculated, the impact of financial aid, support programs, and the value of attending a highly
selective private institutions.
Calculating Socioeconomic Status
The method of calculating SES in a study can have an immense effect on the results and
the significance of any findings. A review of the literature revealed discrepancies not just in the
calculation of SES but also in how students were grouped by SES for comparison. The methods
of calculating SES within the literature reviewed were limited by their reliance on student
reported survey data since students do not always have an accurate understanding of family
income and finances. The process used by multiple authors to standardize the data and create a
normal distribution curve for the analysis of different SES groups also has limitations.
Aries and Seider (2005) began their qualitative study on 30 students using parental
income as the determinant of class but as they learned more about their interview subjects they
decided to also include parental education because they hoped to learn more about the role of
cultural capital. Parental income and education also determined SES for Astin and Oseguera
(2004). They took all of the survey participants that matched the criteria for their study and
grouped the population into quartiles by parental income. They classified parental education as
follows: no college attended; both parents received a degree; or some college attended and/or one
parent with a degree.
IMPACT OF SOCIOECONOMIC STATUS 40
In addition to parental income and educational attainment, the occupations of parents
have also been used to calculate socioeconomic status (Carnevale & Rose, 2003; Goldrick-Rab,
2006; Walpole, 2003). These studies, similar to Astin and Oseguera, rely on financial data
reported by students. The self-reporting of parental income ranges by students is not considered
reliable, but the additional variables of parental education and occupation add more data to the
findings. How capable traditional college aged freshman are when it comes to answering
questions related to the income of their family is not addressed.
Goldrick-Rab (2006) calculated SES for high school seniors and coded the data quintiles
with the lowest quintile considered low SES, the highest quintile high SES, and the three
quintiles in between classified as middle SES. Walpole (2003) used the lowest and highest
quintiles to represent low and high SES. Johnson et al. (2011) also relied on student reported
survey data within their study on a single institution, using students with household income
ranges from the U.S. Census of less than $25,000; $25,001-$40,000; $40,001-$70,000; $70,001-
$90,000; and $90,001 and above. These researchers based socioeconomic status solely on the
range reported within the student survey.
Lee and Kramer (2012) also based their study on a single institution. Utilizing a multi-
method approach, they used data connected to their own institution from a national survey and
then connected it with their own survey and interview data. This allowed them to email all
students with reported family incomes of less than $80,000 per year at their school. They found
29 respondents who were classified as either low-income, working class, or lower-middle class.
The authors considered low-income status to be parental income below $40,000 per year, with
neither parent achieving a four-year college degree. Working class status consisted of parental
income above $40,000 with parent(s) working a blue-collar job and neither parent achieving a
IMPACT OF SOCIOECONOMIC STATUS 41
four-year college degree. The requirements for lower-middle class status were reported family
income below $80,000 with only one parent having a degree from a four-year college. The
authors do not address whether the parental income used is student reported and details regarding
how they received the email addresses for students matching the income criteria is unclear.
Titus (2006) used survey data and defined SES as the composite of standardized parental
income and education. It was measured as a continuous and categorical variable with four
quartiles from lowest SES to highest SES. Similar to the quintiles used by Walpole (2003), the
quartiles used by Titus had a standardized normal distribution. In both of these cases the normal
distribution produced through standardization allowed for a more thorough statistical analysis.
The process of forcing a normal distribution upon the data potentially skews the results,
however.
Walpole (2003) recoded the SES related data into a variable with a normal distribution
and frequency to produce low SES and high SES subsamples each with approximately 2,400
students. The actual distribution of low and high SES students at any tier of four-year
institutions based on Carnevale and Rose (2003) did not have a normal distribution. So, for
Walpole (2003) to achieve approximately equal subsamples, the study would either need to use
only a fraction of the high SES students or the recoded low SES group would need to contain
students from more than just the lowest SES quintile. This is further complicated by the use of
surveys in which the income data comes in the form of a range. As such, it may be advantageous
for future research to explore the collection of actual parental income figures reported to
financial aid offices because the student reported ranges of “under $25,000,”- as well as “more
than $90,000,”- are wide ranging and ambiguous. When also considering the overreliance on
IMPACT OF SOCIOECONOMIC STATUS 42
student reported data, a lesson learned from the literature is that the calculation of SES is
inconsistent and the method of calculation tremendously impacts how the data is analyzed.
Impact of Financial Aid
In general terms, financial aid is calculated by using the full cost of attending the
institution and the expected family contribution. The full cost of attending is estimated by
adding the tuition, housing, meals, and other necessary expenses related to the yearly cost of
enrolling at that college. The expected family contribution is based on a combination of family
variables such as family income, assets, number of children, number of children in college, age
of the older parent, and unusual expenses. Just as the full cost of attending one college or
university can vary immensely with another, so too can a family’s expected contribution. As
such, the expected family contribution can be substantially greater than or less than the full cost
of attending an institution. The financial aid eligibility for a student is determined by subtracting
the expected family contribution from the full cost of attending. Additionally, whereas the
federal calculation of the expected family contribution determines eligibility for federal and state
based aid, an institution can use its own method to determine the amount of institutional need-
based aid that is offered. In both instances, the resulting need-based financial aid amount is
offered via three different and unequal methods.
Grants, loans, and work-study are the three forms of financial aid within higher
education. Students typically receive grants based on merit or financial need from the institution,
federal government, or state government. This form of aid is most advantageous for students
because it does not need to be paid back (Hill, Winston, & Boyd, 2005). It should also be noted
that merit aid differs from need-based aid because it is not associated with the expected family
IMPACT OF SOCIOECONOMIC STATUS 43
contribution of the student. Thus, a student can receive merit aid that reduces the designated
expected family contribution.
Loans are either subsidized or unsubsidized. The United States Department of Education
pays the interest on subsidized loans while the recipient student is enrolled at least half-time or
during a grace period or deferment. Only students with lower expected family contributions are
eligible for subsidized loans. Most all students are eligible for unsubsidized loans, even if their
expected family contribution is greater than the full cost of attending.
Aid in the form of work-study requires a student to work before receiving the funding.
To better understand how this relates to SES, it should be understood that low SES students,
depending on calculation, tend to be eligible for grants, loans, and work-study. High SES
students are traditionally only eligible for merit-based aid in the forms of grants because they are
not eligible for need-based aid grants or work-study funds. Unsubsidized loans would be the
exception for high SES students.
Alon (2005) found that researchers must distinguish additional factors between merit and
need-based aid when examining the resulting graduation rates of students. At face value, merit-
aid can be connected with high graduation rates and need-based aid with lower graduation rates.
The literature reviewed by Alon contained inconsistent findings and the overall interpretation
was that grant-aid had a neutral effect on persistence and graduation. Alon contends that it
cannot be assumed that merit and need-based grant aid recipients are comparable. When
accounting for SES, low-income students who are reliant on grants are more likely to drop out
due to their economic challenges. Increases in the grant money available to low-income students
though, according to Alon (2005) can help promote their persistence to graduation. Chen (2008)
also found that the effects of aid amounts vary because racial/ethnic minorities may be less likely
IMPACT OF SOCIOECONOMIC STATUS 44
to take on more loans. So it appears that grant money is the best stimulator for low SES
students.
Longanecker (2002) found that increases in merit-based aid at public institutions across
the country between 1994 and 1999 did not result in decreases in need-based aid. It is unclear
whether the monetary increases in merit-based aid would have otherwise gone to need-based aid.
Long and Riley (2007) examined 2004 data from the National Center for Education Statistics and
demonstrated that low SES students attending full-time faced higher amounts of unmet financial
need each academic year when compared to higher SES counterparts. They calculated unmet
need as the total cost of attending minus the Expected Family Contribution (EFC) and all grants
received. The EFC is based on a federal formula determined by variables such as family income,
assets, size of family, number of family members attending college, age of the oldest parent, and
the income and assets of the student attending. Long and Riley also discussed how after
adjusting for inflation the average amount of a Pell grant, the largest federal need-based aid
program, has decreased over time. With the purchasing power of the Pell grant subsiding and the
cost of tuition rising, it is not a surprise that low SES students are the ones being squeezed. They
have the largest amount of financial need and most institutions do not have the financial capacity
to meet it.
DesJardins and McCall (2010) examined 12,648 students who entered a large research
university as first-time freshmen in the fall terms of 1984, 1986, and 1991. They tracked these
students for a period of more than six years and sought to explore the effects of financial aid on
stop-out, periods of reenrollment spells, and graduation. A stop-out was defined as an
interruption in enrollment with the potential of the student not returning. An enrollment spell
was listed as time period or episode where a student is alternately enrolled or not enrolled.
IMPACT OF SOCIOECONOMIC STATUS 45
Students in the study who did not stop-out graduated at a rate of 76% compared to the
9.4% rate for students with at least one stop-out, with the initial stop-out potentially followed by
enrollment spells (DesJardins & McCall, 2010). The authors found that all types of aid reduced
the chances of a first stop-out and promoted higher rates of graduation. They also provided a
critique of two financial aid strategies. Front-loading is the process of enticing students with
higher amounts of aid during the first two years and then declining the amount in later years as
the student has become more likely to persist to graduation. The authors found this problematic
and ineffective. Princeton University offers a different approach, which eliminates the need for
loans by offering institutional grants or scholarships. The authors simulated the effects of this
policy and determined it would increase graduation rates. The funding necessary to achieve the
Princeton approach, however, is not realistic at most institutions.
Kim, DesJardins, and McCall (2009) and Van Der Klaauw (2003) focused on the impact
of financial aid on choice of university. The financial aid package offered to the admitted
student greatly impacted whether the student could afford to enroll. Van Der Klaauw (2003)
examined the early enrollment decisions at an East Coast college and found that financial aid is
an effective method of competing with other schools for students. Kim et al. (2009) studied the
expectations students had for aid and how those expectations impacted enrollment at the
University of Iowa during the admission years of 1997 to 1998. The authors found that the same
levels of aid offered to low income Whites and Asians did not lead to equal rates of enrollment
when offered to African Americans and Hispanics. This finding relates to that of Chen's (2008)
in that minority students of similar income levels were less likely to take on similar levels of
loans when compared to their non-minority peers. For each study, the sub-populations within
each racial/ethnic group were not examined. For instance, SES was not examined within a racial
IMPACT OF SOCIOECONOMIC STATUS 46
ethnic group and the ethnic sub-groups within the Asian and Lation/a populations were also not
compared for differences. Either way, it appears that although the amount of financial aid
offered to a student influences whether they enroll in a college or university, the influence can
vary based on the individual characteristics, including race/ethnicity, of each student.
Dwyer, McCloud, and Hodson (2012) examined how debt impacts graduation in
American universities. They used the National Longitudinal Survey of Youth 1997 (NLSY97)
Cohort with individuals born between the years of 1980-1984 and focused on individuals who
were age 25 and higher as of 2007. Fifty percent of the respondents within the sample studied
earned their college degree by 2007. Also examining class status, the authors found that the
educational attainment of parents had a positive effect on college graduation similarly to parental
income. Overall, the authors found that debt taken on in the form of student loans up to ten
thousand dollars over an academic career had a positive effect on graduation rates at public
colleges, but that debt more than ten thousand dollars had a negative effect on graduation rates.
Student debt exceeding ten thousand dollars at private colleges surprisingly did not have a
negative effect on graduation. The authors did not present an explanation why student debt
above ten thousand dollars was not as much of a hindrance on graduation rates at private
institutions, but they did argue that higher education as a whole must be kept affordable if the
nation aims to raise graduation rates and minimize the increasing debt of students.
Hill and Winston (2006) demonstrated how grant aid decisions impact the net price of
attending a highly selective private college. They analyzed financial aid decisions for all
students who applied over a period of 14 years, from 1988-89 to 2001-02, at Williams College in
Massachusetts. Financial aid recipients represented 39-44% of the student population and the net
price was calculated by deducting grant aid from the full, published annual cost of attendance.
IMPACT OF SOCIOECONOMIC STATUS 47
The yearly cost of attending a college is based on the combined costs of tuition, room board, and
fees. In 2001-02, students from the family income range of $0 to $24,000 had a median income
of $15,347 and paid a net price of $931 for the year. Financial aid recipients during the same
year with family incomes of $91,701 or higher (with a median income of $113,689) were
responsible for paying $22,829 of the $32,470 cost of attendance. For the first thirteen years
prior to 2001-02, students from the lowest income bracket paid an average net price of $5,924
after adjusting to 2001-02 dollars, and those receiving aid from the highest bracket paid $22,499.
In providing this detail, they acknowledged that Williams had recently made the college
substantially more affordable for low income students without raising the net cost of attending
for others.
Hill and Winston (2006) claimed that the college offered almost no preferential
packaging, often referred to as merit based aid, so the grant aid amounts were based
predominantly on financial need. The significant changes in net price for low income students
occurred shortly after economist Morton Schapiro was appointed president of Williams College
in 2000. Even though these changes made Williams tremendously more affordable to low SES
students, those from the low and low-middle income ranges of $0 to $24,000 and $24,001 to
$41,000 still only represented 8% of the student body in 2001-02. This finding is similar to
those of other studies (Carnevale and Rose; 2003; Goldrick-Rab, 2006; Walpole, 2003) in that
low SES students are not attending highly selective colleges and universities in great numbers.
Hill, Winston, and Boyd (2005), accessing data from 28 of the 31 colleges and
universities that compose the Consortium on Financing Higher Education, otherwise known as
COFHE, were further able to identify how affordable some of the most highly selective private
institutions in the country were for low SES students. Only 4.7% of the 108,721 students in the
IMPACT OF SOCIOECONOMIC STATUS 48
study had family incomes of $24,000 or less in academic year 2001-02 and only 17.6% had
family incomes less than $61,379. Just over 55% were wealthy enough to pay the full cost of
attending without financial aid and an additional 14.6% had family incomes of greater than
$91,700. So the disparity in access between the low and high SES is evident, but not to the
degree reported by Carnevale and Rose (2003). This is a positive sign that the most highly
selective private institutions are making an effort to be affordable to lower SES students,
although Hill et al. (2005) reveal that the full cost of attending after grant based aid at COFHE
institutions ranges considerably. Among the 28 schools, students with family incomes of
$24,000 or less were charged a net price of attending that ranged from $800 a year to $11,390 for
colleges and universities with an average cost of attendance of $33,831. The average net price
after grant aid of $7,553 for the lowest income students equated to 49% of the median family
income for students from that range. Meanwhile, the students with higher family incomes were
charged a much lower percentage, indicating that the actual cost of attendance for low SES
students equaled a disproportionately higher percentage of their family income in comparison to
their higher income peers.
Impact of Support Programs
Research on the impact that student support programs have on low SES college students’
retention and graduation is limited (Rhoades, 2006). There has been research on the impact of a
summer bridge program (Strayhorn, 2010), the influence of peer support on underrepresented
and first-generation college students (Dennis et al., 2005; Grant-Vallone et al., 2003), and the
success of students in a support program for those considered disadvantaged (Braunstein, Lesser,
& Pescatrice, 2008). Each of these studies had limitations that reduced their significance and
ability to generalize the findings.
IMPACT OF SOCIOECONOMIC STATUS 49
Strayhorn (2010) examined a summer bridge program at a single institution after
acknowledging that similar programs in higher education had become popular despite a lack of
empirical evidence to demonstrate their success. The purpose of the study by Strayhorn was to
measure the impact of the summer bridge program on economically disadvantaged students of
color at a highly selective, predominantly White, research, institution. The focus was on the
influence the program had on students' academic self-efficacy, sense of belonging, academic
skills, and social skills. Students for the five week program were selected and required to
participate because they were considered to be at risk of failure due to criteria such as low
HSGPA and test scores, first-generation status, and/or low income racial/ethnic minority status at
the college of approximately 30,000 undergraduate and graduates. In the summer bridge
program the students took an academic skills/career planning seminar and an English
composition class, with both counting for degree credit. The sample consisted of 55 first year
students in a single cohort. They were surveyed prior to the start of the summer bridge program
and also at the beginning and conclusion of the fall term.
The Summer Institute Survey developed for the study by Strayhorn (2010) consisted of
83 items designed to measure academic self-efficacy, sense of belonging, academic skills, and
social skills. The mean age of the population was 18 and 69.8% were female. The HSGPA of
the sample ranged from 2.45 to 4.57, with a mean of 3.61. The mean ACT scores were 20.92 for
English and 19.13 for math. The group had a mean college GPA of 3.53 after completion of the
summer program, yet only 2.35 after their first term. Strayhorn used a paired-samples t test to
compare whether the survey results changed after participation in the program. Participants were
found to have statistically significant gains in the areas of academic self-efficacy and academic
skills after the program. Additionally, hierarchical linear regression techniques were used to
IMPACT OF SOCIOECONOMIC STATUS 50
examine the relationship between GPA after the fall term and posttest scores for the survey
subsets. High school GPA and academic self-efficacy were found to be significant predictors of
college GPA after the first term.
Although the findings of Strayhorn (2010) are noteworthy, there were limitations.
Strayhorn acknowledged that the data used came from a relatively small sample at a single
institution. The reliance on self-assessments and the assumption that changes in mean scores
between the pretest and posttest were the result solely of the summer bridge program were also
claimed as limitations.
Given the small sample of 55 participants and the wide range of HSGPA from 2.45 to
4.57, with a mean of 3.61 nearly exactly in between, Strayhorn (2010) did not address the
possibility of the sample consisting of more than one subgroup within. For instance, a large
share of the sample could have been at-risk student athletes with lower mean HSGPA, test
scores, and overall levels of academic preparedness when compared to the non-athletes within
the sample. Whether this was the case, or if there were separate differences that led to the
composition of more than one group within the sample, then the groups should have been looked
at separately. The use of a survey questionnaire developed by the author also raised concerns
and would have better been validated had it been tested on more than one cohort sample of
students. The findings might also have been further substantiated had the same surveys been
sent out prior to and after the fall term to a similar group of students at the institution that did not
participate in the summer bridge program. Additionally when it came to the findings, the
positive influence of HSGPA on first-semester GPA in college had already been established
(Reason, 2003; 2009). Strayhorn's finding that gains in academic self-efficacy were positively
related to college GPA after the first semester should not be surprising. A student within an at-
IMPACT OF SOCIOECONOMIC STATUS 51
risk population required to participate in a summer bridge program should have high levels of
academic self-efficacy if they complete their summer course work and first fall semester in
college with a high GPA.
In a literature review of multiple summer bridge program studies and the first generation
college students that participated, Otewalt (2013) noted several flaws within the research.
Similar to the study by Strayhorn (2010), much of the research did not measure outcomes past
the freshman year. Most of the research examined did not contain control groups for comparison
and/or use a quantitative approach to measure the outcomes. In sum, Otewalt (2013) argued that
there is a continued need for longitudinal, quantitative assessments of summer bridge programs
and their impact on participating students.
Braunstein et al. (2007) studied the impact of a student support services program on
entering freshmen at a medium sized college. The authors compared the participants in the
student support services program to the entire freshmen cohort. The combined total over a three
year period were examined. On average, students in the support program had lower SAT scores,
lower HSGPA, and lower family incomes. The groups had very similar first to second year
retention rates though, 76.4% for all freshmen over the three year period compared to 76.2% for
those within the support group. Braunstein et al. (2007) argued that this provided evidence that
the support program was benefitting the most at-risk students at the university. The authors,
however, did not provide details regarding graduation rates, college GPA, units earned, or
racial/ethnic differences between the two groups. So it cannot be determined whether the similar
retention rate for the student support program students led to further successful outcomes.
The positive influence that peer support has on students from low SES backgrounds is
another finding (Dennis et al., 2005; Grant-Vallone et al., 2003) that relates to student support
IMPACT OF SOCIOECONOMIC STATUS 52
programs. Grant-Vallone et al. (2003) sought to measure the impact that supportive relationships
have on individuals that are first generation college students and/or financially disadvantaged.
Two separate surveys were mailed to juniors and seniors that were current members of the Equal
Opportunity Program (EOP), Academic Support Program for Intellectual Rewards and
Enhancement (ASPIRE), or the Faculty Mentoring Program (FMP) at California State
University, San Marcos. In order to participate in these programs members would need to be
financially disadvantaged and/or a first generation college student. The first survey dealt with
questions related to demographics, adjustment to the university, and support networks. The
second survey focused on experiences with the student support service programs they were
members of. The authors received 118 responses to the first survey and 73 for the second. The
authors did not indicate how many students completed both surveys. For students who
completed the demographic questions within the first survey, 76% were female, 59% were
single, 53% lived alone off-campus, 60% worked an average of 20 hours per week, and the
average age of the participants was 32 years old. According to the authors, 33% of the
participants in the study were Mexican-American/Hispanic, 25% were Caucasian, 7% African-
Americans (non-Hispanic), and 6% were Asian/Pacific Islanders. The specifics for the
remaining 29% could not be located within the publication.
Grant-Vallone et al. (2003) reported that participants who reported higher levels of self-
esteem and peer support were more likely to also report high levels of academic and social
adjustment. However, even though the study focused on financially disadvantaged and/or first
generation college students, there were many factors that limit how much the findings could be
generalized. There was a small sample size at a single institution. The methodology and
findings were unclear. Additionally, the average age of the population was 32 years old. A
IMPACT OF SOCIOECONOMIC STATUS 53
question would be how well the findings could be compared to a primarily traditional aged group
of entering freshman at a private, research, university.
Dennis et al. (2005) examined how motivation, parental support, and peer support
impacted the academic success of ethnic minority first generation college students. Similar to
Grant-Vallone et al. (2003), Dennis et al. also found a connection between peer support and
college GPA. Students lacking peer support also had a lower college GPA by the spring of their
second year. There were 100 participants in the short-term longitudinal study by Dennis et al.
(2005), all of whom were ethnic minorities, at a university that predominantly serves ethnic
minority students from lower and lower-middle class backgrounds.
Dennis et al. (2005) collected HSGPA, ethnicity, gender, and SES for the participating
students. The surveys were used to measure motivation, family support, and peer support. The
college GPA of the participants at the end of the spring of their second year was used as the
outcome variable in the study. The authors used t-tests to determine if there were any significant
differences between the ethnic groups within the study and found that the only significant
differences were that the Asian students had a higher average HSGPA and college GPA. Using
regression analysis, Dennis et al. (2005) found evidence showing that career and personal
motivation as well as peer support can predict college grades and adjustment. The limited
sample population and uniqueness of the location for the study, however, make it difficult to
generalize the findings.
Padgett and Reid (2002) used a quasi-experimental design to answer their question
regarding how the at-risk participants in the Student Diversity Program (SDP) at California State
University, Fullerton, were performing academically when compared to a similar group of
students outside of the program. After eliminating one student from the study, they examined 40
IMPACT OF SOCIOECONOMIC STATUS 54
total SDP participants. Nine were from the spring 1994 cohort, ten from fall of 1994, five from
spring of 1995, and 15 from the fall of 1995. All of the participants were considered at-risk, and
many of them were also student-athletes. The authors did not provide details on the specifics,
however, regarding how they were considered at-risk or how many were student-athletes. For
the comparison group in the quasi-experimental design, matches were found for each student in
the SDP based on the following seven questions. What semester they first entered the
university? Ethnic group? Sex? Was there age within two years of the student they would be
matched with? Were they a first time freshmen or a transfer student? Did they have a similar
GPA after one semester at the university? Were they similarly still enrolled in classes? Overall,
Padgett and Reid (2002) used these questions to find between four to 54 comparable students for
each SDP participant. It seems that some of these comparable students though joined SDP after
poor academic performance and thus were removed from the study. The authors then examined
the eventual GPA and graduation rates for the SDP and comparison groups. For each cohort the
SDP students had a higher graduation rate that was statistically significant when all of the
cohorts were combined. The SDP group also had a higher college GPA.
There were flaws in the study, however. Due to the methodology and use of first
semester GPA in the match criteria instead of HSGPA, and the characteristics of unique Student
Diversity Program, Padget and Reid (2002) did not answer their research question in the way
they anticipated. What they found was that at-risk students that either started in the program or
joined the program after struggling academically graduated at a significantly higher rate than
students that achieved a low first semester GPA but did not decide to receive support from the
SDP. Within the discussion section the authors seemingly acknowledged this. What they did
not detail in their findings or discussion, however, was how their methodology could lead to an
IMPACT OF SOCIOECONOMIC STATUS 55
erroneous composed comparison group. For the 39 SDP students in the study, there were 21
African American males, one African male, one Turkish male, twelve African American
females, one Hispanic female, one Anglo-American female, one Anglo-American male, and one
Asian female. Considering the other six criteria used for matching and the acknowledged fact
that there were anywhere from four to 54 matches for each individual SDP student, the final
composites for each group could have been completely different by the time they were
compared. As example, if 38 of the 39 SDP students only had one matching student for the
comparison group, but the 39th student in the SDP had 50 matches then the comparison group is
going to mostly resemble the characteristics of the 39th SDP student rather than the entire SDP
group.
The largest research study examined that pertained to the impact of a student support
program was completed by Angrist, Lang, and Oreopoulos (2009). They created a Student
Achievement and Retention (STAR) project at one of the satellite campuses of a large Canadian
university that was primarily considered a commuter school because around 80 percent of the
students in the sample lived at home. The project randomly assigned 250 entering freshmen into
one of two treatment groups, and an additional 150 students into a third treatment group. The
Student Support Program (SSP) offered students access to peer-advising and supplemental
instruction. The peer advisors emailed the group participants regularly and were available to
meet at the STAR office. The second group, the Student Fellowship Program (SFP), was an
incentive program where students had the opportunity to win merit scholarships of $5,000 or
$1,000 if they were able to reach certain targets with their first year GPA, with the targets
available based on the individual HSGPA of each participant. Lastly, the third treatment group,
referred to as SFSP, offered the students the support of the SSP group and also the financial
IMPACT OF SOCIOECONOMIC STATUS 56
incentives of the SFP group. The remaining entering freshmen that were not randomly assigned
into a treatment group and had no contact with the STAR program other than a baseline survey
sent to all entering freshmen consisted of 1,006 students that became the control group. For each
of the treatment groups, the students that were randomly assigned were asked to provide consent
by signing up for the program. The SFP group received the highest rate of sign-ups, 87 percent,
followed by 79 percent for the SFSP group, and 55 percent for the SSP group. Females agreed to
sign up at a higher rate than males for each of the treatment groups, 61 percent compared to 46
percent.
Angrist et al. (2009) determined the results of their study by looking at the average GPA
of the participants at the end of the first semester and first year. Participants in the SFP and
SFSP fellowship programs, with the opportunity to earn a merit scholarship, achieved a higher
first year GPA than the control group. The SSP group performed similarly to the control group.
When examining the overall results, the authors pointed out that females participating in the
SFSP group benefitted the most, particularly the ones who were first generation college students.
In their conclusion, Angrist et al. (2009) shared that peer advising was more popular than
supplemental instruction for both sexes. They also acknowledged that the higher response rates
of women required attention. Participation in the treatment groups was optional and the pursuit
of support and merit scholarships within each group was also optional. Females were more
likely to participate, receive support, and benefit.
In summary, the literature reviewed in relation to the impact of student support programs
on academic outcomes is indeed limited. The majority of the studies examined had limitations
due to a small sample size, brief time period of the study, reliance on surveys, absence of an
IMPACT OF SOCIOECONOMIC STATUS 57
adequate control group, and/or poor methodology. This demonstrates a need for further research
in the area.
Value of Attending a Highly Selective Private Institution
Brand and Xie (2010) have pointed out that the individuals that could benefit the most
from attending college are also the least likely to attend. From there, very few have the
opportunity to attend a highly selective private college or university. Although the definition of
"highly selective" or "most selective" varies, these institutions admit only a select number of
applicants. According to the U.S. News & World Report (2011) guide to colleges, the 100
public and private schools with the lowest acceptance rates using 2011 data ranged from 3.2% to
33.1% This means that over two thirds of freshmen applicants are denied admission. There is
also the increased cost of a private institution. Based on 2009-10 data, the average price of
attendance before financial aid at a private, non-profit college or university was $32,689 per
year, although those with a yearly family income of $75,000 or less received more than a 50%
discount on average, after accounting for grants and scholarships (Knapp, Kelley-Reid, &
Ginder, 2012).
Fox (1993) examined high school seniors from 1980 who went on to graduate from
college by 1986 in an attempt to determine if students who graduated from an elite private
college had higher earnings than students that graduated from a lesser selective public institution.
The difference in tuition prices invested between the types of schools was also accounted for in
order to get a better sense of which type of school offered the most value. Fox found that
students graduating from an elite private college had a rate of return on their investment that was
comparable, if not higher, in respect to the lesser selective public colleges. This was calculated
by observing net earnings minus the costs of tuition. Students who attended either type of
IMPACT OF SOCIOECONOMIC STATUS 58
institution who did not graduate by 1986 were excluded from the study. Fox acknowledged the
omission but did not elaborate on how future earnings could be impacted by differences in
graduation rate by type of institution. As example, if an elite private university had a six-year
graduation rate of 90% then only 10% would be excluded from the study by Fox. It could be
assumed that the excluded students who did not graduate would lack the earnings minus cost
potential when compared to those that did graduate. Meanwhile, a lesser selective public
university might have a six year graduation of 65%, with 35% representing a larger percentage of
students excluded from the study and also likely having lower earnings minus cost potential. In
sum, the less selective university option would have a higher rate of exclusions and that would
skew the findings. The methodology also did not account for the considerably higher amounts of
aid students receive at private universities or how instances of out-of-state tuition costs at the
lesser selective public universities might have applied. The predominance of high SES students
attending elite private colleges is also not addressed. Even though the author found attendance at
an elite private college to be more valuable, there were many limitations within the study.
Without these limitations, the comparable value of the elite private college could have been even
greater.
Brewer, Eide, and Ehrenberg (1999) similarly researched the rate of return for those
attending an elite private college. The authors examined data from high school seniors from the
class of 1972, college attendance, and professional earnings through 1986. The authors
categorized the four-year colleges as either private or public, as well as by selectivity into the
groups elite, very competitive, and noncompetitive. Differing from Fox (1993), Brewer et al.
(1999) focused on the students who attended a four-year college and did not exclude those that
did not graduate. They accounted for gender, ethnicity, family size, family income, parental
IMPACT OF SOCIOECONOMIC STATUS 59
education, financial aid, test scores, HSGPA, and details regarding the type of high school. Out-
of-state tuition differences were also calculated, as was the likelihood for high ability students to
attend more selective schools. The authors found that attending an elite private institution had a
high rate of return, middle-rated private college and bottom-rated public universities had a lower
rate of return, and attending an elite public university had the weakest rate of return.
Thomas and Zhang (2005) conducted their own study on the earnings of four-year
college graduates. They factored in multiple variables such as race, gender, college quality,
family income, and major. The earnings of students who received their college degrees in 1992
or 1993 were examined in 1994 and in 1997. The authors found that the gap in earnings between
students attending highly selective private colleges versus that of less-selective public
institutions increased from 7% in 1994 to 20% in 1997. This suggested that the earnings of
graduates from better schools increased over time at higher rates on average than those from
less-selective schools. As such, the quality of the school attended positively influenced earnings
later in the career more significantly than the observable differences immediately after
graduation.
So how much of a role does higher education play in the facilitation of social mobility?
Haveman and Smeeding (2006) argue that the number of low SES students qualified to attend
selective institutions is far greater than the amount actually enrolled. Hill et al. (2005) have
already shown that the lowest SES students are charged a net price on average that equates to a
higher percentage of yearly family income than any other group. So it is not a surprise that many
low SES students are choosing to attend colleges and universities that are below what their
abilities could allow. In their review of the literature and national statistics, Haveman and
Smeeding (2006) made multiple recommendations to improve the inequalities. Among these
IMPACT OF SOCIOECONOMIC STATUS 60
recommendations, they called for the tuition at public institutions to be raised to better match the
actual costs. Higher income students at public universities are otherwise benefitting from the
equivalent of a government subsidy and are charged what essentially becomes a discounted rate.
The authors contend that government assistance should go to the individual students with
financial need, rather than the university. Without that type of direct assistance, low SES
students have trouble finding fair value at private and public four-year institutions.
Theoretical Framework
Langhout, et al. (2009) examined classism in the university setting and found that
students who experienced classism were less likely to feel like they belonged on campus. In
addition to promoting access for low SES students, the authors argued that campuses must also
do more to help these students feel welcomed. They also recommended that universities
examine and understand how their infrastructure, policies, and procedures might directly or
indirectly impact campus climate. These de facto types of classism may exist through additional
fees or other expected expenditures for classes and organizations that add even further challenges
to lower SES students.
Highly selective private universities are predominantly attended by higher SES students
whose families pay the full cost of attending (Hill et al., 2005). So the findings discussed in this
chapter so far should not be surprising. Lower SES students are expected to adjust to the upper-
middle class norms and policies of the university while also coping with the prevailing wealth of
the peers around them. This section examines sense of belonging as well as social and cultural
capital.

IMPACT OF SOCIOECONOMIC STATUS 61
Sense of Belonging
Baumeister and Leary (1995) define sense of belonging as a need to form and maintain a
minimum quantity of interpersonal relationships. They reviewed the empirical literature of
social and personality psychology related to belongingness and argued that sense of belonging
should be considered a fundamental human emotion. The research they reviewed repeatedly
included evidence of a desire to form social attachments with others sharing common traits or
experiences. Whereas the forming or solidifying of social attachments would have a positive
effect on emotion and well-being, the opposite could be found for those lacking an adequate
sense of belonging. The ill-effects extended to psychological and physical health problems.
Ostrove and Long (2007) studied social class and sense of belonging at a small liberal
arts college in the Midwest. The authors randomly selected 800 non-international students to
contact for the study. Of the 322 students that participated, 80 were freshmen, 105 were
sophomores, 54 were juniors, and 82 were juniors. An overwhelming majority of participants
were White (267) and female (234). Social class was determined subjectively and objectively
(Ostrove and Long, 2007). For the subjective calculation, students self-selected whether they
were poor (5%), working class (9.4%), lower-middle class (15.8%), middle class (38.8%), upper
middle class (25.2%), upper class (3.6%), or other (2.1%). For the objective calculation of social
class each student selected a range that best matched their family income and indicated the
educational level and occupation of their parents. These yearly income ranges and results were
less than $10,000 (.4%), $10,001 to $20,000 (2.9%), $20,001 to $40,000 (13.8%), $40,001 to
$60,000 (16.3%), $60,001 to $80,000 (18.8%), $80,001 to $100,000 (13.3%), $100,001 to
$150,000 (17.1%), and more than $150,000 (17.6%). For parental education and occupation, the
results found by Ostrove and Long (2007) indicated that an overwhelming majority of the
IMPACT OF SOCIOECONOMIC STATUS 62
participating students came from educated families with parents working administrative, minor
professional, or major professional upper class positions.
In their study, Ostrove and Long (2007) found that both subjective and objective social
class background significantly correlated with sense of belonging, academic adjustment to
college and social adjustment to college. They also used linear regression, path analyses, and
meditational analyses to determine that social class background strongly related to sense of
belonging which then could be found to predict social and academic adjustment to college.
Ostrove and Long (2007) offered several recommendations with their findings. They
encouraged institutions to think about how welcoming their campus, norms, beliefs, and policies
are to underrepresented groups. The authors also recommended that existing programs that
might have a positive effect on low SES students be further assessed. Even though
administrators may not be able to change the social class background of a student, Ostove and
Long (2007) argued that the sense of belonging of a student could be changed.
The widely studied academic and social integration model of Vincent Tinto (1975) relates
well to the sense of belonging of a student (Hausmann, Schofield, & Woods, 2007; Hurtado &
Carter, 1997). Family background, individual attributes, and pre-college schooling inputs impact
the initial levels of commitment of a student. The goal and institutional commitment levels, and
how they change, are vital to integration and whether the student persists or drops out (Tinto,
1975). Terenzini and Pascarella (1977) analyzed the Tinto model and found support for the
predictive validity of both social and academic integration. Tinto (2006-2007) has
acknowledged in recent years that students from different backgrounds often have different
experiences in college and that much more work needs to be dedicated to research on the
persistence of low-income students. Tinto's acknowledgement, in relation to low SES students at
IMPACT OF SOCIOECONOMIC STATUS 63
a highly selective private institution, validates the challenges that are faced. Additional research
(Munro, 1981; Pascarella & Terenzini, 1983) also previously found that the experiences of
students once they arrive on a college campus have a greater impact on their persistence than the
HSGPA and test scores they previously achieved.
Hurtado and Carter (1997) examined sense of belonging in relation to Tinto's social and
academic integration model. The authors raised concerns over the complexity students from
diverse groups face when expected to integrate into the university setting. Examining sense of
belonging at different points along the path to graduation for Latino students, the authors found a
strong positive relationship between students' sense of belonging and reports of frequent out of
class discussion of course content with other students. They also found membership in social
organizations to be positively related to sense of belonging. A potential limitation with the study
is the perspective taken by the authors. They present their findings in a way that implies that the
sense of belonging a student feels is impacted by their academic and social integration. As such,
they discount the possibility that sense of belonging is potentially the cause for the integration
rather than the effect.
Building off the work of Hurtado and Carter (1997), additional researchers examined the
relationship between sense of belonging and diversity. Locks, Hurtado, Bowman, and Oseguera
(2008) looked at survey data from ten public universities and found that positive interactions a
student has with diverse peers resulted in an increased sense of belonging. Once again though,
the authors failed to consider that sense of belonging might be the cause rather than the effect, or
at least that sense of belonging and interactions with diverse peers might have a cyclical
relationship. Locks et al. (2008) also did not examine any role that SES might have played in the
findings.
IMPACT OF SOCIOECONOMIC STATUS 64
Johnson, Soldner, Leonard, Alvarez, Inkelas, Rowan-Kenyon, and Longerbeam (2007)
used survey data from a national sample of 2,967 first-year students and found that African
American, Hispanic/Latino, and Asian Pacific American students reported a sense of belonging
that was lower than that of White/Caucasian students. Only Multiracial/Multiethnic students
reported a higher sense of belonging than White/Caucasian students. The most powerful
predictor of sense of belonging found by the authors came from the college environment.
Students from all of the ethnic groups, except Multiracial/Multiethnic, that claimed their
residence hall was socially supportive reported high levels of sense of belonging. The authors
indicated that SES was examined as part of the student background characteristics, but they did
not indicate how they defined it or whether there were impactful differences in SES between the
racial/ethnic groups.
Freeman, Anderman, and Jensen (2007) distributed a survey to investigate the
relationship between sense of belonging and academic motivation. The 238 first semester
freshmen who participated in the study at a public university in the Southeastern United States
were primarily White and female, and they were registered in non-major biology, psychology,
and English classes. The authors found evidence that sense of belonging within those courses
was associated with their in-class academic motivation and efficacy. Attempts to further
understand sense of belonging at the campus level proved to be more complex for the authors
though.
Pittman and Richmond (2008) studied university belonging, friendship quality, and
psychological adjustment. They collected questionnaire data for 79 freshmen during two
separate time points within their freshman year. The authors found that students who
IMPACT OF SOCIOECONOMIC STATUS 65
experienced positive changes in university belonging also had positive changes in how they
thought about themselves.
Hausmann, Schofield, and Woods (2007) conducted three surveys throughout the
freshman year at a predominantly White university as part of a longitudinal experimental design
where White and African American participants were randomly assigned to either a treatment
group or one of two control groups. Members of the treatment group received messages and
gifts from university administrators designed to help the students feel valued. They were told
that their cooperation and completed surveys would be used to help the campus. Members of
both control groups only received messages from a professor in the Psychology department
requesting them to complete the surveys. While controlling for variables such as race, gender,
financial difficulty, and SAT; Hausmann et al. (2007) found that sense of belonging positively
associated with institutional commitment and intentions to persist.
Furthermore, the findings of Hausmann et al. (2007) suggest that sense of belonging was
most impacted by early interactions with peers and others at the university and did not change as
much as they expected over time. The authors established this by surveying students. Three
surveys were sent out during the first year. The first was mailed during the second week of the
fall semester. The second survey was mailed out during the first week of the spring semester.
The final survey was sent near the conclusion of the spring semester. Once the survey responses
were received, some students were randomly assigned to either an intervention group designed to
enhance their sense of belonging or one of the two control groups. Within the study, sense of
belonging values from the first survey positively predicted intentions to persist and were not
related to race, gender, SAT scores, or financial difficulty. During the first year, sense of
belonging significantly declined for the control group that didn't receive gifts and intentions to
IMPACT OF SOCIOECONOMIC STATUS 66
persist significantly declined for all groups. This suggests that sense of belonging should receive
more attention and with increased attention placed on the period from college admission through
the first couple of weeks on campus. According to Hausmann et al. (2007), their study shows
strong evidence that the pre-college period and first few weeks of enrollment are vital for
establishing high levels of sense of belonging which then predict intention to persist.
Types of Capital
Just as starting a business requires capital, so too does a college education. Bourdieu
(2008) describes three types of capital. Economic capital describes belongings that can
immediately be converted into financial currency. Cultural capital within the field of education
is generally considered the knowledge and skills that can benefit a person within a specific
environment. Social capital generally refers to the potential or actual resources available for a
person to give or receive through their network of connections. This section will focus primarily
on social and cultural capital
According to Coleman (1988), social capital is found in the relationships between
persons and can result from all types of social relations and social structures. Physical capital is
described by Coleman (1988) as the physical tools that can be invested in and used to create
financial capital. An example of physical capital could be the materials and tools necessary to
build a house. Once they are purchased and used properly, the end result is something that can
be more easily exchanged for an amount greater than the original cost. Human capital, similar to
physical capital, is the investment in internal knowledge, skills, and capabilities. Coleman posits
that social capital in the community and the family aids in the formation of human capital that
increases the likelihood of remaining in high school through to graduation.
IMPACT OF SOCIOECONOMIC STATUS 67
Stanton-Salazar (1997) defined social capital as the relationships between people and the
networks that weave them into units. In relationship to education, Stanton-Salazar discussed the
importance of having a relationship with institutional agents that can provide support. Seven
funds of knowledge are described as being available through institutional support.
1. Institutional sanctioned discourses (i.e., socially acceptable ways of using language
and communicating).
2. Academic task-specific knowledge (e.g., subject-area knowledge)
3. Organizational/bureaucratic funds of knowledge (e.g., knowledge of how
bureaucracies operate – chains of command, resources competition among various
branches of bureaucracy).
4. Network development (i.e., knowledge leading to skillful networking behavior; e.g.,
knowledge of how to negotiate with various gatekeepers and agents within and
outside of the school environment; knowledge of how to develop
supportive/cooperative ties with peers who are well integrated in the school’s high-
status academic circles)
5. Technical funds of knowledge (e.g., computer literacy, study skills, test-taking skills,
time-management skills, decision making skills)
6. Knowledge of labor and educational markets (e.g., job and educational opportunities,
requisites and barriers to entrée; knowledge of how to fulfill requisites and how to
overcome barriers)
7. Problem-solving knowledge (i.e., knowing how to integrate the first six knowledge
forms above for the purpose of solving school-related problems, making sound
decisions, and reaching personal or collective goals). (pp. 11-12).
IMPACT OF SOCIOECONOMIC STATUS 68
According to Stanton-Salazar (1997), minorities have problems developing social capital
with institutional agents that are capable of offering support and knowledge. The problem is
blamed on a combination of psychological and institutional forces that limit the development of
essential relationships. These negative forces can be decreased if the minority student is able to
decode the system and learn the rules of the dominant culture. This relates back to the classism
that exists within the upper-middle class culture of higher education (Langhout et al., 2009).
Avery and Daly (2010) similarly define social capital as the investment into social
relations that can lead to an increase in available resources. Within a college setting, social
capital can be accumulated through relationships with fellow students, faculty, administrators,
and others at the institution. In their qualitative study on eight college students, the authors
found that social capital is connected to engagement and that resiliency is connected with self-
efficacy. Resilience, within this study, is defined as the characteristics that allow a student to
persist through adversity.
Lareau and Weininger (2003) review how Bourdieu’s concept of cultural capital has been
interpreted within the literature. According to the authors there are two central premises of the
dominant interpretation that has emerged. The first is that cultural capital "is assumed to denote
knowledge of or competence with "highbrow" aesthetic culture (such as fine art and classical
music,)"; and the second is that "researchers assume the effects of cultural capital must be
partitioned from those of properly educational "skills,"; "ability," or "achievement"" (p. 568).
Lareau and Weininger challenge that this interpretation is inadequate. Upon their literature
review and subsequent study, they define their view of cultural capital to be "the direct or
indirect "imposition" of evaluative norms favoring the children or families of a particular social
milieu" (pp. 597-8). The authors argue that this definition, despite being abstract, is broader and
IMPACT OF SOCIOECONOMIC STATUS 69
more capable of including the impact of academic skills while also allowing for flexibility in that
a particular social milieu is not explicitly favored. In relation to education, their definition
means that certain students would have gained skills from their family and others that would be
useful in a school environment whereas others might not have. As such, low SES students
attending a private, highly selective, research university are not disadvantaged because they are
lacking cultural capital. They are disadvantaged because the cultural capital they possess does
not prepare them for the upper-middle class norms and beliefs of the campus climate they have
entered.
Revisiting the study by Walpole (2003), low SES students work more, study less, are less
involved, and report a lower grade point average than high SES students at four-year schools.
Walpole shows evidence that low SES students possess different types of cultural capital when
compared to high SES peers. The author does not use the words "lesser" or "more" for cultural
capital, but rather chooses to use the word "different." Achieving a college degree allows a low
SES student to reach a higher social status than a low SES student who does not attend college,
but the high SES graduate will continue to be more advantaged. It should be safe to assume,
based on the research of Walpole, that the advantages of a high SES student has more to do with
economic and social capital than that of cultural capital. For instance, the high SES student has
more resources and support at their disposal while seeking to acquire the knowledge and skills
that promote success at a private, highly selective, research university. Meanwhile, the lower
SES student may have to work more, losing opportunities for engagement as result of their lack
of financial resources.
The findings of Aries and Seider (2005) further demonstrate this point. The authors pose
that both economic capital and cultural capital are impacted by social class in higher education.
IMPACT OF SOCIOECONOMIC STATUS 70
They believe that administrators at prestigious colleges and universities need to understand that
low SES students, particularly of first generation status, will not possess the cultural capital
relevant to their campus climate. Once again, assuming that a university’s campus climate
conforms to upper-middle class norms and values, the environmental need to adjust is faced by
students from both the lowest and highest SES backgrounds. Those from high SES backgrounds,
however, have more economic and social capital available to help them adjust to the upper-
middle class norms and nuanced expectations within the collegial environment. That luxury is
not available to low SES students.
Methodology
There is value in understanding how SES impacts academic outcomes at a highly
selective, private, research university. An analysis of existing data could provide a better
description of the issues faced in relation to SES. This type of study could help verify previous
findings that show that students from low SES backgrounds are underrepresented and potentially
struggling to achieve the same academic success when compared to higher SES peers (Walpole,
2003).
There are also existing programs that offer college students support related to the
recommendations suggested in the literature. Learning if the support offered by a program
positively impacts students of various SES backgrounds could be of great value when it comes to
identifying potential best practices. If a support program actively sought to increase the sense of
belonging, social capital, and institutionally relevant cultural capital of its participants then it
would reason that these students would outperform peers of similar SES and other pre-college
characteristics. A study examining the academic outcomes based on this assumption could offer
instruction as to how best to offer support to students from low SES backgrounds.
IMPACT OF SOCIOECONOMIC STATUS 71
Significance
The literature demonstrates that the impact of SES must be further understood in order
for the degree achievement rates of 25 to 34 year olds to be improved in the United States.
Reason (2009) has pointed out that research on persistence is informative, but that improvement
will look different on each campus. From a scholarly standpoint, this study will aim to add
further knowledge to the literature. From a practical standpoint, this study will hope to be
significant when it comes to guiding improvement of the university where the research is
conducted.
Conclusion
In this chapter, the desire to increase the percentage of citizens in this country with an
associate degree or higher has been established (Lee & Rawls, 2010). The relevance of SES in
relation to access and success in higher education has been covered (Carnevale & Rose, 2003;
Walpole, 2003). The relationship between social mobility and attending a highly selective
institution (Haveman & Smeeding, 2006; Hill et al., 2005) has been discussed. The relevance of
sense of belonging (Aries & Seider, 2005; Ostrove & Long, 2007), social capital (Avery & Daly,
2010; Stanton-Salazar, 1997), and cultural capital (Walpole, 2003) has also been identified. Low
SES students also face additional challenges when adjusting to the upper middle class norms of a
highly selective, private, university (Langhout et al., 2009). So it would make sense to wonder
how many low SES students are attending a highly selective, private, university. Information
regarding the performance of these students and whether support programs have bolstered their
academic outcomes would also be useful. The next chapter will detail how a study might acquire
this information.

IMPACT OF SOCIOECONOMIC STATUS 72
Chapter 3: Methodology
The previous chapters offered an in depth look at the concerns faced by lower SES
students at a highly selective, private, research university. Considering the completion agenda
goals of the Obama administration when it comes to increasing the percentage of 25 to 34 year
olds in the country with an associate degree or higher (Lee & Rawls, 2010), it is quite surprising
that SES has not received more attention. According to Brewer, Eide, and Ehrenberg (1999),
students who graduate from an elite private institution were able to achieve the highest rate of
return on their investment. Students from low SES backgrounds have also been found to be
severely underrepresented at these highly selective, private, institutions (Carnevale & Rose,
2003). Given upward social mobility is the primary goal of earning a college degree (Haveman
& Smeeding, 2006; Hill et al., 2005), it makes sense that low SES students have the most to gain
from attending a highly selective university (Brand & Xie, 2010).
As established within the first two chapters, the purpose of this study was to further learn
about the academic progress of entering freshmen of various SES backgrounds at a particular
highly selective, private, research university. Whether there were differences in college GPA,
persistence, and graduation for students of different SES backgrounds was determined. The
study additionally examined whether a student support program had a positive impact on the
academic outcomes of students, especially low SES students. Within the literature, researchers
found that sense of belonging (Aries & Seider, 2005; Ostrove & Long, 2007), social capital
(Avery & Daly, 2010; Stanton-Salazar, 1997), and cultural capital (Walpole, 2003) have an
important influence on the educational outcomes of low SES students. So it would reason that
the educational outcomes of low SES students that participated in a support program could be
compared to similar students who did not participate in the program.
IMPACT OF SOCIOECONOMIC STATUS 73
The goal of this chapter is to describe the research design that was used to analyze the
impact of SES on college GPA, persistence, and graduation at a highly selective, private,
research university. The reasoning behind the research questions and methodology are further
explained within this chapter. The population and sample, data collection procedures, data
analysis, and limitations are also detailed in hopes of establishing a clear rationale for the
methods of this study.
Research Methodology
This exploratory study took place at a single institution and used existing data from
multiple cohorts. The entire first-time freshman cohorts that entered the university in the fall of
2007, 2008, and 2009 were examined. The persistence, grade point average, units earned, and
graduation status through the summer of 2013 were accounted for. Four year graduation rates
were available for each cohort, five year graduation rates were available for the 2007 and 2008
cohorts, and the six year graduation rate was available for the 2007 cohort. The focus for this
study, however, was on four year graduation rates.
The expected family contribution (EFC) calculated internally by the financial aid office
of the institution was used to determine SES for this study. High school grade point average and
standardized test scores were used together through the use of a composite score to measure
academic preparedness. Majors of study were combined into categories such as engineering,
social sciences, and the arts. The specifics regarding the calculation of SES, use of HSGPA and
test scores, and categorization of majors is provided in more detail later in this chapter.
The first focus of the study was to incorporate an exploratory design to measure the
impact of a student support program on academic outcomes. This examination compared
participants of the support program to non-participants. The goal was to determine whether there
IMPACT OF SOCIOECONOMIC STATUS 74
were differences in grade point average, persistence, units earned, and/or graduation. The
analysis had both a general and specific focus. For instance, the study sought to learn how the
students within the program performed academically when compared to a similar group of
students outside of the support program. For each comparison of means between the two groups
analyzed for this part of the study, the single difference was whether students participated in the
support program.
How ranges of SES impacted academic outcomes was then examined. Using the first
year expected family contribution calculated internally (IEFC) by the financial aid office at the
university, students were placed into one of five categories that described their SES background.
The students with an IEFC less than ten percent of the total cost of attending were expected to
contribute the least because they demonstrated the highest financial need. The second and third
ranges of SES included students with progressively higher IEFC amounts, with the IEFC being
less than the full cost of attending. The fourth range consisted of students with an IEFC that was
equal to or greater than the full cost of attending. The final range represented students that did
not apply for financial aid. The assumption regarding this last group was that their SES was high
enough that they were fully capable of paying the full cost of attending.
Use of EFC was not commonly found within the literature. A reason for this was that the
majority of authors studying SES relied on survey data instead of actual financial aid data. This
was likely due to challenges in accessing the data. An exploration for research that specifically
used EFC revealed an example by Bresciani and Carson (2002), who examined EFC as part of
their analysis of financial aid packaging information over a four year period from 1996 to 1999.
They found that unmet need had a greater impact than percentage of gift aid on student
persistence. Cragg (2009) similarly used EFC as part of an affordability match calculation that
IMPACT OF SOCIOECONOMIC STATUS 75
measured unmet need. The total of grants, loans, other aid, and the expected family contribution
were deducted from the institution's full cost of attendance. The result was equal to the unmet
need, or overabundance of funds, for each individual student when entering the university. In
addition to this affordability match, the author also examined an academic match calculation
based on SAT scores to determine how well a student fit in with the rest of the student
population. According to Cragg (2009), the academic match was significantly related to
graduation, but the affordability match was not. Wohlgemuth, Whalen, Sullivan, Nading,
Shelley, and Wang (2007), however, found that financial aid was a positive predictor when they
studied the financial, academic, and environmental influences on retention and graduation at a
Midwestern research university. They particularly found that gift aid and work study were
positively linked with first to second year retention. Their analysis used EFC and types of aid
received in combination with demographic characteristics, high school rank, and test scores.
As result of the multiple ranges of SES within this study, each analysis compared the
means of the five groups. The null hypothesis was that the GPA, persistence, units earned, and
graduation rates would be the same for each of the five groups. The analysis then drilled down
deeper and added additional characteristics such as race/ethnicity and academic preparedness.
For instance, the study sought to find out if the academic outcomes varied for students of
different SES levels when they were from the same race/ethnicity and had similar academic
preparedness scores.
The final part of the SES focused research question measured whether the impact of SES
on academic outcomes for students within the support program was similar to that of students
that did not participate in the support program. As example, if there were large differences in the
graduation rates of low and high SES students that did not participate in the support program,
IMPACT OF SOCIOECONOMIC STATUS 76
would there also be similar differences in graduation rates between low and high SES students
that did participate in the support program? This analysis could indicate whether the support
program helped mitigate some of the challenges faced by low SES students.
Research Questions
There are two primary research questions for this study.
1. For first-time freshmen, what differences exist when the persistence, grade point average,
units earned, and degree completion are compared for students required to participate in a
support program and non-participants of similar socioeconomic status and academic
preparedness?
a. What other differences between the participants in the support program and non-
participants can be found when additionally examining first-generation status,
gender, ethnicity, and pursued major?
2. For the general population of non-participants outside of the support program, how did
socioeconomic status and academic preparedness impact the persistence, grade point
average, units earned, and degree completion of first-time freshmen?
a. When also examining first-generation status, gender, ethnicity, and major, were
differences found when comparing students of similar SES and academic
preparedness?
b. If there were differences in persistence, grade point average, units earned, and
degree completion for students within particular SES and academic preparedness
ranges, were these differences similar to those found for participants of the
support program population?

IMPACT OF SOCIOECONOMIC STATUS 77
Methodology of Other Studies
Research on the impact of SES in higher education has been somewhat limited to date
and the methodologies employed by the researchers have been inconsistent. How SES is defined
and calculated by the researchers has also varied between the studies based on the data available
and the process they were collected. Researchers would often define and categorize SES after
they collected their data, using the variables available to them. Variables most commonly used
were family income and parental education. The differences in how SES is defined and studied
are detailed in chapter two.
Within the research on SES and student support programs, the methodology has also been
inconsistent and limited. The critiques and analysis within chapter two has demonstrated this. In
sum, the underwhelming quantity and quality of the research on SES and student support
programs leads to two conclusions. Accepted standards for researching SES need to be
identified and additional research must be conducted in order to identify standards and best
practices. Thus, this study further examined SES within higher education and attempted to
establish and follow a research methodology appropriate to the questions being asked and the
data that were available.
Reasons for Support Program Methodology
Participants in the student support program were selected by the office of admission and
were required to participate in the program as a condition of their admission. In addition to
participating in the program for two to four semesters on average, students were required to
complete a four unit class that focused on learning strategies and motivation. The course
counted toward graduation and was generally one of the four classes students would enroll in
during their first semester.
IMPACT OF SOCIOECONOMIC STATUS 78
According to the office of admission, students selected for the support program were
identified because they were believed to be someone that could benefit from the additional
support. Even though participants might have been selected for the support program due to
possible academic concerns, the selection process was subjective and did not follow any
established criteria. The assumption based on this practice would be that students selected for
the support program were considered to be more at-risk academically when compared to the
generally admitted first-time freshman population. Within each cohort, however, there existed a
similar if not greater number of students not selected for the program even though they had
similar HSGPA and test scores as those that were selected. Attempts were made to further
identify differences between the participants and non-participants of the program. For instance,
did the support program population have a higher rate of low SES or first-generation college
students when compared to the comparison group of students? Were the support program
students more likely to be in a particular major, or not yet have one at all? These answers and
others were answered.
The support program participants and comparison population were further broken down
into groups based on additional variables, such as academic preparedness, gender, pursued
degree program type, ethnicity, and first generation status. The analysis of these additional
variables created subpopulations. For instance, looking at the academic achievement of low SES
students who also had low HSGPA/test score composites can demonstrate whether students in
the support program performed differently in comparison to similar students not participating.
By starting with a broad grouping and drilling down to more specific and smaller groupings of
students, this study aimed to identify whether there were sub-populations of students that the
support program impacted differently. Within these subpopulations, particular focus was given
IMPACT OF SOCIOECONOMIC STATUS 79
to the differences in SES background. The data analysis section within this chapter provides
further details.
To better understand the differences between two groups of students that are only being
tested once, a t-test for independent samples is the best method (Salkind, 2000). The hypothesis
was that students within the support program will have achieved equal or better grade point
average and units earned in comparison to students outside of the program with similar
characteristics. Similarly, Chi-square analyses compared the means between the two groups for
persistence and graduation. These variables were not continuous and would not be appropriate
for t-tests for independent samples. The Chi-square analyses answered whether the differences
in frequencies would be what you would expect to occur by chance (Salkind, 2000). The null
hypotheses for the t-tests and Chi-square analyses were that the students outside of the program
performed better. This hypothesis was based on the assumption that students were placed into
the program because they were considered to be more at-risk academically in comparison to
students with similar characteristics not selected by the office of admission for the program. As
such, the research hypothesis was directional and a one-tailed test was run (Salkind, 2000).
Reasons for SES Analysis Methodology
There was not a universally accepted process of calculating SES available. This study
used the expected family contribution calculated internally by the university (IEFC) for the first
year of attendance because it was based on institutionally accepted factual data rather than
student reported survey responses. Students with an IEFC less than ten percent of the total cost
of attending the university for the year were considered to be of the lowest SES background. As
example, if the total cost of attending for the first year was $60,000 then the lowest SES group in
this study included students whose families would be expected to be able to contribute up to, but
IMPACT OF SOCIOECONOMIC STATUS 80
not greater, than $6,000 toward the full yearly cost of attending the university. An IEFC of ten
percent to less than 30% was considered to be low SES. Students with an IEFC of 30% to less
than 100% were considered middle SES. An IEFC equal or greater than 100% of the full cost of
attending for the year was considered middle to upper SES. Students that did not apply for
financial aid were classified as upper to highest SES.
The methodology for research question two focusing on SES was different than that used
for research question one. Instead of two groups, there were five groups. Based on the presence
of more than two groups, the best way to test for significant differences between the groups
would be a one-way analysis of variances (Salkind, 2000). Before conducting the one-way
analysis of variance, a multivariate analysis of variance was first ran. A factorial design was also
used to test and examine the variables. For instance, a 5 x 4 factorial design consisted of five
different groups based on ranges of SES and four different groupings based on ranges of
academic preparedness. This allowed for the analysis of more than one treatment variables. The
dependent variables for a factorial design need to be continuous and would include GPA and
units earned after the first and fourth years. Chi-square analyses were also used to measure
differences in graduation rates. Additional details have been provided in the data analysis
section within this chapter.
Population
The proposed site and population were selected for multiple reasons. Attending a highly
selective university has been linked with social mobility (Haveman & Smeeding, 2006; Hill et
al., 2005). The challenges faced by low SES students at highly selective, private, institutions
have also been established (Langhout et al., 2009). Research examining the academic outcomes
IMPACT OF SOCIOECONOMIC STATUS 81
of students of various SES have been limited (Carnevale & Rose, 2003). There is merit in the
examination of a large, highly selective, private university.
The particular university studied matches the criteria listed above. It is a highly selective,
private, research university. The undergraduate student population exceeds 16,000. The total
cost of attending varies by year and is currently about $60,000. The recently calculated six-year
graduation rate at the university is around 90%, similar to other highly selective, private,
research universities.
Participants
The existing data from three entire cohorts of students that entered the university as
freshmen were examined. These fall entering first-time freshman cohorts arrived at the
university in 2007, 2008, and 2009. The size of each fall cohort of first-time freshman ranged
from around 2,700 to 3,000 students. The overwhelming majority of students entering the
university were traditionally aged, first-time, full-time, freshmen. This study did not focus on
the age of the students. Rather than taking a random sample from each cohort, this study started
with the entire cohort and excluded only the students that did not fit the questions being asked.
International students, recruited student athletes, and architecture majors were excluded.
A small number of students with missing HSGPA, adjusted HSGPA, and/or test scores were also
removed. An even smaller amount of students that withdrew from all of their classes during
their first term and/or were listed incorrectly for the wrong cohort or were actually transfer
students instead of entering freshmen were also excluded.
International students were not eligible for financial aid. The majority of international
students attended high school outside the United States. They also faced registration
IMPACT OF SOCIOECONOMIC STATUS 82
requirements necessary to stay in the country. For these reasons, they were excluded from this
study.
Student athletes at this university were uniquely different when it came to average
HSGPA and test scores, time committed to athletics, and aid packages received. Many of the
recruited student athletes at the university had HSGPA and test scores far below the average at
the university. Student athletes also had access to an office that provided academic support and
tutoring not available to the general population. These facts provided enough reason for the
exclusion of these students from the study.
The rigorous five-year architecture degree program presented a challenge with comparing
major and degree completion status. This program, as opposed to the 128 units typically
required for an undergraduate degree at the university, required 158 units and was the only
undergraduate degree program at the university with the additional unit requirements. The
number of students enrolled in the program was small. This is why the students pursuing this
major were also excluded from the study.
For research question one, students in the support program were primarily compared to
other students of similar characteristics. For instance, low SES students that participated in the
program were compared to low SES students outside of the program. Breaking this down
further, low SES program participants were compared to students that did not participate in the
program but had similar SES and academic preparedness ranges. Differences in composition by
first generation status, gender, race/ethnicity, and pursued major were also examined at this
point. This allowed for a comparison of two reasonably similar groups, with participation in the
support program existing as the only difference.

IMPACT OF SOCIOECONOMIC STATUS 83
Data Collection Procedures
For both research questions, the exploratory study used existing university data that the
primary researcher had access to as part of the routine functions of his position at the university.
The researcher gained written permission from his supervisor to analyze the data sets and to use
them for this study. The researcher also awaited authorization from the university institutional
review board (IRB) before beginning the analysis. At that point, once the necessary data had
been collected, a spreadsheet lacking student identifiers was created and used for the remainder
of the study. Data from the spreadsheet were then used to enter into software, SPSS, to conduct
the analysis.
Measurement
In addition to the calculation of SES described in Table 2, there were notable decisions
made within this study. It was assumed that HSGPA and standardized test scores are examined
together for admission decisions. As such, it was expected that an admitted student with a low
HSGPA would likely have a higher test score and vice versa. For this reason, a composite score
for each student was put together to measure HSGPA, adjusted HSGPA, and test scores at the
same time. This composite score measured the academic preparedness of each student. Table
three details the calculation and ranges that were used for this study. Students with a score of
less than .80 were considered to have the lowest range of scores, .80 to less than .90 were
considered the low middle, .90 to less than 1.00 represented the middle high, and scores greater
than or equal to 1.00 were referred to as the highest. Rather than using these broad ranges for
every analysis, more specific ranges of scores were also used based on the subgroups examined.
To better understand how academic preparedness scores were tabulated, table four details
examples. For instance, a student with a HSGPA of 3.5, an adjusted HSGPA of 3.8, and a best
IMPACT OF SOCIOECONOMIC STATUS 84
SAT score of 1900 would produce a score of .8722 for academic preparedness. The maximum
scores of 4.00 for HSGPA, 4.70 for adjusted HSGPA, and 2400 for best SAT score equates to a
score of 1.058 for academic preparedness. Of note, students that completed ACT scores instead
of the SAT had their scores adjusted by the university prior to this study. Students missing either
one of the three scores required to produce the academic preparedness score were excluded from
the full study.
A notable decision regarding differences in the degree program pursued by students were
also made. Rather than looking at each academic major separately, this study collapsed degree
programs into categories. The placing of similar degree programs into categories minimized the
number of levels required for analysis while also increasing the population sizes of each
category. This allowed for a better comparison of students of similar majors. The degree
program pursued used data from the end of the first semester. The possibility for students to
change their major after that point was a limitation of the study, as was the acknowledged
complication of placing uniquely different majors into the same category. As such, these degree
programs were placed into categories based more on their similarities rather than their
differences. Additionally, as previously mentioned, students pursuing the five-year architecture
program were excluded from the full analysis. Table 5 details how certain majors were
categorized.
Data Analysis
Once the data set was collected, the first step was to run descriptive statistics. The
characteristics were provided for each entering fall cohort of students from the fall of 2007
through to the fall of 2009. As detailed further in chapter four, each cohort was examined
separately and a decision was made to examine the three cohorts together as one population.
IMPACT OF SOCIOECONOMIC STATUS 85
This decision was made after conducting a multivariate analysis of variance that detailed there
were no statistically significant differences between the cohorts for first year GPA, first year
units completed, four year GPA, and units completed after the fourth year, F (8, 13102) = 2.352,
p = .016; Wilk’s Λ = .997, partial η
2
= .001. The persistence rates and four year graduation rates
for each cohort were also essentially identical (Table 11).
Research Question One
Once the three cohorts were combined the focus shifted to the research questions.
Independent variables included in this descriptive statistical analysis were status within the
student support program, SES, gender, race/ethnicity, academic preparedness, first generation
status, and pursued degree program type. The majority of students participating in the support
program were expected to be in the lower ranges for academic preparedness, particularly in
comparison to the general admitted population of non-participants.
Support program participants were also expected to have higher rates of minority students
in terms of race/ethnicity and first generation status. Another expected finding was there would
be a higher rate of students in the arts and a lower rate of students pursuing engineering degrees
within the support program participant population. Detailed in chapter four, these expectations
were found.
Correlation coefficients were also examined between the independent and dependent
variables. Rather than the categorical ranges for SES and academic preparedness, the correlation
analyses used the raw scores for these variables. To best determine a precise correlation
coefficient, the continuous variables were examined (Salkind, 2000). This provided a general
picture as to which independent variables were significantly related to GPA, and units earned. A
IMPACT OF SOCIOECONOMIC STATUS 86
correlation can range in value from -1 to +1, with the value indicating the direction and strength
of the relationship (Salkind, 2000).
Students that participated in the support program and those that were not selected for
participation were then examined to determine whether there were differences in composition
and educational outcomes. The dependent variables examined in relationship to educational
outcomes included persistence to the second year, GPA and units earned after the first and fourth
year, and degree completion after four years. The descriptive data were used to show the
differences in SES, academic preparedness, and educational outcomes between the support
program participants and non-participants. This helped guide the analysis toward the
identification of reasonably comparable groups of support program participants and non-
participants in terms of SES, academic preparedness, gender, first generation status,
race/ethnicity, and pursued major. After comparable sub-groups were identified based on similar
ranges of SES and academic preparedness the additional variables such as gender, first
generation status, race/ethnicity, and pursued major were compared. For the groups that were
selected as being reasonably similar, a multivariate analysis of variance was first conducted to
because there were multiple dependent variables. These continuous dependent variables
consisted of GPA and units completed after the first and fourth year. At this point, the academic
outcomes groups were determined to be statistically similar. A t-test for independent samples
was still conducted though for the means between the comparable groups. For example,
participants of the support program with the lowest SES and a particular range of academic
preparedness scores were compared to non-participants of the support program with matching
ranges of SES and academic preparedness.
IMPACT OF SOCIOECONOMIC STATUS 87
The subpopulation means were generated and analyzed. The only difference between
each group compared was whether the student participated in the support program. For each
comparison of means, the research hypothesis was that the support program participants will
have achieved equal or better persistence rates, units earned, GPA, and degree attainment when
compared to non-participants. The null hypothesis was that the non-participants achieved better
academic outcomes than the students required to participate in the support program.
Research Question Two
What differences exist, if any, when the persistence, grade point average, units earned,
and degree completion of first-time freshmen of various socioeconomic backgrounds were
examined? The analysis of the support program participants and non-participants provided many
of the details related to students with lower academic preparedness scores. Instead of a
comparison of students based on their participation or non-participation status in the support
program though, research question two required a comparison of means for students of different
SES ranges. Instead of two groups to compare, there were five groups. Based on the presence of
more than two groups, the best way to test for significant differences between the groups would
be a one-way analysis of variances (Salkind, 2000). This type of ANOVA is only for continuous
variables though, so Chi-square tests were also conducted for graduation rates.
Prior to the one-way analysis of variance, a multivariate analysis of variance was first
conducted. This determined that there were statistically significant differences in the dependent
variable outcomes based on the SES range of students. It also indicated that the dependent
variables consisting of GPA and units completed after the first and fourth year could be
examined further through one way analyses of variance.
IMPACT OF SOCIOECONOMIC STATUS 88
The distribution of students based on their SES ranges demonstrated whether lower or
higher SES students that matriculated into the university were more or less likely to have certain
characteristics. For instance, first generation students were more likely to be lower SES at this
university. Lower SES students were also slightly more likely to have lower levels of academic
preparedness. Additional findings are presented in chapter four.
Similar to the support program analysis, the focus on SES drilled down deeper to find
potential differences in educational outcomes for subpopulations. Academic preparedness, first
generation status, gender, race/ethnicity, and pursued major were examined in relation to SES.
Whereas the support program research question primarily focused on the comparison of similar
students that likely had lower ranges of academic preparedness, the SES analysis allowed for a
more complete determination of the impact of SES because it included a broader range of
academic preparedness and major.
Limitations
All studies have limitations. This particular research study consisted of a great deal of
data. The number of students examined likely increased the power of the study. An array of
limitations existed though.
The study only examined first time freshmen starting in the fall semester at a single
university. Entire freshman cohorts from the fall of 2007, 2008, and 2009 were combined for
analysis in the study. Existing educational outcomes through the summer of 2013 were included.
Significant events such as the peak years of the financial crisis of 2007 and 2008 likely impacted
certain students differently over this period of time.
International students, recruited student-athletes, and other entering freshmen were
excluded from the bulk of the analysis. Freshmen that were purposely excluded from the full
IMPACT OF SOCIOECONOMIC STATUS 89
study, as well as transfer students, spring admits, and graduate students obviously had an impact
on the campus culture. The extent of that impact was not examined.
The degree programs pursued by students examined in the study were categorized based
on the declared major during the first semester of enrollment at the university. Students were
able to change their major after this first semester. For the sake of analysis though, students
were categorized based on their pursued degree program at that time. Additionally, it should be
noted that the categorization of degree programs within this study was done as reasonably as
possible. If a major was included in the same category as another, this did not mean that the two
majors of study were the same.
The population used in the study resulted from students choosing to apply, the university
choosing to admit, and the student deciding to attend. Absent within the study were individuals
that applied to enter the university as freshmen yet did not attend. These individuals were either
accepted and chose to not attend or were denied admission. It should also be known that there
were likely many other potential applicants of similar characteristics that chose not to apply.
Students that left the university also presented another limitation. Their reason for
leaving was not examined. It also was not known whether students that transferred graduated
from other institutions, or if they dropped out of the higher education pool completely.
Participants of the student support program did not have a choice in their participation.
They were admitted into the support program. If they wanted to attend the university then they
needed to agree to the requirements of the support program. Also, even though there were
students outside of the support program with similar pre-college characteristics, the lack of full
understanding as to why certain admitted students were selected for the support program was a
limitation of the study.
IMPACT OF SOCIOECONOMIC STATUS 90
Given the absence of a well-defined calculation of SES, this study added to the variation
within the literature. The data used in this study to calculate SES may not be available or
relevant at another institution. Similarly, the institution examined within this study likely has a
multitude of unique characteristics that minimize how the findings can be generalized.
Academic preparedness within this study was related to the composite of HSGPA,
adjusted HSGPA, and best standardized test score. These three variables do not fully indicate
the true academic preparedness of a student. For instance, students might have a lower HSGPA
or standardized test score because they struggled in math. If the pursued degree program of the
student did not require a math class then their academic preparedness score would not accurately
have indicated their true preparedness.
Additionally, this research study was conducted by one person. Within the literature
examined, the majority of studies were conducted by multiple researchers and often times teams
of researchers. As examined at length, the existing literature was limited and riddled with
limitations. Although the researcher conducting this study did his best, he is still only one
person.
Ethical Considerations
This study used existing data, including that of many students the researcher worked with
in the past. The researcher became director of the support program in 2008, and also assisted
many students outside of the program. Due to the researcher’s connection to this study, certain
decisions were made to minimize bias and protect the confidentiality of students. Identifying
characteristics of students were removed prior to analysis and the use of a quantitative research
methodology helped reduce potential bias. The presentation of the findings also sought to
provide a full, transparent, description in order to minimize any concerns that the researcher was
IMPACT OF SOCIOECONOMIC STATUS 91
seeking to only show the data that supported his dual role as director of the student support
program.
It should also be understood how the variables used only provided a glimpse of the full
picture and that students had potentially different experiences that impacted their lives and
academic progress. Generalizations about the student experience need to acknowledge that
students come from different backgrounds and lead different lives. Correlations should not be
confused with causations. Although there were questions that this study hoped to answer, the
research produced more questions than answers. The researcher conducting this study did his
best to maintain an open and reflective mind. Ideally, any reader of this study will do the same.
Conclusion
This chapter has outlined the methodology used for this study and the rationale behind it.
The methodology of other studies has been examined, the reasons for conducting descriptive and
inferential statistics has been provided, the population and site have been detailed, and the
process for collecting and analyzing the data have been presented. These road maps have
hopefully provided a clear direction toward the results and findings that will be presented in
chapter four and the concluding summary in chapter five.

IMPACT OF SOCIOECONOMIC STATUS 92
Chapter 4: Data and Findings
This chapter provides the data and findings based on the plan established in the previous
chapter. Details regarding the data analysis and decision making are presented. The purpose is
to learn further about SES and the potential benefits of a support program at a private, highly
selective, research university. The data, findings, and challenges of this exploratory study are
shared in this chapter.
Comparing Cohort Years.
The fall semester entering freshman cohorts from 2007, 2008, and 2009 were first
examined separately. There were 2,965 students in the 2007 cohort, 2,755 for the 2008 cohort
and 2,862 for the 2009 cohort. Table six details the breakdown of all students and their SES
range determined by their institutional based expected family contribution. The expected family
contribution for each student was divided by the full yearly cost of attending the university. As
such, a student with an expected family contribution of $10,000 equated to 20% if the full cost of
attending that year was $50,000. The use of a percentage allows for a better comparison between
cohort years because the full cost of attending the university increased during the three year
period.
The total of students that received tuition benefits through a parent or guardian that
worked for the university or another qualified school was also identified. The presence of this
population was not previously considered. The work of Hill et al. (2005) and Hill and Winston
(2006) examined the actual cost of attending for students at highly selective private colleges and
universities, similarly breaking down the totals of students by family income ranges, but they did
not address whether they found a similarly small number of students who receive free or
discounted tuition benefits.
IMPACT OF SOCIOECONOMIC STATUS 93
Exclusions
As previously established, there were certain student populations excluded from the full
analysis for this study. Student athletes, international students, participants in the intensive five-
year architecture program, and others with missing data created challenges that warranted their
exclusion. Table seven details how many of these students were found within each cohort. For
recruited student athletes, there were 150 for the 2007 cohort, 167 for 2008, and 124 for the 2009
cohort. The totals for international students fluctuated from 202 in 2007, 155 in 2008, and 320 in
2009. The number of students pursuing an architecture degree as of the end of the first semester
was relatively stable with a total of 109 for 2007 and 91 for both 2008 and 2009. This amounted
to a total of 532 exclusions for 2007, 427 for 2008, and 562 for 2009. The additional exclusions
not accounted for in the totals for student athletes, international students, and architecture majors
were primarily students with missing data for high school GPA and/or standardized test scores.
There also were students who were excluded because they withdrew from all of their classes
during their first semester.
Overall, the composition of the three cohorts examined appeared to be quite similar. Of
the students excluded from the study, the majority did not apply for financial aid. This was
visible when comparing the results of table six with table eight. The percentage of students who
did not apply for financial aid decreased from nearly 40% to around 30% for each cohort after
removing the excluded populations. Even though international students, not eligible for financial
aid, composed a large share of the total exclusions, there were many other excluded students that
did not apply for financial aid. Between the three cohorts, 73.9% of the students excluded from
the study did not apply for financial aid. For the excluded non-international students, 53% did
not apply for aid. This was notably higher than the roughly 30% that did not apply for aid
IMPACT OF SOCIOECONOMIC STATUS 94
among the students included in the study. This serves as a reminder that the research findings for
students included in the study cannot be generalized across the entire campus.
Upon additional consideration, a decision was made to also remove students receiving
tuition benefits from the study. They represented only a small fraction of the overall population
and would create a challenge. The majority of these students had their entire tuition covered as a
parental benefit. These students and their families, unless they also received merit based
scholarships, were still responsible for paying the remaining costs of attending, such as housing,
meals, books, and fees. The motivation for applying for financial aid and loans, however, would
likely have decreased for many of these students though. For instance, it would be impossible to
determine whether a student and their family did not apply for need based aid because they were
from a high SES or because they knew that any aid they would be eligible for would be negated
by the tuition benefit they were already eligible to receive. Additionally, after the initial
exclusions, of the 208 tuition benefit recipients remaining between the three cohorts, only eight
were participants of the support program. For these reasons, the tuition benefit recipients were
removed and the table detailing the characteristics of the included student population for the
study was revised. Tables 9 and 10 provide further details on the composition of each cohort
year and table eleven displays how each distinct cohort performed academically.
The overall numbers and percentages for SES did not change much after the removal of
the tuition benefit recipients. These are found in Table 9. Also of note, the descriptive ranges
for academic preparedness detailed the students in relation to their peers at a highly selective
university. Categorizing these ranges as "Lowest" and "Low-Middle" is not meant to belittle the
pre-college achievement of these students, but rather is testament to the competition to gain
admission into a highly selective, private, university.
IMPACT OF SOCIOECONOMIC STATUS 95
There was relative consistency between cohorts and the distribution of students by
pursued degree program type, gender, and race/ethnicity. There was not a variable that
fluctuated any more than five percent and most were within three percent of each other. In
general, these descriptive characteristics, as visible in Table 10, showed support for the ability to
combine the three cohorts into one study.
The persistence, four year graduation, grade point average, and units earned for each
student was determined for the student population included in the study. It is worth noting that
students were required to complete at least 128 units to graduate and could enter the university as
freshmen with as many as 32 units from AP, IB, or other coursework taken prior to their
enrollment at the university. This is why the average total units completed after four years was
roughly 20 units more than the number of units completed within the university after four years
(Table 11). Many students brought in AP or IB coursework and were also allowed to take
coursework at other academic institutions over the summer.
The findings for academic outcomes revealed that each had a persistence rate of either
96% or 97% for students returning for their second year. Over 80% of students graduated after
four years. Additionally, the average GPA for students was around 3.30, which relates to a B+
average.
To determine whether the three cohorts were statistically similar, and to provide further
support for combining the three cohorts into one population, multivariate analysis of variance
was conducted. This revealed that each of the three cohorts was statistically similar, F (8,
13102) = 2.352, p = .016; Wilk’s Λ = .997, partial η
2
= .001. To provide further details, a one
way analysis of variance was also run. This is a test that examined differences in the means of
more than two groups (Salkind, 2000). The results of this analysis, provided in chart one and
IMPACT OF SOCIOECONOMIC STATUS 96
two, did not indicate any statistically significant differences between the academic outcomes of
the cohorts. This was true for first year GPA, F (2, 6851) = .174, p = .840; first year units
completed, F (2, 6851) = .086, p = .917; four year GPA, F (2, 6851) = .962, p = .382; and four
year units completed, F (2, 6851) = 2.276, p = .103. The data for first to second year persistence
and four-year graduation were not continuous variables because each student either
accomplished that outcome or they did not. The means for persistence and graduation for each
cohort were very similar, however.
Research Question One
For first freshmen, what differences exist when the persistence, grade point average, units
earned, and degree completion are compared for students required to participate in a support
program and non-participants of similar socioeconomic status and academic preparedness?
Before identifying non-participants of similar SES and academic preparedness, the
overall non-participant group was compared to that of the support program participants. This
comparison made it possible to view each population as its own collective. Visible differences
between the participants and non-participants were identified for academic preparedness, SES,
first generation status, race/ethnicity, and the correlations between variables.
Academic Preparedness
As detailed in Table 12, the most notable difference was found in the average academic
preparedness of the students. More than 92% of the support program participants had academic
preparedness scores in the lowest or low-middle ranges. Fewer than 20% of the non-participants
had scores within those ranges.

IMPACT OF SOCIOECONOMIC STATUS 97
SES and First Generation College Status
There were also substantial differences when examining SES and first-generation college
student status. The support program population had a higher concentration of low SES students.
Over 40% of the program participants were from the lowest or low SES ranges compared to
fewer than 20% of the non-participants. The percentage of first-generation students, 31.3% to
10.8%, was also nearly three times higher in the support program.
Race and Ethnicity
Substantial differences were also found between the composition of the support program
participants and non-participants. These differences, found in Table 13, were particularly
apparent for Race/Ethnicity and degree program type. The percentage of Hispanic/Latino
students in the support program, 27.6%, was more than double that of the non-participant
population, 12.8%. The support program population, with a 24.9% Black/African American
composition, was nearly five times that of the 5.5% in the non-participant population. The
support program had a lower percentage of Asian students, 12.8% to 29.4%, and the 32.3%
compared to 50.1% for White students was also much lower. When it came to the degree
pursued, the largest difference was found within the arts, 38.4% for the support program and
15.6% for the non-participants. Much of that difference was balanced out by the non-participant
group having around 10% more business and engineering majors.
Correlations
Before analyzing the similarities and differences between these two groups, the
correlations between some of the variables warranted attention. Table 14 provides the
correlations between SES, academic preparedness, HSGPA, adjusted HSGPA, best test score,
first-year GPA, and four-year GPA. For SES and academic preparedness the raw scores were
IMPACT OF SOCIOECONOMIC STATUS 98
used as opposed to the categorical ranges. The first finding from this table was the statistically
significant relationships between variables. The academic preparedness variable was found as
having a stronger positive correlation with first-year and four-year grade point average than
HSGPA, adjusted HSGPA, and best test score. This demonstrated the usefulness in combining
these three variables into one.
Additional correlations detailing the units completed after the first year, fourth year, and
total units within the university and outside it after four years were also conducted. A notable
finding was the slightly positive, yet significant, correlation between SES with academic
preparedness, first-year GPA, fourth-year GPA, and totals for units completed. As SES
increased, so too did averages for academic preparedness scores and academic outcomes at the
university (Table 15).
A separate examination of only the support program participants, however, led to a
different finding. For this group, SES did not significantly correlate with any of the same
variables and academic preparedness actually slightly decreased as SES increased. Charts three
and four provide a visual for how SES related to academic preparedness and graduation
differently for the support program participants and the non-participants. For the non-
participants of the support program, their average academic preparedness scores and graduation
rates increased as SES level increased. For all groups it seems the students that did not apply for
financial aid were achieving at slightly lower rates when compared to the students in the middle-
upper SES range that applied for aid and had an expected family contribution equal to or greater
than the full cost of attending the university.

IMPACT OF SOCIOECONOMIC STATUS 99
Overall Differences
Considering how different the support program participants and non-participants groups
were, it was not surprising to find that the academic progress results were also quite different
(Table 16). The average academic preparedness of the support program participants was much
lower than the non-participants. This was expected. The highest academic preparedness score of
.936 for a student within the support program was lower than the .9386 average score for the
non-participants. There were also differences for first year GPA, first year units earned,
persistence to the second year, four year graduation, four year GPA, four year units earned, and
the total of units earned, including transfer, international baccalaureate, and advanced placement
units, after the end of the fourth year. Program participants averaged a 3.00 GPA after four years
and a 65% four year graduation rate. Non-participants averaged a 3.37 GPA and 81% graduation
rate.
The two major trends that emerged from this comparison data were the differences in
academic preparation between the two populations and that the academic outcomes at the
university improved as SES increased. Another noticeable difference was seen when comparing
units earned at the university with the total number of units earned for graduation after four
years. What this means was that the non-participants earned nearly 20 units outside of the
university that counted for graduation, while the support program students only earned an
average of around eight units. These additional units counting for graduation completed outside
of the university were the result of either a combination of AP/IB type units or coursework
completed over the summer. Based on the experience of the researcher, the difference in these
totals was most likely attributed to advanced placement units. More specifically, the support
program participants entered the university with fewer units earned from advanced placement
IMPACT OF SOCIOECONOMIC STATUS 100
test scores when compared to the non-participant population. This difference, in addition to
lower scores on average for academic preparedness, reflected the notion that the students were
placed into the support program because they were considered to be more at-risk academically
than the non-participant students.
In the comparison of support program participants to non-participants, attempts were
made to account for differences in academic preparedness scores. Looking at the distribution
charts for the two populations revealed that there were large differences in scores on average, but
that there were likely enough students within the non-participant population that could be
compared to the support program participants when looking solely at academic preparedness.
The histograms in Chart 5 and 6 reveal that even though the non-participant group had a much
higher mean of .939, there still remained students with similar academic preparedness totals as
the support program. When reviewing the distribution for the support program, the majority of
participants had an academic preparedness score between .800 and .900, and only a small portion
of the non-participant population fell within that range. The total number of students within that
small portion of non-participants though, still exceeded the total number that participated in the
support program.
At the risk of regression toward the mean, the challenge moving forward was establishing
how reasonably similar a group of participants and non-participants would need to be in order to
fairly compare their academic outcomes. Substantial differences between the two groups overall
have already been established for SES, academic preparedness, the correlation between SES and
the other dependent variables, first-generation status, race/ethnicity, and pursued degree type.
Whether these differences remained when looking at groups of similar SES and academic
preparedness required additional analysis. Considering the total number of support program
IMPACT OF SOCIOECONOMIC STATUS 101
participants was less than the non-participant group, this study sought to find non-participant
sub-groups of similar academic preparedness and SES for comparison. The distribution of
academic preparedness scores for the support program population revealed a range of .607 to
.936 with .828 as the mean. The 25th percentile was a score of .801, the 50th percentile equaled
.831, and .864 represented the 75th percentile. Considering the 50th percentile score of .831 was
higher than the .828 mean, this demonstrates how the widely dispersed low end of the population
brought down the average academic preparedness score for the participant population. This can
be seen in the distribution histograms in charts five and six.
Finding Reasonably Similar Comparison Groups
This exploratory study examined different cross sections of the support program
population in comparison to non-participants of reasonably similar SES and academic
preparedness scores. Even though the challenges previously mentioned prevented the groups
from being truly matched by the wide array of variables, this study attempted to show many
examples of how the two populations compared. The percentile scores for the support program
academic preparedness scores were used to identify certain groupings. For the groups that
emerged as being of similar compositions, a comparison of means was analyzed for the paired
groups.
Academic preparedness scores of .80 to less than .90. Since the majority of support
program participants had academic preparedness scores within the range of .800 and .900, this
range was first used to compare the support program students with non-participants. Table 17
provides information for the support program students and non-participants with academic
preparedness scores from .800 to less than .900. The support program students matching that
range averaged an academic preparedness score of .8448 and the non-participants produced a
IMPACT OF SOCIOECONOMIC STATUS 102
.8699 average. This showed that even when students were selected within a common range, the
non-participant group still had a larger average academic preparedness score because most of the
students had scores concentrated near the upper end of the range. For this reason, the results
were also found for a sub-population of non-participants with an average academic preparation
score as closely matched to the support program population as possible. In this particular case,
an exact score of .8448 for 487 students was found to match the support program group. This
was accomplished by lowering the maximum range for academic preparedness from less than
.900 to less than .868 so the averages matched up.
The support program participants achieved lower academic outcomes for first-year GPA,
first-year units earned, four-year graduation, four-year GPA, and total units completed at the end
of the fourth year when compared to non-participants with the same .80 to less than .90 range for
the academic preparedness score. For these two groups though, the support program achieved a
higher rate of first to second year persistence and units earned at the university after four years.
This occurred despite a lower average academic preparedness score. When data for the support
program participants were compared to non-participants with the matching average academic
preparedness score the findings become more comparable.
The outcomes between the groups were even more comparable when they were also
analyzed based on SES in addition to academic preparedness. The sub-populations listed in table
17 reveal that grade point average and four year graduation numbers tended to improve for
students at higher SES ranges. Because the support program participant population had a higher
concentration of lower SES students, it was necessary to also account for SES when comparing
the populations. When this happened and support program participants were compared with
non-participants of similar academic preparedness and SES it became apparent that the
IMPACT OF SOCIOECONOMIC STATUS 103
populations achieved similar results. For graduation rates, it also should be noted that the
support program population achieved similar rates despite a gap in units earned outside of the
university. The support program participants were most likely starting with fewer advanced
placement units on average, essentially giving the non-participants an average head start of
around four to five units toward the 128 necessary for graduation.
Composition of groups. How groups were composed greatly impacted how reasonably
they could be compared. Just as the distribution of students by SES was widely different for the
support program participants and non-participants, there were also other differences in the
composition of these groups. For students with an academic preparedness score between the
range of .80 to less than .90, tables 18 and 19 detail the distribution of certain characteristics.
These tables represent the groups previously discussed in table 17. Immediately visible within
these tables were the differences in gender, first-generation status, race/ethnicity, and major. The
support program participants had substantially higher rates when it came to students that were
female, first-generation, Hispanic/Latino, Black/African American, and/or undecided for their
pursued degree program.
The differences in composition between the overall groups cannot be further understood
without also examining for differences in SES. Tables 20, 21, and 22 provide data that further
examined these three groups to determine how their composition for gender, first-generation
status, race/ethnicity, and pursued degree program type differed by SES level. This analysis
showed that the students with an academic preparedness score of .80 to less than .90 differed
tremendously based on their SES. The rate of first-generation status dropped from 79.6% for the
lowest SES group to 9.4% for students that did not apply for aid in the support program
participant population (Table 20). For the other two comparison groups of non-participants it
IMPACT OF SOCIOECONOMIC STATUS 104
dropped from 48.1% to 3.7% for those with an academic preparedness score of .80 to less than
.90 (Table 21), and from 52.3% to 1.3% for the non-participant group with an average score of
.8448 matched to the support program participants (Table 22).
Accounting for differences in SES for these three groups also led to additional findings.
The distribution of males and females remained relatively similar for the support program group.
For the non-participant groups, the rate of more females to males reversed as SES increased. In
opposition to how the rate of first-generation status decreased, the rate of White students
increased with SES for all three groups. The percentage of students pursuing a major in the arts
was also lowest for students from a lower SES background for all three groups.
After reviewing the composition of students found within tables, 20, 21, and 22 it was
easier to understand the challenges of comparing the support program participants to the non-
participants. Within the five ranges of SES for the three groups, there were only three groups
that were relatively similar. The support program participants with academic preparedness
scores of .80 to less than .90 and SES in the lowest, low, and did not apply for aid ranges related
well to the similar groups with academic preparedness scores of .80 to less than .868 and
identical SES ranges. In addition to SES and academic preparedness, these groups relatively
matched up for gender, race/ethnicity, and pursued degree program type. For the “Lower to
Middle” and “Middle to Upper” ranges for SES, however, there were vast differences between
the support program participant and non-participant groups on demographic composition.
Academic outcomes. The goal of research question one was to compare the academic
outcomes achieved by support programs participants in comparison to similar students that did
not participate in the program. When analyzing the combination of SES and academic
preparedness, it was apparent after analyzing students with academic preparedness scores from
IMPACT OF SOCIOECONOMIC STATUS 105
.80 to less than .90 that some subpopulations could not reasonably be compared. In addition to
differences in first-generation status, some of these sub-populations had a substantially different
rate of males to females, art majors to engineering majors, and students of a different
race/ethnicity.
Academic preparedness scores of less than .83139. In order to further compare support
program participants to students that did not participate in the program, different levels of
academic preparedness scores were also examined. The 50
th
percentile academic preparedness
score for students in the support program population examined was .83139 and students with a
score below that level were compared. Tables 23, 24, 25, 26, 27, and 28 provide data on those
students similar to that of the .80 to less than .90 students.
When comparing support program participants from the lower 50
th
percentile of academic
preparedness score it quickly became apparent that there were very few non-participants with
scores of around or below .80 for academic preparedness. Only 153 non-participants, compared
to the 149 support program participants, were found with academic preparedness scores below
.83139. When attempting to find a group of non-participants with an average academic
preparedness score similar to the .7868 of the support program group, there were only 43 non-
participants identified that would combine to equal a similar, yet still higher, mean of .7872.
Despite these differences in average academic preparedness though, the academic achievement
of the support program participants compared well to the non-participant groupings by SES.
Despite having a lower average academic preparedness score within this range, the
support program participants achieved equal or greater academic outcomes in many of the
categories examined when compared to non-participants. On average, the program participants
in this range of academic preparedness scores earned a higher first year GPA, and completed
IMPACT OF SOCIOECONOMIC STATUS 106
more units after four years. The support program participants in the SES low, mid-upper, and
did not apply ranges had the best academic outcomes, particularly when compared to the
students outside of the program. The findings for all of the outcomes can be found in Table 23.
Composition of groups. Prior to fully accounting for SES ranges, the composition of
these three populations were also compared for differences in gender, first generation status,
race/ethnicity, and pursued degree program type. The data in tables 24 and 25 revealed that the
support program participants with academic preparedness scores of less than .83139 had a higher
rate of students of first-generation status, Hispanic/Latino, and pursuing majors in the arts. As
previously discovered in the analysis of students of academic preparedness scores of .80 to less
than .90 though, the differences in distribution of students by SES greatly changed the overall
averages for a population.
For the subpopulations, many of the same trends as that of the .80 to .90 academic
preparedness groupings were found. As SES increased the rate of first-generation status greatly
declined. The percentage of students pursuing majors within the arts also increased with SES.
Due to substantial differences in gender, degree program, and race/ethnicity there were
challenges finding a non-participant group within this academic preparedness range that closely
resembled the support program groups. The most reasonable group with a similar composition
was found for the students that did not apply for financial aid. For that SES level, the support
program participants and non-participants with academic preparedness scores below .83139 had
a very similar composition when it came to gender, first-generation status, race/ethnicity, and
degree program. Even though the non-participants had an average academic preparedness score
of .8082 compared to.7694, the support program students actually achieved slightly better
IMPACT OF SOCIOECONOMIC STATUS 107
academic outcomes for most of the variables examined, particularly at certain SES levels. Table
23 contains the academic outcomes for these two groups.
Academic outcomes. The support program participants who did not apply for financial
aid had a slightly lower first year GPA, 2.82 to 2.93, and four year GPA, 2.95 to 3.01. The rest
of the numbers matched up well, however. Both completed an average of 30.9 units in the first
year. The support program participants had a slightly higher first to second year persistence rate
of 96% compared to 92%, as well as a four-year graduation rate of 65%, compared to 64%. The
most notable difference was in the number of units earned after four years. The support program
students completed five more units on average, 116.3 to 111.3, when compared to the non-
participants. Meanwhile, the non-participants again had more units earned outside of the
university. They had an average of nearly four more units when compared to the program
participants.
Academic preparedness scores of .80139 to less than .8664. The next range of
students examined were those from the 25
th
to the 75
th
percentile for academic preparedness
score within the support program population. This equated to a range of .80139 to less than
.8664. Tables 29, 30, 31, 32, 33, and 34 examined the performance and composition of these
students, particularly in regards to their level of SES.
Composition of groups. There were notable differences in composition by race/ethnicity
and pursued degree program type between the groups within this range of academic preparedness
scores. There was a lower percentage of White students and a higher rate of Hispanic/Latino
students in the support program population. The distributions of students by pursued degree
program type were similar other than the support program having a lower rate of engineering and
IMPACT OF SOCIOECONOMIC STATUS 108
higher rate of undecided students. Tables 32, 33, and 34 further examined the composition of
these three groups and focused on how they looked differently by SES level.
There was one pair of groups that stood out as being similar when comparing the
composition of support program participants and non-participants. The compositions were
comparable for the support program students who did not apply for financial aid and the non-
participants, both with scores of .80139 to less than .8664. These were visible in Tables 32 and
33. Even though the non-participants had a higher average academic preparedness score, .8448
to .8324, the percentages for all of the other variables were within close proximity of each other.
When attempting to compare other groups of support program participants and non-participants
by SES within the stated range of academic preparedness scores there were greater differences in
composition by gender, race/ethnicity, and pursued degree program type.
Academic outcomes. The academic performance outcomes were examined for the
support program participants, the non-participants within the same range of academic
preparedness scores, and a sub-group of non-participants with a reduced range of scores
identified to create a similar average preparedness score. Among the three groups, the support
program participants produced lower averages in every category except units at the university
after four years. When the results by SES level were compared, the performance indicators of
the support program participants compared much better with the non-participant groups. This
was because the support program students with academic preparedness scores of .80139 to less
than .8664 had a larger composition of students from lower SES levels. Nearly 71% of the
support program students in this analysis had an expected family contribution less than the full
cost of attending, thus putting them in either the lowest, low, or lower-middle SES range. Only
46% of the non-participants within the same range of academic preparedness scores and 48% of
IMPACT OF SOCIOECONOMIC STATUS 109
the matched average score group had expected family contributions less than the full cost of
attending. Table 30 further indicated how the distributions of students by SES for these groups
were different. Additionally, the support program had a slightly higher rate of females and a
first-generation status rate around three times higher than the non-participant groups.
Academic preparedness scores of .83139 to less than .936. Students with academic
preparedness scores from the 50
th
to 100
th
percentile for the support program participants ranged
from .83139 to equal to or lesser than .936 for a total of 147 students and a mean score of .8694.
For the non-participants of the same range, their mean was a much higher .9013. For this reason,
for the additional comparison group of non-participants, the upper range of scores was reduced
to .892 in order to produce a similar mean of .8696. This reduction became necessary because
the non-participant group had a higher number of students concentrated near the high score for
the academic preparedness range.
Composition of groups. For this academic preparedness range the support program
population once again had a higher concentration of lower SES students. Nearly 75% of the
support program group had expected family contributions less than the full cost of attending and
nearly 50% were classified as being from the lowest or low SES ranges. The percentages for the
non-participant groups were much lower. This data can be found in Table 36. The support
program group also had a much higher rate of females and first-generation status students.
Differences in race/ethnicity and pursued degree program type also appeared. The
support program population contained higher rates for Hispanic/Latino and Black/African
American students and lower rates for Asian and White students. The non-participant groups
had nearly double the rate of White students. For degree program type pursued, the most notable
IMPACT OF SOCIOECONOMIC STATUS 110
difference was the higher rate of undecided students in the support program population, 23.8%
compared to 14.6% and 13.4%.
These differences in composition were further analyzed when also accounting for SES.
The support program subgroups had a much higher rate of females when compared to the similar
SES subgroups of the non-participants. This complicated the ability to reasonably compare the
populations. The only groups that were a strong match for exploring a comparison were the 41
lowest SES support program participants in table 38 and the 84 lowest SES non-participants in
Table 40. The academic preparedness score average, gender, race/ethnicity, and pursued major
composition for these groups were noticeably different for the low SES, lower-middle, mid-
upper, and did not apply categories. Despite these differences though, it should be noted that the
performance of the support program participants, when grouped by academic preparedness and
SES, had similar or better academic performance scores than the non-participants for most every
variable.
Academic outcomes. Similar to the previous findings in the study, the support program
participants achieved a higher average number of units completed at the university after four
years in this comparison of outcomes. The program participants also had fewer units on average
earned outside of the university, indicating they likely arrived with fewer advanced placement
units. The support program students also had excellent average results for first-year units earned
and persistence to the second year. These findings can be found in table 35. Overall, the
achievement results were quite similar, especially when also accounting for SES levels.
Six Reasonably Similar Comparison Groupings Found
In total, after examining and comparing all of the subpopulations, six pairings were
identified for further study. For each of these pairs the non-participant group reasonably
IMPACT OF SOCIOECONOMIC STATUS 111
matched that of the support program participants for SES, academic preparedness, gender,
race/ethnicity, and pursued degree program type. Table 41 restates the academic outcome data
for the groups that were most similar in composition. For students that did not apply for aid and
had academic preparedness scores of less than .83139 there were two comparison groups listed
due to challenges in finding enough students with an average academic preparedness score equal
to that of the support program population. Of note, there were not any pairings of groups found
for students from the “Lower-Middle” or “Mid-Upper” SES ranges. All of those groups, when
examined for academic preparedness and SES, had substantially different compositions based on
gender, first-generation status, race/ethnicity, and/or pursued degree program type.
The academic outcomes between these pairs of groups seems to reveal that even though
the non-participant groups tended to have a higher average academic preparedness score, the
support program participants performed equally well or better. For all but one pairing, the
support program group earned more units at the university after four years. The four-year
graduation numbers also indicated the success of the support program despite gaps in units
earned outside of the university. Future research will need to examine how advanced placement
units impact graduation rates for the general population, and how SES relates to that. For this
study, it appeared that the support program participants graduated after four years at rates
comparable with similar non-participants that likely entered the university with a higher average
total of advanced placement units.
A multivariate analysis of variance was performed for each of these pairings. There were
multiple dependent variables that had their means compared. The multivariate analyses helped
indicate whether there were problems continuing with the t-tests for independent samples
because it determined if there were any overall differences between the academic achievements
IMPACT OF SOCIOECONOMIC STATUS 112
of the groups. The outcomes analyzed consisted of first-year GPA, first-year units earned, first
to second year persistence, four-year graduation, GPA after four years, and units earned at the
university after four years.
The first pairing of students from the lowest SES had an academic preparation range of
.80 to less than .90 for the support program participants and a .8488 mean score. The paired
group of non-participants had an academic preparation score range of .80 to less than .868 and a
.8473 mean. The multivariate analysis of variance revealed that there was not a statistically
significant difference between these groups for academic achievement at the university as
determined by GPA and units earned after the first and fourth year, F (6, 86) = 1.08, p = ..380;
Wilk’s Λ = .930, partial η
2
= .070. Similarly, the second pairing of lowest SES participants and
non-participants also showed that there were not any statistically significant differences between
the groups, F (6, 118) = 1.91, p = .084; Wilk’s Λ = .911, partial η
2
= .089. This pairing of lowest
SES students had an academic preparation score range of .83139 to less than .936 for the support
program participants and a .8711 mean. The non-participants ranged from .83139 to less than
.892 with a .8688 mean.
Multivariate analyses of variance were also conducted for the low SES pairing and the
paired groups of students that did not apply for aid. For these four pairings there were not any
statistically significant differences in academic achievement outcomes found. The low SES
support program participants with an academic preparation score range of .80 to less than .90 and
a mean of .8445 were compared to the low SES non-participants with a range of .80 to less than
.868 and a .8444 mean, F (6, 91) = .528, p = .786; Wilk’s Λ = .966, partial η
2
= .034. Support
program participants that did not apply for aid with an academic preparedness score range of .80
to less than .90 and a mean of.8386 were compared to similar non-participants with a range of
IMPACT OF SOCIOECONOMIC STATUS 113
.80 to less than .868 and a mean score of .8451, F (6, 185) = 1.10, p = .362; Wilk’s Λ = .965,
partial η
2
= .035. The support program students that did not apply for aid with academic
preparedness scores of .80139 to less than .8664 were compared to non-participants that also did
not apply for aid and had the same range of scores, F (4, 176) = 1.60, p = .187; Wilk’s Λ = .966,
partial η
2
= .070. In total, the multivariate analysis of variance revealed no statistically
significant differences between these groups when it came to viewing academic achievement
outcomes as a whole.
For the comparison of students that did not apply for aid and had an academic preparation
score of less than .83139, representing the lower 50
th
percentile of the support program
participants, the non-participants within the same range were selected for comparison even
though they had a much higher mean score of .8082 for academic preparedness. There were not
enough students found in the non-participant population with a low enough academic
preparedness score to match with the mean score of .7694 for the support program participants.
Even when searching for students with a range less than .803, the mean for 18 students was
.7856. So, in order to match a similar number of students, the support program students that did
not apply for aid with scores less than .83139, with a mean of .7694, were compared to the non-
participants despite that group having a larger academic preparedness score. Even with this gap
though, the comparison did not reveal any statistically significant differences, F (6, 94) = .504, p
= .804; Wilk’s Λ = .969, partial η
2
= .031. So, for the multivariate analyses of variance of each
of the six pairings of reasonably comparable groups there was not a statistically significant
difference when examining academic achievement as a whole.
Even though the multivariate analysis did not indicate any statistically significant
differences between the groups, t-tests for independent samples were still conducted. A chi
IMPACT OF SOCIOECONOMIC STATUS 114
square analysis for persistence to the second year and four-year graduation rates was also run to
further validate that the support program participants were performing equally well, if not better,
when compared to similar non-participants. The null hypothesis for comparing each the eight
pairings of groups states that the non-participant group is performing better academically than
the support program. If the support program participants were placed into the program instead of
other students of similar SES, academic preparedness, first-generation status, gender,
race/ethnicity, and pursued degree program type then it would reason that the students chosen for
the program were considered to be more at-risk due to reasons not examined in this study.
Furthermore, this reasoning implies that the support program participants would not be able to
perform as well as, or better, than the non-participants if they did not receive the aid from the
support program.
Lowest SES and academic preparedness .80 to less than .90. Support program
participants and non-participants from the lowest SES category with an academic preparedness
score of .80 to less than .90 were compared. Chart seven provides the data from the comparison
of means. The range of SES for the non-participant group was adjusted to that of .80 to less than
.886 so that the averages better matched the program participant average. The results indicated
that none of the academic outcomes for the non-participants were significantly better than the
support program participants. The t-test for independent samples did not show any statistically
significant differences between the support program participants and the non-participants for
academic preparedness (t (91) = -0.34, p = 0.735), first-year GPA (t (91) = 0.24, p = 0.808), first-
year units completed at the university (t (91) = 0.27, p = 0.787), four-year GPA (t (91) = 1.14, p
= 0.258), and units completed at the university after four years (t (91) = 0.20, p = 0.840). These
results provided a trend supporting that the support program participants nearly had a statistically
IMPACT OF SOCIOECONOMIC STATUS 115
significant difference with their higher first to second year persistence and the higher four-year
GPA for the non-participants was nearly statistically significant.
Additionally, a chi square test of independence was conducted to further assess first to
second year persistence and four-year graduation rates. The persistence rate of 100% for the
support program participants was higher than the 95.5% rate of the non-participants, but the
difference was not statistically significant (X
2
(1) = 2.28, p= .131). The four-year graduation
rates of 51% for the participants and 52.3% for the non-participants were very similar and the
minor difference was not statistically significant (X
2
(1) = 0.15, p= .904).
Lowest SES and academic preparedness .83139 to .936. Independent samples tests
were also conducted to compare the means of the lowest SES support program participants with
academic preparedness scores from .83139 to .936 to that of the lowest SES non-participants
with academic preparedness scores of .83139 to less than .892. The results indicated that these
two groups achieved similar academic outcomes at the university. The support program students
earned statistically significantly more units earned after the first year, 34.1 to 31.6 (t (123) = -3.1,
p = 0.003). For the other variables, even though the support program participants outperformed
the non-participants in every area there were no other statistically significant differences for first-
year GPA (t (123) = -1.4, p = 0.152), four-year GPA (t (123) = -0.54, p = 0.587), and units
completed at the university after four years (t (123) = --1.8, p = 0.084). These results indicated
that the differences in first-year GPA, persistence to the second year, and units earned after four
years were almost statistically significant in showing the support program participants
outperformed the non-participants.
When comparing persistence and graduation rates, there were minimal differences
between the two groups. Participants had a persistence rate to the second year of 100%
IMPACT OF SOCIOECONOMIC STATUS 116
compared to 97.6%. This minor difference was not statistically significant (X
2
(1) = 0.992, p=
.319). The differences between four-year graduation rates were also not statistically significant
between the 58.5% for the participants and 57.1% for the non-participants (X
2
(1) = 0.022, p=
.882).
Low SES and academic preparedness .80 to less than .90. The results from the
independent samples test for low SES support program participants with academic preparedness
scores of .80 to less than .90 and low SES non-participants with scores of .80 to less than .868
were provided in chart nine. This comparison of means indicated that even though there was a
trend where the support program participants academically outperformed the non-participant
group in essentially every area analyzed in the study, the differences were not statistically
significant. This was found for first-year GPA (t (96) = -1.2, p = 0.245), first-year units earned (t
(96) = -0.31, p = 0.759), four-year GPA (t (96) = -0.23, p = 0.816), and units earned at the
university after four years (t (96) = -0.97, p = 0.332). Differences in persistence and graduation
rates were also found to not be statistically significant. Participants persisted to the second year
at a rate of 100% compared to 98.2% for the non-participants (X
2
(1) = 0.727, p= .394). The
four-year graduation rates, 70.7% for the participants and 66.7% for the non-participants, also
did not prove to be statistically significant in their difference (X
2
(1) = 0.182, p= .669).
Did not apply for aid and academic preparedness .80 to less than .90. The results for
the independent samples test in chart 10 similarly did not reveal any statistically significant
differences between the support program students and the non-participants. The non-participants
in this analysis had a higher average academic preparedness score, .8451 to .8386, but the
difference was not statistically significant (t (190) = 1.43, p = 0.160). Additionally, the
differences were not statistically significant for first-year GPA (t (190) = 1.07, p = 0.288), units
IMPACT OF SOCIOECONOMIC STATUS 117
completed at the university after the first year (t (190) = 0.133, p = 0.894), four-year GPA (t
(190) = 0.803, p = 0.423), and units earned at the university after four years (t (190) = -1.2, p =
0.216). Persistence to the second year (X
2
(1) = 0.338, p= .561) and four-year graduation rates
were also found to not be statistically significant (X
2
(1) = 0.127, p= .721).
Did not apply for aid and academic preparedness less than .83139. There was a
challenge with comparing support program participants to non-participants that had academic
preparedness scores of less than .83139 and did not apply for financial aid. This was visible in
chart 11. Even when both groups were below a certain cut off score for academic preparedness,
the non-participants had a higher mean score that was statistically significant (t (99) = 5.12, p =
0.000). Despite this significant difference in academic preparedness, the support program
students achieved similar academic results for first-year GPA (t (99) = .859, p = 0.392), units
earned at the university after the first year (t (99) = 0.005, p = 0.996), four-year GPA (t (99) =
0.562, p = 0.576), and units earned at the university after four years (t (99) = -0.83, p = 0.408).
Chi square analysis also revealed that the differences in persistence to the second year rate of
95.8% for the participants and 92.5% for the non-participants was not statistically significant
(X
2
(1) = 0.515, p= .473). The four-year graduate rate of 64.6% for the support program
participants was nearly identical to the 64.2% rate for the non-participants.
Did not apply for aid and academic preparedness of .80139 to less than .8664. A
comparison of means was conducted between support program students and non-participants that
did not apply for aid and had academic preparedness scores of .80139 to less than .8664,
representing the 25th to 75th percentile scores of the support program participants. The results
can be found in chart 12. For this analysis, the non-participants had a higher mean academic
preparedness score, .8448 to .8324, that was statistically significant (t (179) = 3.43, p = 0.001).
IMPACT OF SOCIOECONOMIC STATUS 118
The comparison of academic achievement outcomes revealed similar results between the two
groups for first-year GPA (t (179) = 1.25, p = 0.212), units earned at the university after the first
year (t (179) = -0.45, p = 0.657), four-year graduation (t (179) = 0.386, p = 0.700), and units
completed at the university after four years (t (179) = -1.5, p = 0.125). For persistence to the
second year, the higher rate of the support program participants, 100% to 94.8%, was not
statistically significant (X
2
(1) = 1.404, p= .236). Differences between four-year graduation rates
for the participants, 69.2%, and non-participants, 72.9%, were also not statistically significant
(X
2
(1) = 0.150, p= .698). These findings showed additional evidence that the support program
students achieved similar or better academic outcomes when compared to non-participants of the
same SES and academic preparedness.
Data Analysis Summary for Research Question One
Overall, the support program participants were outperformed academically by their non-
participant peers. When comparisons were made between students of similar SES and academic
preparedness though, the findings changed. When the support program participants were
compared to non-participants of similar SES and academic preparedness they were found to have
performed equally well academically. These support program participants were not chosen
randomly, but selected rather because they were considered to be potentially more at-risk than
the students of similar SES and academic preparedness that were not selected to receive the
additional support. So these findings supported the notion that the extra assistance helped.
Finding similar groups based on academic preparedness, SES, gender, race/ethnicity, and
pursued degree program that could be paired for comparison was a challenge. Only six groups
with relatively similar compositions were identified. When considering the different sub-
sections of the support program population that were examined based on academic preparedness
IMPACT OF SOCIOECONOMIC STATUS 119
scores and SES, there were twenty different sub-groups examined. This means that for fourteen
of the support program sub-groups the non-participants of similar SES and academic
preparedness score had very different compositions of students by gender, race/ethnicity, and
pursued degree program. Unfortunately, there were not any reasonably comparable groups
found for lower-middle or mid-upper SES students.
For the six pairs of groups that had their means compared, the findings verified that the
support program participants achieved similar, if not greater, academic outcomes that were
statistically significant when compared to non-participants of the same SES and academic
preparedness ranges. Without support, the expectation would be that the students identified as
being more at-risk would not perform as well. So it was noteworthy that the support program
participants had comparable results, even during instances when the non-participants had
statistically significantly higher average academic preparedness scores and also likely started at
the university with a head start toward graduation due to additional advanced placement units on
average.
Research Question Two
Trends regarding the impact of SES have already emerged from the analysis of the
support program participants and comparable non-participants. It is important to remember
though that the entire population in the study contained a much larger range of students by
academic preparedness than that used in the comparison of support program participants. The
highest academic preparedness score found within the support program was still lower than the
mean score for the non-participant population. For the second research question the entire group
of non-participants was examined and the data for the support program participants was set
aside.
IMPACT OF SOCIOECONOMIC STATUS 120
The analysis of SES and the impact it had on academic achievement followed some of
the same steps as used for the first research question. The descriptive statistics, primarily
detailing the distribution and achievement outcomes, were examined. Sub-populations and their
distribution and achievements were also identified. Lastly, trends and specific findings were
further explored.
Composition of Non-Participants
The breakdown by SES for the 6,557 non-participants of the support program can be
found in Table 42. Similar to what was seen previously within the support program examination;
there were certain trends that were quite visible. The lowest and low SES groups had the highest
rate of females, 62% and 58.3% when compared to 54.1%, 50.6%, and 51.9%. The percentage
of students that were the first generation in their family to attend college substantially decreased
from 46.6% to 3.4% as SES increased. The percentage of White students increased from 17.3%
to 65.3% as SES increased and the percentage for all other race/ethnicity categories other than
“Unknown” decreased.
Academic Outcomes
The academic outcomes of students when categorized by SES can be found in Table 43.
Also compared were students within the lower and higher 50
th
percentiles for academic
preparedness score. The trend of lower SES students being outperformed academically by their
higher SES peers was clearly visible. To further understand these differences, a multivariate
analysis of variances was conducted. This found that there were statistically different GPA and
units completed outcomes after the first and fourth year for the different ranges of SES, F (16,
20008.163) = 5.84, p < .0005; Wilk’s Λ = .986, partial η
2
= .004. This finding indicated a need
to further examine the differences through a one way analysis of variance.
IMPACT OF SOCIOECONOMIC STATUS 121
The differences in four-year graduation rates by SES range for the 6,557 non-participants
were found to be statistically significant (X
2
(4) = 71.141, p= .000). A means plot for four-year
graduation rates of the 6,557 students in chart 13 revealed the differences in rate by SES level.
Chart 14 shows statistically significant differences resulting from a one way analysis of variance
between the groups for first-year GPA (F (4,6552) = 16.656, p= .000), first-year units completed
(F(4,6552) = 6.799, p= .000), and fourth year GPA (F (4,6552) = 16.667, p= .000). Differences
between groups for units completed after the fourth year were not found to be statistically
significant (F (4,6552) = 1.779, p= .130). The reason why units completed at the university after
the fourth year were not significant is because students only needed to reach 128 units to
graduate and the students who brought in more advanced placement related units did not need to
earn as many units.
A four by five factorial design was used to measure how SES and academic preparedness
levels impacted GPA and units earned after the first and fourth years for the 6,557 non-
participants. For GPA after the first year, the main effect of academic preparedness was
statistically significant (F (1, 6537) = 175.308, p = .000) and the effect of SES approached
significance (F (1, 6537) = 2.048, p = .085). The interaction between academic preparedness
and SES for GPA after the first year was not significant (F (1, 6537) = 1.401, p = .157). The
results for GPA after the fourth year were similar in that the effect of academic preparedness was
significant (F (1, 6537) = 182.041, p = .000), SES did not have a significant effect (F (1, 6537) =
1.146, p = .333), and the interaction between SES and academic preparedness was not significant
(F (1, 6537) = 1.237, p = .251). Much like the effect on GPA, academic preparedness had a
significant impact on units completed after the first (F (1, 6537) = 20.386, p = .000) and fourth
year (F (1, 6537) = 12.729, p = .000). Meanwhile the effect of SES on units completed after the
IMPACT OF SOCIOECONOMIC STATUS 122
first (F (1, 6537) = 0.880, p = .475) and fourth year (F (1, 6537) = 1.069, p = .370) was not
statistically significant. The interaction between SES and academic preparedness on units
completed after the first (F (1, 6537) = 1.106, p = .350) and fourth year (F (1, 6537) = 1.227, p =
.257) was also not statistically significant. Together, when also considering the findings for four
year graduation rates, the data suggests that academic preparedness and SES significantly
impacts four-year graduation, but only academic preparedness has a statistically significant effect
on GPA and units completed after the first and fourth year.
SES and Academic Preparedness
The relationship between academic preparedness score and SES was examined further in
Table 44. The academic outcomes for students with academic preparedness scores from three
distinct ranges, in addition to their SES subgroups, were provided. These findings revealed the
same trends as previously identified. Even when academic preparedness scores were similar,
academic achievement results increased with SES. For the groups within Table 44, the four-year
graduation rates of the lowest SES and mid-upper SES groups differed by as much as 34%.
These differences in four-year graduation rates between SES ranges were found to be statistically
significantly different for students with academic preparedness scores from .80 to less than .90
(X
2
(4) = 13.040, p= .011), scores from .90 to less than 1.0 (X
2
(4) = 41.883, p= .000), and greater
than 1.0 (X
2
(4) = 15.544, p= .004).
These results provided further evidence that academic preparedness scores and SES had a
significant impact on rates of academic success at the university level. The next three tables
further detail the distribution of students within three academic preparedness range scores. Table
45 focuses on students with academic preparedness scores of .80 to less than .90. Students with
scores of .90 to less than 1.00 were detailed in Table 46. The information on highly prepared
IMPACT OF SOCIOECONOMIC STATUS 123
students with scores of 1.00 and higher is provided in Table 47. In addition to details regarding
gender, first-generation status, race/ethnicity, and pursued degree program type, these tables also
list the averages for the expected family contribution calculated as a percentage of the full cost of
attending. The average percentage of the full cost of attending students will have covered
through their expected family contribution, need-based aid, and merit-based aid is also listed.
For instance, an EFC percentage of 5.8% and an EFC + Aid percentage of 88.5% would indicate
that a student received need and merit based aid accounting for 82.7% of the full cost of
attending. After the expected family contribution of 5.8%, the student would be responsible for
paying for the remaining 11.5% through loans and work-study earnings.
For table 45, examining students with an academic preparedness score of .80 to less than
.90, many of the same trends for the overall population were found. Nearly 50% of the lowest
SES group came to the university as a first generation college student. Only 3.7% of the group
that did not apply for aid identified as being the first generation in their family to attend college.
Additionally, the higher rate of females within the lower SES groups and White students in the
higher SES groups were visible. Even more pronounced than the overall population, the upper
SES groups had a higher percentage of students pursuing majors in the arts and the lowest SES
students were around twice as likely to have an undecided major.
In addition to statistically significant differences in four-year graduation rates by SES for
students with academic preparedness scores of .80 to less than .90, there were also statistically
significant differences found for GPA after the first (F(4,1233) = 4.902, p= .001) and fourth year
(F(4,1233) = 5.065, p= .000). Differences between SES ranges for units completed after the first
(F(4,1233) = 0.745, p= .561) and fourth year (F(4,1233) = 0.924, p= .449) were not statistically
significant for students within this preparedness range of .80 to less than .90 (Chart 15).
IMPACT OF SOCIOECONOMIC STATUS 124
The descriptive composition statistics for students within the academic preparedness
score range of .90 to less than 1.00 are listed in Table 46. These findings continue to show that
the lowest SES groups have the highest percentage of females and first-generation students. The
percentage of White students increased with SES as the rates for all of the other groups
decreased. For the students within this range of academic preparedness scores, .90 to less than
1.00, the degree programs pursued by students of different SES were much more similar to each
other. This was not found for students within different academic preparedness ranges.
A one way analysis of variance was conducted for the students outside of the support
program with academic preparedness scores of .90 to less than 1.00. The details in chart 16
show statistically significant differences between the groups for first-year GPA (F (4, 4761) =
5.673, p= .000), first-year units completed (F (4, 4761) = 4.464, p= .001), and fourth year GPA
(F (4, 4761) = 7.536, p= .000). Differences between groups for units completed after the fourth
year were not found to be statistically significant (F (4, 4761) = 0.754, p= .555).
There were fewer students with an exceptional academic preparedness score of 1.00 or
higher, detailed in table 47. The overwhelming majority of these students were not lower SES.
Additionally, most all of the students within this group were either Asian or White. However,
some of the same trends were found. The lower SES groups had the highest percentage of
females and first-generation students. For pursued degree program it was more challenging to
compare students by SES level because there were so few students in the lowest and low SES
ranges. This is why a one-way ANOVA was not conducted.
Data Analysis for Research Question Two
In review of the information presented thus far, there are particular trends that have
repeatedly been identified. Academic preparedness scores and SES seem to both have a
IMPACT OF SOCIOECONOMIC STATUS 125
profound impact on academic success. When academic scores remain consistent, academic
achievement outcomes increased with SES. However, the distribution of race/ethnicity also
changed as SES increased. This pattern is further analyzed with Table 26 detailing the academic
outcomes for all non-participants of the support program. Visible in this table are the lower
achievement outcomes found for Hispanic/Latino and Black/African American students. These
populations also had lower average academic preparedness scores.
Race/Ethnicity. Students of the same race/ethnicity and academic preparedness ranges
were then isolated to determine if academic achievement similarly increased with SES. Students
identified as Asian, Black/African American, Hispanic/Latino, and White represented the largest
race/ethnicity groups within the study. An examination of the academic preparedness scores and
SES distribution revealed differences. The mean academic preparedness score for the 1,930
Asians in the study that did not participate in the support program was .9496 and the majority of
the students had scores between .9 and 1.0 (Chart 17). For Asians, the majority of students were
from middle to upper SES backgrounds (Chart 21). The mean academic preparedness score for
the 359 Black/African American students was .8918 with most students falling between the
range of .85 and .95 (Chart 18). For SES, there was a near bell curve distribution between the
levels for Black/African American students with lower SES representing the dominant side
(Chart 21). Similar to the Black/African American population, most of the 840 Hispanic/Latino
students had academic preparedness scores between .85 and .95 with the mean equaling .9201
(Chart 19). The distribution of SES levels among Hispanic/Latino students fairly even, with the
highest numbers found for the lower to middle and middle to upper levels (Chart 21). The 3,283
White students had a mean academic preparedness score of .9426 and most fell between .90 and
IMPACT OF SOCIOECONOMIC STATUS 126
1.0 (Chart 20). For SES, White students had the most noticeable distribution. As SES level
increased, the number of White students substantially increased (Chart 21).
Considering most Black/African American and Hispanic/Latino students had academic
preparedness scores of .85 to .95, this range was examined more closely. Chart 22 provides a
visual and Chart 23 details the data for the four year graduation rates by SES and race/ethnicity
for students with academic preparedness scores of .85 to less than .95. The visual data are
provided for students identified as American Indian/Alaskan Native and Unknown even though
each SES level for these racial/ethnic groups had fewer than 20 students. For the four other
racial/ethnic groups, it became clear that when academic preparedness scores and SES levels
were similar, the four-year graduation rates of students of different race/ethnicity were relatively
similar. Even without accounting for SES level, only examining the similar range of academic
preparedness score of .85 to less than .95, the four-year graduation rates for these four
racial/ethnic groups were all within five percentage points of each other. When SES is also
factored in then the data shows that the lowest SES group had the lowest four-year graduation
rate. Hence, when comparing students of a different race/ethnicity, this data shows that SES and
academic preparedness need to be accounted for.
Additional variables. Gender, first-generation status, and pursued major were also
individually examined to gain better understanding of their impact on academic achievement.
Females at all SES levels outperformed their male counterparts. Table 51 details the results for
average academic preparedness scores, expected family contribution percentage, GPA and units
earned after the first year, second year persistence, four-year graduation rates, and totals for
GPA, units, and total units after four years.
IMPACT OF SOCIOECONOMIC STATUS 127
Gender. Females slightly outperformed the males for every variable except second year
persistence. Average academic preparedness scores for the females and males were fairly similar
overall and at each SES level. The female and male populations were not examined further for
differences in first-generation status, pursued major, and race/ethnicity. However, this cursory
examination shows evidence that females achieved slightly better academic outcomes at the
university. These results also further support the evidence showing the impact of SES. For
males and females, achievement outcomes increased as did SES.
First generation status. The academic outcomes for all first-generation college students
in the study were compared to that of their non-first-generation peers. Table 52 provides the data
from this analysis. Overall, the first-generation students did not perform as well. They had a
lower average GPA after the first year and fourth year, did not earn as many units, and graduated
at a lower rate after four years. However, these disparate outcomes were likely the result of
different distributions of SES and unequal average academic preparedness scores between the
groups. The first-generation college student population had a higher concentration of lower SES
students, 62.9% compared to 15% were identified as lowest or low SES, and also had lower
average academic preparedness scores, .9167 to .9361, when compared to students who were not
the first generation in their family to attend college. Despite the differences in academic
preparedness, the first-generation population achieved similar academic outcomes when SES
was accounted for.
Pursued major type. For all majors, the lowest SES group had the lowest four year
graduation rate. Chart 24 illustrates the differences in four year graduation rates for students
pursuing different majors, and details the results by SES. For the most part, graduation rates
climbed as SES increased, with only a slight dip for the students that did not apply for financial
IMPACT OF SOCIOECONOMIC STATUS 128
aid. Overall, Business, 87%, and Communication majors, 88%, had the highest four year
graduation rates. Engineering, at 74%, had the lowest rate. These majors were identified at the
conclusion of the first semester.
For students from the lowest SES range, graduation rates were concerning for
Engineering at 57%, and Natural Sciences at 65%. Students that were Undecided at the end of
their first semester, with a 62% four year graduation rate, were also below average. For all
students from the lowest SES range, the average four year graduation rate was 67%. Students
from the lowest SES level pursuing Business achieved a four-year graduation rate of 78%, as did
Communication majors. The gap of 21% between the lowest graduation rate, Engineering, and
highest rates for Business and Communication was alarming. Similarly, for students in the low
SES range, the gap between Engineering at 64% and Humanities at 69% when compared to
Business at 92% and Social Sciences at 89% was also concerning. For all other SES levels and
pursued majors, the four year graduation rate was between 75% and 90%. So, when it came to
differences in the graduation rate by major, SES appears have played a vital role. This is
particularly true in that the gap in four year graduation rates between lowest SES and middle to
upper SES students in the same major was 10% or greater for all majors, with the gap exceeding
20% for Engineering, Natural Sciences, and Undecided.
Summary for Research Question Two
In summary of the findings for research question two, SES and academic preparedness
appear to have had the greatest effect on academic achievement at the university. This was
particularly true for four year graduation rates. When all pre-college variables were similar,
other than SES, the higher SES students performed better on average academically than their
lower SES peers. This was also found for students that participated in the support program, but
IMPACT OF SOCIOECONOMIC STATUS 129
to a lesser degree. Chart 25 provides a visual for how support program participants with
academic preparedness scores of .80 to less than .90 compared to non-participants with scores of
.80 to less than .868. The chart shows the fluctuation in graduation rates by SES level.
Even though the higher SES support program participants tended to outperform their
lower SES peers, there were not statistically significant differences found for first-year GPA (F
(4, 292) = 0.773, p= .543), units completed after the first year (F (4, 292) = 1.519, p= .197),
fourth year GPA (F (4, 292) = 1.505, p= .201), and units completed after the fourth year (F (4,
292) = 0.995, p= .410). The differences in graduation rates between the support program
participants of different SES ranges were statistically significant (X
2
(4) = 9.489, p= .050). When
focusing on students with academic preparedness scores of .80 to less than .90, the differences
in four-year graduation rates by SES levels was not statistically significant for the support
program population (X
2
(4) = 7.621, p= .107), meanwhile the differences by SES for the non-
participants were statistically significant (X
2
(4) = 13.040, p= .011). Similarly, for support
program students with the same range of preparedness scores of .80 to less than .90, there were
no statistical differences between SES ranges for first-year GPA (F (4, 197) = 0.961, p= .430),
units completed after the first year (F (4, 197) = 0.427, p= .789), fourth year GPA (F (4, 197) =
1.971, p= .100), and units completed after the fourth year (F (4, 197) = 0.919, p= .454). For the
non-participants though, the differences in first year GPA (F (4, 1233) = 4.902, p= .001) and
fourth year GPA (F (4, 1233) = 5.065, p= .000) by SES were found to be statistically significant,
but the units completed after the first (F (4, 1233) = 0.745, p= .561) and fourth year were not (F
(4, 1233) = 0.924, p= .449). This provides some evidence supporting the notion that differences
in academic achievement by SES could have been partially reduced via the assistance provided
IMPACT OF SOCIOECONOMIC STATUS 130
by the support program. The differences in population size, however, also could have resulted in
these differences.
Overall, when academic preparedness was adjusted, it appeared that the lowest and low
SES groups were struggling the most academically. Charts 26 and 27 provide the visual image
and data on the four-year graduation rates by SES and academic preparedness levels for non-
participants of the support program. These charts clearly show the impact of SES and academic
preparedness levels. Whether the support program helped minimize the effects of SES on
academic outcomes, particularly four-year graduation, is inconclusive. The data does, however,
show that lower SES students could likely benefit from extra support. Chapter five will further
address the entirety of findings presented in this chapter.

IMPACT OF SOCIOECONOMIC STATUS 131
Chapter 5: Conclusions and Implications
The first four chapters introduced the need to learn more about the impact of SES,
provided background information on research and literature, detailed the methodology of this
study, and examined the data. The goal of this chapter is to tie it all together. This will require a
summary of the study, a discussion about what the findings mean, conclusions related to the
literature, and implications for the future.
Summary
Literature related to assessment practices, financial aid, support programs, highly
selective institutions, and theoretical perspectives related to SES were examined in the previous
chapters. The purpose of this study was to identify how and why the impact of SES matters.
With the goal of President Obama to increase the percentage of citizens in the United States
between the ages of 25 and 34 with a college degree to 55% by the year 2025 (Lee & Rawls,
2010), it is important to understand that low SES students struggle the most with access and
success in higher education (Carnevale & Rose, 2003; Walpole, 2003). Identifying practices that
increase graduate rates for low SES students is imperative.
Degree achievement percentages, rates of return on investment, and social mobility are
highest at the most selective colleges and universities (Haveman & Smeeding, 2006; Hill et al.,
2005). More specifically, the most highly selective, private, colleges and universities have the
highest rate of return (Brewer, Eide, & Ehrenberg, 1999; Thomas & Zhang, 2005). There is
evidence to suggest the benefits of increasing the access and degree attainment for low SES
students at highly selective, private, institutions.
Researchers have already provided details regarding the academic achievement gaps
between low and high SES students (Carnevale & Rose, 2003; Walpole, 2003). The findings of
IMPACT OF SOCIOECONOMIC STATUS 132
this study further clarify the struggles of low SES students. When all things were similar except
for SES, the academic outcomes of lower SES students were less than that of higher SES peers.
Research Question One
For first time freshmen, what differences exist when the persistence, grade point average,
units earned, and degree completion are compared for students required to participate in a
support program and non-participants of similar socioeconomic status and academic
preparedness? In search of an answer to this question, the academic achievement of students
required to participate in a support program were examined and compared to that of non-
participants. This proved to be challenging. There were 297 potentially at-risk support program
participants compared to the general population of 6,557 students. As indicated in table 24, the
program participants had the same first to second year persistence rate of 97% as the non-
participants. This finding was similar to that of Braunstein et al. (2007), who also found that
students receiving support achieved similar first to second year persistence rates as their peers
that had higher HSGPA and test scores. However, a similar first to second year persistent rate
did not produce a similar four year graduation rate. For this study, the non-participants achieved
an 81% four year graduation rate and the support program participants combined for 65% even
though both populations had a 97% first to second year persistence rate. Braunstein et al. (2007)
did not report academic achievement past the first year.
Even when students of similar academic preparedness were found, there remained a
question. Why were certain students selected for the support program and others were not? The
assumption was that the students required to participate in the support program were selected by
the office of admission because they provided non-quantitative information in their admission
IMPACT OF SOCIOECONOMIC STATUS 133
applications that presented them as being more at-risk than other students of similar academic
preparedness.
Based on this reasoning, it would be expected that students in the support program would
not be able to achieve the academic outcomes of their peers without support. So this shaped the
research question. Were support program participants achieving similar, or better, academic
success when compared to non-participants of similar academic preparedness and SES? The
findings reveal that participants in the support program achieved similar educational outcomes
when compared to non-participants of similar academic preparedness and SES. This appeared to
happen despite the support program participants having what appeared to be fewer AP units
transferring in on average. Beyond this general finding though, little more could be found.
Students in the support program proved to be different when examined closely in
comparison with the non-participant population. First of all, there were challenges when finding
students of similar academic preparedness. The program participants had a fairly large range of
scores considering the relatively small size of the population, 297 students. This wide range was
particularly noticed for the lower end of scores. For instance, 202 of the 297 students in the
support program had academic preparedness scores of .80 to less than .90 and they averaged
.8448. Meanwhile, 1,238 of the 6,557 non-participants were found within the same academic
preparedness range with an average score of .8699. Charts five and six provide visual
illustrations for the academic preparedness score distribution for support program participants
and non-participants. The bell-curve reaches a peak at around .85 for the support program
participants, but does not reach a peak until around .95 for the non-participants. This is why the
non-participants had a higher average. The majority of non-participants with scores of .80 to less
than .90 were concentrated near the upper range.
IMPACT OF SOCIOECONOMIC STATUS 134
The participant population also had a much higher rate of low SES students. Over 41%
of the support program participants belonged to either the lowest or low SES ranges. Less than
20% of the non-participants were from the lowest or low SES ranges. Additionally, by the time
participants were split up by academic preparedness score ranges and SES, there were not many
students left to compare to the larger non-participant sub-groups. The tables in chapter four
detail these complications.
The challenge of comparing the subgroups of program participants and non-participants
became even more difficult when also factoring in differences in the distribution of gender, first-
generation status, race/ethnicity, and major. There were only six pairings of groups that
appeared to be relatively similar. The comparison of these matched groups revealed that, when
the variables examined were relatively similar, the program participants achieved similar or
better academic outcomes. This was despite non-participants earning a larger number of units
completed outside of the university. Within the data for sub-groups, the non-participants
regularly achieved around four to five more units outside of the university. This was found by
taking the total of units completed after four years and subtracting the number of units completed
at the university. Based on experience, this shows that the non-participant group likely
transferred in more advanced placement units than the participant group. When considering the
program participants completed more units after four years at the university, it seems the non-
participants relied on the additional units outside of the university in order to achieve similar four
year graduation results.
In summary, few conclusions or generalizations can be pulled from research question
one. Support program participants were assumedly considered to be more at-risk than their non-
participant peers of similar SES and academic preparedness. The additional support of the
IMPACT OF SOCIOECONOMIC STATUS 135
program hypothetically made it more possible for the participant group to achieve equal, if not
slightly better, academic outcomes. Without a true experimental design or the presence of a
substantial population size though, these findings are not very reliable or conclusive.
Research Question Two
What differences exist, if any, when examining the persistence, grade point average, units
earned, and degree completion of first-time freshmen of various socioeconomic backgrounds?
Research addressing gaps in access and success for low SES students has already been addressed
(Carnevale & Rose, 2003; Walpole, 2003). The primary goal of this study was to further clarify
the problem, particularly at a highly selective, private, research university. In addition to
identifying potential differences in the educational outcomes of students of various SES levels,
the data was further disaggregated by academic preparedness, first generation status, gender,
race/ethnicity, and pursued major. As the findings show, SES was found to have had a profound
impact on academic achievement.
Lower SES students, on average, arrived with lower academic preparedness scores and
went on to earn lower first and fourth year cumulative GPAs, had fewer units completed after the
first and fourth year, and lower four year graduation rates. The first to second year persistence
rates of low SES students were comparable or better to higher SES peers, but that did not
translate into comparable graduation rates. Even when grouping students with similar academic
preparedness scores, academic outcomes remained better for the higher SES students. For the
entire study, academic preparedness scores and SES emerged as overwhelmingly significant
predictors of success. Low SES students with lower academic preparedness scores were
substantially outperformed by upper SES students with high academic preparedness scores.
IMPACT OF SOCIOECONOMIC STATUS 136
In addition to having lower average academic preparedness scores, lower SES students
were also more likely to be the first in their family to attend college, female, and from a
racial/ethnic minority group. There were vast differences in student composition found for
students of various SES ranges and the stark uniformity was surprisingly consistent. For
instance, as SES increased the percentage of White students rose from 17.3% to 35.6% to 42.8%
to 55.0% to 65.3%. Meanwhile, the percentage of first-generation college students dropped from
46.6% to 3.4%, Hispanic/Latino students dropped from 28.9% to 7.6%, Asians from 39.0% to
23.6%, and Black/African Americans from 11.8% to 1.5%. This demonstrates resounding
differences between the lower and upper SES levels.
When the differences between racial/ethnic groups were further analyzed, Asian
American and White students had achieved better academic outcomes when compared to
Hispanic/Latino and Black/African American students. The academic achievement gaps,
however, were primarily the result of differences in SES and academic preparedness. The Asian
and White populations had higher average academic preparedness scores and a larger
concentration of upper SES students. The outcome results were different when SES levels and a
common range of .85 to less than .95 for academic preparedness scores were additionally
accounted for. Asian American and White students still had higher average academic
preparedness scores within this range, but the numbers were more similar to that of
Hispanic/Latino and Black/African American students. The differences in GPA, units earned,
and graduation rates between the racial/ethnic groups became minimal when factoring in SES
within this range of academic preparedness.
Females at all SES levels performed slightly better academically when compared to
males. Both male and female students from the lowest SES range were the most at risk, as seen
IMPACT OF SOCIOECONOMIC STATUS 137
in table 36. Lowest SES males achieved a 63% four year graduation rate compared to the 83%
rate of mid-upper SES males. Lowest SES females achieved a 70% four year graduation rate
compared to the 87% rate of mid-upper SES females.
First generation college students were outperformed academically by their non-first
generation college peers. This was found for all outcome variables except first to second year
persistence. The primary causes of these achievement gaps seem to be SES and academic
preparedness. The gap appeared to be minimized when students were disaggregated by SES. In
reviewing the academic preparedness scores for each group, it also appears that first generation
college students also had much lower average academic preparedness scores. In other words, if
SES and academic preparedness were both accounted for, then the differences in academic
achievement between first generation and non-first generation students would likely be minimal
or non-existent based on the data.
The lowest SES students in all majors achieved the lowest four year graduation rate.
Overall, the findings illustrate a need for institutions to disaggregate their data by pursued major.
This need for disaggregation was previously called for by Bensimon (2005). Certain majors in
this study, such as business and communication, achieved much higher rates when compared to
engineering and undecided students. There were not tremendous differences in the distribution
of pursued major between SES levels. This means that lower and upper SES students were not
more or less likely to have pursued a particular major. It was still important to examine if
students of a certain SES were underperforming in a particular major. An achievement gap
between the lowest and higher SES students was found for students that were undecided or
pursuing engineering, humanities, natural sciences, and social sciences.

IMPACT OF SOCIOECONOMIC STATUS 138
Limitations and Delimitations
Due to the exploratory nature of this study and the research design there were multiple
limitations and delimitations. These were primarily the result of the time period, population, and
other decisions that were made. Many of these were expected. As the research was conducted
though, additional limitations were discovered.
First-time freshmen that entered a single university in the fall of 2007, 2008, and 2009
were examined. The researcher combined these cohorts into a single population for the study
and made decisions related to the calculation of SES, academic preparedness, and pursued
degree program. The expected family contribution produced internally by the university was
used to determine SES within the study. It is not known how the outcomes of the research would
have differed had other variables been used in addition to or instead of EFC. Even though this
measurement for SES proved to be significantly linked with academic success, this cannot be
generalized across other calculations for SES or other institutions. Similarly, the method of
calculating academic preparedness was found to be significantly related to academic success.
Different determinations for academic preparedness and/or studies at other universities would
like produce different outcomes. For the categorization of majors into pursued degree program
types, efforts were made to organize majors in a reasonable fashion. The types of major, as well
as the respect and prestige of each major, varies at different colleges and universities. Just as one
major can be very different with another in the same category, the same major can also be very
different at another university. So, attempts to compare how students pursuing majors in the arts
at one university to that of another would be challenging. Thus, the findings of this study are
limited to the university where it was conducted.
IMPACT OF SOCIOECONOMIC STATUS 139
An additional limitation resulted from the fact that participation was required for students
in the support program. If they wanted to attend the university then they needed to accept the
admission office decision and placement into the program. Some students embraced the extra
support available to them. Others resented the decision, viewed it as a stigma, and/or were non-
cooperative with the additional requirements. Considering that students outside of the program
were capable of seeking and receiving comparable support from the same office and staff, there
very well could have been unhappy program participants that utilized the support far less than
the eager non-participants wanting to take advantage of everything available to them. This was a
known complication that did not receive much attention during the study.
In addition to discovering the complication presented by tuition benefit recipients, there
were other limitations discovered during the process of the research. Advanced placement units
seemed to play a role in the comparison of support program participants and non-participants.
This variable was not included in the study so it was not known why participants of the support
program had fewer units earned outside the university when compared to non-participants. It
was assumed that this was due to non-participants arriving at the university with a larger number
of transferrable advanced placement units, even when academic preparedness levels were
similar. This assumption was made solely based on the experience of the researcher and would
require additional research to determine whether this was true. Even then, this very specific
finding could likely not be generalized across other programs or universities.
The study also found that the percentages of female and first generation college students
were higher within the support program. These differences, particularly for gender, made it
challenging to find reasonably comparable subpopulations of support program participants and
non-participants. Additional steps could have been made for females in the support program to
IMPACT OF SOCIOECONOMIC STATUS 140
be compared to non-participant females, and the same for males to males, but that was not
pursued. As result, these smaller subpopulations, that could have perhaps been reasonably
comparable, were not examined.
The higher rate of low SES students found within the support program was also a newly
discovered complication. As result, there were challenges finding enough low SES students of
similar academic preparedness in the non-participant population. Another reason for this
challenge was the discovered relationship between SES and academic preparedness for the
different populations. As SES levels increased within the support program population the
average academic preparedness scores declined. Meanwhile, academic preparedness scores
increased with the rise of SES levels within the non-participant population. These differences
were not expected and created additional challenges when trying to reasonably compare
participants to non-participants.
Discussion
The statistically significant correlation between academic preparedness scores and
academic achievement was somewhat unexpected. The academic preparedness variable was
created out of the assumption that high school GPA and standardized test scores are examined by
admission offices in connection with each other. For instance, a student with a lower high school
GPA might warrant consideration for admission if their test scores are high. For this study, the
primary goal of using an academic preparedness variable that would combine the results for
HSGPA, adjusted HSGPA, and best SAT score was to more easily group students into levels of
academic preparedness. The finding that this variable also was a more statistically significant
predictor of academic success than either of its three component variables was unexpected.
IMPACT OF SOCIOECONOMIC STATUS 141
Based on previous research, it was expected that lower SES students would likely have
lower achievement outcomes than their upper SES peers. Both Walpole (2003) and Carnevale
and Rose (2003) had previously shown the achievement gap between lower and upper SES
students. Langhout, et al. (2009) provided examples of de facto types of classism and how they
impact campus culture sense of acceptance for lower SES students. Ostrove and Long (2007)
also found that class background impacted sense of belonging, academic adjustment, and social
adjustment. Different types of capital have also been connected to the challenges faced by low
SES college students. Economic capital deficits are obvious. According to Stanton-Salazar
(1997), a combination of psychosocial and institutional forces limits the development of
relationships and social capital that minorities need with institutional agents. According to
Lareau and Weininger (2003), the types of cultural capital acquired by low SES students do not
typically transfer over to the upper-middle class norms and beliefs of a college campus.
Essentially, lower SES students face a multitude of challenges when they attend college. This is
especially true at a highly selective, private, research university.
The findings regarding the impact of the support program were somewhat expected.
Published research on support programs was scarce and the literature that was found had
limitations and results that did not show the full picture. Braunstein et al. (2007) showed that
participants in a student support services program achieved similar first to second year
persistence rates despite having lower SAT scores, lower HSGPA, and lower family incomes.
Academic achievement outcomes other than persistence were not detailed though. Angrist et al.
(2009) examined the impact of support on entering freshman and found that participants that
received additional opportunities achieved a higher first year GPA than the control group. The
researchers did not examine results past the first year though.
IMPACT OF SOCIOECONOMIC STATUS 142
Due to the absence of information in the literature detailing the academic achievement of
support program participants past the first year, the researcher decided to report as much data as
possible in the study. This choice was also made because the research was conducted by the
director of the support program. The goal was to be as transparent as possible in order to reduce
the risk of bias. As such, table 24 provides an example of these details. Students in the support
program from the lowest and low SES ranges, despite much lower average academic
preparedness scores, were more likely to persist to the second year, completed the same number
of units at the university after the first year, and earned a similar number of units at the university
after four years. The lowest and low SES participants of the support program persisted to the
second year at a rate of 100%, completed an average of 32.4 and 32.6 units after the first year,
and averaged 115.0 and 121.8 units respectively after the fourth year. This compared well to the
97% persistence rate of the non-participants, 32.4 and 32.5 units after the first year, and 118.8
and 119.7 units completed after the fourth year. This data alone does not tell the whole story.
The lowest and low SES non-participants achieved substantially higher GPA and four year
graduation rates (Table 24). The outcomes were also varied for students of different SES levels.
This example illustrates why the complete data was provided in this study and also why there
was a need to compare support program participants to non-participants of similar academic
preparedness and SES.
When participants were compared to non-participants of similar SES and academic
preparedness the academic achievement results between the groups were quite similar (Tables
45, 48, 49, and 28a). These tables detail how participants of similar SES and academic
preparedness achieved academically when compared to non-participants and especially non-
participants that more closely matched the average academic preparedness scores of the support
IMPACT OF SOCIOECONOMIC STATUS 143
program participant group. A question that remained however was how reasonably similar
should groups be in order to compare them. Many of the participant and non-participant groups
of similar academic preparedness and SES had differences in composition when examining
gender, race/ethnicity, and pursued major. Academic outcomes were found to be quite similar
for the six pairings of participant and non-participant groups with similar academic preparedness
scores, SES, and composition (Table 52). Even though the outcomes for the other fourteen
pairings had similar academic outcomes, the composition by gender, race/ethnicity, and pursued
major were quite different.
Conclusions
Academic preparedness and the SES of a student profoundly impacted academic success
at the university level. The comparison of support program participants to non-participants
proved to be complicated. Furthermore, whether the support program minimized the impact of
SES could not be determined. This section will provide further details on the key findings of this
study.
Academic Preparedness
The academic preparedness score variable created for this study used a combination of
best test score, HSGPA on a 4.0 scale, and adjusted HSGPA on a 4.7 scale. The initial
correlations showed that academic preparedness scores had a statistically significant positive
correlation with GPA earned after the first (.406**) and fourth year (.410**) at the university
(Table 42). This positive correlation was stronger than that of HSGPA (.363** and .376**),
adjusted HSGPA (.376** and .393**), and best test score (.240** and .221**). Academic
preparedness scores had a statistically significant positive correlation with units completed at the
university after the first (.172**) and fourth year (.126**), in addition to total degree eligible
IMPACT OF SOCIOECONOMIC STATUS 144
units completed after the fourth year (.297**). These findings were statistically significant at the
0.01 level (Table 43). The descriptive statistics throughout the study provided examples of how
higher academic preparedness scores equated to higher academic achievement scores on average.
The academic achievement outcomes increased for students of increasingly better academic
preparedness score ranges. Additionally, these differences were maintained when SES was
accounted for.
The statistical significance of this finding was unexpected. Previous research examining
SES did not fully account for the impact of academic preparedness on academic outcomes at the
college level. Carnevale and Rose (2003) identified the connection of SES to SAT-equivalent
scores and also to graduation rates, but they did not examine whether the relationship between
SES and graduation rates was also influenced by differences in SAT-equivalent scores or other
types of academic preparedness. As such, the findings in this study show that there is a need for
future research on SES to account also for academic preparedness. Low SES students in this
study tended to also have lower levels of academic preparedness. For this reason, it is best to
account for potential differences in academic preparedness before examining for differences in
academic outcomes by SES level.
SES
The expected family contribution (EFC) for the first year calculated internally by the
university was converted to a percentage of the full cost of attendance for the first year. The
resulting output was used to quantify SES. The range and specific number of this variable and
its relationship with academic outcomes was the central focus of the study.
Overall, SES had a statistically significant positive correlation with university GPA and
units earned (Table 43). The differences in four year graduation rates by SES range were also
IMPACT OF SOCIOECONOMIC STATUS 145
statistically significant for the 6,557 non-participants of the support program. A one way
analysis of variance by SES groups was also examined for the non-participants of the support
program. The findings showed that differences in first year GPA, first year units completed at
the university, and fourth year GPA, were statistically significant based on SES level.
These findings for SES connected well with previous research. Walpole (2003) found
differences in graduation rates by SES quartile. What separated this study from previous
research was that the impact of SES on academic outcomes was further isolated. Other variables
such as academic preparedness, gender, and race/ethnicity were also accounted for.
Gender
Descriptive statistics showed that female students at all SES levels performed slightly
better than males. This was found for all outcome variables except persistence to the second
year. The 3,162 males in the study averaged a four year graduation rate of 78% and a GPA after
the fourth year of 3.28. The 3,692 female students averaged 83% and 3.41. Academic
preparedness scores on average were relatively similar for males and females. When also
accounting for SES, it was discovered that females of similar academic preparedness and SES
performed slightly better than males. Although this finding was revealing, it was not further
examined because the primary focus of the study was on SES.
First Generation Status
First generation college students had a substantially higher rate of students from lower
SES backgrounds and were expected to have lower academic achievement outcomes. This was
found to be true, but most of the gap in achievement was the result of SES (Table 52). The 6,053
students in the study that were not the first generation in their family to attend college achieved a
four year graduation rate of 82% and a GPA after the fourth year of 3.37. These outcomes were
IMPACT OF SOCIOECONOMIC STATUS 146
higher than the 73% four year graduation rate and 3.22 GPA after the fourth year for first
generation college students (Table 52). Additionally, the first generation college students had
lower average academic preparedness scores. Although steps were not taken to further compare
first generation college students to non-first generation students of similar academic
preparedness, the data found was enough to assume that much of the difference in achievement
outcomes resulted from differences in academic preparedness and SES.
Race/Ethnicity
White and Asian American students achieved higher GPA and four year graduation rates
when compared to Hispanic/Latino and Black/African American students. When these
populations were examined more closely, however, the achievement gaps were greatly
minimized after accounting for differences in the distribution of students by SES and academic
preparedness. The Hispanic/Latino and Black/African American populations had lower
academic preparedness scores on average. There also were substantial differences in the
distribution of students by SES between these racial/ethnic groups.
The majority of Hispanic/Latino and Black/African American students had academic
preparedness scores between .85 and .95. Most White and Asian students had scores between
.90 and 1.0. In order to better compare the academic achievement of these four largest
racial/ethnic groups, focus was given to students with academic preparedness scores of .85 to
less than .95. Hispanic/Latino and Black/African American students with scores within this
range would have an average of around .90. Due to a higher number of students with scores near
the high point of this range, .95, the White and Asian populations still had higher average
academic preparedness scores. The average scores between groups, however, were much more
similar when examining this range of scores.
IMPACT OF SOCIOECONOMIC STATUS 147
By focusing solely on students with academic preparedness scores of .85 to less than .95,
it became possible to more reasonably compare the academic achievement of the four largest
racial/ethnic populations (Tables 49 and 50). Both SES and academic preparedness were
accounted for. As result, there were no statistically significant differences in four-year
graduation rates. In sum, the overall differences in degree completion between racial/ethnic
populations were primarily the result of differences in the distribution of SES and academic
preparedness.
Due to the exploratory nature of this study, these findings related to race/ethnicity were
not expected. It is not known how or if these findings connected to prior research conducted by
others because the literature reviewed prior to this study was primarily related to SES. The
findings, however, do suggest a need for future research.
Pursued Major Type
An analysis of academic achievement for students of different pursued degree programs
revealed disparities. Students pursuing majors in business, communication, and social sciences
at the conclusion of their first semester would go on to achieve higher four year graduation rates
(Chart 24). Engineering majors had the lowest four year graduation rates. For all pursued
majors, lowest SES students had the lowest four-year graduation rates. Even though there are
differences between specific majors and categorical types of majors, it is troubling to see that
students from the lowest SES struggled.
Support Program
Participants of the support program achieved similar academic outcomes when compared
to non-participants of reasonably similar SES, academic preparedness, gender, first generation
status, race/ethnicity, and pursued major. This leads to the conclusion that the support program
IMPACT OF SOCIOECONOMIC STATUS 148
successfully helped students that were considered to be more at risk than their peers of similar
academic preparedness and SES. However, there were challenges identifying reasonably
comparable students to compare with the program participants. Only six of the twenty groupings
examined were deemed reasonably comparable when examining for SES, academic
preparedness, gender, first generation status, and pursued major. There also remained the
question as to why the support program participants were considered to be more at-risk than
peers that were found to be reasonably similar.
The overall findings and challenges related to the analysis of the support program were
inconclusive. Rather than detailing only the findings that favored the support program, all of the
data was presented within this study with the aim of being as transparent as possible. Evidence
was found that suggested that the extra support was valuable. In relation to the overall focus on
SES, it seems that extra support and guidance could best benefit low SES students and others
with lower academic preparedness scores.
Persistence
The findings of this study repeatedly demonstrated that persistence from the first year to
the second was not predictive of four year graduation rates. Low SES subpopulations regularly
had higher first to second year persistence rates. This did not relate to higher graduation rates.
Considering persistence research has received a lot of attention (Reason, 2003; 2009), it should
be important to remember that degree attainment is the primary goal.
Implications
Throughout the review of the research literature and the findings within this study there
were themes that continued to emerge. The three most prominent themes were acceptance,
belonging, and capital. Each of these three areas intersects and nurtures each other. The
IMPACT OF SOCIOECONOMIC STATUS 149
following recommendations for action and understanding have been developed by the researcher
based on what was learned through the completion of this study.
Acceptance
Acceptance relates to more than simply the opportunity for low SES students to attend
highly selective, private, colleges and universities. It also equates to an understanding that the
goals of the completion agenda likely cannot be met in a meaningful way if only a select few
token low SES students make it through the filter of the education system and into a highly
selective, private, college or university. The 55% goal of the completion agenda (Lee & Rawls,
2010) could be met without increasing the percentage of qualified low SES students at highly
selective institutions, but it is doubtful that would improve the economic stability or disparities
between the rich and the poor within this country.
Low SES children need to feel that attending a highly selective, private, university is
possible. Information and success stories need to be shared. Colleges and universities have an
opportunity to form partnerships and coalitions to help facilitate this. Successful examples
include the Posse Foundation, Questbridge, and the Neighborhood Academic Initiative program.
To further accomplish the goal of increasing socioeconomic diversity, Williams College
was one of a handful of schools to adopt a need-blind admission policy (Hill & Winston, 2006).
The financial aid costs associated with doing this and the challenges with balancing the overall
budget, however, are de-motivators for doing this. For this reason, there needs to be additional
motivations for colleges and universities to pursue socioeconomic diversity and a commitment to
meeting the financial needs of students. University ranking publications and systems provide
ample motivation. If the more popular ranking publications were to reward schools for enrolling
IMPACT OF SOCIOECONOMIC STATUS 150
a socioeconomically diverse student body and for graduating high rates of low SES students then
that would be a step in the right direction.
Belonging
Low SES students are already underrepresented in higher education (Haveman &
Smeeding, 2006) and face a multitude of challenges when they are admitted. In addition to
needing to conform to the upper-class norms of the university culture, there are also policies and
fees that serve as a de facto form of classism (Langhout et al, 2009). More should be done to
develop and nurture the sense of belonging of low SES students. Just as Aries and Seider (2005)
suggested, additional programs and services could benefit low SES students. Careful
consideration should be given during the creation of any support program designed to assist low
SES students. For instance, a diversity office at a smaller college might be able to integrate the
goals of promoting and supporting socioeconomic diversity. This type of integration may not be
possible at larger universities where diversity initiatives have typically been decentralized into a
variety of programs and services related to areas such as race/ethnicity, LGBTQ, and ability.
Additionally, if students were required to participate in a socioeconomic diversity
program then that too may create a negative reaction. However, if staff members are not able to
form valuable relationships or generate motivation for students to utilize their support then many
low SES students might continue to slip through the cracks. This complexity requires that each
university individually explore how best to provide support to low SES students until best
practices have been identified. Here is one example that could potentially improve these
concerns.
Merit-based aid often comes in the form of presidential, trustee, dean, or other
prestigiously titled scholarships. Each of these titled examples produces a sense of pride and
IMPACT OF SOCIOECONOMIC STATUS 151
prestige. Recipients of this titled merit aid feel wanted. Meanwhile the need-based university
grants are typically nameless. To improve this, a university could take the need-based university
grant money received by the lowest SES students and title this aid after an accomplished person
affiliated with the university that overcame their low SES. Furthermore, this naming could even
be used as an opportunity to seek out donors willing to dedicate funding to the financial aid
office earmarked specifically for low SES students. Then, as admitted students are offered this
titled need-based aid, they will feel more wanted and curious about the unique name. As they
learn more about the name they will become even more invested in the possibility of similarly
achieving success at the university. This is where each of these students could also learn about a
program that offers support specifically to recipients of this named need-based aid for low SES
students. Instead of feeling stigmatized, the students will feel more wanted and likely have more
motivation to take advantage of the support available to them. Additional peer mentoring and
networking with upperclassmen receiving the same need-based aid can also be made available as
new students transition into and begin their first year. As such, this example would work to
improve the acceptance, belonging, and capital of low SES students.
The timing of additional support offered to low SES students is also of importance. Once
low SES students are admitted there needs to be additional communication coming from the
university. Prior to the certification deadline and throughout the summer, staff members
representing various programs have the opportunity to proactively reach out to incoming students
in order to nourish the development of positive relationships. Low SES students would greatly
benefit from this outreach because they might not be able to afford a campus visit prior to move-
in day. Additionally, once low SES students arrive on campus there needs to be organized
activities and/or meetings to help minimize the culture shock and spreading of misinformation
IMPACT OF SOCIOECONOMIC STATUS 152
among nervous freshmen. The first month or two of college can be overwhelming. Welcome
weeks often kick off the beginning of the year at many colleges and universities. Purposeful
follow-ups and outreach must be maintained though. These can consist of guest speakers,
special events, support group meetings, field trips, and one-on-one guidance appointments. The
main thing is for each student to feel accepted and like they belong. And, if they do not, then
they should know a staff person that they can turn to for support.
Social and Cultural Capital
Relationships formed with student peers and staff at the college level lead to more than
just an increased sense of belonging; they also help promote social and cultural capital. Low
SES students often arrive with social and cultural capital that is not transferrable to the upper
middle class norms of the college campus. This is especially true at highly selective, private,
universities. For these reasons, it is important for institutional agents to reach out to help bridge
the gap. This support can come in many forms, but typically entails a staff member providing
the student with information, guidance, intervention, and/or simply just someone trustworthy to
talk to.
Information. When information is distributed to new students there needs to be
additional consideration given to how the information is received. Otherwise students will be
more likely have challenges understanding and applying it. In relation to social and cultural
capital, students might not have the right information or know who or where to turn to for advice
and guidance. Paper forms and mailings are often lost or thrown away. Email messages are
regularly not opened, deleted, misplaced, and/or misunderstood. Accurate information posted
online is often hard to find and when offices and programs are decentralized then there are often
incorrect links or rabbit holes leading to old information. These issues are especially
IMPACT OF SOCIOECONOMIC STATUS 153
complicating to low SES students because they are least likely to have someone to turn to for
correct information and often have the most to lose.
The distribution of information needs to be collaborative and multifaceted. Colleges and
universities should assess and discover more about the success rate of information being
received. Rather than operating within silos, each launching their own email blasts, more should
be accomplished in unison. For instance, approved updates could be emailed out each week
from a single address and posted online on a unique page for updates. Social media could also
be used with potentially five feeds created for potential undergraduates, current undergraduates,
potential graduate students, current graduate students, and faculty/staff. Anyone with questions
regarding critical information could peruse this site to learn new things or refresh their
understanding. Everything would be in one place. Once each month, students in the residence
halls could also discuss the meaning of the information distributed with their resident advisor.
As example, perhaps there is a financial aid or housing deadline approaching or possibly a class
registration period. The resident advisor, in a group setting, could inform residents of the
importance of the deadlines and information. They could also read through prepared examples to
further illustrate the intended message. The main point from this example is that critical
information can and should be more successfully distributed.
Guidance. Once a student receives critical information they then can benefit from
receiving additional guidance before decisions and actions are made. For instance, timely group
workshops and/or individual meetings typically occur before students register for classes. These
exist so students can learn more about what they should do and also how to do it. Considering
what has been learned about low SES students, similar outreach leading to group or individual
meetings could be organized during other important time periods. For instance, during welcome
IMPACT OF SOCIOECONOMIC STATUS 154
week there could be workshops providing guidance and advice regarding the purchasing of
textbooks. Similar offerings later in the year could also address housing options and costs.
These examples would be especially helpful to low SES students, particularly if the guidance
was provided by a trusted person on campus, or a referral from one was given.
Intervention. No matter the information and guidance available, some students will still
have challenges. These struggles might be personal, academic, social, and/or environmental. To
help minimize these struggles, and to intervene when necessary, it is important that relationships
with support staff be established early. If outreach and additional support is not delivered until
after a student goes on academic probation then that would be too late. Considering the
challenges faced by low SES students, a proactive approach is much more appropriate.
Trusted partners. Low SES students would benefit the most from having at least one
person on campus that they can turn to for support and guidance. Preferably, they have a
network of people they can turn to. This network of trusted partners would consist of peers,
faculty, and staff. Ideally, these networks would collaborate and receive information from an
office familiar with the issues at hand. Furthermore, this type of office could work to facilitate
the matching of low SES students to individuals and networks based on common interests and
pursued majors.
Economic capital
When considering the full cost of attending, all potential fees and charges need to be
accounted for. This includes both the costs for the student, as well as those that are
recommended for parents and guardians. Low SES students might not be familiar with what
costs can or cannot be covered by financial aid. More simplified information and examples
should be distributed to students and staff. Expenses that cannot be included into the full cost of
IMPACT OF SOCIOECONOMIC STATUS 155
attending and covered by financial aid require further thought. For instance, when family
members are encouraged to attend orientation but are charged an amount that they cannot afford
then the implicit message sent is that they do not belong at the orientation. Instead of sending
this message, there are opportunities to do things differently. Financial aid offices are typically
aware of low SES students by the time the orientation sessions begin. If the university were to
set aside funds or raise money so that at least one family member could attend orientation for
students meeting a certain threshold then that would send a more appropriate message to low
SES families. And this same process could likely be duplicated for other costs faced by the
student or their families.
Re-envisioning work-study jobs. Most every low SES student has the opportunity to
receive work-study funds. This creates a further challenge. In addition to adapting to the culture
shock of the campus and academic expectations, they also need to find a job. Even with the
presence of work-study job fairs, much of the responsibility is on the student. Additionally,
students that successfully find a position will be at their work-study job an upwards of 10, 15, or
even 20 hours each week. Whether a student is able to find a position, as well as any
relationships they form as result of it, likely produce profound effects. In most cases, too much
of this is left to chance. This process could be overhauled to produce better results.
A university could integrate the hiring of new work-study students with the admission
timeline. Similar to how many graduate students pursuing higher education programs seek out
internships and campus jobs related to their studies before they select which college they commit
to, some of the same processes can be adopted on the undergraduate level. If a work-study
location with an expected open position had access to information on potential incoming work-
study students then they would have the ability to proactively contact students. When
IMPACT OF SOCIOECONOMIC STATUS 156
considering the findings of this and other research, it should be accepted that a stand-alone single
day work-study job fair is insufficient.
Future Research
The findings of this study point to directions where additional research could be
beneficial. The relationship and examination of SES and academic preparedness needs to be
further examined at other institutions, particularly at other highly selective, private, colleges and
universities. Even if the calculation of SES and academic preparedness might be inconsistent
within other studies, the importance is figuring out whether there are differences in the
educational outcomes of students of similar academic preparedness by SES level. Only then can
the problem be fleshed out. For instance, the educational outcome gaps found within this study
might be minimal compared to that of other universities, or the opposite. Until additional
research has been conducted it will be challenging to identify best practices.
There was enough evidence within this study to suggest the further examination of the
impact of Advanced Placement (AP) examination credit. It would be interesting to see whether
students of similar academic preparedness, SES, and pursued major graduate at similar rates
when they arrive with higher or lower amounts of AP credit. Further, considering admission
offices internally calculate the adjusted HSGPA of applicants based on the rigor of their
coursework, it would be revealing to see if students of certain high schools achieved low scores
on their AP examinations despite being rewarded during the admission process. This
examination could reveal the possibility of high school grade inflation. An example of potential
grade inflation could be found if a high number of students from a particular high school tended
to have favorably high adjusted HSGPA based on the rigor of their coursework but low success
rates on the actual AP examinations in comparison to students from other high schools.
IMPACT OF SOCIOECONOMIC STATUS 157
When accounting for both academic preparedness and SES, there were noteworthy
findings within this study for race/ethnicity, gender, first generation status, and pursued major.
Although the findings for these variables were not as profound as that for academic preparedness
and SES, they do warrant further consideration. Additional research could expand the knowledge
regarding these findings.
Closing
In summary, colleges and universities need to make decisions regarding how low SES
students fit into their missions and strategic plans. Acceptance, belonging, and capital are three
crucial areas that could benefit low SES students. As colleges and universities identify how or if
they will be providing support to low SES students, additional information will need to be
appropriately shared with the families of potential applicants. Only then can the completion
agenda be delivered in a way that minimizes the economic inequalities in this country.
IMPACT OF SOCIOECONOMIC STATUS 158
References
Alon, S. (2005). Model mis-specification in assessing the impact of financial aid on academic
outcomes. Research in Higher Education, 46(1), 109-125. doi:10.1007/s11162-004-
6291-x
Alon, S. & Tienda, M. (2005). Assessing the "Mismatch" hypothesis: Differences in college
graduation rates by institutional selectivity. Sociology of Education, 78 (4), 294-315.
doi:10.1177/003804070507800402
Angrist, J., Lang, D., & Oreopoulos, P. (2009). Incentives and services for college achievement:
Evidence from a randomized trial. American Economic Journal: Applied
Economics, 1(1), 136-163. doi:10.1257/app.1.1.136
Aries, E., & Seider, M. (2005). The interactive relationship between class identity and the
college experience: The case of lower income students. Qualitative Sociology, 28(4),
419-443. doi:10.1007/s11133-005-8366-1
Astin, A. W. (1993). What matters in college? Four critical years revisited. San Francisco:
Jossey-Bass.
Astin, A. W. (1997). How ‘good’ is your institution’s retention rate? Research in Higher
Education, 38, 647–658. doi:10.1023/A:1024903702810
Astin, A. W., & Oseguera, L. (2004). The declining" equity" of American higher education. The
Review of Higher Education, 27(3), 321-341. doi:10.1353/rhe.2004.0001
Avery, C. M., & Daly, A. J. (2010). Promoting equitable educational outcomes for high-risk
college students: The roles of social capital and resilience. Journal of Equity in
Education, 1(1), 46-70. Retrieved from http://ucsd.academia.edu/AlanDaly/
IMPACT OF SOCIOECONOMIC STATUS 159
Baumeister, R. F., & Leary, M. R. (1995). The need to belong: desire for interpersonal
attachments as a fundamental human motivation. Psychological bulletin, 117(3), 497.
doi:10.1037/0033-2909.117.3.497
Bensimon, E. M. (2005). Closing the achievement gap in higher education: An organizational
learning perspective. New Directions for Higher Education, 131, 99-111.
doi:10.1002/he.190
Bensimon, E. M. (2007). The underestimated significance of practitioner knowledge in the
scholarship on student success. Review of Higher Education, 30, 441-469. Retrieved
from:
http://libproxy.usc.edu/login?url=http://search.proquest.com.libproxy.usc.edu/docview/22
0823027?accountid=14749
Bourdieu, P. (2008). 15 The Forms of Capital. Readings in economic sociology, 280.
Brand, J. E., & Xie, Y. (2010). Who benefits most from college? Evidence for negative selection
in heterogeneous economic returns to higher education. American Sociological
Review, 75(2), 273-302. doi:10.1177/0003122410363567
Braunstein, A. W., Lesser, M. H., & Pescatrice, D. R. (2008). The impact of a program for the
disadvantaged on student retention. College student journal,42(1), 36. Retrieved from:
http://www.alliedacademies.org/public/proceedings/Proceedings20/AEL-
Proceedings.pdf#page=12
Bresciani, M. J., & Carson, L. (2002). A study of undergraduate persistence by unmet need and
percentage of gift aid. NASPA Journal, 40(1), 104-123. doi:10.2202/1949-6605.1193
IMPACT OF SOCIOECONOMIC STATUS 160
Brewer, D. J., Eide, E. R., & Ehrenberg, R. G. (1999). Does it pay to attend an elite private
college? Cross-cohort evidence on the effects of college type on earnings. Journal of
Human Resources, 104-123. doi:10.1016/S0272-7757(97)00037-X
Brewer, D. J., Gates, S. M., & Goldman, C. A. (2002). In pursuit of prestige. New Brunswick,
NJ: Transaction Publishers.
Carnevale, A. P., & Rose, S. J. (2003). Socioeconomic status, race/ethnicity, and selective
college admissions. New York: Century Foundation.
Chen, R. (2008). Financial aid and student dropout in higher education: A heterogeneous
research approach. Higher Education, 209-239. doi:10.1007/978-1-4020-6959-8_7
Coleman, J. S. (1988). Social capital in the creation of human capital. American Journal of
Sociology, 95-120. doi:10.2307/2780243
Cragg, K. M. (2009). Influencing the probability for graduation at four-year institutions: A
multi-model analysis. Research in Higher Education, 50(4), 394-413.
doi:10.1007/s11162-009-9122-2
Crellin, M., Kelly, P., & Prince, H. (2012). Increasing College Attainment in the United States:
Variations in Returns to States and Their Residents. Change: The Magazine of Higher
Learning, 44(4), 35-41. doi:10.1080/00091383.2012.691863
Dennis, J. M., Phinney, J. S., Chuateco, L.I. (2005). The role of motivation, parental support,
and peer support in the academic success of ethnic minority first-generation college
students. Journal of College Student Development 46(3), 223-236.
doi:10.1353/csd.2005.0023
IMPACT OF SOCIOECONOMIC STATUS 161
DesJardins, S. L., & McCall, B. P. (2010). Simulating the effects of financial aid packages on
college student stopout, reenrollment spells, and graduation chances. The Review of
Higher Education, 33(4), 513-541. doi:10.1353/rhe.0.0169
Driscoll, E., Comm, C. L., & Mathaisel, D. F. (2013). A Lesson Plan For Sustainability In
Higher Education. American Journal of Business Education (AJBE), 6(2), 255-266.
Retrieved from:
http://www.journals.cluteonline.com/index.php/AJBE/article/viewFile/7691/7756
Dwyer, R. E., McCloud, L., & Hodson, R. (2012). Debt and graduation from American
universities. Social forces, 90(4), 1133-1155. doi:10.1093/sf/sos072
Evenbeck, S. & Johnson, K. E. (2012). Students must not become victims of the completion
agenda. Liberal Education, 98(1). Retrieved from:
http://www.aacu.org/liberaleducation/le-wi12/evenbeck.cfm
Fox, M. (1993). Is it a good investment to attend an elite private college?.Economics of
Education Review, 12(2), 137-151. doi:10.1016/0272-7757(93)90025-C
Freeman, T. M., Anderman, L. H., & Jensen, J. M. (2007). Sense of belonging in college
freshmen at the classroom and campus levels. The Journal of Experimental Education,
75(3), 203-220. doi:10.3200/JEXE.75.3.203-220
Geiser, S., & Santelices, M. V. (2007). Validity of high-school grades in predicting student
success beyond the freshman year: High-school record vs. standardized tests as
indicators of four-year college outcomes (Research and Occasional Paper Series No.
CSHE.6.07). Berkeley, CA: Center for Studies in Higher Education. Retrieved from
http://cshe.berkeley.edu/publications/docs/ROPS.GEISER._SAT_6.12.07.pdf
IMPACT OF SOCIOECONOMIC STATUS 162
Giroux, H. A. (2002). Neoliberalism, corporate culture, and the promise of higher education:
The university as a democratic public sphere. Harvard Educational Review, 72(4), 425-
464. Retrieved from: http://geoggingclub.files.wordpress.com/2012/03/2002-giroux-
neoliberalism-corporate-culture-and-the-promise-of-higher-education.pdf
Goenner, C. F., & Snaith, S. M. (2003). Predicting graduation rates: An analysis of student and
institutional factors at doctoral universities. Journal of College Student Retention, 5(4),
409-420. doi:10.2190/LKJX-CL3H-1AJ5-WVPE
Goldrick-Rab, S. (2006). Following their every move: An investigation of social-class
differences in college pathways. Sociology of Education, 79(1), 67-79.
doi:10.1177/003804070607900104
Grant-Vallone, E., Reid, K., Umali, C., & Pohlert, E. (2003). An analysis of the effects of self-
esteem, social support, and participation in student support services on students’
adjustment and commitment to college. Journal of College Student Retention: Research,
Theory and Practice, 5(3), 255-274. doi:10.2190/C0T7-YX50-F71V-00CW
Harper, S. R. & Kuh, G. D. (2007). Myths and misconceptions about using qualitative methods
in assessment. New Directions for Institutional Research, 136, 5-14. doi:10.1002/ir.227
Hausmann, L. R., Schofield, J. W., & Woods, R. L. (2007). Sense of belonging as a predictor of
intentions to persist among African American and White first-year college students.
Research in Higher Education, 48(7), 803-839. doi:10.1007/s11162-007-9052-9
Haveman, R. H., & Smeeding, T. M. (2006). The role of higher education in social mobility.
The Future of Children, 16(2), 125-150. doi:10.1353/foc.2006.0015
IMPACT OF SOCIOECONOMIC STATUS 163
Hill, C. B., Winston, G. C., & Boyd, S. A. (2005). Affordability Family Incomes and Net Prices
at Highly Selective Private Colleges and Universities. Journal of Human
Resources, 40(4), 769-790. doi:10.3368/jhr.XL.4.769
Hill, C. B., & Winston, G. C. (2006). Access: Net prices, affordability, and equity at a highly
selective college. Economics of Education Review, 25(1), 29-41.
doi:10.1016/j.econedurev.2004.11.004
Hoffman, J. L. & Lowitzki, K. E. (2005). Predicting college success with high school grades
and test scores: Limitations for minority students. The Review of Higher Education 28
(4), 455-74. doi:10.1353/rhe.2005.0042
Humphreys, D. (2012). What’s wrong with the completion agenda -- and what we can do about
it. Liberal Education, 98(1). Retrieved from: https://www.aacu.org/liberaleducation/le-
wi12/humphreys.cfm
Hurtado, S., & Carter, D. F. (1997). Effects of college transition and perceptions of the campus
racial climate on Latino students’ sense of belonging. Sociology of Education, 70, 324-
345. doi:10.1353/csd.2007.0054
Johnson, S. E., Richeson, J. A., & Finkel, E. J. (2011). Middle class and marginal?
Socioeconomic status, stigma, and self-regulation at an elite university. Journal of
personality and social psychology, 100(5), 838. doi:10.1037/a0021956
Johnson, D. R., Soldner, M., Leonard, J. B., Alvarez, P., Inkelas, K. K., Rowan-Kenyon, H. T.,
& Longerbeam, S. D. (2007). Examining sense of belonging among first-year
undergraduates from different racial/ethnic groups. Journal of College Student
Development, 48(5), 525-542. doi:10.1353/csd.2007.0054
IMPACT OF SOCIOECONOMIC STATUS 164
Jones-White, D. R., Radcliffe, P. M., Huesman, R. L., & Kellogg, J. P. (2010). Redefining
student success: Applying different multinomial regression techniques for the study of
student graduation across institutions of higher education. Research in Higher Education,
51(2), 154-174. doi:10.1007/s11162-009-9149-4
Kelderman, E. (2011). Shifts in politics and policies complicate college- completion agenda for
states. The Chronicle of Higher Education, August 14, 2011. Retrieved from:
http://chronicle.com/article/Shifts-in-Politics-and/128641
Kim, J., DesJardins, S. L., & McCall, B. P. (2009). Exploring the effects of student expectations
about financial aid on postsecondary choice: A focus on income and racial/ethnic
differences. Research in Higher Education, 50(8), 741-774. doi:10.1007/s11162-009-
9143-x
Knapp, L. G., Kelly-Reid, J. E., & Ginder, S. A. (2012, March). Enrollment in postsecondary
institutions, fall 2010; financial statistics, fiscal year 2010; and graduation rates,
selected cohorts, 2002 –2007: First look (NCES 2012-280). Washington, DC: U.S.
Department of Education, Institute of Education Sciences, National Center for Education
Statistics. Retrieved from: http://files.eric.ed.gov/fulltext/ED530514.pdf
Kobrin, J. L., Patterson, B. F., Shaw, E. J., Mattern, K. D., & Barbuti, S. M. (2008). "Validity of
the SAT for Predicting First-Year College Grade Point Average." The College Board 5:
1-10. Retrieved from:
http://research.collegeboard.org/sites/default/files/publications/2012/7/researchreport-
2008-5-validity-sat-predicting-first-year-college-grade-point-average.pdf
Kokkelenberg, E. C., Sinha, E., Porter, J. D., & Blose, G. L. (2008). The efficiency of private
universities as measured by graduation rates [Electronic version]. Retrieved [1/19/13],
IMPACT OF SOCIOECONOMIC STATUS 165
from Cornell University, School of Industrial and Labor Relations site:
http://digitalcommons.ilr.cornell.edu/workingpapers/135/
Kuh, G. D., and Pascarella, E. T. (2004). What Does Institutional Selectivity Tell Us About
Educational Quality? Change, 36(5): 52-58. doi:10.1080/00091380409604986
Langhout, R. D., Drake, P., & Rosselli, F. (2009). Classism in the university setting: Examining
student antecedents and outcomes. Journal of Diversity in Higher Education; Journal of
Diversity in Higher Education, 2(3), 166. doi:10.1037/a0016209
Lareau, A., & Weininger, E. B. (2003). Cultural capital in educational research: A critical
assessment. Theory and society, 32(5), 567-606.
doi:10.1023/B:RYSO.0000004951.04408.b0
Lee, E. M., & Kramer, R. (2012). Out with the Old, In with the New? Habitus and Social
Mobility at Selective Colleges. Sociology of Education.
doi:10.1177/0038040712445519
Lee Jr, J. M., & Rawls, A. (2010). The College Completion Agenda: 2010 Progress Report.
College Board Advocacy & Policy Center. doi:10.1787/664024334566
Locks, A. M., Hurtado, S., Bowman, N. A., & Oseguera, L. (2008). Extending notions of
campus climate and diversity to students' transition to college. The Review of Higher
Education, 31(3), 257-285. doi:10.1353/rhe.2008.0011
Long, B. T., & Riley, E. (2007). Financial aid: A broken bridge to college access? Harvard
Educational Review, 77(1), 39-63. Retrieved from:
http://isites.harvard.edu/fs/docs/icb.topic1233004.files/Long%20Riley%202007%20Fina
ncial%20Aid%20-%20A%20Broken%20Bridge%20to%20Access%20-%20HER.pdf
IMPACT OF SOCIOECONOMIC STATUS 166
Longanecker, D. (2002). Is Merit-Based Student Aid Really Trumping Need-Based
Aid? Change: The Magazine of Higher Learning, 34(2), 30-37.
doi:10.1080/00091380209601843
Munro, B. H. (1981). Dropouts from higher education: Path analysis of a national sample.
American Educational Research Journal, 18(2), 133-141.
doi:10.3102/00028312018002133
Niu, S. X., & Tienda, M. (2011). Test Scores, Class Rank and College Performance: Lessons for
Broadening Access and Promoting Success. The College Board. Retrieved from:
http://theop.princeton.edu/reports/forthcoming/NiuTienda_2011.pdf
Ostrove, J. M., & Long, S. M. (2007). Social class and belonging: Implications for college
adjustment. Review of Higher Education, 30(4), 363. Retrieved from:
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.332.4099&rep=rep1&type=pdf
Otewalt, E. C. (2013). An Examination of Summer Bridge Programs for First-Generation
College Students. Retrieved from:
http://digitalcommons.calpoly.edu/cgi/viewcontent.cgi?article=1043&context=psycdsp
Padgett, V. R., & Reid, Jr, J. F. (2002). Five year evaluation of the Student Diversity Program:
A Retrospective quasi-experiment. Journal of College Student Retention: Research,
Theory and Practice, 4(2), 135-145. doi:10.2190/25T7-3BBF-6HYB-NHAY
Park, J. J., Denson, N., & Bowman, N. A. (2012). Does socioeconomic diversity make a
difference? Examining the effects of racial and socioeconomic diversity on the campus
climate for diversity. American Educational Research Journal.
doi:10.3102/0002831212468290
IMPACT OF SOCIOECONOMIC STATUS 167
Pascarella, E. T., & Terenzini, P. T. (1983). Predicting voluntary freshman year
persistence/withdrawal behavior in a residential university: A path analytic validation of
Tinto's model. Journal of educational psychology, 75(2), 215-226. doi:10.1037/0022-
0663.75.2.215
Pittman, L. D., & Richmond, A. (2008). University belonging, friendship quality, and
psychological adjustment during the transition to college. The Journal of Experimental
Education, 76(4), 343-362. doi:10.3200/JEXE.76.4.343-362
Reason, R. D. (2003). Student Variables that Predict Retention: Recent Research and New
Developments. NASPA Journal, 40(4). doi:10.2202/1949-6605.5022
Reason, R. D. (2009). An examination of persistence research through the lens of a
comprehensive conceptual framework. Journal of College Student Development, 50 (6),
659-682. doi:10.1353/csd.0.0098
Rhoades, G. (2006). The higher education we choose: A question of balance. The Review of
Higher Education, 29(3), 381-404. doi:10.1353/rhe.2006.0015
Ryan, J. F. (2004). The relationship between institutional expenditures and degree attainment at
baccalaureate colleges. Research in Higher Education, 45(2), 97-114.
doi:10.1023/B:RIHE.0000015691.02545.61
Sadler, P. M., & Tai, R. H. (2007). Weighting for recognition: Accounting for advanced
placement and honors courses when calculating high school grade point average. NASSP
Bulletin, 91(1), 5-32. doi:10.1177/0192636506298726
Salkind, N. J. (2000). Statistics for people who (think they) hate statistics (p. 4). London: Sage.
IMPACT OF SOCIOECONOMIC STATUS 168
Schneider, C. G. (2012). Where completion goes awry: The metrics for “success” mask
mounting problems with quality. Liberal Education, 98(1). Retrieved from:
https://www.aacu.org/liberaleducation/le-wi12/president.cfm
Stanton-Salazar, R. D. (1997). A social capital framework for understanding the socialization of
racial minority children and youths. Harvard educational review, 67(1), 1-41. Retrieved
from:
http://zb5lh7ed7a.scholar.serialssolutions.com/?sid=google&auinit=RD&aulast=Stanton-
Salazar&atitle=A+social+capital+framework+for+understanding+the+socialization+of+r
acial+minority+children+and+youths&title=Harvard+educational+review&volume=67&
issue=1&date=1997&spage=1&issn=0017-8055
Strayhorn, T. L. (2011). Bridging the pipeline: Increasing underrepresented students’
preparation for college through a summer bridge program. American Behavioral
Scientist, 55(2), 142-159. doi:10.1177/0002764210381871
Terenzini, P. T., & Pascarella, E. T. (1977). Voluntary freshman attrition and patterns of social
and academic integration in a university: A test of a conceptual model. Research in
Higher Education, 6(1), 25-43. doi:10.1007/BF00992014
Thomas, S. L., & Zhang, L. (2005). Post-baccalaureate wage growth within four years of
graduation: The effects of college quality and college major. Research in Higher
Education, 46(4), 437-459. doi:10.1007/s11162-005-2969-y
Tinto, V. (1975). Dropout from higher education: A theoretical synthesis of recent research.
Review of educational research, 45(1), 89-125. doi:10.3102/00346543045001089
Tinto, V. (2006-2007). Research and practice of student retention: What next? Journal of
College Student Retention, 8 (1), 1-19. doi:10.2190/4YNU-4TMB-22DJ-AN4W
IMPACT OF SOCIOECONOMIC STATUS 169
Titus, M. A. (2006). Understanding college degree completion of students with low
socioeconomic status: The influence of the institutional financial context. Research in
Higher Education, 47(4), 371-398. doi:10.1007/s11162-005-9000-5
U.S. News & World Report (2012). Top 100 lowest acceptance rates. Retrieved from:
http://colleges.usnews.rankingsandreviews.com/best-colleges/rankings/lowest-
acceptance-rate/spp+50
Van der Klaauw, W. (2003). Estimating the Effect of Financial Aid Offers on College
Enrollment: A Regression–Discontinuity Approach. International Economic
Review, 43(4), 1249-1287. doi:10.1111/1468-2354.t01-1-00055
Walpole, M. B. (2003). Socioeconomic status and college: How SES affects college experiences
and outcomes. The Review of Higher Education, 27(1), 45-73.
doi:10.1353/rhe.2003.0044
Walters, G. (2012). It’s not so easy: The completion agenda and the states. Liberal Education,
98(1). Retrieved from: http://www.aacu.org/liberaleducation/le-wi12/walters.cfm
Webber, D. A., & Ehrenberg, R. G. (2010). Do expenditures other than instructional
expenditures affect graduation and persistence rates in American higher education?
Economics of Education Review, 29(6), 947-958. doi:10.1016/j.econedurev.2010.04.006
Wellman, J., Johnson, N., & Steele, P. (2012). Measuring (and Managing) the Invisible Costs of
Postsecondary Attrition. Policy Brief. Delta Cost Project at American Institutes for
Research. Retrieved from: http://files.eric.ed.gov/fulltext/ED536120.pdf
Wohlgemuth, D., Whalen, D., Sullivan, J., Nading, C., Shelley, M., & Wang, Y. (2007).
Financial, academic, and environmental influences on the retention and graduation of
IMPACT OF SOCIOECONOMIC STATUS 170
students. Journal of College Student Retention: Research, Theory and Practice, 8(4),
457-475. doi:10.2190/86X6-5VH8-3007-6918
Zwick, R., & Sklar, J. C. (2005). Predicting college grades and degree completion using high
school grades and SAT scores: The role of student ethnicity and first language. American
Educational Research Journal, 42(3), 439-464. doi:10.3102/00028312042003439

IMPACT OF SOCIOECONOMIC STATUS 171
Tables
Table 1
Prominent Studies and Their Connection to SES

Authors

Equity

Inst.Eff

PCV

DataCol

FinAid

Belong

Capital

Support
Astin (1997) Yes Yes Yes Both No No No No
Brewer et al.
(1999)
Yes Yes Yes Survey Yes No No No
Carnevale & Rose
(2003)
Yes No Yes Survey Yes No No No
Goldrick-Rab
(2006)
Yes No Yes Both No No No No
Hausmann et al.
(2007)
Yes No Yes Survey Yes Yes No No
Hill et al. (2005) No No No Actual Yes No No No
Johnson et al.
(2011)
Yes No No Survey No Yes No No
Langhout et al.
(2009)
No No No Survey No Yes Yes No
Ostrove & Long
(2007)
No No No Survey No Yes No No
Ryan (2004) Avg.'s Yes Avg.'s Actual No No No Unk
Walpole (2003) Yes No Yes Both No No Yes No
Proposed Study
(2014)
Yes Yes Yes Actual Yes *No *No Yes
Note. Inst.Eff = Institutional Efficiency. PCV =Pre-College variables. DataCol = Data
Collection methods. Belong = Sense of Belonging. Capital = Economic, Social, and/or Cultural
Capital. Support = Support Programs. Avg.'s = Cohort averages used as opposed to specifics for
individual students. Unk = Unknown because Student Support Programs were either combined
together with "Academic Support Services" which includes curriculum development and
libraries, or they were combined with "Student Services" offices such as Admission, Financial
Aid, and Student Affairs. *No = The study will not measure belonging or capital, but will be
influenced by these theories and will examine a support program that ideally helps promote
belonging and social/cultural capital.
IMPACT OF SOCIOECONOMIC STATUS 172
Table 2
Ranges of SES as Calculated by Institutional Expected Family Contribution (IEFC)

Range of IEFC
Lowest SES Less than 10% of the yearly full cost of attending
Low SES 10% to less than 30%
Middle SES 30% to less than 100%
Middle to upper SES 100% and greater
Upper to highest SES Did not apply for financial aid
Note: These ranges were decided on prior to the review of the data. These amounts do not
directly relate to the net cost of attending after aid has been applied because scholarships and
other grant-based aid may vary from student to student. An equal distribution was not expected.

IMPACT OF SOCIOECONOMIC STATUS 173
Table 3
Academic Preparedness Composite Score Calculation

Composite Score range
Lowest Less than .80
Low middle .80 to less than .90
Middle high .90 to less than 1.00
Highest Equal to or greater than 1.00
Note: These ranges were decided on prior to the review of the data and an equal distribution was
not expected. The HSGPA and adjusted HSGPA data were previously internally adjusted by the
office of admission so that the HSGPA is out of a total of 4.0 and the adjusted HSGPA has a
maximum of ~4.7. The composite score will be based on the following calculation:
((HSGPA/4)+(Adj. HSGPA/4)+(Test Score/Max))/3

IMPACT OF SOCIOECONOMIC STATUS 174
Table 4
Academic Preparedness Score Examples
HSGPA Adj. HSGPA Best SAT Academic
Preparedness score
2.80 3.10 2100 .7833
3.50 3.80 1900 .8722
3.40 3.40 2300 .8861
4.00 4.10 1860 .9333
3.80 4.10 2250 .9708
4.00 4.70 2400 1.058
Note: Students that were missing a score for their HSGPA, adjusted HSGPA, and/or best SAT
score were excluded from the full analysis of this study.

IMPACT OF SOCIOECONOMIC STATUS 175
Table 5
Degree Program Categories

Examples of Majors Included
Architecture* Architecture
Arts Animation, Cinema, Fine Arts, Music, Theatre
Business Accounting, Business
Communication Communication, Journalism, Public Relations
Engineering Aerospace, Astronautics, Biomedical, Chemical, Civil, Computer Science,
Electrical, Industrial
Humanities American Studies, Classics, Comparative Literature, East Asian Studies, English,
Health and Humanities, History, Philosophy
Natural Sciences Biochemistry, Biology, Chemistry, Environmental Studies, Global Health, Health
Promotion, Mathematics, Neuroscience, Physics
Social Sciences Anthropology, Economics, International Relations, Occupational Therapy,
Political Science, Psychology, Public Policy, Sociology
Undecided Undecided, Undeclared
Note: *Students in the Architecture degree program were removed from the study.

IMPACT OF SOCIOECONOMIC STATUS 176
Table 6
All Students by SES
2007 2008 2009
Entire Cohort 2965 2755 2862
- Lowest (<10%) 181 (6.7%) 232 (8.4%) 185 (6.5%)
- Low (10% to <30%) 346 (11.7%) 303 (11.0%) 323 (11.3%)
- Lower to Middle (30% to
<100%)
597 (20.1%) 611 (22.2%) 563 (19.7%)
- Middle to Upper (100% +) 585 (19.7%) 544 (19.7%) 662 (23.1%)
- Did Not Apply 1167 (39.4%) 976 (35.4%) 1089 (38.1%)
- Tuition Benefits 89 (3.0%) 89 (3.2%) 40 (1.4%)
Note: These totals were prior to the exclusion of any students.

IMPACT OF SOCIOECONOMIC STATUS 177
Table 7
Cohort Details
2007 2008 2009
Entire Cohort 2965 2755 2862
- Student Athletes 150 167 124
- International 202 155 320
- Architecture 109 91 91
- Total Exclusions 532 426 562
- Included in Study 2433 2329 2300
Note: After the exclusion of student athletes, international students, and architecture majors,
there were additional students excluded solely based on missing information such as HSGPA and
test scores. Some overlap between the Student Athletes, International Students, Architecture
majors, and other exclusions existed.

IMPACT OF SOCIOECONOMIC STATUS 178
Table 8
Students Examined for Study
2007 2008 2009
SES
- Lowest 153 (6.3%) 210 (9%) 171 (7.4%)
- Low 322 (13.2%) 275 (11.8%) 292 (12.7%)
- Lower to Middle 565 (23.2%) 566 (24.3%) 517 (22.5%)
- Middle to Upper 565 (23.2%) 497 (21.3%) 613 (26.7%)
- Did Not Apply 744 (30.6%) 695 (29.8%) 669 (29.1%)
- Tuition Benefits 84 (3.5%) 86 (3.5%) 38 (1.7%)
Academic Preparedness
- Lowest 36 (1.5%) 51 (2.2%) 27 (1.2%)
- Low-Middle 491 (20.2%) 481 (20.7%) 519 (22.6%)
- Middle-High 1754 (72.1%) 1603 (68.8%) 1568 (68.2%)
- Highest 152 (6.2%) 194 (8.3%) 186 (8.1%)
First Generation Status
- No 2194 (90.2%) 2061 (88.5%) 1999 (86.9%)
- Yes 239 (9.8%) 268 (11.5%) 301 (13.1%)
Support Program
Participation

- No 2336 (96%) 2221 (95.4%) 2200 (95.7%)
- Yes 97 (4.0%) 108 (4.6%) 100 (4.3%)
Note: The ranges for SES were established based on the institutional derived EFC. Academic
preparedness ranges were based on the composite scores of HSGPA, adjusted HSGPA, and test
scores.

IMPACT OF SOCIOECONOMIC STATUS 179
Table 9
Initial Data on Students in Study
2007 2008 2009
SES 2349 total 2243 total 2262 total
- Lowest 153 (6.5%) 210 (9.4%) 171 (7.6%)
- Low 322 (13.7%) 275 (12.3%) 292 (12.9%)
- Lower to Middle 565 (24.1%) 566 (25.2%) 517 (22.9%)
- Middle to Upper 565 (24.1%) 497 (22.2%) 613 (27.1%)
- Did Not Apply 744 (31.7%) 695 (31.0%) 669 (29.6%)
Academic Preparedness
- Lowest 33 (1.4%) 49 (2.2%) 26 (1.1%)
- Low-Middle 467 (19.9%) 464 (20.7%) 509 (22.5%)
- Middle-High 1701 (72.4%) 1546 (68.9%) 1542 (68.2%)
- Highest 148 (6.3%) 184 (8.2%) 185 (8.2%)
First Generation Status
- No 2112 (89.9%) 1976 (88.1%) 1965 (86.9%)
- Yes 237 (10.1%) 267 (11.9%) 297 (13.1%)
Support Program
Participation

- No 2256 (96%) 2138 (95.3%) 2163 (95.6%)
- Yes 93 (4.0%) 105 (4.7%) 99 (4.4%)
Note: The ranges for SES were established based on the institutional derived EFC. Academic
preparedness ranges were based on the composite scores of HSGPA, adjusted HSGPA, and test
scores as detailed in table three.

IMPACT OF SOCIOECONOMIC STATUS 180
Table 10
Additional Data on Students in Study
2007 2008 2009
Pursued Degree 2349 total 2243 total 2262 total
- Arts 370 (15.8%) 373 (16.6%) 395 (17.5%)
- Business 425 (18.1%) 423 (18.9%) 387 (17.1%)
- Communication 130 (5.5%) 149 (6.6%) 127 (5.6%)
- Engineering 360 (15.3%) 377 (16.8%) 380 (16.8%)
- Humanities 116 (4.9%) 77 (3.4%) 82 (3.6%)
- Natural Sciences 340 (14.5%) 306 (13.6%) 315 (13.9%)
- Social Sciences 239 (10.2%) 215 (9.6%) 226 (10.0%)
- Undecided 369 (15.7%) 323 (14.4%) 350 (15.5%)
Gender
- Female 1256 (53.5%) 1225 (54.6%) 1211 (53.5%)
- Male 1093 (46.5%) 1018 (45.4%) 1051 (46.5%)
Race/Ethnicity
- Hispanic or Latino 307 (13.1%) 316 (14.1%) 299 (13.2%)
- American Indian or
Alaskan Native
41 (1.7%) 42 (1.9%) 28 (1.2%)
- Asian 629 (26.8%) 660 (29.4%) 680 (30.1%)
- Black or African
American
119 (5.1%) 146 (6.5%) 168 (7.4%)
- White 1229 (52.3%) 1066 (47.5%) 1084 (47.9%)
- Unknown 24 (1.0%) 12 (0.5%) 3 (0.1%)
Mixed Race/Ethnicity 244 (10.4%) 243 (10.8%) 234 (10.3%)
Note: Regarding race/ethnicity, students that classified themselves as belonging to more than
racial/ethnic group were considered to belong to the group first listed within their admission
application. Students belonging to more than one racial/ethnic group were also included within
this table as “Mixed Race/Ethnicity” so their total numbers can be accounted for, however, this
study did not focus on that group. Native Hawaiians and Pacific Islanders were included in the
Asian population.

IMPACT OF SOCIOECONOMIC STATUS 181
Table 11
Outcomes for Students in Study
2007 2008 2009
Persistence
- 2
nd
semester
registration
99% 99% 99%
- 2
nd
year registration 97% 96% 97%
- 3
rd
year registration 94% 94% 95%
- 4
th
year registration 90% 89% 92%
Graduation
- 4 years 82% 80% 80%
Grade point Average
- 1rst Semester 3.27 3.28 3.29
- 1rst Year 3.29 3.29 3.30
- 2
nd
Year 3.32 3.32 3.32
- 4
th
Year 3.36 3.35 3.34
Units earned
-1rst Semester 16.6 16.6 16.6
-1rst Year 32.9 32.9 32.9
-2
nd
Year 64.4 64.1 64.6
-4
th
Year 119.5 118.8 120.3
-Total Units after 4 yrs 139.3 139.2 138.1
Note: A minimum of 128 units were required to graduate. Students were able to transfer in up
to 32 units of Advanced Placement (AP) and International Baccalaureate (IB) units.

IMPACT OF SOCIOECONOMIC STATUS 182
Table 12
Support Program Totals for SES, Preparedness, and First Generation Status
Support Program Participants Non-Participants
Total students 297 6557
SES
- Lowest 60 (20.2%) 474 (7.2%)
- Low 62 (20.9%) 827 (12.6%)
- Lower to Middle 67 (22.6%) 1581 (24.1%)
- Middle to Upper 38 (12.8%) 1637 (25.0%)
- Did Not Apply for
Aid
70 (23.6%) 2038 (31.1%)
Academic Preparedness
- Lowest 72 (24.2%) 36 (0.5%)
- Low-Middle 202 (68.0%) 1238 (18.9%)
- Middle-High 23 (7.7%) 4766 (72.7%)
- Highest 0 517 (7.9%)
First Generation Status
- No 204 (68.7%) 5849 (89.2%)
- Yes 93 (31.3%) 708 (10.8%)
Note: The majority of students that participated in the support program were expected to have
academic preparedness ranges of “Lowest” and “Low-Middle.”

IMPACT OF SOCIOECONOMIC STATUS 183
Table 13
Support Program Results for Degree Program Categories
Support Program Participants Non-Participants
Total students 297 6557
Gender
-Female 174 (58.6%) 3518 (53.7%)
-Male 123 (41.4%) 3039 (46.3%)
Race/Ethnicity
-Hispanic/Latino 82 (27.6%) 840 (12.8%)
-American Indian/
Alaskan Native
5 (1.7%) 106 (1.6%)
-Asian 38 (12.8%) 1931 (29.4%)
-Black/African
American
74 (24.9%) 359 (5.5%)
-White 96 (32.3%) 3283 (50.1%)
-Unknown 2 (0.7%) 38 (0.6%)
Degree Program Type
- Arts 114 (38.4%) 1024 (15.6%)
- Business 26 (8.8%) 1209 (18.4%)
- Communication 14 (4.7%) 392 (6.0%)
- Engineering 18 (6.1%) 1099 (16.8%)
- Humanities 8 (2.7%) 267 (4.1%)
- Natural Sciences 23 (7.7%) 938 (14.3%)
- Social Sciences 39 (13.1%) 641 (9.8%)
- Undecided 55 (18.5%) 987 (15.1%)
Note: Differences in the distribution by race/ethnicity and by major were expected. For
instance, the support program participant population was expected to have a higher percentage of
Black/African-American, Hispanic/Latino, and students pursuing majors in the arts.

IMPACT OF SOCIOECONOMIC STATUS 184
Table 14
Correlations for Independent and Dependent Variables
SES AP HSGPA Adj
HSGPA
Best
Test
1yr
GPA
4yr
GPA
SES X .069** -.004 .049** .111** .078** .074**
Academic
Preparedness (AP)
.069** X .826** .906** .673** .406** .410**
HSGPA -.004 .826** X .815** .201** .363** .376**
ADJ HSGPA .049** .906** .815** X .377** .376** .393**
Best Test score .111** .673** .201** .377** X .240** .221**
1rst yr GPA (1GPA) .078** .406** .363** .376** .240** X .845**
GPA after yr 4 .074** .410** .376** .393** .221** .845** X
Note: ** Correlation is significant at the 0.01 level (2-tailed). * Correlation is significant at the
0.05 level (2-tailed).

IMPACT OF SOCIOECONOMIC STATUS 185
Table 15
Correlations for Independent and Dependent Variables
SES AP 1GPA 1Units 4GPA 4Units 4TUnits
SES X .069** .078** .045** .074** .032* .049**
Academic
Preparedness (AP)
.069** X .406** .172** .410** .126** .297**
1rst yr GPA (1GPA) .078** .406** X .492** .845** .330** .385**
1rst yr Units (1Units) .045** .172** .492** X .440** .485** .481**
GPA after yr 4 .074** .410** .845** .440** X .417** .457**
Units after yr 4 .032* .126** .330** .485** .417** X .916**
Total units after yr 4 .049** .297** .385** .481** .457** .916** X
Note: ** Correlation is significant at the 0.01 level (2-tailed). * Correlation is significant at the
0.05 level (2-tailed). “1GPA” is equal to GPA after the summer of the first year. “1Units” is
equal to units completed after the summer of the first year. “4GPA” and “4Units” are equal to
the GPA and units earned at the completion of the summer of the fourth year. “4TUnits” is equal
to the total number of units counting toward graduation and also includes units taken elsewhere,
such as AP/IB and transfer units.

IMPACT OF SOCIOECONOMIC STATUS 186
Table 16
Comparison of Support Program Participant and Non-Participant Outcomes
AP 1GPA 1Units 1-2 4Grad 4GPA 4Units 4TUnits
Support Program
Participation (SPP)

- No .9386 3.31 32.9 97% 81% 3.37 119.7 139.5
- Yes .8282 2.98 31.9 97% 65% 3.00 116.4 124.2
SES
-Lowest (SPP no) .9279 3.21 32.4 97% 69% 3.25 118.8 138.1
-Lowest (SPP yes) .8488 2.92 32.4 100% 50% 2.89 115.0 123.1
- Low (SPP no) .9357 3.24 32.5 97% 78% 3.32 119.7 138.6
- Low (SPP yes) .8358 2.97 32.6 100% 68% 2.96 121.8 129.7
- Middle (SPP no) .9393 3.30 32.9 96% 82% 3.35 118.6 138.9
- Middle (SPP yes) .8320 3.05 30.8 91% 64% 3.10 112.9 122.4
- Mid-Upper (SPP
no)
.9457 3.37 33.2 97% 85% 3.42 120.6 141.7
- Mid-Upper (SPP
yes)
.8269 3.03 32.8 97% 79% 3.04 114.8 123.9
- Did Not Apply
(SPP no)
.9360 3.31 33.0 97% 82% 3.38 120.0 139.1
- Did Not Apply
(SPP yes)
.8008 2.93 31.3 96% 67% 3.02 117.2 122.3
Note: “SPP no” and “SPP yes” are in regards to whether the student participated in the support
program. Four year graduation, “4Grad”, was focused on within this study due to the assumption
that most students and families plan on graduating within four years. 1GPA = grade point
average after the first year, 1Units = units earned after the first year, 1-2 = persistence to the
second year, 4GPA = grade point average after the fourth year, 4Units = units earned after the
fourth year, 4TUnits = total units earned after the fourth year.

IMPACT OF SOCIOECONOMIC STATUS 187
Table 17
Comparison of Similar Preparedness Groups (AP .80 to <.90)
AP EFC 1GPA 1Units 1-2 4Grad 4GPA 4Units 4TUnits
SPP Yes (.80-.90 AP)
– 202 students
.8448 X 2.97 31.7 97% 66% 3.02 116.4 125.0
+SES (Lowest) – 49 .8488 5.1% 2.88 32.0 100% 51% 2.88 113.4 121.4
+SES (Low) – 41 .8445 18.1% 2.92 32.0 100% 71% 2.97 121.8 129.8
+SES (Lower-Middle)
– 55
.8460 54.7% 3.06 30.9 89% 67% 3.15 113.7 123.6
+SES (Mid-Upper) –
25
.8425 242% 3.02 32.2 100% 80% 3.05 113.7 123.7
+SES Did not Apply –
32
.8386 N/A 2.96 31.8 97% 69% 3.07 120.7 128.0
SPP No (.80-.90) –
1238 students
.8699 X 3.04 32.2 96% 72% 3.12 115.9 129.7
+SES (Lowest) –106 .8695 5.8% 2.92 31.8 97% 58% 2.99 116.8 130.4
+SES (Low) – 146 .8699 18.2% 2.95 31.9 98% 69% 3.06 118.1 131.2
+SES (Lower-Middle)
– 294
.8675 56.8% 3.01 32.0 95% 73% 3.08 113.6 127.8
+SES (Mid-Upper) –
263
.8709 257% 3.06 32.3 97% 75% 3.13 115.9 131.1
+SES Did not Apply –
429
.8710 N/A 3.11 32.4 95% 75% 3.19 116.4 129.4
SPP No (.80-.868) –
487 students
.8448 X 2.97 31.9 95% 71% 3.06 115.0 127.1
+SES (Lowest) –44 .8473 5.8% 2.91 32.3 95% 52% 3.02 114.7 127.5
+SES (Low) – 57 .8444 18.5% 2.80 31.7 98% 67% 2.95 117.7 130.0
+SES (Lower-Middle)
– 128
.8432 58.6% 2.92 31.5 95% 74% 3.01 114.5 125.8
+SES (Mid-Upper) –
98
.8453 257% 3.00 32.2 96% 76% 3.06 116.0 129.6
+SES Did not Apply –
160
.8451 N/A 3.07 32.0 94% 72% 3.14 113.9 125.4
Note: SPP no = non-participant in the support program, SPP yes = participant, AP = academic
preparedness, EFC = expected family contribution as a percentage of the full cost of attending,
1GPA = grade point average after the first year, 1Units = units earned after the first year, 1-2 =
persistence to the second year, 4GPA = grade point average after the fourth year, 4Units = units
earned after the fourth year, 4TUnits = total units earned after the fourth year.

IMPACT OF SOCIOECONOMIC STATUS 188
Table 18
Comparison Totals for SES, Gender, and First Generation Status
Support Program
Participants (.80-.90)
Non-Participants(AP
.80-.90)
Non-Participants (AP
.80-.868)
Total students 202 1238 487
SES
- Lowest 49 (24.3%) 106 (8.6%) 44 (9.0%)
- Low 41 (20.3%) 146 (11.8%) 57 (11.7%)
- Lower to Middle 55 (27.2%) 294 (23.7%) 128 (26.3%)
- Middle to Upper 25 (12.4%) 263 (21.2%) 98 (20.1%)
- Did Not Apply 32 (15.8%) 429 (34.7%) 160 (32.9%)
Academic Preparedness .8448 avg. .8699 avg. .8448 avg.
Gender
-Female 122 (60.4%) 619 (50%) 245 (50.3%)
-Male 80 (39.6%) 619 (50%) 242 (49.7%)
First Generation Status 77 (38.1%) 156 (12.6%) 52 (10.7%)
Note: AP = academic preparedness, .80-.90 = academic preparedness score of .80 to less than
.90, 80-.868 = academic preparedness score of .80 to less than .868

IMPACT OF SOCIOECONOMIC STATUS 189
Table 19
Comparison Results for Race/Ethnicity and Degree Program Categories
Support Program
Participants (.80-.90)
Non-Participants(AP
.80-.90)
Non-Participants (AP
.80-.868)
Total students 202 1238 487
Academic Preparedness .8448 avg. .8699 avg. .8448 avg.
Race/Ethnicity
-Hispanic/Latino 65 (32.2%) 237 (19.1%) 85 (17.5%)
-American Indian/
Alaskan Native
5 (2.5%) 34 (2.7%) 8 (1.6%)
-Asian 26 (12.9%) 231 (18.7%) 76 (15.6%)
-Black/African
American
54 (32.2%) 182 (14.7%) 106 (21.8%)
-White 51 (25.2%) 544 (43.9%) 206 (42.3%)
-Unknown 1 (0.5%) 10 (0.8%) 6 (1.2%)
Mixed Race/Ethnicity 28 (13.9%) 160 (12.9%) 57 (11.7%)
Degree Program Type
- Arts 61 (30.2%) 360 (29.1%) 173 (35.5%)
- Business 18 (8.9%) 150 (12.1%) 48 (9.9%)
- Communication 12 (5.9%) 88 (7.1%) 29 (6.0%)
- Engineering 11 (5.4%) 193 (15.6%) 66 (13.6%)
- Humanities 7 (3.5%) 50 (4.0%) 19 (3.9%)
- Natural Sciences 20 (9.9%) 102 (8.2%) 41 (8.4%)
- Social Sciences 26 (12.9%) 134 (10.8%) 42 (8.6%)
- Undecided 47 (23.3%) 161 (13.0%) 69 (14.2%)
Note: AP = academic preparedness, .80-.90 = academic preparedness score of .80 to less than
.90, 80-.868 = academic preparedness score of .80 to less than .868

IMPACT OF SOCIOECONOMIC STATUS 190
Table 20
Composition for Participants by SES for Additional Variables (AP .80-.90)
SES for SPP with (AP
.80-.90)
Lowest Low Lower to
Middle
Middle to
Upper
Did not Apply
Total students 49 41 55 25 32
Academic Preparedness .8488 avg. .8445 avg. .8460 avg. .8425 avg. .8386 avg.
Gender
-Female 26 (53.1%) 25 (61.0%) 37 (67.3%) 15 (60.0%) 19 (59.4%)
-Male 23 (46.9%) 18 (43.9%) 18 (32.7%) 10 (40.0%) 13 (40.6%)
First Generation Status 39 (79.6%) 18 (43.9%) 11 (20.0%) 6 (24.0%) 3 (9.4%)
Race/Ethnicity
-Hispanic/Latino 30 (61.2%) 9 (22.0%) 17 (30.9%) 7 (28.0%) 2 (6.3%)
-American Indian/
Alaskan Native
0 2 (4.9%) 1 (1.8%) 0 2 (6.3%)
-Asian 2 (4.1%) 8 (19.5%) 7 (12.7%) 6 (24.0%) 3 (9.4%)
-Black/African
American
15 (30.6%) 16 (39.0% 14 (25.5%) 4 (16.0%) 5 (15.6%)
-White 2 (4.1%) 6 (14.6%) 15 (27.3%) 8 (32.0%) 20 (62.5%)
-Unknown 0 0 1 (1.8%) 0 0
Mixed Race/Ethnicity 4 (8.2%) 6 (14.6%) 10 (18.2%) 4 (16.0%) 4 (12.5%)
Degree Program Type
-Arts 6 (12.2%) 6 (24.4%) 24 (43.6%) 11 (44.0%) 10 (31.3%)
-Business 4 (8.2%) 6 (14.6%) 0 5 (20.0%) 3 (9.4%)
-Communication 4 (8.2%) 3 (7.3%) 1 (1.8%) 2 (8.0%) 2 (6.3%)
-Engineering 2 (4.1%) 6 (14.6%) 1 (1.8%) 0 2 (6.3%)
-Humanities 3 (6.1%) 0 1 (1.8%) 0 3 (9.4%)
-Natural Sciences 4 (8.2%) 4 (9.8%) 10 (18.2%) 1 (4.0%) 1 (3.1%)
-Social Sciences 8 (16.3%) 5 (12.2%) 5 (9.1%) 4 (16.0%) 4 (12.5%)
-Undecided 18 (36.7%) 7 (17.1%) 13 (23.6%) 2 (8.0%) 7 (21.9%)
Note: AP = academic preparedness, .80-.90 = academic preparedness score of .80 to less than
.90

IMPACT OF SOCIOECONOMIC STATUS 191
Table 21
Composition for Non-Participants by SES for Additional Variables (AP .80-.90)
SES for non-SPP with
(AP .80-.90)
Lowest Low Lower to
Middle
Middle to
Upper
Did not Apply
Total students 106 146 294 263 429
Academic Preparedness .8695 avg. .8699 avg. .8675 avg. .8709 avg. .8710 avg.
Gender
-Female 62 (58.5%) 79 (54.1%) 156 (53.1%) 121 (46.0%) 201 (46.9%)
-Male 44 (41.5%) 67 (45.9%) 138 (46.9%) 142 (54.0%) 228 (53.1%)
First Generation Status 51 (48.1%) 39 (26.7%) 33 (11.2%) 17 (6.5%) 16 (3.7%)
Race/Ethnicity
-Hispanic/Latino 45 (42.5%) 31 (21.2%) 66 (22.4%) 48 (18.3%) 47 (11.0%)
-American Indian/
Alaskan Native
2 (1.9%) 3 (2.1%) 15 (5.1%) 7 (2.7%) 7 (1.8%)
-Asian 26 (24.5%) 35 (24.0%) 46 (15.6%) 47 (17.9%) 77 (17.9%)
-Black/African
American
26 (24.5%) 39 (26.7%) 68 (23.1%) 29 (11.0%) 20 (4.7%)
-White 7 (6.6%) 37 (25.3%) 96 (32.7%) 130 (49.4%) 274 (63.9%)
-Unknown 0 1 (0.7%) 3 (1.0%) 2 (0.8%) 4 (0.9%)
Mixed Race/Ethnicity 12 (11.3%) 18 (12.3%) 55 (18.7%) 37 (14.1%) 38 (8.9%)
Degree Program Type
-Arts 15 (14.2%) 38 (26.0%) 92 (31.3%) 91 (34.6%) 124 (28.9%)
-Business 7 (6.6%) 15 (10.3%) 29 (9.9%) 34 (12.9%) 65 (15.2%)
-Communication 8 (7.5%) 6 (4.1%) 27 (9.2%) 14 (5.3%) 33 (7.7%)
-Engineering 15 (14.2%) 21 (14.4%) 45 (15.3%) 43 (16.3%) 69 (16.1%)
-Humanities 5 (4.7%) 3 (2.1%) 14 (4.8%) 7 (2.7%) 21 (4.9%)
-Natural Sciences 9 (8.5%) 24 (16.4%) 28 (9.5%) 22 (8.4%) 19 (4.4%)
-Social Sciences 19 (17.9%) 20 (13.7%) 32 (10.9%) 23 (8.7%) 40 (9.3%)
-Undecided 28 (26.4%) 19 (13.0%) 27 (9.2%) 29 (11.0%) 58 (13.5%)
Note: AP = academic preparedness, .80-.90 = academic preparedness score of .80 to less than
.90

IMPACT OF SOCIOECONOMIC STATUS 192
Table 22
Composition for Non-Participants by SES for Additional Variables (AP.80-.868)
SES for SPP with
(AP.80-.868)
Lowest Low Lower to
Middle
Middle to
Upper
Did not Apply
Total students 44 57 128 98 160
Academic Preparedness .8473 avg. .8444 avg. .8432 avg. .8453 avg. .8451 avg.
Gender
-Female 27 (61.4%) 30 (52.6%) 67 (52.3%) 38 (38.8%) 83 (51.9%)
-Male 17 (38.6%) 27 (47.4%) 61 (47.7%) 60 (61.2%) 77 (48.1%)
First Generation Status 23 (52.3%) 11 (19.3%) 13 (10.2%) 3 (3.1%) 2 (1.3%)
Race/Ethnicity
-Hispanic/Latino 22 (50%) 8 (14.0%) 27 (21.1%) 13 (13.3%) 15 (9.4%)
-American Indian/
Alaskan Native
1 (2.3%) 1 (1.8%) 4 (3.1%) 0 2 (1.3%)
-Asian 9 (20.5%) 10 (17.5%) 15 (11.7%) 14 (14.3%) 28 (17.5%)
-Black/African
American
9 (20.5%) 24 (42.1%) 47 (36.7%) 16 (16.3%) 10 (6.3%)
-White 3 (6.8%) 13 (22.8%) 33 (25.8%) 55 (56.1%) 102 (63.8%)
-Unknown 0 1 (1.8%) 2 (1.6%) 0 3 (1.9%)
Mixed Race/Ethnicity 5 (11.4%) 9 (15.8%) 21 (16.4%) 12 (12.2%) 10 (6.3%)
Degree Program Type
-Arts 7 (15.9%) 16 (28.15) 47 (36.7%) 41 (41.8%) 62 (38.8%)
-Business 1 (2.3%) 4 (7.0%) 16 (12.5%) 10 (10.2%) 17 (10.6%)
-Communication 5 (11.4%) 1 (1.8%) 10 (7.8%) 3 (3.1%) 10 (6.3%)
-Engineering 6 (13.6%) 8 (14.0%) 15 (11.7%) 16 (16.3%) 21 (13.1%)
-Humanities 2 (4.5%) 2 (3.5%) 5 (3.9%) 3 (3.1%) 7 (4.4%)
-Natural Sciences 4 (9.1%) 13 (22.8%) 10 (7.8%) 6 (6.1%) 8 (5.0%)
-Social Sciences 5 (11.4%) 5 (8.8%) 12 (9.4%) 7 (7.1%) 13 (8.1%)
-Undecided 14 (31.8%) 8 (14.0%) 13 (10.2%) 12 (12.2%) 22 (13.8%)
Note: AP = academic preparedness, .80-.868 = academic preparedness score of .80 to less than
.868

IMPACT OF SOCIOECONOMIC STATUS 193
Table 23
Lower 50
th
Percentile Preparedness Comparison (AP less than .83139)
AP EFC 1GPA 1Units 1-2 4Grad 4GPA 4Units 4TUnits
SPP Yes (<.83139
AP) – 149 students
.7868 X 2.86 30.7 94% 56% 2.88 111.3 117.8
+SES (Lowest) –19 .8007 4.7% 2.58 28.1 100% 32% 2.51 98.2 105.2
+SES (Low) –29 .7850 18.6% 2.92 32.3 100% 59% 2.85 120.2 127.0
+SES (Lower-Middle)
– 31
.8003 53.2% 2.90 29.1 81% 42% 2.90 100.6 109.8
+SES (Mid-Upper) –
22
.7962 203% 3.06 32.8 95% 77% 3.05 114.9 123.8
+SES Did not Apply –
48
.7694 N/A 2.82 30.9 96% 65% 2.95 116.3 119.8
SPP No (<.83139) –
153 students
.8106 X 2.85 31.2 93% 59% 2.94 110.4 117.7
+SES (Lowest) –11 .8078 5.0% 2.86 34.9 100% 27% 2.89 125.1 130.6
+SES (Low) – 21 .8086 16.6% 2.49 31.4 100% 62% 2.78 113.2 122.9
+SES (Lower-Middle)
– 43
.8134 56.2% 2.89 30.9 91% 63% 2.93 110.9 116.9
+SES (Mid-Upper) –
25
.8136 200% 2.92 30.6 88% 52% 2.95 99.1 107.2
+SES Did not Apply –
53
.8082 N/A 2.93 30.9 92% 64% 3.01 111.3 118.6
SPP No (<.803 AP) –
43 students
.7872 X 2.75 31.0 95% 49% 2.80 108.0 113.2
+SES (Lowest) –4 .7874 5.3% 2.89 35.3 100% 25% 2.92 129.8 135.5
+SES (Low) – 7 .7877 16.6% 2.45 32.4 100% 57% 2.57 109.4 116.0
+SES (Lower-Middle)
– 9
.7858 49.4% 2.96 31.2 100% 56% 2.95 118.7 125.3
+SES (Mid-Upper) – 5 .7945 261% 2.58 30.2 100% 40% 2.65 95.2 97.8
+SES Did not Apply –
18
.7856 N/A 2.77 29.6 89% 50% 2.82 100.9 105.3
Note: SPP no = non-participant in the support program, SPP yes = participant, AP = academic
preparedness, EFC = expected family contribution as a percentage of the full cost of attending,
1GPA = grade point average after the first year, 1Units = units earned after the first year, 1-2 =
persistence to the second year, 4GPA = grade point average after the fourth year, 4Units = units
earned after the fourth year, 4TUnits = total units earned after the fourth year.

IMPACT OF SOCIOECONOMIC STATUS 194
Table 24
Comparison totals for SES, Gender, and First Generation Status (AP < .83139)
Support Program
Participants
(AP<.83139)
Non-Participants
(AP<.83139)
Non-Participants (AP <
.803)
Total students 149 153 43
SES
- Lowest 19 (12.8%) 11 (7.2%) 4 (9.3%)
- Low 29 (19.5%) 21 (13.7%) 7 (16.3%)
- Lower to Middle 31 (20.8%) 43 (28.1%) 9 (20.9%)
- Middle to Upper 22 (14.8%) 25 (16.3%) 5 (11.6%)
- Did Not Apply 48 (32.2%) 53 (34.6%) 18 (41.9%)
Academic Preparedness .7868 avg. .8106 avg. .7872 avg.
Gender
-Female 73 (49.0%) 73 (47.7%) 20 (46.5%)
-Male 76 (51.0%) 80 (52.3%) 23 (53.5%)
First Generation Status 30 (20.1%) 17 (11.1%) 5 (11.6%)
Note: AP = academic preparedness, AP<.83139 = academic preparedness score less than .83139,
AP < .803 = academic preparedness score less than .803

IMPACT OF SOCIOECONOMIC STATUS 195
Table 25
Comparison Totals for Race/Ethnicity and Degree Program Categories (AP < .83139)
Support Program
Participants
(AP<.83139)
Non-Participants
(AP<.83139)
Non-Participants (AP <
.803)
Total students 149 153 43
Race/Ethnicity
-Hispanic/Latino 32 (21.5%) 20 (13.1%) 4 (9.3%)
-American Indian/
Alaskan Native
2 (1.3%) 4 (2.6%) 0
-Asian 17 (11.4%) 11 (7.2%) 3 (7.0%)
-Black/African
American
34 (22.8%) 46 (30.1%) 12 (27.9%)
-White 62 (41.6%) 70 (45.8%) 24 (55.8%)
-Unknown 2 (1.3%) 2 (1.3%) 0
Mixed Race/Ethnicity 13 (8.7%) 14 (9.2%) 4 (9.3%)
Degree Program Type
- Arts 80 (53.7%) 62 (40.5%) 17 (39.5%)
- Business 9 (6.0%) 8 (5.2%) 3 (7.0%)
- Communication 8 (5.4%) 8 (5.2%) 2 (4.7%)
- Engineering 3 (2.0%) 18 (11.8%) 4 (9.3%)
- Humanities 3 (2.0%) 7 (4.6%) 2 (4.7%)
- Natural Sciences 10 (6.7%) 11 (7.2%) 2 (4.7%)
- Social Sciences 16 (10.7%) 16 (10.5%) 5 (11.6%)
- Undecided 20 (13.4%) 23 (15.0%) 8 (18.6%)
Note: AP = academic preparedness, AP<.83139 = academic preparedness score less than .83139,
AP < .803 = academic preparedness score less than .803

IMPACT OF SOCIOECONOMIC STATUS 196
Table 26
Composition for Participants by SES for Additional Variables (AP<.83139)
SES for SPP with
(AP<.83139)
Lowest Low Lower to
Middle
Middle to
Upper
Did not Apply
Total students 19 29 31 22 48
Academic Preparedness .8007 avg. .7850 avg. .8003 avg. .7962 avg. .7694 avg.
Gender
-Female 7 (36.8%) 14 (48.3%) 18 (58.1%) 12 (54.5%) 22 (45.8%)
-Male 12 (63.2%) 15 (51.7%) 13 (41.9%) 10 (45.5%) 26 (54.2%)
First Generation Status 13 (68.4%) 12 (41.4%) 2 (6.5%) 2 (9.1%) 1 (2.1%)
Race/Ethnicity
-Hispanic/Latino 12 (63.2%) 8 (27.6%) 7 (22.6%) 2 (9.1%) 3 (6.3%)
-American Indian/
Alaskan Native
0 1 (3.4%) 1 (3.2%) 0 0
-Asian 1 (5.3%) 1 (3.4%) 3 (9.7%) 9 (40.9%) 3 (6.3%)
-Black/African
American
6 (31.6%) 14 (48.3%) 8 (25.8%) 1 (4.5%) 5 (10.4%)
-White 0 5 (17.2%) 11 (35.5%) 10 (45.5%) 36 (75.0%)
-Unknown 0 0 1 (3.2%) 0 1 (2.1%)
Mixed Race/Ethnicity 2 (10.5%) 3 (10.3%) 3 (9.7%) 2 (9.1%) 3 (6.3%)
Degree Program Type
-Arts 4 (21.1%) 10 (34.5%) 21 (67.7%) 17 (77.3%) 28 (58.3%)
-Business 2 (10.5%) 4 (13.8%) 1 (3.2%) 2 (9.1%) 0
-Communication 2 (10.5%) 2 (6.9%) 1 (3.2%) 1 (4.5%) 2 (4.2%)
-Engineering 1 (5.3%) 1 (3.4%) 0 0 1 (2.1%)
-Humanities 1 (5.3%) 0 0 0 2 (4.2%)
-Natural Sciences 3 (15.8%) 3 (10.3%) 3 (9.7%) 0 1 (2.1%)
-Social Sciences 1 (5.3%) 2 (6.9%) 1 (3.2%) 2 (9.1%) 10 (20.8%)
-Undecided 5 (26.3%) 7 (24.1%) 4 (12.9%) 0 4 (8.3%)
Note: AP = academic preparedness, AP<.83139 = academic preparedness score less than .83139

IMPACT OF SOCIOECONOMIC STATUS 197
Table 27
Composition for Non-Participants by SES for Additional Variables (AP<.83139)
SES for non-SPP with
(AP <.83139)
Lowest Low Lower to
Middle
Middle to
Upper
Did not Apply
Total students 11 21 43 25 53
Academic Preparedness .8078 avg. .8086 avg. .8134 avg. .8136 avg. .8082 avg.
Gender
-Female 7 (63.6%) 12 (57.1%) 24 (55.8%) 6 (24.0%) 24 (45.3%)
-Male 4 (36.4%) 9 (42.9%) 19 (44.2%) 19 (76.0%) 29 (54.7%)
First Generation Status 6 (54.5%) 4 (19.0%) 4 (9.3%) 1 (4.0%) 2 (3.8%)
Race/Ethnicity
-Hispanic/Latino 6 (54.5%) 4 (19.0%) 5 (11.6%) 3 (12.0%) 2 (3.8%)
-American Indian/
Alaskan Native
1 (9.1%) 0 3 (7.0%) 0 0
-Asian 1 (9.1%) 1 (4.8%) 3 (7.0%) 1 (4.0%) 5 (9.4%)
-Black/African
American
3 (27.3%) 14 (66.7%) 23 (53.5%) 5 (20.0%) 1 (1.9%)
-White 0 2 (9.5%) 8 (18.6%) 16 (64.0%) 44 (83.0%)
-Unknown 0 0 1 (2.3%) 0 1 (1.9%)
Mixed Race/Ethnicity 1 (9.1%) 3 (14.3%) 6 (14.0%) 3 (12.0%) 1 (1.9%)
Degree Program Type
-Arts 2 (18.1%) 1 (4.8%) 16 (37.2%) 16 (64.0%) 27 (50.9%)
-Business 0 1 (4.8%) 6 (14.0%) 0 1 (1.9%)
-Communication 0 2 (9.5%) 4 (9.3%) 0 2 (3.8%)
-Engineering 0 3 (14.3%) 3 (7.0%) 6 (24.0%) 6 (11.3%)
-Humanities 1 (9.1%) 2 (9.5%) 1 (2.3%) 1 (4.0%) 2 (3.8%)
-Natural Sciences 0 4 (19.0%) 4 (9.3%) 0 3 (5.7%)
-Social Sciences 2 (18.2%) 2 (9.5%) 6 (14.0%) 0 6 (11.3%)
-Undecided 6 (54.5%) 6 (28.6%) 3 (7.0%) 2 (8.0%) 6 (11.3%)
Note: AP = academic preparedness, AP<.83139 = academic preparedness score less than .83139

IMPACT OF SOCIOECONOMIC STATUS 198
Table 28
Composition for Non-Participants by SES for Additional Variables (AP < .803)
SES for SPP with (AP <
.803)
Lowest Low Lower to
Middle
Middle to
Upper
Did not Apply
Total students 4 7 9 5 18
Academic Preparedness .7874 avg. .7877 avg. .7858 avg. .7945 avg. .7856 avg.
Gender
-Female 2 (50.0%) 5 (71.4%) 6 (66.7%) 3 (60.0%) 4 (22.2%)
-Male 2 (50.0%) 2 (28.6%) 3 (33.3%) 2 (40.0%) 14 (77.8%)
First Generation Status 3 (75.0%) 1 (14.3%) 0 0 1 (5.6%)
Race/Ethnicity
-Hispanic/Latino 3 (75.0%) 1 (14.3%) 0 0 0
-American Indian/
Alaskan Native
0 0 0 0 0
-Asian 0 0 2 (22.2%) 0 1 (5.6%)
-Black/African
American
1 (25.0%) 6 (85.7%) 4 (44.4%) 1 (20.0%) 0
-White 0 0 3 (33.3%) 4 (80.0%) 17 (94.4%)
-Unknown 0 0 0 0 0
Mixed Race/Ethnicity 0 0 4 (44.4%) 0 0
Degree Program Type
-Arts 1 (25.0%) 1 (14.3%) 3 (33.3%) 3 (60.0%) 9 (50.0%)
-Business 0 1 (14.3%) 1 (11.1%) 0 1 (5.6%)
-Communication 0 1 (14.3%) 1 (11.1%) 0 0
-Engineering 0 0 1 (11.1%) 2 (40.0%) 1 (5.6%)
-Humanities 1 (25.0%) 1 (14.3%) 0 0 0
-Natural Sciences 0 0 0 0 2 (11.1%)
-Social Sciences 0 0 2 (22.2%) 0 3 (16.7%)
-Undecided 2 (50.0%) 3 (42.9%) 1 (11.1%) 0 2 (11.1%)
Note: AP = academic preparedness, AP < .803 = academic preparedness score less than .803

IMPACT OF SOCIOECONOMIC STATUS 199
Table 29
25
th
to 75
th
Percentile Preparedness Comparison (AP .81039 to <.8664)
AP EFC 1GPA 1Units 1-2 4Grad 4GPA 4Units 4TUnits
SPP Yes (.80139-
.8664 AP) – 151
students
.8343 X 2.94 31.5 96% 64% 3.00 117.0 125.1
+SES (Lowest) –34 .8369 5.3% 2.79 31.6 100% 47% 2.81 112.7 119.7
+SES (Low) –28 .8313 17.7% 2.91 30.8 100% 71% 2.95 122.3 130.8
+SES (Lower-Middle)
–45
.8379 55.3% 3.05 30.9 87% 62% 3.11 112.9 121.9
+SES (Mid-Upper) –
18
.8274 216% 3.00 32.1 100% 78% 3.04 115.6 126.4
+SES Did not Apply –
26
.8324 N/A 2.92 32.5 100% 69% 3.05 123.2 130.8
SPP No (.80139-
.8664) –456 students
.8437 X 2.98 32.0 95% 72% 3.07 115.4 127.2
+SES (Lowest) –41 .8458 5.6% 2.89 32.6 95% 54% 3.01 115.6 127.6
+SES (Low) – 53 .8440 18.3% 2.79 31.6 98% 68% 2.95 117.2 129.2
+SES (Lower-Middle)
– 118
.8417 58.7% 2.95 31.6 95% 75% 3.03 115.1 126.1
+SES (Mid-Upper) –
89
.8431 264% 3.02 32.5 96% 76% 3.09 116.6 129.5
+SES Did not Apply –
155
.8448 N/A 3.07 32.1 95% 73% 3.14 114.4 125.8
SPP No (.80139-.854)
– 289 students
.8342 X 2.98 32.1 96% 71% 3.07 116.1 126.7
+SES (Lowest) –26 .8379 5.4% 2.95 33.8 96% 58% 3.11 120.4 130.1
+SES (Low) – 30 .8318 18.0% 2.64 30.8 100% 60% 2.88 114.9 125.2
+SES (Lower-Middle)
– 82
.8336 56.4% 2.95 31.7 95% 77% 3.03 117.6 127.3
+SES (Mid-Upper) –
57
.8338 273% 3.04 32.3 95% 74% 3.09 115.1 127.1
+SES Did not Apply –
94
.8347 N/A 3.09 32.1 96% 70% 3.16 114.7 125.5
Note: SPP no = non-participant in the support program, SPP yes = participant, AP = academic
preparedness, EFC = expected family contribution as a percentage of the full cost of attending,
1GPA = grade point average after the first year, 1Units = units earned after the first year, 1-2 =
persistence to the second year, 4GPA = grade point average after the fourth year, 4Units = units
earned after the fourth year, 4TUnits = total units earned after the fourth year.

IMPACT OF SOCIOECONOMIC STATUS 200
Table 30
Comparison Totals for SES, Gender, and First Generation Status (AP .80139-.8664)
Support Program
Participants (AP .80139-
.8664)
Non-Participants (AP
.80139-.8664)
Non-Participants (AP
.80139-.854)
Total students 151 456 289
SES
- Lowest 34 (22.5%) 41 (9.0%) 26 (9.0%)
- Low 28 (18.5%) 53 (11.6%) 30 (10.4%)
- Lower to Middle 45 (29.8%) 118 (25.9%) 82 (28.4%)
- Middle to Upper 18 (11.9%) 89 (19.5%) 57 (19.7%)
- Did Not Apply 26 (17.2%) 155 (34.0% 94 (32.5%)
Academic Preparedness .8343 avg. .8437 avg. .8342 avg.
Gender
-Female 87 (57.6%) 229 (50.2%) 151 (52.2%)
-Male 64 (42.4%) 227 (49.8%) 138 (47.8%)
First Generation Status 46 (30.5%) 47 (10.3%) 26 (9.0%)
Note: AP = academic preparedness, AP. 80139-.8664 = academic preparedness score of .80139
to less than .83139, AP .80139-.854 = academic preparedness score of .80139 to less than .854

IMPACT OF SOCIOECONOMIC STATUS 201
Table 31
Comparison Results for Race/Ethnicity and Degree Program Categories (AP .80139-.8664)
Support Program
Participants (AP .80139-
.8664)
Non-Participants (AP
.80139-.8664)
Non-Participants (AP
.80139-.854)
Total students 151 456 289
Race/Ethnicity
-Hispanic/Latino 44 (29.1%) 76 (16.7%) 45 (15.6%)
-American Indian/
Alaskan Native
3 (2.0%) 8 (1.8%) 6 (2.1%)
-Asian 19 (12.6%) 70 (15.4%) 34 (11.8%)
-Black/African
American
42 (27.8%) 101 (22.1%) 75 (26.0%)
-White 42 (27.8%) 195 (42.8%) 125 (43.3%)
-Unknown 1 (0.7%) 6 (1.3%) 4 (1.4%)
Mixed Race/Ethnicity 17 (11.3%) 53 (11.6%) 35 (12.1%)
Degree Program Type
- Arts 50 (33.1%) 164 (36%) 112 (38.8%)
- Business 13 (8.6%) 44 (9.6%) 19 (6.6%)
- Communication 8 (5.3%) 28 (6.1%) 21 (7.3%)
- Engineering 4 (2.6%) 63 (13.8%) 37 (12.8%)
- Humanities 4 (2.6%) 18 (3.9%) 13 (4.5%)
- Natural Sciences 14 (9.3%) 35 (7.7%) 19 (6.6%)
- Social Sciences 22 (14.6%) 41 (9.0%) 27 (9.3%)
- Undecided 36 (23.8%) 63 (13.8%) 41 (14.2%)
Note: AP = academic preparedness, AP. 80139-.8664 = academic preparedness score of .80139
to less than .83139, AP .80139-.854 = academic preparedness score of .80139 to less than .854

IMPACT OF SOCIOECONOMIC STATUS 202
Table 32
Composition for Participants by SES for Additional Variables (AP .80139-.8664)
SES for SPP with
(AP.80139-.8664 )
Lowest Low Lower to
Middle
Middle to
Upper
Did not Apply
Total students 34 28 45 18 26
Academic Preparedness .8369 avg. .8313 avg. .8379 avg. .8274 avg. .8324 avg.
Gender
-Female 17 (50.0%) 17 (60.7%) 27 (60.0%) 10 (55.6%) 16 (61.5%)
-Male 17 (50.0%) 11 (39.3%) 18 (40.0%) 8 (44.4%) 10 (38.5%)
First Generation Status 25 (73.5%) 10 (35.7%) 7 (15.6%) 2 (11.1%) 2 (7.7%)
Race/Ethnicity
-Hispanic/Latino 19 (55.9%) 7 (25.0%) 13 (28.9%) 3 (16.7%) 2 (7.7%)
-American Indian/
Alaskan Native
0 2 (7.1%) 1 (2.2%) 0 0
-Asian 1 (2.9%) 3 (10.7%) 6 (13.3%) 6 (33.3%) 3 (11.5%)
-Black/African
American
12 (35.3%) 12 (42.9%) 10 (22.2%) 4 (22.2%) 4 (15.4%)
-White 2 (5.9%) 4 (14.3%) 14 (31.1%) 5 (27.8%) 17 (65.4%)
-Unknown 0 0 1 (2.2%) 0 0
Mixed Race/Ethnicity 4 (11.8%) 3 (10.7%) 5 (11.1%) 3 (16.7%) 2 (7.7%)
Degree Program Type
-Arts 4 (11.8%) 9 (32.1%) 22 (48.9%) 7 (38.9%) 8 (30.8%)
-Business 3 (8.8%) 5 (17.9%) 0 3 (16.7%) 2 (7.7%)
-Communication 3 (8.8%) 1 (3.6%) 1 (2.2%) 2 (11.1%) 1 (3.8%)
-Engineering 1 (2.9%) 1 (3.6%) 0 0 2 (7.7%)
-Humanities 1 (2.9%) 0 1 (2.2%) 0 2 (7.7%)
-Natural Sciences 3 (8.8%) 2 (7.1%) 7 (15.6%) 1 (5.6%) 1 (3.8%)
-Social Sciences 6 (17.6%) 4 (14.3%) 4 (8.9%) 4 (22.2%) 4 (15.4%)
-Undecided 13 (38.2%) 6 (21.4%) 10 (22.2%) 1 (5.6%) 6 (23.1%)
Note: AP = academic preparedness, AP. 80139-.8664 = academic preparedness score of .80139
to less than .83139

IMPACT OF SOCIOECONOMIC STATUS 203
Table 33
Composition for Non-Participants by SES for Additional Variables (AP .80139-.8664)
SES for non-SPP with
(AP.80139-.8664 )
Lowest Low Lower to
Middle
Middle to
Upper
Did not Apply
Total students 41 53 118 89 155
Academic Preparedness .8458 avg. .8440 avg. .8417 avg. .8431 avg. .8448 avg.
Gender
-Female 26 (63.4%) 27 (50.9%) 60 (50.8%) 35 (39.3%) 81 (52.3%)
-Male 15 (36.6% 26 (49.1%) 58 (49.2%) 54 (60.7%) 74 (47.7%)
First Generation Status 22 (53.7%) 10 (18.9%) 11 (9.3%) 2 (2.2%) 2 (1.3%)
Race/Ethnicity
-Hispanic/Latino 20 (48.8%) 7 (13.2%) 23 (19.5%) 12 (13.5%) 14 (9.0%)
-American Indian/
Alaskan Native
1 (2.4%) 1 (1.9%) 4 (3.4%) 0 2 (1.3%)
-Asian 8 (19.5%) 9 (17.0%) 14 (11.9%) 12 (13.5%) 27 (17.4%)
-Black/African
American
9 (22.2%) 23 (43.4%) 43 (36.4%) 16 (18.0%) 10 (6.5%)
-White 3 (7.3%) 12 (22.6%) 32 (27.15) 49 (55.1%) 99 (63.9%)
-Unknown 0 1 (1.9%) 2 (1.7%) 0 3 (1.9%)
Mixed Race/Ethnicity 5 (12.2%) 8 (15.1%) 19 (16.1%) 11 (12.4%) 10 (6.5%)
Degree Program Type
-Arts 6 (14.6%) 14 (26.4%) 44 (37.3%) 39 (43.8%) 61 (39.4%)
-Business 1 (2.4%) 4 (7.5%) 15 (12.7%) 9 (10.1%) 15 (9.7%)
-Communication 5 (12.2%) 1 (1.9%) 9 (7.6%) 3 (3.4%) 10 (6.5%)
-Engineering 6 (14.6%) 8 (15.1%) 15 (12.7%) 14 (15.7%) 20 (12.9%)
-Humanities 2 (4.9%) 2 (3.8%) 4 (3.4%) 3 (3.4%) 7 (4.5%)
-Natural Sciences 3 (7.3%) 13 (24.5%) 7 (5.9%) 5 (5.6%) 7 (4.5%)
-Social Sciences 5 (12.2%) 5 (9.4%) 12 (10.2%) 6 (6.7%) 13 (8.4%)
-Undecided 13 (31.7%) 6 (11.3%) 12 (10.2%) 10 (11.2%) 22 (14.2%)
Note: AP = academic preparedness, AP. 80139-.8664 = academic preparedness score of .80139
to less than .83139

IMPACT OF SOCIOECONOMIC STATUS 204
Table 34
Composition for Non-Participants by SES for Additional Variables (AP .80139-.854)
SES for Non-SPP (AP
.80139-.854)
Lowest Low Lower to
Middle
Middle to
Upper
Did not Apply
Total students 26 30 82 57 94
Academic Preparedness .8379 avg. .8318 avg. .8336 avg. .8338 avg. .8347
Gender
-Female 20 (76.9%) 14 (46.7%) 41 (50.0%) 23 (40.4%) 53 (56.4%)
-Male 6 (23.1%) 16 (53.3%) 41 (50.0%) 34 (59.6%) 41 (43.6%)
First Generation Status 13 (50.0%) 5 (16.7%) 5 (6.1%) 1 (1.8%) 2 (2.1%)
Race/Ethnicity
-Hispanic/Latino 12 (46.2%) 4 (13.3%) 15 (18.3%) 5 (8.8%) 9 (9.6%)
-American Indian/
Alaskan Native
1 (3.8%) 1 (3.3%) 3 (3.7%) 0 1 (1.1%)
-Asian 3 (11.5%) 4 (13.3%) 7 (8.5%) 7 (12.3%) 13 (13.8%)
-Black/African
American
9 (34.6%) 16 (53.3%) 33 (40.2%) 13 (22.8%) 4 (4.3%)
-White 1 (3.8%) 5 (16.7%) 22 (26.8%) 32 (56.1%) 65 (69.1%)
-Unknown 0 0 2 (2.4%) 0 2 (2.1%)
Mixed Race/Ethnicity 5 (19.2%) 5 (16.7%) 13 (15.9%) 9 (15.8%) 3 (3.2%)
Degree Program Type
-Arts 4 (15.4%) 5 (16.7%) 30 (36.6%) 30 (52.6%) 43 (45.7%)
-Business 1 (3.8%) 1 (3.3%) 10 (12.2%) 3 (5.3%) 4 (4.3%)
-Communication 4 (15.4%) 1 (3.3%) 7 (8.5%) 2 (3.5%) 7 (7.4%)
-Engineering 0 5 (16.7%) 11 (13.4%) 10 (17.5%) 11 (11.7%)
-Humanities 2 (7.7%) 1 (3.3%) 2 (2.4%) 2 (3.5%) 6 (6.4%)
-Natural Sciences 2 (7.7%) 7 (23.3%) 5 (6.1%) 2 (3.5%) 3 (3.2%)
-Social Sciences 3 (11.5%) 5 (16.7%) 9 (11.0%) 4 (7.0%) 6 (6.4%)
-Undecided 10 (38.5%) 5 (16.7%) 8 (9.8%) 4 (7.0%) 14 (14.9%)
Note: AP = academic preparedness, AP .80139-.854 = academic preparedness score of .80139
to less than .854

IMPACT OF SOCIOECONOMIC STATUS 205
Table 35
50
th
to 100
th
Percentile Preparedness Comparison (AP .83139-.936)
AP EFC 1GPA 1Units 1-2 4Grad 4GPA 4Units 4TUnits
SPP Yes (.83139-.936
AP) – 147 students
.8694 X 3.09 33.0 99% 73% 3.12 121.7 130.7
+SES (Lowest) –41 .8711 5.2% 3.08 34.4 100% 59% 3.06 122.7 131.3
+SES (Low) – 32 .8787 18.0% 3.00 32.9 100% 75% 3.04 123.8 132.2
+SES (Lower-Middle)
– 36
.8594 56.6% 3.18 32.4 100% 83% 3.27 123.5 133.4
+SES (Mid-Upper) –
16
.8690 248% 2.99 32.8 100% 81% 3.01 114.8 124.1
+SES Did not Apply –
22
.8695 N/A 3.15 32.0 95% 73% 3.17 119.0 127.8
SPP No (.83139-.936)
– 2744 students
.9013 X 3.15 32.4 96% 77% 3.22 117.4 133.9
+SES (Lowest) –240 .8997 5.8% 3.07 32.0 97% 66% 3.15 117.7 134.0
+SES (Low) – 362 .9039 18.4% 3.09 32.2 97% 71% 3.17 118.1 134.1
+SES (Lower-Middle)
– 646
.9016 58.3% 3.14 32.5 96% 79% 3.20 116.8 134.2
+SES (Mid-Upper) –
614
.9019 246% 3.18 32.7 97% 82% 3.25 118.4 135.6
+SES Did not Apply –
882
.9000 N/A 3.17 32.4 96% 78% 3.26 116.7 132.5
SPP No (.83139-
.892)- 882 students
.8696 X 3.04 32.3 96% 72% 3.11 116.1 130.2
+SES (Lowest) –84 .8688 6.0% 2.94 31.6 98% 57% 3.01 116.1 130.3
+SES (Low) – 101 .8708 18.7% 3.00 32.0 97% 68% 3.08 118.2 131.5
+SES (Lower-Middle)
– 202
.8686 57.1% 3.00 32.2 96% 73% 3.08 114.1 129.0
+SES (Mid-Upper) –
194
.8700 273% 3.07 32.7 98% 77% 3.13 118.0 133.4
+SES Did not Apply –
301
.8698 N/A 3.09 32.3 94% 73% 3.16 115.6 128.4
Note: SPP no = non-participant in the support program, SPP yes = participant, AP = academic
preparedness, EFC = expected family contribution as a percentage of the full cost of attending,
1GPA = grade point average after the first year, 1Units = units earned after the first year, 1-2 =
persistence to the second year, 4GPA = grade point average after the fourth year, 4Units = units
earned after the fourth year, 4TUnits = total units earned after the fourth year.

IMPACT OF SOCIOECONOMIC STATUS 206
Table 36
Comparison Totals for SES, Gender, and First Generation Status (AP .83139-.936)
Support Program
Participants (.83139-
.936)
Non-Participants(AP
.83139-.936)
Non-Participants (AP
.83139-.892)
Total students 147 2744 882
SES
- Lowest 41 (27.9%) 240 (8.7%) 84 (9.5%)
- Low 32 (21.8%) 362 (13.2%) 101 (11.5%)
- Lower to Middle 36 (24.5%) 646 (23.5%) 202 (22.9%)
- Middle to Upper 16 10.9%) 614 (22.4%) 194 (22.0%)
- Did Not Apply 22 (15.0%) 882 (32.1%) 301 (34.1%)
Academic Preparedness .8694 avg. .9013 avg. .8696 avg.
Gender
-Female 100 (68.0%) 1423 (51.9%) 444 (50.3%)
-Male 47 (32.0%) 1321 (48.1%) 438 (49.7%)
First Generation Status 63 (42.9%) 393 (14.3%) 111 (12.6%)
Note: AP = academic preparedness, AP. 83139-.936 = academic preparedness score of .83139 to
less than . 936, AP .83139-.892 = academic preparedness score of .80139 to less than .892

IMPACT OF SOCIOECONOMIC STATUS 207
Table 37
Comparison Results for Race/Ethnicity and Degree Program Categories (AP .83139-.936)
Support Program
Participants (.83139-
.936)
Non-Participants(AP
.83139-.936)
Non-Participants (AP
.83139-.892)
Total students 147 2744 882
Race/Ethnicity
-Hispanic/Latino 50 (34.0%) 522 (19.0%) 171 (19.4%)
-American Indian/
Alaskan Native
3 (2.0%) 55 (2.0%) 22 (2.5%)
-Asian 20 (13.6%) 660 (24.1%) 170 (19.3%)
-Black/African
American
40 (27.2%) 247 (9.0%) 125 (14.2%)
-White 34 (23.1%) 1235 (45.0%) 387 (43.9%)
-Unknown 0 25 (0.9%) 7 (0.8%)
Mixed Race/Ethnicity 21 (14.3%) 352 (12.8%) 108 (12.2%)
Degree Program Type
- Arts 34 (23.1%) 571 (20.8%) 259 (29.4%)
- Business 17 (11.6%) 441 (16.1%) 106 (12.0%)
- Communication 6 (4.1%) 186 (6.8%) 64 (7.3%)
- Engineering 15 (10.2%) 435 (15.9%) 137 (15.5%)
- Humanities 5 (3.4%) 112 (4.1%) 32 (3.6%)
- Natural Sciences 12 (8.2%) 295 (10.8%) 75 (8.5%)
- Social Sciences 23 (15.6%) 304 (11.1%) 91 (10.3%)
- Undecided 35 (23.8%) 400 (14.6%) 118 (13.4%)
Note: AP = academic preparedness, AP. 83139-.936 = academic preparedness score of .83139 to
less than . 936, AP .83139-.892 = academic preparedness score of .80139 to less than .892

IMPACT OF SOCIOECONOMIC STATUS 208
Table 38
Composition for Participants by SES for Additional Variables (AP .83139-.936)
SES for SPP with
(AP.83139-.936)
Lowest Low Lower to
Middle
Middle to
Upper
Did not Apply
Total students 41 32 36 16 22
Academic Preparedness .8711 avg. .8787 avg. .8594 avg. .8690 avg. .8695 avg.
Gender
-Female 25 (61.0%) 22 (68.8%) 28 (77.8%) 10 (62.5%) 15 (68.2%)
-Male 16 (39.0%) 10 (31.3%) 8 (22.2%) 6 (37.5%) 7 (31.8%)
First Generation Status 31 (75.6%) 17 (53.1%) 9 (25.0%) 4 (25.0%) 2 (9.1%)
Race/Ethnicity
-Hispanic/Latino 21 (51.2%) 9 (28.1%) 13 (36.1%) 5 (31.3%) 2 (9.1%)
-American Indian/
Alaskan Native
0 1 (3.1%) 0 0 2 (9.1%)
-Asian 2 (4.9%) 8 (25.0%) 6 (16.7%) 2 (12.5%) 2 (9.1%)
-Black/African
American
16 (39.0%) 10 (31.3%) 9 (25.0%) 3 (18.8%) 2 (9.1%)
-White 2 (4.9%) 4 (12.5%) 8 (22.2%) 6 (37.5%) 14 (63.6%)
-Unknown 0 0 0 0 0
Mixed Race/Ethnicity 4 (9.8%) 3 (9.4%) 8 (22.2%) 3 (18.8%) 3 (13.6%)
Degree Program Type
-Arts 6 (14.6%) 6 (18.8%) 13 (36.1%) 5 (31.3%) 4 (18.2%)
-Business 5 (12.2%) 5 (15.6%) 0 3 (18.8%) 4 (18.2%)
-Communication 2 (4.9%) 1 (3.1%) 1 (2.8%) 1 (6.3%) 1 (4.5%)
-Engineering 3 (7.3%) 7 (21.9%) 1 (2.8%) 1 (6.3%) 3 (13.6%)
-Humanities 2 (4.9%) 0 1 (2.8%) 0 2 (9.1%)
-Natural Sciences 1 (2.4%) 3 (9.4%) 7 (19.4%) 1 (6.3%) 0
-Social Sciences 9 (22.0%) 6 (18.8%) 4 (11.1%) 2 (12.5%) 2 (9.1%)
-Undecided 12 (31.7%) 4 (12.5%) 9 (25.0%) 3 (18.8%) 6 (27.3%)
Note: AP = academic preparedness, AP. 83139-.936 = academic preparedness score of .83139 to
less than . 936

IMPACT OF SOCIOECONOMIC STATUS 209
Table 39
Composition for Non-Participants by SES for Additional Variables (AP .83139-.936)
SES for non-SPP with
(AP.83139-.936)
Lowest Low Lower to
Middle
Middle to
Upper
Did not Apply
Total students 240 362 646 614 882
Academic Preparedness .8997 avg. .9039 avg. .9016 avg. .9019 avg. .9000 avg.
Gender
-Female 142 (59.2%) 205 (56.6%) 356 (55.1%) 296 (48.2%) 424 (48.1%)
-Male 98 (40.8%) 157 (43.4%) 290 (44.9%) 318 (51.8%) 458 (51.9%)
First Generation Status 128 (53.3%) 105 (29.0%) 82 (12.7%) 39 (6.4%) 39 (4.4%)
Race/Ethnicity
-Hispanic/Latino 93 (38.8%) 91 (25.1%) 142 (22.0%) 103 (16.8%) 93 (10.5%)
-American Indian/
Alaskan Native
4 (1.7%) 10 (2.8%) 18 (2.8%) 10 (1.6%) 13 (1.5%)
-Asian 67 (27.9%) 104 (28.7%) 162 (25.1%) 143 (23.3%) 184 (20.9%)
-Black/African
American
44 (18.3%) 53 (14.6%) 86 (13.3%) 39 (6.4%) 25 (2.8%)
-White 31 (12.9%) 102 (28.2%) 232 (35.9%) 315 (51.3%) 555 (62.9%)
-Unknown 1 (0.4%) 2 (0.6%) 6 (0.9%) 4 (0.7%) 12 (1.4%)
Mixed Race/Ethnicity 28 (11.7%) 52 (14.4%) 109 (16.9%) 86 (14.0%) 77 (8.7%)
Degree Program Type
-Arts 21 (8.8%) 69 (19.1%) 155 (24.0%) 150 (24.4%) 176 (20.0%)
-Business 35 (14.6%) 55 (15.2%) 76 (11.8%) 99 (16.1%) 176 (20.0%)
-Communication 14 (5.8%) 22 (6.1%) 46 (7.1%) 43 (7.0%) 61 (6.9%)
-Engineering 34 (14.2%) 50 (13.8%) 107 (16.6%) 107 (17.4%) 137 (15.5%)
-Humanities 9 (3.8%) 14 (3.9%) 33 (5.1%) 17 (2.8%0 39 (4.4%)
-Natural Sciences 29 (12.1%) 50 (13.8%) 84 (13.0%) 65 (10.6%) 67 (7.6%)
-Social Sciences 40 (16.7%) 42 (11.6%) 66 (10.2%) 66 (10.7%) 90 (10.2%)
-Undecided 58 (24.2%) 60 (16.6%) 79 (12.2%) 67 (10.9%) 136 (15.4%)
Note: AP = academic preparedness, AP. 83139-.936 = academic preparedness score of .83139 to
less than . 936

IMPACT OF SOCIOECONOMIC STATUS 210
Table 40
Composition for Non-Participants by SES for Additional Variables (AP .83139-.892)
SES for non-SPP with
(AP.83139-.892)
Lowest Low Lower to
Middle
Middle to
Upper
Did not Apply
Total students 84 101 202 194 301
Academic Preparedness .8688 avg. .8708 avg. .8686 avg. .8700 avg. .8698 avg.
Gender
-Female 47 (56.0%) 55 (54.5%) 105 (52.0%) 91 (46.9%) 146 (48.5%)
-Male 37 (44.0%) 46 (45.5%) 97 (48.0%) 103 (53.1%) 155 (51.5%)
First Generation Status 40 (47.6%) 27 (26.7%) 24 (11.9%) 11 (5.7%) 9 (3.0%)
Race/Ethnicity
-Hispanic/Latino 37 (44.0%) 19 (18.8%) 49 (24.3%) 33 (17.0%) 33 (11.0%)
-American Indian/
Alaskan Native
0 3 (3.0%) 9 (4.5%) 5 (2.6%) 5 (1.7%)
-Asian 20 (23.8%) 26 (25.7%) 36 (17.8%) 35 (18.0%) 53 (17.6%)
-Black/African
American
22 (26.2%) 25 (24.8%) 40 (19.8%) 24 (12.4%) 14 (4.7%)
-White 5 (6.0%) 27 (26.7%) 66 (32.7%) 96 (49.5%) 193 (64.1%)
-Unknown 0 1 (1.0%) 2 (1.0%) 1 (0.5%) 3 (1.0%)
Mixed Race/Ethnicity 7 (8.3%) 13 (12.9%) 38 (18.8%) 26 (13.4%) 24 (8.0%)
Degree Program Type
-Arts 12 (14.3%) 29 (28.7%) 64 (31.75) 63 (32.5%) 91 (30.2%)
-Business 6 (7.1%) 13 (12.9%) 18 (8.9%) 26 (13.4%) 43 (14.3%)
-Communication 8 (9.5%) 3 (3.0%) 14 (6.9%) 12 (6.2%) 27 (9.0%)
-Engineering 13 (15.5%) 11 (10.9%) 37 (18.3%) 31 (16.0%) 45 (15.0%)
-Humanities 4 (4.8%) 2 (2.0%) 10 (5.0%) 6 (3.1%) 10 (3.3%)
-Natural Sciences 7 (8.3%) 18 (17.8%) 18 (8.9%) 16 (8.2%) 16 (5.3%)
-Social Sciences 11 (13.1%) 12 (11.9%) 19 (9.4%) 19 (9.8%) 30 (10.0%)
-Undecided 23 (27.4%) 13 (12.9%) 22 (10.9%) 21 (10.8%) 39 (13.0%)
Note: AP = academic preparedness, AP .83139-.892 = academic preparedness score of .80139
to less than .892

IMPACT OF SOCIOECONOMIC STATUS 211
Table 41
Comparison of Reasonably Similar Groups
AP EFC 1GPA 1Units 1-2 4Grad 4GPA 4Units 4TUnits
Lowest SES
+SPP (AP .80-.90) –
49 students
.8488 5.1% 2.88 32.0 100% 51% 2.88 113.4 121.4
+non-SPP (AP .80-
.868) – 44 students
.8473 5.8% 2.91 32.3 95% 52% 3.02 114.7 127.5
+SPP (AP .83139-
.936) – 41 students
.8711 5.2% 3.08 34.4 100% 59% 3.06 122.7 131.3
+non-SPP (AP .83139-
.892) – 84 students
.8688 6.0% 2.94 31.6 98% 57% 3.01 116.1 130.3
Low SES
+SPP (AP .80-.90) –
41 students
.8445 18.1% 2.92 32.0 100% 71% 2.97 121.8 129.8
+non-SPP (AP. 80-
.868) – 57 students
.8444 18.5% 2.80 31.7 98% 67% 2.95 117.7 130.0
SES Did not Apply
+SPP (AP .80-.90) –
32 students
.8386 N/A 2.96 31.8 97% 69% 3.07 120.7 128.0
+non-SPP (.80-.868) –
160 students
.8451 N/A 3.07 32.0 94% 72% 3.14 113.9 125.4
+SPP (AP <.83139) –
48 students
.7694 N/A 2.82 30.9 96% 65% 2.95 116.3 119.8
+SPP (AP <.803) – 18
students
.7856 N/A 2.77 29.6 89% 50% 2.82 100.9 105.3
+SPP (AP <.83139) –
53 students
.8082 N/A 2.93 30.9 92% 64% 3.01 111.3 118.6
+SPP (AP .80139-
.8664) – 26 students
.8324 N/A 2.92 32.5 100% 69% 3.05 123.2 130.8
+non-SPP (AP .80139-
.8664) – 155students
.8448 N/A 3.07 32.1 95% 73% 3.14 114.4 125.8
Note: SPP no = non-participant in the support program, SPP yes = participant, AP = academic
preparedness, EFC = expected family contribution as a percentage of the full cost of attending,
1GPA = grade point average after the first year, 1Units = units earned after the first year, 1-2 =
persistence to the second year, 4GPA = grade point average after the fourth year, 4Units = units
earned after the fourth year, 4TUnits = total units earned after the fourth year.

IMPACT OF SOCIOECONOMIC STATUS 212
Table 42
Research Question Two Composition Totals by SES
Non-SPP students Lowest Low Lower to
Middle
Middle to
Upper
Did not Apply
Total students 474 827 1581 1637 2038
EFC percentage 6.0% 18.7% 59.1% 241% X
EFC+Aid percentage 90.4% 91.5% 98.7% 261% X
Academic Preparedness .9279 avg. .9357 avg. .9393 avg. .9457 avg. .9360 avg.
-Lowest 4 (0.8%) 5 (0.6%) 6 (0.4%) 5 (0.3%) 16 (0.8%)
-Low-Middle 106 (22.4%) 146 (17.7%) 294 (18.6%) 263 (16.1%) 429 (21.1%)
-Middle-High 352 (74.3%) 640 (77.4%) 1155 (73.1%) 1175 (71.8%) 1444 (70.9%)
-Highest 12 (2.5%) 36 (4.4%) 126 (8.0%) 194 (11.9%) 149 (7.3%)
Gender
-Female 294 (62.0%) 482 (58.3%) 856 (54.1%) 829 (50.6%) 1057 (51.9%)
-Male 180 (38.0%) 345 (41.7%) 725 (45.9%) 808 (49.4%) 981 (48.1%)
First Generation Status 221 (46.6%) 210 (25.4%) 145 (9.2%) 62 (3.8%) 70 (3.4%)
Race/Ethnicity
-Hispanic/Latino 137 (28.9%) 143 (17.3%) 224 (14.2%) 182 (11.1%) 154 (7.6%)
-American Indian/
Alaskan Native
11 (2.3%) 17 (2.1%) 32 (2.0%) 22 (1.3%) 24 (1.2%)
-Asian 185 (39.0%) 289 (34.9%) 507 (32.1%) 470 (28.7%) 480 (23.6%)
-Black/African
American
56 (11.8%) 81 (9.8%) 135 (8.5%) 57 (3.5%) 30 (1.5%)
-White 82 (17.3%) 294 (35.6%) 676 (42.8%) 900 (55.0%) 1331 (65.3%)
-Unknown 2 (0.4%) 3 (0.4%) 7 (0.4%) 6 (0.4%) 19 (0.9%)
Mixed Race/Ethnicity 52 (11.0%) 97 (11.7%) 208 (13.2%) 184 (11.2%) 146 (7.2%)
Degree Program Type
-Arts 46 (9.7%) 114 (13.8%) 280 (17.7%) 285 (17.4%) 299 (14.7%)
-Business 79 (16.7%) 122 (14.8%) 236 (14.9%) 304 (18.6%) 468 (23.0%)
-Communication 19 (4.0%) 48 (5.8%) 108 (6.8%) 87 (5.3%) 130 (6.4%)
-Engineering 66 (13.9%) 138 (16.7%) 290 (18.3%) 311 (19.0%) 294 (14.4%)
-Humanities 18 (3.8%) 36 (4.4%) 69 (4.4%) 53 (3.2%) 91 (4.5%)
-Natural Sciences 75 (15.8%) 150 (18.1%) 244 (15.4%) 238 (14.5%) 231 (11.3%)
-Social Sciences 70 (14.8%) 87 (10.5%) 153 (9.7%) 136 (8.3%) 195 (9.6%)
-Undecided 101 (21.3%) 132 (16.0%) 201 (12.7%) 223 (13.6%) 330 (16.2%)
Note: EFC+Aid percentage = EFC plus merit and need aid divided by full cost of attending
IMPACT OF SOCIOECONOMIC STATUS 213
Table 43
Outcome Totals by SES
AP EFC 1GPA 1Units 1-2 4Grad 4GPA 4Units 4TUnits
All (6557 students) .9386 X 3.31 32.9 97% 81% 3.37 119.7 139.5
+SES (Lowest) –474 .9279 6.0% 3.21 32.4 97% 69% 3.25 118.8 138.1
+SES (Low) – 827 .9357 18.7% 3.24 32.5 97% 78% 3.32 119.7 138.6
+SES (Lower-Middle)
– 1581
.9393 59.1% 3.30 32.9 96% 82% 3.35 118.6 138.9
+SES (Mid-Upper) –
1637
.9457 241% 3.37 33.2 97% 85% 3.42 120.6 141.7
+SES Did not Apply –
2038
.9360 N/A 3.31 33.0 97% 82% 3.38 120.0 139.1
AP lower 50
th

percentile (<.94278) –
3287 students
.9016 X 3.15 32.4 96% 77% 3.23 117.3 133.8
+SES (Lowest) –287 .9012 5.7% 3.08 32.2 97% 66% 3.16 118.3 134.9
+SES (Low) – 433 .9033 18.6% 3.08 32.1 97% 71% 3.16 117.8 134.0
+SES (Lower-Middle)
– 780
.9012 58.3% 3.15 32.5 96% 79% 3.21 116.5 133.4
+SES (Mid-Upper) –
726
.9033 246% 3.18 32.7 97% 82% 3.26 117.8 135.1
+SES Did not Apply –
1061
.9001 N/A 3.17 32.4 96% 78% 3.26 117.1 132.9
AP upper 50
th

percentile (>=.94278)
– 3270 students
.9758 X 3.47 33.5 97% 86% 3.51 122.1 145.3
+SES (Lowest) –187 .9688 6.3% 3.40 32.6 97% 75% 3.40 119.4 143.1
+SES (Low) –394 .9714 18.8% 3.41 33.0 98% 85% 3.48 121.8 143.6
+SES (Lower-Middle)
– 801
.9764 59.9% 3.45 33.4 96% 85% 3.49 120.6 144.2
+SES (Mid-Upper) –
911
.9796 238% 3.51 33.7 97% 88% 3.55 122.9 146.9
+SES Did not Apply –
977
.9750 N/A 3.47 33.6 98% 87% 3.51 123.1 145.8
Note: AP = academic preparedness, EFC = expected family contribution as a percentage of the
full cost of attending, 1GPA = grade point average after the first year, 1Units = units earned after
the first year, 1-2 = persistence to the second year, 4GPA = grade point average after the fourth
year, 4Units = units earned after the fourth year, 4TUnits = total units earned after the fourth
year.
IMPACT OF SOCIOECONOMIC STATUS 214
Table 44
Outcomes for AP ranges by SES
AP EFC 1GPA 1Units 1-2 4Grad 4GPA 4Units 4TUnits
AP Low-Middle (.80-
.90) – 1238 students
.8699 X 3.04 32.2 96% 72% 3.12 115.9 129.7
+SES (Lowest) –106 .8695 5.8% 2.92 31.8 97% 58% 2.99 116.8 130.4
+SES (Low) –146 .8699 18.2% 2.95 31.9 98% 69% 3.06 118.1 131.2
+SES (Lower-Middle)
– 294
.8675 56.8% 3.01 32.0 95% 73% 3.08 113.6 127.8
+SES (Mid-Upper) –
263
.8709 257% 3.06 32.3 97% 75% 3.13 115.9 131.1
+SES Did not Apply –
429
.8710 N/A 3.11 32.4 95% 75% 3.19 116.4 129.4
AP Middle-High (.90-
1.00) – 4766 students
.9494 X 3.34 33.0 97% 83% 3.40 120.1 140.7
+SES (Lowest) –352 .9442 6.0% 3.28 32.5 97% 74% 3.33 119.0 139.8
+SES (Low) – 640 .9477 18.9% 3.29 32.7 97% 79% 3.36 120.1 140.0
+SES (Lower-Middle)
– 1155
.9502 59.3% 3.34 33.1 96% 83% 3.39 119.3 140.4
+SES (Mid-Upper) –
1175
.9519 241% 3.38 32.3 97% 87% 3.44 120.5 141.5
+SES Did not Apply –
1444
.9489 N/A 3.35 33.1 97% 84% 3.42 120.6 140.8
AP Highest (1.00+) –
517 students
1.014 X 3.67 34.3 98% 90% 3.68 126.5 154.5
+SES (Lowest) –12 1.011 7.1% 3.57 34.2 92% 58% 3.61 125.8 158.3
+SES (Low) – 36 1.012 17.4% 3.57 33.1 97% 86% 3.61 121.8 147.7
+SES (Lower-Middle)
– 126
1.015 62.8% 3.66 34.3 97% 90% 3.68 123.6 151.3
+SES (Mid-Upper) –
194
1.014 221% 3.72 34.3 98% 92% 3.71 128.8 157.9
+SES Did not Apply –
149
1.014 N/A 3.65 34.6 98% 91% 3.65 127.0 154.0
Note: Only 36 students had academic preparedness scores below .80. SPP no = non-participant
in the support program, SPP yes = participant, AP = academic preparedness, EFC = expected
family contribution as a percentage of the full cost of attending, 1GPA = grade point average
after the first year, 1Units = units earned after the first year, 1-2 = persistence to the second year,
4GPA = grade point average after the fourth year, 4Units = units earned after the fourth year,
4TUnits = total units earned after the fourth year.

IMPACT OF SOCIOECONOMIC STATUS 215
Table 45
Composition Totals by SES for AP .80 < .90
Non-SPP students with
AP of .80 to <.90
Lowest Low Lower to
Middle
Middle to
Upper
Did not Apply
Total students 106 146 294 263 429
EFC Percentage 5.8% 18.2% 56.8% 257% X
EFC+Aid Percentage 88.8% 87.8% 90.8% 263% X
Academic Preparedness .8695 avg. .8699 avg. .8675 avg. .8709 avg. .8710 avg.
Gender
-Female 62 (58.5%) 79 (54.1%) 156 (53.1%) 121 (46.0%) 201 (46.9%)
-Male 44 (41.5%) 67 (45.9%) 138 (46.9%) 142 (54.0%) 228 (53.1%)
First Generation Status 51 (48.1%) 39 (26.7%) 33 (11.2%) 17 (6.5%) 16 (3.7%)
Race/Ethnicity
-Hispanic/Latino 45 (42.5%) 31 (21.2%) 66 (22.4%) 48 (18.3%) 47 (11.0%)
-American Indian/
Alaskan Native
2 (1.9%) 3 (2.1%) 15 (5.1%) 7 (2.7%) 7 (1.8%)
-Asian 26 (24.5%) 35 (24.0%) 46 (15.6%) 47 (17.9%) 77 (17.9%)
-Black/African
American
26 (24.5%) 39 (26.7%) 68 (23.1%) 29 (11.0%) 20 (4.7%)
-White 7 (6.6%) 37 (25.3%) 96 (32.7%) 130 (49.4%) 274 (63.9%)
-Unknown 0 1 (0.7%) 3 (1.0%) 2 (0.8%) 4 (0.9%)
Mixed Race/Ethnicity 12 (11.3%) 18 (12.3%) 55 (18.7%) 37 (14.1%) 38 (8.9%)
Degree Program Type
-Arts 15 (14.2%) 38 (26.0%) 92 (31.3%) 91 (34.6%) 124 (28.9%)
-Business 7 (6.6%) 15 (10.3%) 29 (9.9%) 34 (12.9%) 65 (15.2%)
-Communication 8 (7.5%) 6 (4.1%) 27 (9.2%) 14 (5.3%) 33 (7.7%)
-Engineering 15 (14.2%) 21 (14.4%) 45 (15.3%) 43 (16.3%) 69 (16.1%)
-Humanities 5 (4.7%) 3 (2.1%) 14 (4.8%) 7 (2.7%) 21 (4.9%)
-Natural Sciences 9 (8.5%) 24 (16.4%) 28 (9.5%) 22 (8.4%) 19 (4.4%)
-Social Sciences 19 (17.9%) 20 (13.7%) 32 (10.9%) 23 (8.7%) 40 (9.3%)
-Undecided 28 (26.4%) 19 (13.0%) 27 (9.2%) 29 (11.0%) 58 (13.5%)
Note: AP = academic preparedness, AP .80 < .90 = academic preparedness score ranges from
.80 to less than .90

IMPACT OF SOCIOECONOMIC STATUS 216
Table 46
Composition Totals by SES for AP .90 < 1.0
Non-SPP students with
AP of .90 to < 1.0
Lowest Low Lower to
Middle
Middle to
Upper
Did not Apply
Total students 352 640 1155 1175 1444
EFC Percentage 6.0% 18.9% 59.3% 241% X
EFC+Aid Percentage 90.4% 91.9% 98.3% 258% X
Academic Preparedness .9442 avg. .9477 avg. .9502 avg. .9519 avg. .9489 avg.
Gender
-Female 219 (62.2%) 378 (59.1%) 620 (53.7%) 609 (51.8%) 775 (53.7%)
-Male 133 (37.8%) 262 (40.9%) 535 (46.3%) 566 (48.2%) 669 (46.3%)
First Generation Status 165 (46.9%) 164 (25.6%) 110 (9.5%) 41 (3.5%) 53 (3.7%)
Race/Ethnicity
-Hispanic/Latino 89 (25.3%) 112 (17.5%) 155 (13.4%) 126 (10.7%) 101 (7.0%)
-American Indian/
Alaskan Native
9 (2.6%) 13 (2.0%) 15 (1.3%) 14 (1.2%) 17 (1.2%)
-Asian 149 (42.3%) 239 (37.3%) 401 (34.7%) 348 (29.6%) 355 (24.6%)
-Black/African
American
29 (8.2%) 37 (5.8%) 63 (5.5%) 27 (2.3%) 9 (0.6%)
-White 73 (20.7%) 238 (37.2%) 517 (44.8%) 657 (55.9%) 948 (65.7%)
-Unknown 2 (0.6%) 1 (0.2%) 4 (0.3%) 3 (0.3%) 14 (1.0%)
Mixed Race/Ethnicity 40 (11.4%) 78 (12.2%) 134 (11.6%) 130 (11.1%) 104 (7.2%)
Degree Program Type
-Arts 30 (8.5%) 71 (11.1%) 169 (14.6%) 173 (14.7%) 154 (10.7%)
-Business 67 (19.0%) 101 (15.8%) 186 (16.1%) 233 (19.8%) 369 (25.6%)
-Communication 11 (3.1%) 41 (6.4%) 77 (6.7%) 67 (5.7%) 93 (6.4%)
-Engineering 51 (14.5%) 109 (17.0%) 217 (18.8%) 217 (18.5%) 192 (13.3%)
-Humanities 12 (3.4%) 32 (5.0%) 50 (4.3%) 40 (3.4%) 65 (4.5%)
-Natural Sciences 63 (17.9%) 119 (18.6%) 190 (16.5%) 171 (14.6%) 184 (12.7%)
-Social Sciences 48 (13.6%) 64 (10.0%) 111 (9.6%) 108 (9.2%) 141 (9.8%)
-Undecided 70 (19.9%) 103 (16.1%) 155 (13.4%) 166 (14.1%) 246 (17.0%)
Note: AP = academic preparedness, AP .90 < 1.0 = academic preparedness score ranges from
.90 to less than 1.0

IMPACT OF SOCIOECONOMIC STATUS 217
Table 47
Composition Totals by SES for AP 1.0+
Non-SPP students with
AP of 1.0 and higher
Lowest Low Lower to
Middle
Middle to
Upper
Did not Apply
Total students 12 36 126 194 149
EFC Percentage 7.1% 17.4% 62.8% 221% X
EFC+Aid Percentage 102% 98.9% 122% 271% X
Academic Preparedness 1.011 avg. 1.012 avg. 1.015 avg. 1.014 avg. 1.014
Gender
-Female 11 (91.7%) 22 (61.1%) 75 (59.5%) 96 (49.5%) 77 (51.7%)
-Male 1 (8.3%) 14 (38.9%) 51 (40.5%) 98 (50.5%) 72 (48.3%)
First Generation Status 2 (16.7%) 7 (19.4%) 2 (1.6%) 4 (2.1%) 0
Race/Ethnicity
-Hispanic/Latino 0 0 3 (2.4%) 8 (4.1%) 6 (4.0%)
-American Indian/
Alaskan Native
0 1 (2.8%) 2 (1.6%) 1 (0.5%) 0
-Asian 10 (83.3%) 15 (41.7%) 58 (46.0%) 75 (38.7%) 47 (31.5%)
-Black/African
American
0 0 1 (0.8%) 0 1 (0.7%)
-White 2 (16.7%) 19 (52.8%) 62 (49.2%) 109 (56.2%) 94 (63.1%)
-Unknown 0 1 (2.8%) 0 1 (0.5%) 1 (0.7%)
Mixed Race/Ethnicity 0 1 (2.8%) 16 (12.7%) 17 (8.8%) 4 (2.7%)
Degree Program Type
-Arts 0 5 (13.9%) 17 (13.5%) 18 (9.3%) 13 (8.7%)
-Business 5 (41.7%) 5 (13.9%) 21 (16.7%) 37 (19.1%) 34 (22.8%)
-Communication 0 0 4 (3.2%) 6 (3.1%) 4 (2.7%)
-Engineering 0 8 (22.2%) 27 (21.4%) 49 (25.3%) 32 (21.5%)
-Humanities 0 0 5 (4.0%) 6 (3.1%) 5 (3.4%)
-Natural Sciences 3 (25.0%) 7 (19.4%) 26 (20.6%) 45 (23.2%) 26 (17.4%)
-Social Sciences 3 (25.0%) 3 (8.3%) 8 (6.3%) 5 (2.6%) 11 (7.4%)
-Undecided 1 (8.3%) 8 (22.2%) 18 (14.3%) 28 (14.4%) 24 (16.1%)
Note: AP = academic preparedness, AP 1.0+= academic preparedness scores of 1.0 and higher

IMPACT OF SOCIOECONOMIC STATUS 218
Table 48
Outcomes by Race/Ethnicity
AP EFC 1GPA 1Units 1-2 4Grad 4GPA 4Units 4TUnits
Hispanic/Latino –840
students
.9201 X 3.19 32.5 96% 77% 3.26 118.4 136.6
American
Indian/Alaskan Native
- 106
.9249 X 3.17 32.3 99% 80% 3.33 123.1 140.4
Asian –1930 .9496 X 3.33 33.3 98% 84% 3.38 121.5 145.2
Black/African
American - 359
.8918 X 3.08 32.0 97% 75% 3.14 117.6 130.5
White – 3283 .9426 X 3.35 33.0 96% 82% 3.41 119.1 138.0
Unknown – 38 .9212 X 3.32 33.5 92% 82% 3.40 118.3 136.6
Note: AP = academic preparedness, EFC = expected family contribution as a percentage of the
full cost of attending, 1GPA = grade point average after the first year, 1Units = units earned after
the first year, 1-2 = persistence to the second year, 4GPA = grade point average after the fourth
year, 4Units = units earned after the fourth year, 4TUnits = total units earned after the fourth
year.

IMPACT OF SOCIOECONOMIC STATUS 219
Table 49
Outcomes by Race/Ethnicity Part One (AP .85 < .95)
AP EFC 1GPA 1Units 1-2 4Grad 4GPA 4Units 4TUnits
Hispanic/Latino –600
students
.9099 X 3.12 32.2 96% 77% 3.20 117.1 134.4
+SES (Lowest) – 105 .9094 5.8% 3.01 32.0 99% 67% 3.12 120.0 137.8
+SES (Low) – 111 .9140 18.2% 3.11 32.2 97% 71% 3.22 120.1 136.2
+SES (Lower-Middle)
– 159
.9089 57.3% 3.12 31.9 95% 79% 3.20 115.1 134.2
+SES (Mid-Upper) –
121
.9096 221% 3.19 32.7 97% 81% 3.26 115.3 132.8
+SES Did not Apply –
104
.9080 N/A 3.14 32.0 92% 84% 3.23 116.2 131.4
Asian –888 students .9177 X 3.16 32.8 97% 79% 3.24 119.1 140.1
+SES (Lowest) – 97 .9171 6.2% 3.14 32.0 97% 72% 3.20 118.1 138.6
+SES (Low) – 142 .9185 18.0% 3.14 32.7 97% 74% 3.23 120.6 142.1
+SES (Lower-Middle)
– 222
.9201 57.8% 3.17 32.9 97% 80% 3.20 118.2 139.7
+SES (Mid-Upper) –
192
.9175 226% 3.19 33.1 96% 85% 3.28 119.0 140.7
+SES Did not Apply –
235
.9155 N/A 3.16 32.8 98% 79% 3.26 119.3 139.6
Note: AP = academic preparedness, EFC = expected family contribution as a percentage of the
full cost of attending, 1GPA = grade point average after the first year, 1Units = units earned after
the first year, 1-2 = persistence to the second year, 4GPA = grade point average after the fourth
year, 4Units = units earned after the fourth year, 4TUnits = total units earned after the fourth
year.

IMPACT OF SOCIOECONOMIC STATUS 220
Table 50
Outcomes by Race/Ethnicity Part Two (AP .85 < .95)
AP EFC 1GPA 1Units 1-2 4Grad 4GPA 4Units 4TUnits
Black/African
American – 243
students
.8995 X 3.09 31.7 97% 74% 3.14 117.3 129.6
+SES (Lowest) – 43 .9034 5.6% 3.11 32.0 100% 60% 3.11 119.6 131.7
+SES (Low) – 52 .8973 18.7% 3.01 31.5 96% 79% 3.03 116.5 127.3
+SES (Lower-Middle)
– 90
.9025 58.0% 3.11 31.3 97% 78% 3.19 114.9 128.1
+SES (Mid-Upper) –
34
.8996 193% 3.17 32.7 97% 79% 3.25 121.5 133.7
+SES Did not Apply –
24
.8856 N/A 3.05 32.3 96% 71% 3.06 117.8 131.0
White – 1561
students
.9147 X 3.24 32.6 95% 79% 3.31 117.1 133.5
+SES (Lowest) – 41 .9213 5.6% 3.34 32.5 88% 66% 3.30 108.3 126.3
+SES (Low) – 134 .9158 19.4% 3.19 32.1 94% 71% 3.25 114.4 130.3
+SES (Lower-Middle)
– 305
.9167 60.0% 3.26 33.0 95% 80% 3.33 116.4 133.1
+SES (Mid-Upper) –
395
.9156 275% 3.26 32.8 97% 82% 3.32 119.1 136.0
+SES Did not Apply –
686
.9127 N/A 3.23 32.5 95% 80% 3.32 117.4 133.2
Note: AP = academic preparedness, EFC = expected family contribution as a percentage of the
full cost of attending, 1GPA = grade point average after the first year, 1Units = units earned after
the first year, 1-2 = persistence to the second year, 4GPA = grade point average after the fourth
year, 4Units = units earned after the fourth year, 4TUnits = total units earned after the fourth
year.

IMPACT OF SOCIOECONOMIC STATUS 221
Table 51
Outcomes for All Students by Gender
AP EFC 1GPA 1Units 1-2 4Grad 4GPA 4Units 4TUnits
Male – 3162 students .9324 X 3.23 32.7 97% 78% 3.28 118.7 138.3
+SES (Lowest) – 208 .9143 5.8% 3.08 31.8 98% 63% 3.08 116.5 134.8
+SES (Low) – 370 .9281 19.1% 3.14 32.1 98% 75% 3.23 119.5 138.4
+SES (Lower-Middle)
– 746
.9359 59.7% 3.23 32.7 96% 78% 3.26 117.5 137.2
+SES (Mid-Upper) –
824
.9409 241% 3.32 33.1 97% 83% 3.35 119.6 140.6
+SES Did not Apply –
1014
.9283 N/A 3.22 32.7 97% 78% 3.28 119.1 137.8
Female – 3692
students
.9350 X 3.35 33.1 96% 83% 3.41 120.3 139.4
+SES (Lowest) – 326 .9220 5.9% 3.23 32.8 97% 70% 3.30 119.5 137.5
+SES (Low) – 519 .9293 18.4% 3.27 32.9 97% 78% 3.33 120.2 137.6
+SES (Lower-Middle)
– 902
.9342 58.3% 3.35 33.0 96% 84% 3.41 119.1 139.0
+SES (Mid-Upper) –
851
.9451 241% 3.39 33.3 96% 87% 3.46 121.4 142.0
+SES Did not Apply –
1094
.9345 N/A 3.38 33.1 97% 86% 3.45 120.7 139.1
Note: AP = academic preparedness, EFC = expected family contribution as a percentage of the
full cost of attending, 1GPA = grade point average after the first year, 1Units = units earned after
the first year, 1-2 = persistence to the second year, 4GPA = grade point average after the fourth
year, 4Units = units earned after the fourth year, 4TUnits = total units earned after the fourth
year.

IMPACT OF SOCIOECONOMIC STATUS 222
Table 52
Outcomes for All Students by First Generation Status
AP EFC 1GPA 1Units 1-2 4Grad 4GPA 4Units 4TUnits
1rst Gen – 801
students
.9167 X 3.16 32.6 98% 73% 3.22 118.5 136.8
+SES (Lowest) – 265 .9091 5.4% 3.09 32.6 98% 65% 3.14 118.8 136.0
+SES (Low) – 239 .9217 16.9% 3.19 32.8 97% 75% 3.26 119.9 138.0
+SES (Lower-Middle)
– 156
.9189 51.6% 3.20 32.7 96% 81% 3.26 117.4 138.0
+SES (Mid-Upper) –
68
.9186 208% 3.14 32.7 99% 76% 3.24 117.4 134.6
+SES Did not Apply –
73
.9209 N/A 3.19 32.1 97% 78% 3.29 116.7 135.8
Non 1rst Gen – 6053
students
.9361 X 3.31 32.9 97% 82% 3.37 119.7 139.1
+SES (Lowest) – 269 .9288 6.3% 3.25 32.2 96% 70% 3.28 117.9 136.9
+SES (Low) – 650 .9314 19.3% 3.23 32.4 97% 78% 3.30 119.9 138.0
+SES (Lower-Middle)
– 1492
.9366 59.7% 3.30 32.9 96% 81% 3.35 118.5 138.2
+SES (Mid-Upper) –
1607
.9441 242% 3.37 33.3 97% 85% 3.42 120.6 141.6
+SES Did not Apply –
2035
.9319 N/A 3.31 33.0 97% 82% 3.37 120.0 138.6
Note: AP = academic preparedness, EFC = expected family contribution as a percentage of the
full cost of attending, 1GPA = grade point average after the first year, 1Units = units earned after
the first year, 1-2 = persistence to the second year, 4GPA = grade point average after the fourth
year, 4Units = units earned after the fourth year, 4TUnits = total units earned after the fourth
year.

IMPACT OF SOCIOECONOMIC STATUS 223
Charts and Other Graphics
Chart 1
One Way Analysis of Variance for Means between Cohorts
ANOVA

Sum of Squares df Mean Square F Sig.
Yr1GPA Between Groups .085 2 .042 .174 .840
Within Groups 1670.071 6851 .244

Total 1670.156 6853

Yr1Units Between Groups 2.850 2 1.425 .086 .917
Within Groups 112994.809 6851 16.493

Total 112997.659 6853

Yr4GPA Between Groups .395 2 .198 .962 .382
Within Groups 1407.126 6851 .205

Total 1407.521 6853

Yr4Units Between Groups 2594.879 2 1297.440 2.276 .103
Within Groups 3904705.222 6851 569.947

Total 3907300.101 6853

Note: Yr1GPA = grade point average after the first year, Yr1Units = units earned after the first
year, Yr4GPA = grade point average after the fourth year, Yr4Units = units earned after the
fourth year. Groups compared are the three cohort years of 2007, 2008, and 2009.

IMPACT OF SOCIOECONOMIC STATUS 224
Chart 2
Additional Comparison of Means between Cohorts
Multiple Comparisons
Tukey HSD
Dependent
Variable (I) Cohort (J) Cohort
Mean
Difference
(I-J)
Std.
Error Sig.
95% Confidence
Interval
Lower
Bound
Upper
Bound
Yr1GPA Fall 2007 cohort Fall 2008 cohort .00086 .01458 .998 -.0333 .0350
Fall 2009 cohort -.00702 .01454 .879 -.0411 .0271
Fall 2008 cohort Fall 2007 cohort -.00086 .01458 .998 -.0350 .0333
Fall 2009 cohort -.00788 .01471 .854 -.0424 .0266
Fall 2009 cohort Fall 2007 cohort .00702 .01454 .879 -.0271 .0411
Fall 2008 cohort .00788 .01471 .854 -.0266 .0424
Yr1Units Fall 2007 cohort Fall 2008 cohort .0314 .1199 .963 -.250 .312
Fall 2009 cohort -.0183 .1196 .987 -.299 .262
Fall 2008 cohort Fall 2007 cohort -.0314 .1199 .963 -.312 .250
Fall 2009 cohort -.0497 .1210 .911 -.333 .234
Fall 2009 cohort Fall 2007 cohort .0183 .1196 .987 -.262 .299
Fall 2008 cohort .0497 .1210 .911 -.234 .333
Yr4GPA Fall 2007 cohort Fall 2008 cohort .00696 .01338 .861 -.0244 .0383
Fall 2009 cohort .01837 .01335 .354 -.0129 .0497
Fall 2008 cohort Fall 2007 cohort -.00696 .01338 .861 -.0383 .0244
Fall 2009 cohort .01140 .01350 .675 -.0203 .0431
Fall 2009 cohort Fall 2007 cohort -.01837 .01335 .354 -.0497 .0129
Fall 2008 cohort -.01140 .01350 .675 -.0431 .0203
Yr4Units Fall 2007 cohort Fall 2008 cohort .6552 .7048 .621 -.997 2.307
Fall 2009 cohort -.8577 .7033 .442 -2.506 .791
Fall 2008 cohort Fall 2007 cohort -.6552 .7048 .621 -2.307 .997
Fall 2009 cohort -1.5130 .7114 .085 -3.181 .155
Fall 2009 cohort Fall 2007 cohort .8577 .7033 .442 -.791 2.506
Fall 2008 cohort 1.5130 .7114 .085 -.155 3.181
Note: Yr1GPA = grade point average after the first year, Yr1Units = units earned after the first
year, Yr4GPA = grade point average after the fourth year, Yr4Units = units earned after the
fourth year. Groups compared are the three cohort years of 2007, 2008, and 2009.

IMPACT OF SOCIOECONOMIC STATUS 225
Chart 3
Relationship between SES and Academic Preparedness Scores
Note: SPP = support program participants, Non-SPP = non-participants of the support program

0.79
0.81
0.83
0.85
0.87
0.89
0.91
0.93
0.95
SES Lowest Low Lower-Middle Middle-Upper Did not Apply
SPP
Non-SPP
IMPACT OF SOCIOECONOMIC STATUS 226
Chart 4
Relationship between SES and Four Year Graduation Rates

Note: SPP = support program participants, Non-SPP = non-participants of the support program
0.45
0.55
0.65
0.75
0.85
0.95
SES Lowest Low Lower-Middle Middle-Upper Did not Apply
SPP
non-SPP
IMPACT OF SOCIOECONOMIC STATUS 227
Chart 5
Distribution of Academic Preparedness Scores for Non-Participants
Note: AcadPrep = academic preparedness scores

IMPACT OF SOCIOECONOMIC STATUS 228
Chart 6
Distribution of Academic Preparedness Scores for Participants

Note: AcadPrep = academic preparedness scores

IMPACT OF SOCIOECONOMIC STATUS 229
Chart 7
RQ1: Comparison of Means: Lowest SES (AP .80 < .90)
Independent Samples Test

Levene's Test
for Equality
of Variances t-test for Equality of Means
F Sig. t df
Sig.
(2-
tailed)
Mean
Difference
Std. Error
Difference
95% Confidence
Interval of the
Difference
Lower Upper
AcadPrep Equal variances
assumed
16.70 .000 -.33 91 .742 -.001526 .004619 -.01070 .00765
Equal variances
not assumed

-.34 77.84 .735 -.001526 .004493 -.01047 .00742
Yr1GPA Equal variances
assumed
1.68 .198 .244 91 .808 .02979 .12201 -.2126 .2721
Equal variances
not assumed

.247 89.48 .805 .02979 .12045 -.2095 .2691
Yr1Units Equal variances
assumed
.224 .637 .271 91 .787 .3613 1.3345 -2.289 3.012
Equal variances
not assumed

.273 90.60 .785 .3613 1.3219 -2.265 2.987
Yr4GPA Equal variances
assumed
1.45 .233 1.14 91 .258 .13725 .12047 -.1021 .3766
Equal variances
not assumed

1.15 90.79 .254 .13725 .11946 -.1001 .3745
Yr4Units Equal variances
assumed
.002 .964 .203 91 .840 1.2101 5.9732 -10.65 13.08
Equal variances
not assumed

.202 88.67 .840 1.2101 5.9906 -10.69 13.11
Note: AcadPrep = academic preparedness scores, Yr1GPA = grade point average after the first
year, Yr1Units = units earned after the first year, Yr4GPA = grade point average after the fourth
year, Yr4Units = units earned after the fourth year

IMPACT OF SOCIOECONOMIC STATUS 230
Chart 8
RQ1: Comparison of Means: Lowest SES (AP .83139 < .936)
Independent Samples Test

Levene's Test
for Equality
of Variances t-test for Equality of Means
F Sig. t df
Sig.
(2-
tailed)
Mean
Difference
Std. Error
Difference
95% Confidence
Interval of the
Difference
Lower Upper
AcadPrep Equal variances
assumed
10.63 .001 -.62 123 .538 -.002302 .003730 -.0097 .00508
Equal variances
not assumed

-.54 58.18 .591 -.002302 .004257 -.0108 .00622
Yr1GPA Equal variances
assumed
.56 .454 -1.4 123 .152 -.13865 .09623 -.3291 .0518
Equal variances
not assumed

-1.5 87.10 .139 -.13865 .09292 -.3233 .0460
Yr1Units Equal variances
assumed
1.45 .231 -3.1 123 .003 -2.8307 .9293 -4.67 -.991
Equal variances
not assumed

-3.4 102.3 .001 -2.8307 .8435 -4.50 -1.158
Yr4GPA Equal variances
assumed
2.95 .088 -.54 123 .587 -.05383 .09897 -.2497 .1421
Equal variances
not assumed

-.59 99.69 .554 -.05383 .09076 -.2339 .1262
Yr4Units Equal variances
assumed
4.36 .039 -1.5 123 .130 -6.6127 4.3338 -15.19 1.966
Equal variances
not assumed

-1.8 111.5 .084 -6.6127 3.7877 -14.12 .893
Note: AcadPrep = academic preparedness scores, Yr1GPA = grade point average after the first
year, Yr1Units = units earned after the first year, Yr4GPA = grade point average after the fourth
year, Yr4Units = units earned after the fourth year

IMPACT OF SOCIOECONOMIC STATUS 231
Chart 9
RQ1: Comparison of Means: Low SES (AP .80 < .90)
Independent Samples Test

Levene's Test
for Equality
of Variances t-test for Equality of Means
F Sig. t df
Sig.
(2-
tailed)
Mean
Difference
Std. Error
Difference
95% Confidence
Interval of the
Difference
Lower Upper
AcadPrep Equal variances
assumed
15.52 .000 -.02 96 .987 -.000082 .005181 -.01037 .01020
Equal variances
not assumed

-.02 61.97 .988 -.000082 .005574 -.01122 .01106
Yr1GPA Equal variances
assumed
.117 .734 -1.2 96 .245 -.11441 .09773 -.3084 .0796
Equal variances
not assumed

-1.2 89.91 .239 -.11441 .09654 -.3062 .0774
Yr1Units Equal variances
assumed
1.27 .263 -.31 96 .759 -.3051 .9932 -2.277 1.667
Equal variances
not assumed

-.30 79.51 .764 -.3051 1.0139 -2.323 1.713
Yr4GPA Equal variances
assumed
.956 .331 -.23 96 .816 -.02050 .08784 -.1949 .1539
Equal variances
not assumed

-.23 78.73 .820 -.02050 .08987 -.1994 .1584
Yr4Units Equal variances
assumed
.639 .426 -.97 96 .332 -4.0066 4.1116 -12.168 4.155
Equal variances
not assumed

-1.0 95.66 .303 -4.0066 3.8649 -11.679 3.666
Note: AcadPrep = academic preparedness scores, Yr1GPA = grade point average after the first
year, Yr1Units = units earned after the first year, Yr4GPA = grade point average after the fourth
year, Yr4Units = units earned after the fourth year
IMPACT OF SOCIOECONOMIC STATUS 232
Chart 10
RQ1: Comparison of Means: Did Not Apply for Aid (AP .80 < .90)
Independent Samples Test

Levene's Test
for Equality
of Variances t-test for Equality of Means
F Sig. t df
Sig.
(2-
tailed)
Mean
Difference
Std. Error
Difference
95% Confidence
Interval of the
Difference
Lower Upper
AcadPrep Equal variances
assumed
11.41 .001 1.80 190 .074 .006530 .003633 -.00064 .013695
Equal variances
not assumed

1.43 37.44 .160 .006530 .004560 -.00271 .015765
Yr1GPA Equal variances
assumed
.020 .888 1.07 190 .288 .10887 .10223 -.09277 .31052
Equal variances
not assumed

1.04 43.17 .306 .10887 .10513 -.10312 .32087
Yr1Units Equal variances
assumed
.158 .691 .133 190 .894 .1250 .9377 -1.7246 1.9746
Equal variances
not assumed

.144 47.89 .886 .1250 .8698 -1.6240 1.8740
Yr4GPA Equal variances
assumed
.210 .647 .803 190 .423 .07456 .09284 -.10857 .25769
Equal variances
not assumed

.758 42.07 .453 .07456 .09838 -.12396 .27309
Yr4Units Equal variances
assumed
.249 .618 -1.2 190 .216 -6.7906 5.4749 -17.590 4.0088
Equal variances
not assumed

-1.2 44.07 .224 -6.7906 5.5044 -17.884 4.3022
Note: AcadPrep = academic preparedness scores, Yr1GPA = grade point average after the first
year, Yr1Units = units earned after the first year, Yr4GPA = grade point average after the fourth
year, Yr4Units = units earned after the fourth year

IMPACT OF SOCIOECONOMIC STATUS 233
Chart 11
RQ1: Comparison of Means: Did Not Apply for Aid (AP < .80139)
Independent Samples Test

Levene's Test
for Equality
of Variances t-test for Equality of Means
F Sig. t df
Sig.
(2-
tailed)
Mean
Difference
Std. Error
Difference
95% Confidence
Interval of the
Difference
Lower Upper
AcadPrep Equal variances
assumed
23.38 .000 5.32 99 .000 .038810 .007295 .024336 .05329
Equal variances
not assumed

5.12 58.81 .000 .038810 .007575 .023651 .05397
Yr1GPA Equal variances
assumed
.188 .666 .859 99 .392 .10976 .12779 -.14381 .36333
Equal variances
not assumed

.861 98.82 .391 .10976 .12743 -.14309 .36260
Yr1Units Equal variances
assumed
.037 .849 .005 99 .996 .0059 1.2279 -2.4305 2.4423
Equal variances
not assumed

.005 96.11 .996 .0059 1.2114 -2.3987 2.4105
Yr4GPA Equal variances
assumed
.092 .762 .562 99 .576 .06371 .11340 -.16131 .28873
Equal variances
not assumed

.565 99.00 .574 .06371 .11282 -.16015 .28758
Yr4Units Equal variances
assumed
1.062 .305 -.83 99 .408 -4.9898 6.0010 -16.897 6.9174
Equal variances
not assumed

-.84 97.35 .402 -4.9898 5.9329 -16.764 6.7848
Note: AcadPrep = academic preparedness scores, Yr1GPA = grade point average after the first
year, Yr1Units = units earned after the first year, Yr4GPA = grade point average after the fourth
year, Yr4Units = units earned after the fourth year

IMPACT OF SOCIOECONOMIC STATUS 234
Chart 12
RQ1: Comparison of Means: Did Not Apply for Aid (AP.80139 < .8664)
Independent Samples Test

Levene's Test
for Equality
of Variances t-test for Equality of Means
F Sig. t df
Sig.
(2-
tailed)
Mean
Difference
Std. Error
Difference
95% Confidence
Interval of the
Difference
Lower Upper
AcadPrep Equal variances
assumed
.272 .603 3.43 179 .001 .012421 .003623 .005272 .019570
Equal variances
not assumed

3.27 32.76 .003 .012421 .003800 .004687 .020154
Yr1GPA Equal variances
assumed
.081 .776 1.25 179 .212 .14261 .11373 -.08182 .36704
Equal variances
not assumed

1.16 32.10 .255 .14261 .12300 -.10791 .39312
Yr1Units Equal variances
assumed
.648 .422 -.45 179 .657 -.4355 .9785 -2.3663 1.4954
Equal variances
not assumed

-.61 48.29 .547 -.4355 .7189 -1.8807 1.0097
Yr4GPA Equal variances
assumed
.334 .564 .875 179 .383 .08979 .10262 -.11271 .29230
Equal variances
not assumed

.805 32.00 .427 .08979 .11151 -.13734 .31693
Yr4Units Equal variances
assumed
1.188 .277 -1.5 179 .125 -8.8115 5.7118 -20.083 2.4597
Equal variances
not assumed

-1.8 38.17 .087 -8.8115 5.0166 -18.966 1.3425
Note: AcadPrep = academic preparedness scores, Yr1GPA = grade point average after the first
year, Yr1Units = units earned after the first year, Yr4GPA = grade point average after the fourth
year, Yr4Units = units earned after the fourth year

IMPACT OF SOCIOECONOMIC STATUS 235
Chart 13
Four Year Graduation Means Plot for Non-Participants (n= 6,557)

Note: Grad4 = four year graduation rate

IMPACT OF SOCIOECONOMIC STATUS 236
Chart 14
ANOVA by SES: All Non-Participants
ANOVA

Sum of Squares df Mean Square F Sig.
Yr1GPA Between Groups 14.757 4 3.689 15.656 .000
Within Groups 1543.893 6552 .236

Total 1558.649 6556

Yr1Units Between Groups 430.137 4 107.534 6.799 .000
Within Groups 103622.758 6552 15.815

Total 104052.895 6556

Yr4GPA Between Groups 13.012 4 3.253 16.667 .000
Within Groups 1278.726 6552 .195

Total 1291.738 6556

Yr4Units Between Groups 4000.103 4 1000.026 1.779 .130
Within Groups 3682774.021 6552 562.084

Total 3686774.124 6556

Note: Yr1GPA = grade point average after the first year, Yr1Units = units earned after the first
year, Yr4GPA = grade point average after the fourth year, Yr4Units = units earned after the
fourth year

IMPACT OF SOCIOECONOMIC STATUS 237
Chart 15
ANOVA by SES: Non-Participants (AP .80 < .90)
ANOVA

Sum of Squares df Mean Square F Sig.
Yr1GPA Between Groups 5.041 4 1.260 4.902 .001
Within Groups 316.990 1233 .257

Total 322.031 1237

Yr1Units Between Groups 57.284 4 14.321 .745 .561
Within Groups 23706.368 1233 19.227

Total 23763.652 1237

Yr4GPA Between Groups 4.556 4 1.139 5.065 .000
Within Groups 277.300 1233 .225

Total 281.857 1237

Yr4Units Between Groups 2475.535 4 618.884 .924 .449
Within Groups 825496.885 1233 669.503

Total 827972.420 1237

Note: Yr1GPA = grade point average after the first year, Yr1Units = units earned after the first
year, Yr4GPA = grade point average after the fourth year, Yr4Units = units earned after the
fourth year

IMPACT OF SOCIOECONOMIC STATUS 238
Chart 16
ANOVA by SES: Non-Participants (AP .90 < 1.0)
ANOVA

Sum of Squares df Mean Square F Sig.
Yr1GPA Between Groups 4.643 4 1.161 5.673 .000
Within Groups 974.092 4761 .205

Total 978.735 4765

Yr1Units Between Groups 263.491 4 65.873 4.464 .001
Within Groups 70260.329 4761 14.757

Total 70523.820 4765

Yr4GPA Between Groups 5.052 4 1.263 7.536 .000
Within Groups 797.887 4761 .168

Total 802.938 4765

Yr4Units Between Groups 1608.518 4 402.129 .754 .555
Within Groups 2538581.807 4761 533.203

Total 2540190.325 4765

Note: Yr1GPA = grade point average after the first year, Yr1Units = units earned after the first
year, Yr4GPA = grade point average after the fourth year, Yr4Units = units earned after the
fourth year

IMPACT OF SOCIOECONOMIC STATUS 239
Chart 17
Asian American Distribution of Academic Preparedness Scores

Note: AcadPrep = academic preparedness score

IMPACT OF SOCIOECONOMIC STATUS 240
Chart 18
Black/African American Distribution of Academic Preparedness Scores

Note: AcadPrep = academic preparedness score

IMPACT OF SOCIOECONOMIC STATUS 241
Chart 19
Hispanic/Latino Distribution of Academic Preparedness Scores
Note: AcadPrep = academic preparedness score

IMPACT OF SOCIOECONOMIC STATUS 242
Chart 20
White Distribution of Academic Preparedness Scores
Note: AcadPrep = academic preparedness score

IMPACT OF SOCIOECONOMIC STATUS 243
Chart 21
Distribution of Students by SES Category for Racial/Ethnic Groups (.85 < .95)

Note: This chart shows the number of students by race/ethnicity at each SES level. .85 < .95 =
students with academic preparedness scores of .85 to less than .95

0
40
80
120
160
200
240
280
320
360
400
440
480
520
560
600
640
680
Asian
Black/African American
Hispanic/Latino
White
IMPACT OF SOCIOECONOMIC STATUS 244
Chart 22
Four Year Graduation Rates by SES and Race/Ethnicity (AP .85 < .95)

Note: AP .85 < .95 = students with academic preparedness scores of .85 to less than .95, Grad4
= four year graduation rate

IMPACT OF SOCIOECONOMIC STATUS 245
Chart 23
Data for Graduation Rates by SES and Race/Ethnicity (AP .85 < .95)
Descriptive Statistics
Dependent Variable:Grad4
RaceEthnic SEScategory Mean Std. Deviation N
Asian/Pacific Islander Lowest SES .72 .451 97
Low SES .74 .440 142
Lower to Middle SES .80 .403 222
Middle to Upper SES .85 .359 192
Did Not Apply .79 .410 235
Total .79 .409 888
Black/African American Lowest SES .60 .495 43
Low SES .79 .412 52
Lower to Middle SES .78 .418 90
Middle to Upper SES .79 .410 34
Did Not Apply .71 .464 24
Total .74 .437 243
Hispanic/Latino/a Lowest SES .67 .474 105
Low SES .71 .455 111
Lower to Middle SES .79 .407 159
Middle to Upper SES .81 .394 121
Did Not Apply .84 .372 104
Total .77 .423 600
White Lowest SES .66 .480 41
Low SES .71 .456 134
Lower to Middle SES .80 .403 305
Middle to Upper SES .82 .382 395
Did Not Apply .80 .400 686
Total .79 .405 1561
Total Lowest SES .68 .468 295
Low SES .73 .444 451
Lower to Middle SES .79 .408 800
Middle to Upper SES .83 .379 757
Did Not Apply .80 .401 1078
Total .78 .412 3381
Note: The 64 students identified as American Indian/Alaskan Native and the 25 students
identified as Unknown are included in the total, but now shown individually.

IMPACT OF SOCIOECONOMIC STATUS 246
Chart 24
Differences in Four Year Graduation Rates by Major and SES

Note: Grad4 = four year graduation rate

IMPACT OF SOCIOECONOMIC STATUS 247
Chart 25
Differences in Graduation Rates by Participation and SES (AP .80 < .90)
Note: Grad4 = four year graduation rate, AP .80 < .90 = students with academic preparedness
scores of .80 to less than .90

IMPACT OF SOCIOECONOMIC STATUS 248
Chart 26
Impact of SES and Academic Preparedness on Graduation

Note: Grad4 = four year graduation rate, AcadPrepRange = academic preparedness range. There
were very few students in the lowest academic preparedness range and this resulted in the
inconsistent graduation rate when compared to other ranges.

IMPACT OF SOCIOECONOMIC STATUS 249
Chart 27
Data for Impact of SES and Academic Preparedness on Graduation
Descriptive Statistics
Dependent Variable:Grad4
SEScategory AcadPrepRange Mean Std. Deviation N
Lowest SES Lowest .25 .500 4
Low-Middle .58 .495 106
Middle-High .74 .442 352
Highest .58 .515 12
Total .69 .461 474
Low SES Lowest .60 .548 5
Low-Middle .69 .463 146
Middle-High .79 .407 640
Highest .86 .351 36
Total .78 .418 827
Lower to Middle SES Lowest .67 .516 6
Low-Middle .73 .442 294
Middle-High .83 .372 1155
Highest .90 .305 126
Total .82 .385 1581
Middle to Upper SES Lowest .40 .548 5
Low-Middle .75 .434 263
Middle-High .87 .341 1175
Highest .92 .268 194
Total .85 .355 1637
Did Not Apply Lowest .44 .512 16
Low-Middle .75 .436 429
Middle-High .84 .363 1444
Highest .91 .283 149
Total .82 .380 2038
Total Lowest .47 .506 36
Low-Middle .72 .447 1238
Middle-High .83 .374 4766
Highest .90 .298 517
Total .81 .389 6557
Note: Grad4 = four year graduation rate

Abstract (if available)

Abstract There will be challenges resulting from the goals of the completion agenda (Lee & Rawls, 2010), underrepresentation of low socioeconomic status (SES) students at highly selective colleges (Carnevale & Rose, 2003), and relationship between institution type and social mobility (Haveman & Smeeding, 2006). If rates of access and success for low SES students are not improved then the economic intentions behind the completion agenda may be compromised. This study measured the impact of SES and academic preparedness on academic outcomes at a highly selective, private, research university. Academic outcome data consisted of grade point average (GPA) and completed units after the first and fourth year, persistence to the second year, and graduation after the fourth year for the 2007, 2008, and 2009 freshmen cohorts. A composite score of high school GPA and test scores was used to determine the academic preparedness of students and the variable was statistically significantly for all academic outcomes measured. The comparison of students of similar academic preparedness revealed SES was statistically significant for GPA after the first and fourth year, first year units completed, and four year graduation. When further examining the effectiveness of a student support program, the results were inconclusive. Although the potentially at-risk students required to participate in the program achieved similar outcomes when compared to non-participants of similar SES and academic preparedness, the support program did not minimize the effects of SES. The findings of this study further advance previous research pointing to the challenges faced by low SES students in the areas of acceptance, belonging, and capital in higher education. The identification of potential best practices to respond to this will require future research examining the impact of SES at other universities, especially when academic preparedness is factored.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The academic integration and retention of Latino community college transfer students at a highly selective private four-year institution

PDF

The impact of learning communities on the retention and academic integration of Latino students at a highly selective private four-year institution

PDF

Unlocking the potential: improving persistence and graduation of Black female students at select historically Black colleges and universities: a case study

PDF

The impact of learning communities on the retention and academic integration of African American students at a highly selective four-year private institution

PDF

The impact of learning community involvement and campus climate on student satisfaction and the retention of Latino students at a highly selective private institution

PDF

The impact of learning communities on the retention and social integration of Latino students at a highly selective private four-year institution

PDF

Studying the impact of a career academy on Chicano/Latino students in a specific high school: a quantitative case study

PDF

Attracting and retaining talent: improving the impact of workplace mentorship

PDF

The impact of work study on the integration and retention of undergraduate students at the University of California

PDF

A case study on readmitted students: the impact of social and academic involvement on degree completion

PDF

Supporting the transition and academic success of transfer students at a large research university

PDF

Evaluating the effects of diversity courses and student diversity experiences on undergraduate students' democratic values at a private urban research institution

Asset Metadata

Creator Mattson, Christopher Erik (author)

Core Title Acceptance, belonging, and capital: the impact of socioeconomic status at a highly selective, private, university

School Rossier School of Education

Degree Doctor of Education

Degree Program Education (Leadership)

Publication Date 09/10/2014

Defense Date 06/17/2014

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag academic preparedness,acceptance,belonging,Capital,GPA,Graduation,Higher education,highly selective,OAI-PMH Harvest,private,retention,SES,social mobility,socioeconomic status,student support,test scores,university progress

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Tambascia, Tracy Poon (committee chair), Schafrik, Janice (committee member), Tobey, Patricia Elaine (committee member)

Creator Email cemattso@usc.edu,christopher.e.m@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-474506

Unique identifier UC11287014

Identifier etd-MattsonChr-2911.pdf (filename),usctheses-c3-474506 (legacy record id)

Legacy Identifier etd-MattsonChr-2911.pdf

Dmrecord 474506

Document Type Dissertation

Format application/pdf (imt)

Rights Mattson, Christopher Erik

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA