University of Southern California Dissertations and Theses

Input-adjusted transfer scores as an accountability model for California community colleges

(USC Thesis Other)

Input-adjusted transfer scores as an accountability model for California community colleges

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Running head: COMMUNITY COLLEGE ACCOUNTABILITY 1
INPUT-ADJUSTED TRANSFER SCORES AS AN ACCOUNTABILITY MODEL FOR
CALIFORNIA COMMUNITY COLLEGES

by

Josephine Jones

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

August 2015

Copyright 2015 Josephine Jones
COMMUNITY COLLEGE ACCOUNTABILITY 2
Dedication
“Here’s to the crazy ones. The misfits. The rebels. Because the people who are crazy enough to
think they can change the world are the ones who actually do”- Steve Jobs

COMMUNITY COLLEGE ACCOUNTABILITY 3
Acknowledgments
I would like to give thanks to God who allowed me to have this journey, I am thankful
for the doors that were closed so that this opportunity could be born! I would also like to
acknowledge my dissertation committee Dr. Hocevar, Dr. Pacheco, and Dr. Keim. Dr. Hocevar I
could not have selected a better person to work with your dedication to students and commitment
to learning is beyond any academic experience I have had and I am forever grateful for your
support. I would like to acknowledge my family and friends whom without your ear to listen in
times of pain and joy I could not have made it through.

COMMUNITY COLLEGE ACCOUNTABILITY 4
Table of Content
Dedication 2

Acknowledgments 3

List of Tables 6

List of Figures 7

Abstract 8

CHAPTER ONE: OVERVIEW OF THE STUDY 10

Background 11

Statement of the Problem 12

Purpose of the Study 13

Significance of the Study 14

Definitions and Terms 15

Organization of Study 17

CHAPTER TWO: LITERATURE REVIEW 18

Accountability and Higher Education 18

Current Accountability Methods 19

Accountability in Community Colleges 23

Student Learning Outcomes 24

Wages and Employment 25

Progression and Completion 26

California Community Colleges 27

Current Measures of Accountability in CCC 28

Barriers Students Face 30

Transfer 30

Socioeconomic Status 32

Developmental Education 36

Academic Preparedness 37

Ethnicity 38

Discussion Barriers Students Face 39

Problem 40

Conceptual Framework 42

Summary 44

CHAPTER THREE: METHODOLOGY 46

Research Design 46

Population 47

Instrumentation 47

Predictor Variables 48

Outcome Variables 51

Data Sources 52

Input-Adjusted Transfer Success Scores 53

Limitations 53

Delimitations 53

CHAPTER FOUR: RESULTS 55

Overview 55

Findings 55

COMMUNITY COLLEGE ACCOUNTABILITY 5
Transfer Rates by Ethnic Classification 55

URM vs. Non-URM 56

Stability of Annual Changes in Transfer Rates 58

Input-Adjusted Scores 59

Stability of Input-Adjusted Scores 63

CHAPTER FIVE: DISCUSSION 67

Discussion 68

Implications and Conclusions 75

Implications 75

Conclusion 78

Future Research 80

References 82

Appendix: Data Collection Methodology 90

COMMUNITY COLLEGE ACCOUNTABILITY 6
List of Tables
Table 1: Year to year Correlation Between Prior and Present Change Scores 59

Table 2: Control Variables and Annual Transfer 60

Table 3: Correlations Among Predictor Variables 61

Table 4: Correlations Among Outcome Variables 62

Table 5: Correlations Among Outcome and Control Variables 63

Table 6: Regression Coefficients for Total 2013 Transfer Rate 64

Table 7: Highest Preforming Community Colleges 64

Table 8: 2009-2012 Regression Analysis 65

Table 9: 2009-2012 Regression Analysis 65

Table 10: Stability of Input-Adjusted Scores 2009-2013 66

Table 11: Over-Performing CSU Transfer Colleges 71

Table 12: Over Performing Colleges’ Characteristics 71

Table 13: Underperforming CSU Transfer Colleges 72

Table 14: Underperforming Colleges’ Characteristics 72

Table 15: Additional Information on Underperforming Institutions 73

COMMUNITY COLLEGE ACCOUNTABILITY 7
List of Figures
Figure 1: Average Level of Education 50

Figure 2: Median Income 51

Figure 3: Number of Transfers Disaggregated by Race 56

Figure 4: Number of Transfers for URM and Non-URM Students 57

Figure 5: Percent Transfer Gap for URM Students Versus Non-URM 57

Figure 6: Math Success Rate 1992-2013 77

COMMUNITY COLLEGE ACCOUNTABILITY 8
Abstract
In recent years, due to issues of the increased cost of education, challenges with student
retention, and significant increases in the number of students defaulting on federal student loans,
accountability within higher education has gained national attention. Currently, the United
States ranks 14
th
overall among developed nations in educational attainment, and the Obama
Administration has charged the United States to work back to the top by increasing the number
of college graduates by the year 2020. With the increased calls for accountability, the Obama
Administration has proposed the post-secondary educational rating system (PIRS) to assess and
rank institutional accountability as it relates to access, affordability, and student outcomes. These
rankings will be tied to federal financial aid awards and thus impact funding received by both
institutions and students. The goal of the rating system is to improve institutional performance
and increase educational outcomes. Being the leading access point to higher education,
community colleges play a significant role in increasing the number of college graduates via the
transfer route.
California has the largest community college system in the nation and is one the most
diverse states. The purpose of this twenty-year study was to examine the relationship between
ethnicity and transfer rates from California Community Colleges to California State Universities.
Input-adjusted scores will be used as an accountability measure to adjust for external factors that
have been known to impact student success. This study aimed to provide an accountability
measure that allows for fair institutional comparison of transfer rates and provides increased
awareness on equity in student outcomes as it relates to transfer in California Community
Colleges.
COMMUNITY COLLEGE ACCOUNTABILITY 9
In the last twenty years, there were no significant changes in the annual transfer rates
from California Community Colleges to California State Universities across ethnic lines. Both
Hispanic and African American students consistently trailed behind their peers in annual
transfers to California State Universities. Annual changes in transfer rates were not found to be a
stable measure enough measure for accountability purposes. However, input-adjusted transfer
rates were found to be stable from year to year, and are recommended as a tool to accurately
measure transfers rates and allow for fair institutional comparisons.

COMMUNITY COLLEGE ACCOUNTABILITY 10
CHAPTER ONE: OVERVIEW OF THE STUDY
The need for increased accountability in higher education has gone beyond an
institutional issue and is now being debated by both state and federal officials. In 2009, the
Obama Administration introduced the American Graduation Initiative (AGI) with a goal of
increasing the number of certificates and post-secondary degrees awarded by the year 2020
(Kotamraju & Blackman, 2011). This has increased the need for states to examine the
effectiveness of post- secondary institutions (Ewell, 2011; Kotamraju & Blackman, 2011).
Community colleges are at the forefront of this debate as they are open enrollment institutions
and provide access to post-secondary education for a significant number of underserved students
(Melguizo, 2007). Because community colleges serve a disproportionate number of
underrepresented students the increased push for individuals to earn post-secondary degrees will
fall disproportionately on community colleges (Kotamraju & Blackman, 2011). Increased
accountability efforts at both the state and local levels make it essential to examine student
outcomes as it relates to community colleges.
More than ever, performance is being tied to funding. While community colleges serve
more of the states underserved students, they receive less funding than their state university
counterparts (Harbour, 2003; Porchea, Allen, Robbins, & Phelps, 2010). With the increased
demand for post-secondary degree completion, there will be increased enrollment in community
colleges, which, can ultimately lead to overcrowding, and a decline in student completion rates
(Goldrick-Rab, 2010). This will ultimately decrease access to higher education and impact
educational outcomes for many of our nation’s most disadvantaged students.
In this study, I aim to examine the correlation between transfer rates and ethnicity in
California Community Colleges over the last twenty-nine years. In addition, using input-adjusted
COMMUNITY COLLEGE ACCOUNTABILITY 11
scores, I will examine if transfer rates are stable enough as a measurement for institutional
accountability.
Background
Community colleges began with a strong mission for students to transfer to four-year
institutions (Eagan & Jaeger, 2008). Over time, community colleges have developed into
comprehensive institutions with diverse missions; providing vocational training, certificate
programs, developmental coursework, and lifelong learning programs (Eagan & Jaeger, 2008;
Long & Kurlaender, 2009). The diverse institutional missions of community college have made
it difficult to measure institutional accountability (Bailey, Carlos, Calcagno, Jenkins, Kienzl, &
Leinbach, 2005; Ewell, 2011). Additionally, some believe the variability of student goals serves
as a diversion for students and weakens their academic progress leaving them less likely to
transfer to a four-year university (Long & Korlaender 2009). Such debates have made it
increasingly difficult to identify accountability measures that support both the institutional
mission of community colleges and individual student goals.
For California Community Colleges, the push for accountability is likely to have a major
impact, with 46% of all students in the state attending community college (Melguizo, 2007). In
addition, more than 70% of students enter California community colleges unprepared for
college-level work (Shaffer, 2008; Student Success Task Force, 2012). With community colleges
primarily serving students from underrepresented backgrounds, outcomes based on equity are an
important part of institutional accountability (Melguizo, Hagedorn, & Cypers, 2008). It would be
helpful for institutions and policy makers to examine factors that hinder underrepresented groups
from reaching equitable outcomes at the community college level.

COMMUNITY COLLEGE ACCOUNTABILITY 12
Statement of the Problem
Community colleges enroll over 40% of undergraduate students in the United States and
serve as the leading point of access to higher education for low-income and URM (Goldrick-Rab
& Pfeffer, 2009; Jenkins, 2011). Being open access institutions, the growing demand for
increased college graduates in the United States will fall disproportionately on community
colleges (Ewell, 2011; Kotamraju, & Blackman, 2011). The call for improved accountability
measures is demanding community colleges shift from open access institutions to emphasizing
completion (Mullin, 2010). Measuring completion within community college is a complex task.
Simply measuring transfer rates does not account for the characteristics of the students enrolled
or the resources available at a particular institution, which are each fundamental in understanding
student outcomes (Bailey & Xu, 2012). In order to meet the growing demand for college
graduates, it is essential to have appropriate accountability systems that are representative of the
diverse student body and that appropriately measure student outcomes.
Community colleges are essential in providing increased access to four-year degrees via
the transfer path (Bailey & Xu, 2012; Goldrick-Rab, 2010;) and despite being open access
institutions there are too few minority students transferring to four-year universities (Goldrick-
Rab & Pfeffer, 2009; Shullock & Moore, 2011). Current transfer policies are not believed to
address the unique needs of low-income and URM (Chase, Dowd, Pazich, & Bensimon, 2014).
Outcomes should be consistently disaggregated by race and regularly reviewed as a means of
combating the pervasive racial inequalities in student outcomes (Harris & Bensimon, 2007). The
push for equity in student outcomes is a recent part of the accountability discussion. Tools are
needed to accurately measure transfer success in a way that is comprehensive, and that accounts
for the unique student body served at each institution.
COMMUNITY COLLEGE ACCOUNTABILITY 13
Purpose of the Study
The purpose of this study is to examine the relationship between ethnicity and transfer
rates in California Community Colleges by using input-adjusted scores in an accountability
model. Previous research has demonstrated that there are differences in community college
transfer rates for minority students (Chase et al., 2014; Goldrick-Rab & Pfeffer, 2009; Melguizo
et al., 2008; Moore & Shulock, 2010). Student outcomes are joint products of both institutional
and individual student characteristics; input-adjusted scores allow an institution to adjust for pre-
college student characteristics that are beyond the control of the institution (Bailey & Xu, 2012).
With the new accountability measures urging that institutional performance be tied to
funding, input-adjusted scores ensure that outcomes are reflective of the student body, i.e.
colleges are then not targeting only high performing students who have significantly higher
chances of graduating to improve student outcomes (Bailey & Xu, 2012). When controlling for
external and institutional characteristics input-adjustment also allow for fair comparison of like
institutions (Bailey & Xu, 2012: NASFAA, 2014). In this study input-adjustment will be used to
examine transfer rates in California Community Colleges over the last twenty-nine years and
adjust for student characteristics, which include the average parental education and income in
each community college locale, as well as student ethnicity and whether or not a student is
younger than age 25.
The research questions for this study are:
1. Is there a relationship between ethnicity and transfer rates to four-year institutions?
2. Over the past two decades, what changes in transfer rates have occurred in California
Community Colleges for underrepresented minority students and non-underrepresented
minority students?
COMMUNITY COLLEGE ACCOUNTABILITY 14
3. Are annual changes in transfer rates stable enough to be used for institutional
accountability purposes?
4. Are input-adjusted transfer scores stable enough to be used for institutional accountability
purposes?
Significance of the Study
The post-secondary institutional rating system (PIRS) proposed by the Obama
administration in 2009, has called for increased accountability within higher education. PIRS
aims to increase access, accountability, and improved outcomes for students in both community
colleges and four-year institutions (US Department of Education, 2014). The push for increased
accountability in higher education has been largely driven by the rising cost of higher education
and the mounting federal student loan debt, which is estimated to be at over one trillion dollars
(Forbes, 2013). The proposed Obama Administration rating plan would be tied to performance-
based funding in which each institution would be rated in regard to financial aid, employment,
transfer, and completing a four-year degree (US Department of Education, 2014).
Acknowledging the unique needs of both four-year and two-year institutions, the US
Department of Education (2014) has identified multiple variables that will be considered as they
develop the rating system. Transfer is acknowledged to be an essential component of the
community college mission, however the complexities of measuring transfer rates pose a unique
barrier. The majority of data to be collected by PIRS will be gathered from the Integrated Post
Secondary Data System (IPEDS) and the National Student Loan Data System (NSLDS)
however; each of these data sources has its limitations. Transfer rates are currently reported to
IPEDS on a voluntary basis and NSLDS only tracks students who are receiving federal financial
aid (US Department of Education, 2014). Furthermore, NSLDS tracks subset demographic
COMMUNITY COLLEGE ACCOUNTABILITY 15
information (socioeconomic status and race) as reported on student financial aid applications and
IPEDS does not provide such information limiting the ability for transfer rates to be
disaggregated by race. In addition, when measuring transfer rates it is difficult to account for
lateral transfers (transfers to other community colleges) and measuring transfers that do not
include completion of an associate’s degree (US Department of Education, 2014). It is evident
that transfer is an important aspect of accountability for community colleges. This study will
contribute to the growing body of knowledge on the use of input-adjusted transfer as a potential
reliable tool to effectively measure transfer rates in a way that accounts for racial and ethnic
disparities and allows for comparison across institutions.
California is of specific interest being one of the largest community college systems
representing 25% of the nation's community college student population and serving
approximately 2.6 million students (Student Success Task Force, 2012). California is also one of
the most diverse states in the nation and has a growing Latino population that is predicted to be
at 50% of the total state population by the year 2040 (Melguizo, 2007; Moore & Shulock, 2010).
With community colleges being the primary access point for higher education for the majority of
low-income and minority students, issues of equity as it relates to student transfer are a top
concern (Chase et al., 2014; Melguizo et al., 2008; Moore & Shulock, 2010). The results of this
study would yield information on the transfer patterns of minority students in California
Community Colleges and possibly provide a framework for utilizing transfer data as an
accountability measure in a way that is equitable for cross-institutional comparison.
Definitions and Terms
Completion - successfully transferring to a four-year university with or without an associate's
degree.
COMMUNITY COLLEGE ACCOUNTABILITY 16
Ethnicity - The California Community College Chancellor’s Office (CCCCO) tracks eight
different ethnicities, but for the purposes of this study I focused on six African American,
Hispanic, White, Asian, Filipino, and “Other”.
Input-adjusted scores - Multi-regression analysis involving the examination of residual scores,
which are the actual achievement scores minus expected achievement scores to get the predicted
score was used to compute input-adjusted scores. Input adjustment involves controlling for
"inputs" that are beyond the control of an institution and impact performance scores (Bailey &
Xu, 2012). Throughout this study, the term input-adjusted and residual scores are used
interchangeably.
Socioeconomic status - defined by parental educational attainment and income.
1. Student success - is defined by the student success task force (2012):
2. Percentage of community college students completing their goals
3. Percentage of community college students earning a certificate or a degree, transferring,
or achieving transfer readiness
4. Number of students transferring to a four-year institutions
5. Number of degrees and certificates awarded
For the purposes of this study, number three-number of students transferring to four-year
institutions will be used to define student success.
Transfer Rate - Students who have successfully transferred to California State Universities
divided by the number of students enrolled in California Community colleges.
Underrepresented minority students - Refers to students who self-identify as Hispanic or African
American.

COMMUNITY COLLEGE ACCOUNTABILITY 17
Organization of Study
The dissertation is organized into five chapters. Chapter One presents an introduction,
statement of the problem, the purpose of the study, the research questions, the significance of the
study, and the definitions of key terms. Chapter Two provides a literature review on research on
the topics of the of California Community Colleges in higher education, impact of institutional
accountability, the barriers students face in the transfer process, and the role of support services
in aiding students to meet transfer goals. Chapter Three describes the methodology that includes
the research design utilized in this case study, population and sample, instrumentation, data
collection, data analysis, and ethical considerations. Chapter Four represent the finding for the
data collected as well as a description and analysis of the data. Finally, Chapter Five summarizes
and presents conclusions, which include implications for further research as well as
recommendations.

COMMUNITY COLLEGE ACCOUNTABILITY 18
CHAPTER TWO: LITERATURE REVIEW
The push for increased accountability has served to be a critical challenge within
community colleges, with diverse missions and varying student bodies, community colleges have
grappled with developing the means to provide accurate outcomes that are reflective of their
unique missions and diverse student interest (Eagan & Jaeger, 2008). While there is much debate
over the varying missions of community colleges, there is no doubt that transfer success is a
significant part of their mission. Being the leading access point to higher education community
colleges predominately serve low-income, first-generation, and minority students (Long &
Kurlaender, 2009; Melguizo et al., 2008). This review seeks to explore the issues of
accountability within higher education and specifically community colleges, issues of transfer
success and how race and ethnicity impact student transfer.
To gain an increased understanding of the role of accountability in higher education, this
literature review will begin with a broad overview of accountability and provide a current
political climate on accountability as it relates to higher education. Given the vast differences
between four-year and two-year institutions the literature will further explore the issues of
accountability within the community college. Finally, exploration of barriers students face in
community college, theoretical frameworks, and finally the use of input-adjusted scores as an
institutional accountability measure for transfer in community colleges will also be explored.
Accountability and Higher Education
The role of institutional accountability in higher education began in the 1970’s and
progressed over time. Institutional accountability began as the government’s way to ensure
federal and state dollars were being used effectively, however defining accountability was
largely left up to the institutions themselves (Burke, 2005; Shin, 2010). In recent years the role of
COMMUNITY COLLEGE ACCOUNTABILITY 19
accountability has shifted, federal and state governments have increased the demand for
accountability and tying institutional performance to funding (Espinosa, Crandall, &
Tukibayeva, 2014; NASFAA, 2014). Current trends in accountability and the concept of
accreditation will be explored. The literature will explore how institutional data is collected for
accountability purposes, and the current climate regarding institutional accountability and higher
education.
Current Accountability Methods
The primary way that states and the federal government ensure institutional quality and
accountability is through accreditation. Accreditation serves as an arm of the federal government
to ensure that federal and state dollars are properly managed, and it provides the official seal of
quality for institutions of higher education (Ewell, 2011). There are seven regional accreditation
agencies across the nation that set out to measure educational quality and institutional
effectiveness; this process happens every 5-10 years (ACCJC, 2014). However, community
colleges receive nearly 60% of their funding from local state revenue, giving states an increased
stake in the institutional effectiveness of community colleges to ensure they meet the needs of
the local communities they serve (Goldrick-Rab, 2010). Community colleges have many
stakeholders, meeting their needs as well as, the individual, institutional needs make identifying
appropriate accountability systems a challenge.
Data serves an integral part of accountability. The Integrated Post Secondary Data
System (IPEDS) provides data for all public institutions of higher education and one of the
primary sources of where the US Department of Education will receive their data for the Obama
Administration's proposed post-secondary institutional rating system (PIRS). IPEDS is a national
database that collects data from all public institutions of higher education on student enrollment,
COMMUNITY COLLEGE ACCOUNTABILITY 20
completion, persistence and success, and institutional resources (IPEDS, 2014). Participation in
IPEDS related surveys are mandatory for institutions to receive federal funding (IPEDS, 2014).
However, there is controversy regarding IPEDS ability to provide reliable data as it relates to
community colleges. The data provided by IPEDS does not count part-time students; students
who have delayed enrollment, or students who have taken some time off and are now returning
to school (Espinosa et al., 2014). This gap in data excludes 6.7 million part-time undergraduate
students; this is a significant group of students in public two-year and four-year institutions
(Espinosa et al., 2014). In community colleges, 31% of students enroll full time and 26% enroll
less than part-time, leaving IPEDS data unable to account for a significant number of community
college students (Goldrick-Rab, 2010). The US Department of Education (2014) has announced
that beginning in the fall of 2015 information for part-time and students who transfer into the
institution students will be required, but these metrics will not be publicly available until early
2017.
Transfer data is also a metric of concern within the IPEDS database currently this
information is reported by community colleges on a voluntary basis (US Department of
Education, 2014). Finally, there are concerns regarding IPEDS focus on inputs and inability to
provide outcome data as it relates to equity in student outcomes (Espinosa et al., 2014). IPEDS
has been able to collect a significant amount of data on institutional and student performance, but
the lack of reliable data impacts community colleges ability to assess and address disparities in
student outcomes.
In recent years accountability in higher education has become a national issue, in 2009
the current Obama administration proposed the American Graduation Initiative (AGI), which
seeks to increase the number of post-secondary degrees and certificates awarded in the United
COMMUNITY COLLEGE ACCOUNTABILITY 21
States by five million by the year 2020 (US Department of Education, 2014). To meet the AGI
goal it is estimated that the number of post-secondary degrees received would have to be
increased by 50%, with community colleges being the primary access point for higher education
it is believed the increased number of post-secondary degrees earned will fall disproportionately
on them (Kotamraju & Blackman, 2011). AGI has heightened the pressure on community
colleges to examine student outcomes and design accountability systems that will help them
improve student completion rates.
Kotamraju and Blackman (2011), conducted a study using data from the Integrated
Postsecondary Education System (IPEDS) to explore factors that impact completion and
simulate what community college completion rates would have to look like in order to
significantly contribute to the AGI goal proposed by President Obama. Currently, across the
nation community colleges produce 1.5 million graduates per year and to meet AGI this would
have to increase to 1.75 million (Kotamraju & Blackman, 2011). Their sample included 1,013
community colleges from across the United States. The independent variables were 2005
graduation rates for full-time students and the three-year average between 2005 and 2007 and the
predictor variables were a) student characteristics and behavior b) external workforce
development variables c) intuitional environmental factors. Using multiple regression Kotamraju
and Blackman (2011), found that when adjusting for these factors community college completion
rates were predicted to improve by 50% this would be very close to producing 1.75 million
graduates at 1,595.903 needed by community colleges to meet the stated goals of AGI. Being
the gateway to higher education it is known that AGI will disproportionately call on community
colleges to increase graduation rates. Kotamraju and Blackman (2011), believe this is possible if
community colleges focus on improving graduation outcomes for underrepresented students and
COMMUNITY COLLEGE ACCOUNTABILITY 22
first time post-secondary students, the development of central governance structures to meet the
needs of two-year institutions and increased global competitiveness.
Following the development of AGI, in August 2013 the Obama administration proposed
a Post Secondary Institution Rating System (PIRS), as a national rating system for institutions of
higher education across America (White House Office of Press Secretary, 2013). The goal of the
proposed national rating system is to increase institutional accountability for student outcomes
and to provide information to consumers on institutional performance, as a means of informing
student choice (NASFAA, 2014; White House Office of Press Secretary, 2013). The PIRS
system would rate institutions in relation to three areas a) access, as defined by the number of
students receiving Pell Grants, b) affordability, defined by the net price of the institution and the
percentage loan debt incurred by students, and c) outcomes, defined by graduation rates, transfer
rates, earnings of graduates, and completion of advanced degrees (US Department of Education,
2014). With PIRS, highly rated institutions would receive more federal financial aid in exchange
for improved outcomes in the above areas. Opponents of PIRS believe it is a "one size fits all"
approach to solving a complex problem of institutional accountability in higher education
(Espinosa et al., 2014). Additionally, it is believed that tying ratings to institutional funding will
cause greater barriers to low-income and minority students by decreasing their access to financial
aid and limiting college choice (California Community College Chancellor’s Office, 2014;
Espinosa et al., 2014). PIRS is one attempt to generate movement in the accountability debate
by developing a single rating system, but it is evident that a single rating system would be
problematic for institutions who have varying missions and roles within higher education, and it
may also cause unintended barriers for students in accessing financial aid.
COMMUNITY COLLEGE ACCOUNTABILITY 23
It is evident that accountability within higher education is an evolving process, for public
institutions this involves state, federal, and institutional mandates. The Obama Administration
has put forth AGI and PIRS in hopes of increasing institutional accountability and graduation
rates for students at both four-year and two-year institutions. Community colleges appear to be at
a disadvantage in obtaining reliable data to capture student outcomes as participation IPEDS is
voluntary, and the system also does not account for part-time students. Such limitations would
make it difficult for community colleges to measure student outcomes in a way that is
comprehensive.
Accountability in Community Colleges
Given community colleges varying institutional missions the role of accountability is a
multifaceted challenge. Many community colleges do not feel the present Obama
Administrations Post Secondary Institutional Rating System (PIRS) is inclusive of the diverse
accountability measures currently used in community colleges to define student success.
Furthermore, with the proposed American Graduation Initiative to increase the number of
college graduates in the United States, community colleges will be charged with increasing
access to degree achievement via the transfer route. One of the primary objectives of PIRS is to
allow for institutional comparison, to achieve that measures of institutional accountability must
be universal. Bailey and Xu (2012) conducted a meta-analysis of research aimed at improving
the quality and effectiveness colleges and universities and examined proxies of college
performance. They found that accountability models within higher education focused on three
areas student learning, employment, completion, and progression. Each of these will be
examined further to explore their role in institutional accountability within community colleges.

COMMUNITY COLLEGE ACCOUNTABILITY 24
Student Learning Outcomes
While not specifically unique to community colleges student learning outcomes (SLO’s)
play a significant role in institutional accountability within higher education. SLO’s were first
implemented in a handful of accrediting bodies in the 1980’s and became widespread in the
1990’s as a result of a federal mandate by the US Department of Education (Ewell, 2001). This
widespread advancement of SLO’s was largely to provide accreditation entities the ability to
define and provide sufficient evidence of student learning to stakeholders, to develop a common
assessment language amongst accrediting bodies, and to develop common resources to assess
institutions and programs (Ewell, 2001).
SLO’s were broadly defined by Ewell (2001), who provided the framework for
accrediting bodies to incorporate SLO’s into institutional evaluation, as “ …the particular levels
of knowledge, skills, and abilities, that a student has attained at the end (or as a result) of his or
her engagement in a particular set of collegiate experiences” (p. 14). SLO’s serve as a substantial
part of institutional accountability however, it is left to individual institutions to identify and
determine their own SLO’s within the context of their mission and programmatic content (Beno,
2004). SLO’s are both institution and program specific they often serve as a proxy for program
performance, effective teaching, and student learning but are not easily able to be aggregated for
institutional comparison (Bailey & Xu, 2012). Student learning is fundamental to institutional
assessment, however, given the variability of SLO's at community colleges it is not a sufficient
measure of institutional accountability, and it does not allow for comparison amongst
institutions.

COMMUNITY COLLEGE ACCOUNTABILITY 25
Wages and Employment
Increased employability and earnings are a significant by-product of completing some
form of higher education. Most states have a way of reporting student employment and earnings
post graduation to stakeholders using data from unemployment insurance (UI) benefits (Cunha &
Miller, 2012; US Department of Education, 2014). Conversely, Cunha and Miller, (2012),
stipulate there are limitations with data gathered from UI benefits, as it will be only able to track
jobs that offer UI, which, does not account for people who are self-employed and certain job
earnings are excluded, i.e., federal government and military jobs. These limitations may cause
individual earnings to be under-reported. Data gathered from UI benefits is limited to state
boundaries and unable to account for anyone who may have moved out of the state (Cunha &
Miller, 2012; US Department of Education, 2014). Given the variability of the job market in each
state there is no sound way to know if a student's employment is a result of the institution they
attended or intrinsic motivation (Bailey & Xu, 2012). Using UI data can provide insight to
student earnings post graduation but there are several barriers that prohibit this data from being a
reliable metric for performance evaluation.
The US Department of Education (2014), has received a wealth of feedback from the
higher education community regarding the use of post graduation student earnings. Some are
concerned with the adverse effects this could have on certain majors causing students to steer
clear of majors with lower earning potential. Contrary to that, there is also a concern of students
receiving well-paid jobs immediately after school based on job market demands and industry,
which would skew wages to be higher for students from certain schools. To combat some of
these concerns the US Department of Education (2014), has proposed they would use federal
data to capture employment and wages; this would allow employment outcomes to be examined
COMMUNITY COLLEGE ACCOUNTABILITY 26
across state boundaries. They are also examining ways to categorize data so that it is not simply
earnings being associated with particular occupations to avoid negatively impacting certain
majors. There is no specific mention to how this could impact community colleges, but one
could imagine that given that community college students are more likely to go on to some other
form of education tracking post-graduation employment and earnings would be a challenge.
Progression and Completion
Student progression and completion data are the most comprehensive data available that
are widely used across various institutions (Bailey & Xu, 2012). Boeke and Ewell (2007),
proposed a tool, Core Indicators of Effectiveness for Community Colleges to measure student
achievement by completion of identified milestones, which included intermediate progression
i.e. gateway courses (developmental math and English) or reaching terminal milestone like
transfer or degree completion. Ewell (2011), noted several challenges with progression measures
that do not make them the most consistent form of measurement for accountability purposes. The
first being leadership and planning, given the ever-changing environment of community colleges
internally evaluating institutional performance should be an on-going internal practice. Another
challenge with the core indicators is enrollment management. The broad range of students from
varied backgrounds enrolled in community colleges poses unique challenges with student
retention and degree completion (Ewell, 2011). An additional measure of student progression is
provided through IPEDS, Gradation Rate Survey. However, this survey only tracks the
progression of an entering student who is a first-time and full-time student (Ewell, 2011). This
poses the limitation of the variability in student’s enrolment at community colleges and does not
account for the significant number of students that transfer schools and often begin community
college enrolled part time (Ewell, 2011). Although the indicators for progression provide great
COMMUNITY COLLEGE ACCOUNTABILITY 27
insight into both student and institutional performance, the current shift in accountability is
calling for greater focus on student outcomes.
The focus on transfer rates as an accountability measure has been further emphasized as a
result of the AGI and PIRS rating system proposed by the Obama Administration. In addition,
long-term degree expectations among community college students are high, with 70% first-year
students expecting to earn a bachelor's degree or higher (Goldrick-Rab, 2010). In terms of
institutional accountability, transfer is also believed to be the most stable outcome to measure
student success (Bailey, Calcagno, Kienzl, & Leinbach, 2005). Finally, transfer rates provide the
most economic value to the state and the country. Research has demonstrated the inherent
economic value of degree attainment for the economy and an individual's lifelong earnings,
which cannot be measured by certificated or other courses that support lifelong learning at the
community college level (Bailey et al., 2005; Myers, 2007). While transfer is not viewed as the
primary role of community colleges, it has a significant role in the debate on institutional
accountability and student success.
Many of the accountability measures for community colleges are institutionally specific
and limit the ability of cross-institutional comparison. Student transfer rates and graduation rates
are the most consistent level of measurement across all community colleges that will allow for
such comparisons. In order to meet the global demand for increased accountability and improved
student outcomes, there will be a need for consistent measures of accountability.
California Community Colleges
California Community Colleges (CCC) is the largest community college system in the
nation. For CCC the push for increased accountability is likely to have a major impact, with
estimates of 46% of all students in the state attending community college and where more than
COMMUNITY COLLEGE ACCOUNTABILITY 28
70% of students enter unprepared for four-year college level work (Melguizo, 2007; Shaffer,
2008; Student Success Task Force, 2012). Given the variation in measuring accountability within
community colleges a further exploration of the accountability measures and barriers faced by
students within California Community Colleges was conducted.
CCC’s enroll approximately 73% of students pursing higher education in comparison to
18% that attend the California State University System (Melguizo et al., 2008). In California,
there is a growing Hispanic population, estimated at 38% in 2012 and predicted to be at 50% by
2040 (Moore & Shulock, 2010; US Census Bureau, 2014). The number of students enrolling in
CCC’s is continuously increasing and becoming progressively diverse. The increasingly diverse
student body has shifted efforts of community colleges to look beyond student outcomes and to
focus on equity in student outcomes.
Current Measures of Accountability in CCC
The California Community College Chancellor's Office (CCCCO) oversees the
performance of CCC's, via the Board of Governors (BOG). In 2010, the California Legislature
grew concerned about the low completion rates of CCC students and required the BOG to adopt
a plan to improve student success (Taylor, 2014). In 2012, the Student Success Task Force was
developed (SSTF) (Taylor, 2014). SSTF set out with twenty-two recommendations centered on
the following: improvement of placement test to accurately reflect skill level of entering
students, use of educational plans, hiring additional counselors, developing educational
roadmaps for students, enhancing professional development for faculty and staff, revising
financial systems to ensure resources are aligned with student success, increasing statewide
coordination, and increased alignment between CCC and local school districts (Student Success
Task Force, 2012). More importantly they have developed and began implementation of the
COMMUNITY COLLEGE ACCOUNTABILITY 29
student success scorecard, an accountability tool aimed at closing equity and achievement gaps
in student outcomes in CCC's. The implementation of the student success scorecard began in
2008-09 school year and tracked full-time students over a period of six years. The student
success scorecard measures progress toward momentum points and completion outcomes. The
momentum points are aimed at increasing the likelihood of student persistence toward
educational goals, and completion outcomes are degree completion, certificate or achieving
transfer readiness. The CCC has welcomed such plans, in order to ensure equitable outcomes for
all students and efficient use of resources to increase the number of students successfully
transferring to four-year universities.
The SSTF has made progress in fulfilling additional proposed recommendations
specifically they established policies regarding mandatory assessment and orientation for new
students entering community college (Taylor, 2014). The BOG has established deadlines for
students to define educational goals, established minimum academic standards for low-income
students to receive waivers, and established the student success program (Taylor, 2014). No
specific data exist yet to determine the impact if any, the above changes have on student
outcomes in CCC.
The SSTF is leading the charge to improve accountability within CCC, and it evident that
that as a result growing diversity in the state equity in student outcomes equity in student
outcomes is a one of the primary concern. Other than the development of the scorecard in which
community colleges share student progression toward educational goals and outcomes there is no
mention of the development of accountability measures that allow for fair institutional
comparison. The current measures do not account for the unique student population served at
each institution and how that may impact student outcomes.
COMMUNITY COLLEGE ACCOUNTABILITY 30
Barriers Students Face
In California, educational outcomes are consistent with national trends, in that we are in
danger of having a generation of young adults that are less educated than previous generations
(Moore & Shulock, 2010). Community colleges provide all students the opportunity to pursue
higher education regardless of their academic or individual background. In addition, it provides
an opportunity for students pursue a wide range of academic goals (transfer, degree completion,
or certificate completion). Given their open enrollment policies and the wide range of
educational goals, students enter community colleges with varying academic abilities. A
significant numbers of students who enter community college are academically unprepared for
college-level work (Goldrick-Rab & Pfeffer 2009). The primary challenges that student face in
reaching their educational goals at community colleges will be further explored.
Transfer
There is much debate about how to measure transfer rates and to determine if a student is
actively seeking the transfer route. According to CCCCO students demonstrating a "behavioral
intent" to transfer as demonstrated by a) having completed at least 12 units and b) enrolled in
transfer-level math or English (Bahr, Hom, & Perry, 2005). In California, 25% of degree-
seeking students achieve a degree or transfers to a four-year institution (Melguizo, 2007). To
meet the increased demands of the AGI proposed by President Obama community colleges will
have to increase the number of students successfully transferring to four-year institutions and
with a large portion of CCC students being low-income and underrepresented minorities it is
essential to examine data that examines equity in student outcomes.
Transfer is viewed as a primary role of community colleges, and while community
colleges provide increased access to higher education, it has not necessarily produced increased
COMMUNITY COLLEGE ACCOUNTABILITY 31
outcomes for all students (Goldrick-Rab, 2010). Previous research has consistently identified that
minority students fall behind their peers in transfer rates (Brown & Niemi, 2007; Jenkins, 2011;
Melguizo, 2007; Melguizo et al., 2008). Moore and Shulock (2010), conducted a study with
CCC's they tracked 255,253 first year students over a 6-year period, beginning in 2003-2004
school year. Students selected were defined as "degree-seeking" on the basis of enrolling in six
or more units in their first year. Student data was disaggregated into four broad racial categories;
White, Asian-Pacific Islander, Latino, African American. By the second year, 56% of the cohort
still attending indicating that almost half of the cohort had dropped out, with. In terms of transfer
outcomes 23% transferred to a four-year university, when disaggregated by race, 20% of
African-American and 14% Latino students transferred compared to 29% of White students and
24% Asian American and Pacific Islander students. Underrepresented minority students (URM)
were observed to have lower completion rates of significant milestones (i.e. remedial course
work and number of units completed). If community colleges continue to disaggregate data by
race, it will provide context to the barriers that URM face and aid in the development of
appropriate supports to increase transfer success among underrepresented groups.
Even given the modifications in the way transfer is defined, and measured URM still trail
behind their peers in transfer rates. Certain community colleges are more likely than others to
have a larger population of minority students based on community demographics. This would
likely have a direct impact on their transfer rates. Given the increasingly diverse population in
California understanding the trends of transfer rates and the transfer outcomes for minority
students will be beneficial to aid in the accountability efforts to close the achievement gap for
underserved students.

COMMUNITY COLLEGE ACCOUNTABILITY 32
Socioeconomic Status
While it is known that minority students trail behind their peers in overall transfer rates
there are several factors that impact student performance. Wassmer, Moore, and Shulock (2004),
utilizing transfer data from the first time freshman study conducted by the CCCCO they
examined institutional data based on two cohorts of first time freshman that were tracked over a
six-year period in 108 CCC’s. Students selected for the study previously demonstrated
“behavioral intent” to transfer. They found that after six years, even when they controlled for
academic preparation and socioeconomic status, institutions with higher percentages of Latino
and African American students had lower overall transfer rates. This study signified that the
student population impacts institutional outcomes, and underrepresented students face unique
barriers in their academic journey.
Porchea et al. (2010), conducted a national study of 21 community colleges and tracked
4,481 students over a five-year period between 2003 and2008. Data were gathered using the
Student Readiness Inventory a self-report Likert scale instrument that was given to students
during freshman orientation and in courses with heavy freshman enrollment. Porchea et al.,
analyzed the data using multinomial logic regression, which allowed them to estimate the
likelihood academic preparation, psychosocial, socio-demographic, situational, and institutional
variables would impact student's academic outcomes. Results indicated that parental income
(.23) and education level (.11) had a significant impact on student's ability to transfer or earn a
degree. In addition, situational factors (enrollment patterns, student expectations, and distance
from home to college) were found to be predictive of a student reaching transfer success or
degree completion. Enrollment patterns had a regression value of .60, thus full time students had
a greater likelihood of transferring. Limitations of this study are the intuitions that participated
COMMUNITY COLLEGE ACCOUNTABILITY 33
were located in 13 states, concentrated mostly in the Midwest, the average enrollment size was 2,
293 and the mean enrollment for African American and Hispanic students was 23. However, it
does provide an indication of how various factors impact student success in community colleges.
Bailey, Calcagno, Jenkins, Kienzl, and Leinbach (2005) explain that both institutional
and student factors impact student outcomes. Jenkins (2007) conducted a study in Florida to
measure student success, as defined by their ability to persist at their starting institution over the
course of three years or transfer to the four-year institution. Jenkins (2007) analyzed transcript
data of 150,000-degree seeking students who enrolled for the first time in one of Florida's 28
community colleges during the fall of 1998-2001. Researchers controlled for risk factors that
had been previously identified in the literature (enrollment status during the first semester,
socioeconomic status, and placement test). Jenkins (2007) found that community colleges with
greater institutional supports designed for minority students were more likely to succeed. While
this research was conducted in another state, it implicates the importance of both institutional
and student factors on completion rates.
An additional study that examines the impact institutional and student factors on student
transfer is Hayward (2011) transfer velocity project. This study explored how institutional
characteristics interact with student characteristics and behavior, and how those factors enhance
or impede student progression. Over a 9-year period, Hayward (2011) analyzed data for 147,207
students who entered CCC during the 1999-2000 school year. Hayward (2011), retrieved data
from three primary sources, the CCCCO, a central data repository for CCC, and transfer center
data. This study sought out to identify educational practices or policies that could be adjusted to
support student progression to reach transfer. Hayward (2011) found that several institutional
factors in CCC impact student transfer rates a.) The number of California State University's that
COMMUNITY COLLEGE ACCOUNTABILITY 34
the community college has established transfer admission agreements with b) number of students
that sign a transfer agreement with University of California (U.C. schools) and c) having a full
time transfer center director. Also, the number of transferable courses available at a particular
community college increased the likelihood of a student transferring. Due to community colleges
often being over-crowded course availability poses a significant barrier in the transfer process.
In examining student-level factors, Hayward (2011) found that race/ethnicity, academic
performance, and developmental coursework were associated with decreased chances of
transferring. Latino students were 27% less likely to transfer, African American 15% less likely
and White students were 18% less likely. Asian students were omitted as they were found to be
13% more likely to transfer because they were found to be positively associated with
transferring. There were also other student level factors associated with transfer like
developmental coursework and GPA. Students enrolled in basic skills math were 49% less likely
to transfer and students who held GPA above a 3.0 were 85% more likely to transfer. Hayward
(2011) identified that both institutional factors and student characteristics impact student
progression. However, he concludes that there are also several institutional factors that can be
remedied to close the achievement gap as it relates to transfer success for underrepresented
students such as, increasing the transfer culture on campus, building strong relationships with
high schools, strong relationships with four-year institutions, and having effective support
services on-campus.
Pacheco (2012) indicates that in an effort to identify institutional measures for
accountability purposes within community colleges, such measures must be reflective of the
unique student population that each community college serves. Pacheco (2012) utilized data
from the Accountability Reporting for Community Colleges (ARCC), a system that was
COMMUNITY COLLEGE ACCOUNTABILITY 35
approved by the California Assembly in 2004 to allow CCC to develop rigorous accountability
measures for institutional performance. The ARCC developed six metrics that were tied to the
mission of the community college system. These included: student progress and achievement
rate (SPAR), percentage of students who earned 30 units, persistence rate, annual successful
course completion rate for credit based vocational courses, annual successful course completion
rate for credit based basic skills courses, and basic skills improvement rate. Pacheco (2012)
gathered data on 106 community colleges to analyze the temporal stability of the six
performance indicators identified by the ARCC. Using multivariate correlation analysis, Pacheco
(2012) found that explanatory variables (students’ age, gender, ethnicity, race, college size, and
educational attainment of the community) are strongly correlated with institutional performance.
For example, SPAR had a multiple correlation of .861, which indicates a positive relationship
between student progress, completion, and explanatory variables. Pacheco (2012) concluded that
due to socioeconomic factors the performance measures identified by the ARCC did contain a
level of systematic error because it did not account for the effect of explanatory variables on
student outcomes. This speaks to the need for institutional accountability measurements within
community colleges to control for both institutional and student level input characteristics.
It is evident that both institutional and student level characteristics impact transfer
outcomes. In regard to transfer rates underrepresented students have been known to trail behind
their peers, with appropriate institutional supports it is believed that their transfer rates can be
improved. However, in order to reach such outcomes accountability metrics have to be aligned
with the identified factors that impact student performance socioeconomic status which includes
(income and educational attainment) and the unique barriers faced by URM.

COMMUNITY COLLEGE ACCOUNTABILITY 36
Developmental Education
Developmental education is a barrier for students pursuing transfer. According to Bailey,
Jeong, and Cho (2010), developmental education is lower division math and English courses that
students are required to take if their placement test reveals their academic skills are below
college level. Providing developmental coursework and English as a second language (ESL) is a
part of the community college mission however, these courses are non-credit bearing and slow
students’ academic progress (Howell, 2011). Bailey, Jeong, and Cho, (2010) conducted a meta-
analysis of data collected through the Achieving the Dream Initiative, a nationwide network of
community colleges focused on closing the achievement gap of low-income and minority
students. They examined student records for 256, 672 first-time degree-seeking students and
tracked them over three years between 2003 and 2006, during this time only 20% of students
completed the developmental math sequence, and 37% of students passed the developmental
English course. Failure to progress through developmental coursework creates a significant
barrier in student progression toward degree completion.
Developmental education also impacts the probability of students transferring to a four-
year university. Melguizo, Hagedorn, and Cypers, (2008), conducted a study in the Los Angeles
Community College District, which is one of the largest districts in the country. They surveyed
5,011 students who were enrolled in remedial coursework, 411 students from the initial sample
successfully transferred. On average students within the sample spent 50% of their time
(equivalent to one year), making up remedial courses. Their average length of enrollment in
community college was 5-years, but students subsequently transferred with the equivalent of one
year of full time credit. Once student transferred to four-year institutions they spent an
additional three to four to reach degree completion, thus totaling eight years to reach degree
COMMUNITY COLLEGE ACCOUNTABILITY 37
completion (Melguizo et al., 2008). Community college is often touted as the affordable option
for students, however facing the barriers in the transfer process overtime causes students incur
more cost and it decreases the likelihood they will transfer or reach degree completion.
In addition to analyzing transfer rates Moore and Shulock (2010) also examined the
impact of developmental education on student completion. Within two years 36% of students
complete college level English and 29% complete college level math. When disaggregated by
race White and Asian students were found to have higher completion rates, with 39% of White
student completing college level English and 30% completing college level math. For Asian
students 39% completed college level English and 38% completed college level Math. URM
were found to have lower completion rates, 33% of Hispanic students complete college level
English and 25% complete college level math. For African American students 26% complete
college level English and 17% college level math. Completion of college level coursework is
essential for both degree attainment and transfer. Within California completing college level
coursework is a barrier for URM, without addressing this barrier it is unlikely that California will
be able to close the achievement gap for URM.
Academic Preparedness
Community college students are more likely to enter college academically unprepared
and this leads to their placement in remedial coursework, 61% of students in community colleges
take a minimum of one developmental course and 25% take two or more developmental courses
(Goldrick-Rab & Pfeffer 2009). California Community colleges have a higher proportion of
students in remedial education with 70% of all students requiring developmental math (Brown &
Niemi, 2007). For students who enter one level below transfer level math, 46.2% achieve an
associate’s degree and for students who enter four levels below transfer level math this decreases
COMMUNITY COLLEGE ACCOUNTABILITY 38
to 25.5% (Student Success Task Force, 2012). When students enter college taking below college
level courses their chances of transferring significantly decrease, which decreases their chances
of obtaining a bachelor’s degree.
Minority students are more likely to have experienced substandard elementary and high
school environments, leading to increased chances of enrolling in developmental courses
(Melguizo et al., 2008). Academic preparation has been directly correlated with first year
performance and overall retention rates (Porchea et al., 2010). Using data from the National
Student Clearing House, Porchea et al. (2010) tracked 4,481 students over five years between
2003 and 2008 and found that 48% of students dropped out. The high number of students
entering college academically unprepared speaks to the need of more coordination between K-12
and community colleges.
Ethnicity
Several factors impact a student’s ability to transfer; lack of academic preparedness,
socioeconomic status, and institutional factors, but an underlying factor within all of these is
student ethnicity. Low income, minority, and first generation college students are more likely to
attend community college than four-year institutions (Long & Kurlaender, 2009). From the start
there are disparities in the college going process for minority students. Latino and African
American students are less likely to take SAT or ACT test, and scores from these test could be
used to aid in entering transfer level courses at the community college level (Melguizo, 2007).
Instead students opt to take the placement exam given at the local community college to
determine placement for their math and English courses. In California each community college
selects the type of placement test it uses, therefore there is no consistency in how students are
assessed (Melguizo, 2007). Poor performance on the placement test point to the larger issue of
COMMUNITY COLLEGE ACCOUNTABILITY 39
the lack academic preparation often experienced by Latino and African American students
(Melguizo, 2007; Melguizo et al., 2008; Moore & Shulock, 2010). Academic preparation plays a
significant role in the college going process for any student, but in particularly minority students
whose lack of preparation can cause barriers in reaching their academic goals.
At community colleges that have high rates of success in student transfers to highly
selective public four-year institutions (U.C. Berkeley & U.C. Los Angeles), Asian and Caucasian
students overwhelmingly represent the majority of students with a combined transfer rate of
70%, and less than 10% of Latino and less than 5% of African American successfully transfer
from these top feeder community colleges (Melguizo, 2007). While there are institutions known
for having high rates of transfers it still does not correlate with improved outcomes for African
American and Latino students. This speaks to the need for accountability measures that allow
institutions to examine outcomes that are disaggregated by race to ensure increased equity in
student outcomes and close achievement gap of underrepresented groups within higher
education.
Discussion Barriers Students Face
California community colleges are one of the largest systems in the nation, and there
have been recent developments to improve access and increase equity in student outcomes
(Student Success Taskforce, 2012), but there is still significant work to be done to close the
achievement gap faced by underrepresented students. In terms of transfer and academic
preparedness underrepresented students in CCC students are trailing behind their peers. By the
year 2040 California is destined be a majority minority state with the growing Hispanic
population estimated to be at 50%, closing the achievement gap of underrepresented students is a
matter of public self-interest (Dowd, 2003; Long & Kurlaender, 2009). Several studies have
COMMUNITY COLLEGE ACCOUNTABILITY 40
concluded there are strong correlation behind student outcomes, social economic status, and
racial/ethnic background (Melguizo, 2007; Moore & Shulock, 2010; Porchea et al., 2010). It is
important to develop accountability measures that take into account of the unique student
populations served with community colleges to allow for development of fair accountability
systems (Hayward, 2011; Pacheco, 2012). Continuous research has to be done to disaggregate
student outcomes by race in effort to address disparities in educational outcomes by
underrepresented students.
Problem
Institutional comparison is a part of the proposed PIRS rating system on which
institutional performance would be linked to student financial aid. This is likely to create a high
stakes environment within institutions of higher education. Many within in the higher education
community are growing concerned with how institutions will be selected and grouped for peer
comparison. In a recent reply to stakeholders within the higher education community the US
Department of Education (2014) gave no indication as to how institutions will be selected for
grouping other than surface institutional characteristics: degrees available, programs available,
and selectivity of the institution. This type of grouping is subjective in terms of outcomes
because it does not account for the diversity in the student body at each institution and how that
impacts institutional performance (Bailey & Xu; NAFAA, 2014). Institutional groupings should
be based on performance relative to their peer institutions (Miller, 2013; NAFAA, 2014).
Utilizing input-adjustment allows would support the development of peer groupings with like
institutions by controlling inputs that account for both student and institutional characteristics.
Input-adjusted scores allow institutions to adjust for external factors that impact student
success, but are beyond the control of the institution i.e. socioeconomic status and parental
COMMUNITY COLLEGE ACCOUNTABILITY 41
education (Bailey & Xu, 2012; NAFAA, 2014; Bahr, Hom, & Perry, 2005; Pacheco, 2012). The
use of input-adjusted scores as an accountability measure allows community colleges to be
evaluated in way that is fair and will also allow for institutional comparison with peer
institutions. In comparing like institutions it enables educational leaders and policy makers to
gain further understanding as to why a particular institution is underperforming or over
preforming in regard to a particular educational outcome.
Understanding institutional performance will ultimately allow community colleges to
utilize benchmarking to improve student outcomes. Baldwin, Bensimon, Dowd, and Kleiman
(2011), shared how the use of bench marking tools in a series work groups within community
colleges across the nation allowed institutions to engage in the process of creating institutional
change and equity in student outcomes. Their study was based on the work of Dowd (2005) the
various forms of benchmarking (performance, diagnostic, and process benchmarking) were used
as a means for community colleges to identify performance issues, set goals, and observe
practice of institutions identified as over preforming in regard to an area of practice. Having the
ability to employ similar strategies with transfer rates could be a useful tool to address issues of
equity in student outcomes.
As explored previously, a multitude of factors impact student outcomes (socio economic
status, ethnicity, academic preparedness, and institutional factors), however completion and
progression are believed to be the best measures available to compare performance among
institutions (Bailey & Xu, 2012). Pacheco (2012), measured the test-retest reliability of residual
scores (aka input-adjusted scores) of student progression as measured by completion of 30 units
and found the residual scores to be highly reliable with r>.69, p=.01. As it relates to completion
raw graduation rates do not account for the multiple factors that impact student graduation rates
COMMUNITY COLLEGE ACCOUNTABILITY 42
(Bailey & Xu, 2012), thus using input-adjusted scores would be a reliable measurement tool to
effectively measure student completion in a way that is fair.
Raw graduation rates merely explain the number of students that graduate from a
particular institution, but it does not address the fact that any one college may have more
academically prepared students and it is unfair to evaluate community colleges without
accounting for their unique student and institutional factors (Bailey & Xu, 2012). In addition,
using raw gain scores to measure annual changes in transfer rates are unreliable as they are
subject to sampling error due to constant changes in student enrollment and variability individual
student abilities (Cornbach & Furby, 1970). Input-adjusted scores will allow transfer rates to be
adjusted for individual student and institutional characteristics.
Conceptual Framework
In an effort to explain the disparities in educational outcomes experienced by students
from underrepresented minorities within higher education several researchers have explored
underlying theoretical constructs. Wassmer, Moore, and Shulock (2004), posits that such
outcomes are explained by Bourdieu’s (1973) Theory of Cultural Capital, Bourdieu’s Theory
explains how social class affects educational attainment due to parental expectations and parental
education. Under-represented students are more likely to be the first in their family to attend
college and may lack access to the social capital needed to navigate college or the transfer
process. Wassmer, Moore, and Shulock (2004) also highlight that minority students also are
more likely to be impacted by cultural factors that place a high emphasis on family and
community that may deter them from advancing academically due to familial responsibilities.
This is evidence that when examining student outcomes it is essential to examine how culture
impacts student progress.
COMMUNITY COLLEGE ACCOUNTABILITY 43
Beyond the challenges that students face individually, there are also institutional factors
that impact student outcomes. Argyris and Shon (1996) discuss how organizational cultures and
structures impact student individual student learning. Bensimon (2005) shares that closing the
achievement gap of minority student lies within “institutional actors” (faculty, counselors, and
administrators) and their ability to develop an equity mindset. Institutional actors must move
beyond viewing diversity as a general characteristic of the institution and take into consideration
the unique experiences and outcomes that minority students face. To become equity minded
institutional actors must take ownership of institutional barriers that impact student performance
(i.e. lack of resources or lack of transfer culture) and be accountable to eradicating such
outcomes (Bensimon, 2005; Harris & Bensimon, 2007). Support programs and transfer centers
have been shown to have a positive impact on student persistence (Melguizo et al., 2008).
However, Harris and Bensimon (2007) indicate that excessive development of support programs
can become indicators that equity in student outcomes are solely a student issue and beyond the
control of the institution.
Through Bordeau’s Theory of Cultural Capital (1973) it could be understood that the
outcomes of underrepresented students is a result of their lack social capital and that there are
unique cultural norms that prevent them from reaching educational goals within community
colleges. However, the role of institutions to understand the student population they serve and
utilize data to address inequities in student outcomes for underrepresented students is believed to
within the grasp of community colleges through the use of data. Community colleges can utilize
data disaggregated by race to gain and understanding of achievement gaps faced by
underrepresented students and take action to address these gaps by changing practice or policies
that impede the needs of the unique student population they serve.
COMMUNITY COLLEGE ACCOUNTABILITY 44
Summary
Accountability in higher education is center stage as the national effort to increase the
number of college graduates in the United States has called institutions of higher education
across the nation to action. Obama’s proposed Post Secondary Institutional Rating System is one
step to move toward increased accountability efforts and a focus on student outcomes within
higher education. While this is not perfect system there is much research that supports the need
for improved educational outcomes in particularly, for low-income and URM (Bensimon, 2005;
Chase et al., 2014; Harris & Bensimon, 2007). In California the development of the Student
Success Task Force (2012) is aiming to improve outcomes for underrepresented students in
community colleges but many of the recommendations from this task force are still in progress
of being implemented.
Without adequate intervention the achievement gap for underrepresented students in
California is slated to continue to increase (Moore & Shulock, 2010). The disparities in
educational outcomes for URM are apparent (Melguizo, 2007, 2008; Moore & Shulock, 2010;
Porchea et al., 2010; Student Success Taskforce, 2012), both Latino and African American
students have transfer rates below that of White students (Moore & Shulock, 2010). This is
evidence of a continued need to develop accountability measures within community colleges to
close the achievement gap as it relates to student completion.
Educational outcomes within community colleges are a result of both institutional and
student characteristics. It is essential for accountability measures to evaluate community
colleges in a way that is fair and accounts for the unique student population served (Bailey & Xu,
2012). Input-adjusted scores allow extraneous factors that impact student completion to be
COMMUNITY COLLEGE ACCOUNTABILITY 45
adjusted for. This study aims to utilize input-adjusted transfer rates as an accountability measure
within California Community Colleges.

COMMUNITY COLLEGE ACCOUNTABILITY 46
CHAPTER THREE: METHODOLOGY
In this study, the relationship between transfer rates and ethnicity will be explored as it
relates to institutional accountability in California Community Colleges. The purpose of this
study is to answer the following research questions:
1. Is there a relationship between ethnicity and transfer rates to four-year institutions?
2. Over the past two decades, what changes in transfer rates have occurred in California
Community Colleges for underrepresented minority students and non-underrepresented
minority students?
3. Are annual changes in transfer rates stable enough to be used for institutional
accountability purposes?
4. Are input-adjusted transfer scores stable enough to be used for institutional accountability
purposes?
Research Design
This study is a correlational study aimed at analyzing the relationship between transfer
rates and ethnicity in California Community Colleges. In this study, data were analyzed at the
institutional level, and predictor variables were identified to determine the rate of transfer of
minority students over the last twenty years. This information will then be used to examine the
stability of transfer rates as a measure of institutional accountability in California Community
Colleges.
The predictor variables in this study are student ethnicity, college size, percent non-
traditional students, percent Non-URM, socioeconomic status as measured by educational
attainment and income, and the institutions in this study, which were the California Community
Colleges during 1993-2013. The outcome variables for this study are transfer success, year-to-
COMMUNITY COLLEGE ACCOUNTABILITY 47
year change in transfer success, and input-adjusted transfer success. Additional details on the
measurement of each of these variables will be provided in the sections below.
Population
The unit of analysis for this study is institutional, and the target population for this study
is the California Community College System (CCCS). CCCS is the largest system of higher
education in the nation with 112 colleges throughout the state and serving more than 2 million
students (Student Success Taskforce, 2012). California community colleges serve three primary
purposes for students: transfer, career technical education, and basic skills (California
Community Colleges Chancellor's Office [CCCCO], 2014. For the purposes of this study, the
researcher is focusing on the goal of transfer.
In California, community colleges serve as a primary point of access to higher education
for many students, in particularly minority low-income students (Goldrick-Rab & Pfeffer, 2009).
During the 2013-2014 school year, the student body consisted of: Hispanic 38%, African
American 7%, Asian 11%, Filipino 4%, White 30% and other 5%. None of the colleges in this
study elected to be a part of the sample. This study is based solely on publicly available data
through the CCCCO’s website www.cccco.org and the California State University (CSU)
analytic studies website at www.csu.edu. This study is non-experimental in that no treatment or
random assignments were completed.
Instrumentation
Transfer data for the last twenty-nine years were collected and retrieved from the
CCCCO website (www.californiacommunitycolleges.cccco.edu) and through the CSU Analytic
Studies website (www.calstate.edu) and disaggregated by ethnicity. These raw transfer scores
COMMUNITY COLLEGE ACCOUNTABILITY 48
were adjusted to account for student socioeconomic status and parental level of educational
attainment. See Appendix (A) for data collection methodology.
Predictor Variables
The predictor variables in this study are student ethnicity, college size, percent non-
traditional students, percent non-URM, educational attainment, income, population density, and
the institutions in this study, the California Community Colleges during 1993-2013. Five
predictor variables were taken directly from a data set created by (Pacheco, 2012); college size,
percent non- URM, educational attainment, income, and population density.
Student ethnicity. CCCCO tracks race and ethnicity, and it is self-declared by students
in their admissions application. CCCCO tracks eight different ethnicities. This study focused on
five: Hispanic, African American, Asian, White and “other”. The ethnic classification was then
categorized by Non-URM (White, Asian, and Filipino) and URM (Hispanic and African
American students).
College Size. College size tracked by the CCCCO and defined by the number of full-time
equivalents (FTE) at an institution. FTE's are tracked by the Management Information Systems
(MIS) through CCCCO using enrollment data reported by individual colleges. For this study
college size was determined by a) retrieving FTE's for each college in the sample for the years
that Accountability Reporting System for California Community Colleges (ARCCC) has been in
place b) Computing the mean FTE's figure for the years under consideration. The college size
data were continuous, ratio data on the number of FTE's (rounded to the nearest hundredths
decimal place) (Pacheco, 2012). To enhance interpretation, college size was recoded into
quintiles.
COMMUNITY COLLEGE ACCOUNTABILITY 49
Non-traditional students. Non-traditional students were a category derived from
gathering student's age, which is self-reported based on their date of birth on their application for
admission. For this study categories were determined by a) retrieving the age figures for colleges
in the sample for the years that (ARCCC) has been in place; b) computing the mean for each
category; and c) placing the age groups into two categories "Traditional" and "Non-traditional"
students. Traditional students were defined as students under the age of 24, and non-traditional
students were considered students over the age of 25 (Pacheco, 2012).
Population density. Population density is an established measure through the USCB
(2010) and it is defined by the population of an area divided by the number of square miles of
land in the area (US Census Bureau, 2010). For this study the population density was determined
by (a) identifying the zip code associated with the principal address for the community college as
reported to CCCCO (b) matching the identified population density for the zip code area as
reported by USCB and (c) recoding and assigning the population density figure to each college
in the sample (Pacheco, 2012).
Percent non-URM. Percent non-URM was determined by using enrollment information
as reported to California State University Analytics Office and Asian, White, and Filipino
students. The total number of Non-URM students was then divided by the number of students
enrolled at each college.
Educational attainment. Educational attainment as defined by the USCB (2010), is
defined as the highest level of scholastic achievement that an individual has received. For this
study, educational attainment was determined by: 1) identifying the zip code associated with the
principal address for the college as reported to the CCCCO; 2) matching the zip code with the
identified levels of education achievement as noted by the USCB; and 3) recording and assigning
COMMUNITY COLLEGE ACCOUNTABILITY 50
the educational level figure to each college in the sample (Pacheco, 2012). In this study zip-
code education was scaled by:
1=less than 9
th
grade
2=9-12
th
grade (non-grad)
3= high school graduate
4= some college
6= associate degree
7= bachelor’s degree
The average level of education for neighborhoods in this study was 4.21 indicating some
college (See Figure 1). The figure for level of education was slightly positively skewed with
more neighborhoods having educational levels below “some college”.

Figure 1. Average Level of Education
Income. For this study the income variable was established by 1) identifying the zip code
identified with the principal address for the college as reported to the CCCCO; 2) matching the
COMMUNITY COLLEGE ACCOUNTABILITY 51
zip code with the median income for that area as reported by the USCB; and 3) recording and
assigning the population density figure to each college in the sample. The range of income within
this sample was $25,000 to $150,000, with an average income of $66,531.26. The figure for
income is positively skewed with disproportionately more people earning below the median
income of $66, 531.66 (see Figure 2).

Figure 2. Median Income
Outcome Variables
Success as defined by the Student Success Taskforce (2012) is the percentage of community
college students completing their academic goals, percentage of community college students
earning a certificate or a degree, transferring, or achieving transfer readiness, number of students
transferring to four-year institutions, and number of degrees and certificates awarded. For the
purposes of this study, I am focusing on the number of students transferring to California State
Universities.
The outcome variables in this study are transfer success rate, year-to-year change in transfer
rates, and input-adjusted transfer rates. The formula for transfer rates is: Number of transfers
COMMUNITY COLLEGE ACCOUNTABILITY 52
divided by number of students enrolled. Year-to-year change to transfer success is the annual
change in the number of students successfully transferring and is calculated by: previous year
transfers minus current year transfers.
In addition, student success scores will be adjusted using input-adjusted scores to account for
the six aforementioned control variables. Computation for input-adjusted scores will be as
follows:
1. Regress the raw transfer rates on socioeconomic status.
2. Compute the residual (actual - expected transfer rate), the residual is the input-adjusted
score.
Data Sources
The primary source of data for this study was the Management and Information Systems
(MIS) on the CCCCO website. MIS data are organized into five categories: student/headcounts,
student services, faculty and staff, courses/calendar, and outcomes. Demographic and enrollment
data were gathered from the student/headcounts portal that provides current enrollment data for
all students as recently as winter 2013. The outcomes category provided information on nine
outcomes that are tracked by the community college system: Basic skills tracker,
retention/success rate, system wage tracker, grade distribution, program awards, college wage
tracker, student success score card metrics, transfer velocity and transfer volume.
Within the outcomes, category transfer volume was used to retrieve data used in this
study. Transfer volume refers to number of students transferring from a two-year to a four-year
university (CCCCO, 2014). Data for in state transfers to public four-year universities were
retrieved through the California State University Analytic Studies (CSU Analytic Studies), and
these data spanned over the last twenty years.
COMMUNITY COLLEGE ACCOUNTABILITY 53
Input-Adjusted Transfer Success Scores
Student progression and degree completion remain to be the primary measure of
institutional effectiveness at the community college level (Bailey & Xu, 2012). Input-adjusted
rates control for individual student and community characteristics, controlling for such factors
allows reasonable comparisons to be made between institutions (Bailey & Xu, 2012; Pacheco,
2012). Utilizing input-adjusted rates at the institutional level allows for community colleges to
account for challenges that are “beyond their control”, in a way that is fair and representative of
the student population and the community it serves.
The adjusted transfer rates are the outputs once the raw transfer scores are adjusted to
account for extraneous factors. This study explored how these extraneous factors impact a
student’s ability to transfer. The input-adjusted transfer rates were then examined in relation to
institutional accountability.
Limitations
This study is limited in that it cannot provide a causal explanation for the transfer rates
identified at the various institutions in the study. The data used can only provide speculation as
to what the causes are, and much further experimental research would have to be completed at
the institutional level to gain understanding of specific causes.
Delimitations
There are also delimitations with this study. The data used in this study are limited to
community colleges in California, which limits this studies ability to provide insight or
generalizations to transfer rates in other states or nations. California is also one of the largest
community college systems, which could also limit the applicability of findings in states with
smaller community college systems. Another delimitation is that this study examines the transfer
COMMUNITY COLLEGE ACCOUNTABILITY 54
rates of all students in California Community Colleges and is not disaggregated students by an
academic goal of having the "intent to transfer". Colleges are working to clarify the definition of
a student having the “intent to transfer” and it should be considered in other studies that examine
transfer rates. Finally, the use of the treatment (input-adjusted scores) as an identified
accountability measure for transfer rates is limited to the schools in which it was applied to in
California. Further research would have to be conducted in community colleges in other states
using input-adjusted scores as a measurement of accountability to know if community colleges in
other states yield similar outcomes.

COMMUNITY COLLEGE ACCOUNTABILITY 55
CHAPTER FOUR: RESULTS
Overview
The purpose of this study is to examine the relationship between ethnicity and transfer
rates in California Community Colleges over the last twenty years and examine the use of input-
adjusted scores as an accountability model. To address research questions related to this study
the following statistical procedures were conducted: a) summary descriptive statistics b)
regression analysis c) the degree, directionality, and strength of relationships between the
outcome and control variables were assessed in a correlation analysis.
Findings
Transfer Rates by Ethnic Classification
Research Question One: Is there a relationship between ethnicity and transfer rates to
California State Universities?
Figure 3 shows the average transfer rate from California Community Colleges (CCC) to
California State Universities (CSU) over the last twenty years disaggregated by ethnicity. It is
evident that both Hispanic and African American students trail behind their peers, the average
transfer rates for Hispanics was 1.5% and for African American students 1.3%. Filipinos
outperformed both Asians and Whites with an average transfer rate of 2.2%. Asian students
transfer rates nearly mirrored that of Whites with an average of nearly 2% whereas Whites
averaged 1.9%.

COMMUNITY COLLEGE ACCOUNTABILITY 56

Figure 3. Number of Transfers Disaggregated by Race
URM vs. Non-URM
Research Question Two: Over the past two decades, what changes in transfer rates have
occurred in California Community colleges for underrepresented minority students and non-
underrepresented minority students?
Ethnic groups were combined to form URM and Non-URM groups. URM consist of
African American and Hispanic students and non-URM consists of White, Asian, and Filipino
students who have been known to consistently achieve above average educational outcomes.
Figure 4 shows the percentage of transfers for URM and Non-URM students from CCC to
CSU’s over the last twenty years. It is evident that URM students consistently transfer at lower
rates than non-URM students with their highest transfer rate being 1.9% in 1997 and the highest
rate of transfer for Non-URM students at 2.3% in 2011. There was a significant drop in the rate
of transfers by all students in 2010. Interestingly, the 2013 transfer rates for non-URM students
are almost identical to the 1993 rates.
0
0.005
0.01
0.015
0.02
0.025

White
Filip
Asian
PI
AI
Hisp
AA

COMMUNITY COLLEGE ACCOUNTABILITY 57

Figure 4. Number of Transfers for URM and Non-URM Students
Figure 5 depicts the gap annual transfer rates experienced between URM and non-URM
students. The transfer rate for URM students typically trails behind non-URM students at rates
that range from .3% to .6%.

Figure 5. Percent Transfer Gap for URM Students Versus Non-URM

0

0.005

0.01

0.015

0.02

0.025

1993
1995
1997
1999
2001
2003
2005
2007
2009
2011
2013

Transfer
Rate

Years

URM
Non-‐URM

0

0.001

0.002

0.003

0.004

0.005

0.006

0.007

URM
transfer
gap

COMMUNITY COLLEGE ACCOUNTABILITY 58
Stability of Annual Changes in Transfer Rates
Research Question Three: Are annual changes in transfer rates stable enough to be used for
institutional accountability purposes?
Annual total transfer rates were aggregated by year; the annual transfer mean scores were
then used to subtract the one-year mean transfer rates from the previous year to compute change
scores. The change scores were then correlated from year to year (Table 1). Table 1 shows that
the majority of change scores for various CCC’s had a negative relationship with the previous
year’s score and only one year, 2000-2001, had a gain score that was moderately correlated with
the previous year r=.380. Change scores have been noted to be unreliable for several reasons
Cronbach and Furby (1970), report that change scores or “gain scores” are subject to random
error of measurement. Within the context of community colleges there is measurement error due
to the variability in individual abilities of cohorts and student enrollment. Using such scores can
lead to the false conclusion that the community colleges caused the change in scores. Overall the
results show that annual change scores are clearly not stable enough for accountability purposes.

COMMUNITY COLLEGE ACCOUNTABILITY 59
Table 1
Year to year Correlation Between Prior and Present Change Scores
Year r
2012-13 -0.613
2011-12 -0.337
2010-11 -0.526
2009-10 -0.275
2008-09 -0.080
2007-08 -0.158
2006-07 -0.326
2005-06 -0.392
2004-05 -0.266
2003-04 -0.115
2002-03 -0.680
2001-02 -0.691
2000-01 0.380
1999-00 0.005
1998-99 -0.363
1997-98 -0.128
1996-97 -0.278
1995-96 -0.503
1994-95 -0.397

Input-Adjusted Scores
Research Question Four: Are input-adjusted transfer scores stable enough to be used for
institutional accountability purposes?
Using raw graduation rates to determine student success within community colleges does
not take into the account the “inputs” that impact such outcomes. Student outcomes are largely a
result of student characteristics, institutional resources, and instructional processes (Bailey & Xu,
2012). Pacheco (2012) used residual scores as a measure to explore if systematic error had an
influence on accountability reporting for California community colleges and found that input-
adjusted scores were found to adequately control for the external variables that impact student
completion.
COMMUNITY COLLEGE ACCOUNTABILITY 60
Descriptive statistics were analyzed to examine both predictor and outcome variables.
The predictor variables in this study are educational attainment, percent Non-URM, population
density, income, percentage non-traditional students, and college size. The outcomes variables
were total transfer rates. To determine the stability of input-adjusted scores as an accountability
measure for CCC transfers rates input-adjusted scores were analyzed for the last five years 2009-
2013. Table 2 displays the mean and SD for each of the predictor and outcome variables.
Table 2
Control Variables and Annual Transfer
Descriptive Statistics
Mean SD N
Educ Attainment 4.2049 0.74657 106
pct_nonurm2013 0.4949 0.17481 118
Pop Dens 2010 4149.3 4809.377 106
Income 2010 66531.26 25224.415 106
Non-traditional 0.4793 0.10174 106
College Size 3.0000 1.26491 106
Total2009rate 0.016844 0.0073846 109
Total2010rate 0.012685 0.0059307 110
Total2011rate 0.020082 0.0091966 110
Total2012rate 0.019863 0.0091269 111
Total2013rate 0.01762 0.0082775 113

To examine the relationship among predictor and control variables correlations were
computed using Pearson’s r. Table 3 indicates the correlations between predictor variables.
Positive correlations were found between educational attainment and percent on non-URM
students r=.483, indicating schools with larger non-URM student populations are in locations
that have higher rates of educational attainment. A positive correlation was found between
educational attainment income r=.756, which indicates schools in higher socio economic
locations have higher levels of educational attainment in the same location. The location’s
income was also moderately correlated with percent non-URM, r=.328.
COMMUNITY COLLEGE ACCOUNTABILITY 61
Table 3
Correlations Among Predictor Variables

Table 4 displays Pearson’s r correlations among outcome variables (transfer rates 2009-
2013). The correlations range from .766 to .906. As expected these rates are highly and
positively correlated indicating that institutional transfer rates are very stable from year to year.

Correlations

Educational
Attainment
Percent
Non-URM
2013
Pop Dens
2010
Income
2010
Non-
traditional
students
College
Size
Educ Attainment
Pearson
Correlation 1 .483** -0.037 .756** -0.034 .193*

Sig. (2-tailed)

0 0.706 0 0.729 0.048
Pct_nonurm2013
Pearson
Correlation .483** 1 -.209* .328** 0.179 -0.032

Sig. (2-tailed) 0

0.032 0.001 0.066 0.746
Pop Density
2010
Pearson
Correlation -0.037 -.209* 1 -0.156 -0.072 .201*

Sig. (2-tailed) 0.706 0.032

0.111 0.461 0.038
Income 2010
Pearson
Correlation .756** .328** -0.156 1 -0.069 .211*

Sig. (2-tailed) 0 0.001 0.111

0.483 0.03
Non-Traditional
Pearson
Correlation -0.034 0.179 -0.072 -0.069 1 -.387**

Sig. (2-tailed) 0.729 0.066 0.461 0.483

0
College Size
Pearson
Correlation .193* -0.032 .201* .211* -.387** 1

Sig. (2-tailed) 0.048 0.746 0.038 0.03 0
N 106 106 106 106 106 106
COMMUNITY COLLEGE ACCOUNTABILITY 62
Table 4
Correlations Among Outcome Variables
Correlations

Total2009
rate
Total2010
rate
Total2011
rate
Total2012
rate
Total2013
rate
Total2009rate
Pearson
Correlation 1 .848** .860** .890** .856**

Sig. (2-tailed)

0 0 0 0

N 109 109 109 109 109
Total2010rate
Pearson
Correlation .848** 1 .827** .766** .837**

Sig. (2-tailed) 0

0 0 0

N 109 110 110 110 110
Total2011rate
Pearson
Correlation .860** .827** 1 .869** .906**

Sig. (2-tailed) 0 0

0 0

N 109 110 110 110 110
Total2012rate
Pearson
Correlation .890** .766** .869** 1 .889**

Sig. (2-tailed) 0 0 0

0

N 109 110 110 111 111
Total2013rate
Pearson
Correlation .856** .837** .906** .889** 1

Sig. (2-tailed) 0 0 0 0
N 109 110 110 111 113
Note *p=.05 **p=0.01

Table 5 displays the Pearson’s r correlations for both predictor and outcome variables.
Educational attainment and percent of Non-URM students were not found to be significantly
correlated with transfer rates to CSU’s. Having large numbers of non-traditional students was
negatively correlated with transfer rates r=-.682, thus schools with larger number of non-
traditional students are more likely to have lower transfer rates. In addition, college size was
positively correlated with transfer rates r= .438, larger CCC’s are likely to have higher transfer
rates to CSU’s.

COMMUNITY COLLEGE ACCOUNTABILITY 63
Table 5
Correlations Among Outcome and Control Variables
Correlations

Total2009
rate
Total2010
rate
Total2011
rate
Total2012
rate
Total2013
rate
Educ Attainment Pearson Correlation 0.089 0.015 0.146 0.111 0.13

Sig. (2-tailed) 0.366 0.881 0.137 0.26 0.186

N 105 105 105 105 105
pct_nonurm2013 Pearson Correlation 0.087 0.045 0.109 0.067 .186*

Sig. (2-tailed) 0.369 0.64 0.255 0.486 0.05

N 109 110 110 111 112
Pop Dens 2010 Pearson Correlation -0.055 -0.078 -0.034 -0.097 -0.095

Sig. (2-tailed) 0.576 0.429 0.729 0.327 0.333

N 105 105 105 105 105
Income 2010 Pearson Correlation 0.17 0.102 .237* 0.158 .197*

Sig. (2-tailed) 0.083 0.299 0.015 0.107 0.044

N 105 105 105 105 105
Non-Traditional Pearson Correlation -.738** -.724** -.706** -.713** -.682**

Sig. (2-tailed) 0.00 0.00 0.00 0.00 0.00

N 105 105 105 105 105
College Size Pearson Correlation .433** .386** .502** .457** .438**

Sig. (2-tailed) 0.00 0.00 0.00 0.00 0.00
N 105 105 105 105 105
Note *p=.05 **p=0.01

Stability of Input-Adjusted Scores
Input-adjusted scores are the difference between a community colleges actual transfer
rates and it’s predicted transfer rates. Predicted transfer rates were obtained by regressing the
transfer rates on each of the six-predictor variables: Educational Attainment, percent of Non-
URM, population density, income, non-traditional students (age>25), and college size. For 2013
the multiple correlation (R=.776) was statistically significant, F(6, 98)=24.67, p=.001. The
adjusted R-square was .577; thus 57.7 percent of the variance in the 2013 transfer rates was
explained by the six-predictor variables.
COMMUNITY COLLEGE ACCOUNTABILITY 64
Table 6 shows the 2013 results for each of the predictor variables. The significant
predictors of community college transfer rates are percentage of Non-URM, non-traditional
students, and college size. Specifically higher predicted transfer rates are associated with larger
percentage Non-URM students, fewer non-traditional students, and larger community colleges.
Table 6
Regression Coefficients for Total 2013 Transfer Rate

Unstandardized
Coefficients
Standardized
Coefficients

B
Std.
Error Beta t Sig.
Educ Attainment -0.002 0.001 -0.155 -1.450 0.150
pct_nonurm2013 0.013 0.003 0.294 3.901 0.000
Pop Dens 2010 -2.04E-07 0.000 -0.122 -1.763 0.081
Income 2010 3.53E-08 0.000 0.111 1.089 0.279
Non-Traditional -0.051 0.006 -0.650 -9.203 0.000
College Size 0.001 0.000 0.230 3.143 0.002

For illustrative purposes Table 7 shows the computation of input-adjusted scores for the five
highest preforming community colleges. It displays the schools actual transfer rates during the
2012-2013 school year, their predicted transfer rate and the residual scores (aka input-adjusted
scores).
Table 7
Highest Preforming Community Colleges
Residual Scores
School Actual Predicted Residual
Hartnell 0.0268 0.0117 0.0150
Orange Coast 0.0430 0.0298 0.0131
Butte 0.0368 0.0248 0.0118
Napa Valley 0.0241 0.0130 0.0110
Cuesta 0.0338 0.0237 0.0100

Table 8 and 9 show the regression results for the remaining four years. These results
clearly parallel those reported for 2013.
COMMUNITY COLLEGE ACCOUNTABILITY 65
Table 8
2009-2012 Regression Analysis
R Adjusted R-Square F (df
1
,df
2
) Observed Prob.
2009 Transfer Rate 0.802 0.622 29.52(6,98) 0.001
2010 Transfer Rate 0.770 0.569 23.86(6,98) 0.001
2011 Transfer Rate 0.798 0.614 28.59 (6,98) 0.001
2012 Transfer Rate 0.775 0.576 24.54(6,98) 0.001

Table 9
2009-2012 Regression Analysis
2012

2011
Beta t Sig Beta t Sig
Educ Attainment -0.057 -0.527 0.599 -0.190 -1.836 0.069
pct_nonurm 0.159 2.083 0.04 0.255 3.495 0.001
Pop Dens 2010 -0.161 -2.326 0.022 -0.060 -0.906 0.367
Income 2010 0.030 0.292 0.771 0.189 1.947 0.054
Non-Traditional -0.654 -9.249 0.001 -0.642 -9.507 0.001
College Size 0.248 3.383 0.001 0.268 3.845 0.001

2010 2009
Beta t Sig Beta t Sig
Educ Attainment -0.216 -1.979 0.051 -0.188 -1.871 0.064
pct_nonurm 0.210 2.707 0.008 0.263 3.713 0.001
Pop Dens 2010 -0.107 -1.535 0.128 -0.074 -1.136 0.259
Income 2010 0.095 0.927 0.356 0.130 1.352 0.179
Non-Traditional -0.705 -9.893 0.001 -0.716 -10.691 0.000
College Size 0.151 2.041 0.044 0.179 2.591 0.011

Table 10 depicts the stability of residual scores for 2009-2013. Residuals scores re
moderately stable with year over year correlations ranging from r=.590 to .748.

COMMUNITY COLLEGE ACCOUNTABILITY 66
Table 10
Stability of Input-Adjusted Scores 2009-2013

Residual
2009
Residual
2010
Residual
2011
Residual
2012
Residual
2013
Residual 2009 Pearson Correlation 1 .590** .652** .773** .667**

Sig. (2-tailed)

0 0 0 0
Residual 2010 Pearson Correlation .590** 1 .603** .476** .667**

Sig. (2-tailed) 0

0 0 0
Residual 2011 Pearson Correlation .652** .603** 1 .681** .777**

Sig. (2-tailed) 0 0

0 0
Residual 2012 Pearson Correlation .773** .476** .681** 1 .748**

Sig. (2-tailed) 0 0 0

0
Residual 2013 Pearson Correlation .667** .667** .777** .748** 1

Sig. (2-tailed) 0 0 0 0
N 105 105 105 105 105

COMMUNITY COLLEGE ACCOUNTABILITY 67
CHAPTER FIVE: DISCUSSION
The American Graduation Initiative (AGI) would require the number of bachelorette
degrees conferred to increase by 16% annually (Kotamraju & Blackman, 2011). To meet the
goals proposed through AGI, the Obama Administration proposed a national ranking system for
institutions of higher education, the Post-Secondary Institutional Rating System (PIRS). PIRS
would rank institutions as it relates to access, affordability, and student outcomes. Institutions
would be rated in comparison with like institutions, thus the development of peer groups will be
essential.
Community colleges lie at the center of the debate of accountability as they serve as open
access institutions and serve more than 6 million students across the United States (The White
House, 2015). Community colleges provide 50% of the post-secondary degrees awarded via the
transfer route, and to meet the proposed goals of AGI community colleges would have increase
completion rates from 1.5 million to 1.75 million per year (Kotamraju & Blackman, 2011). To
increase the number of graduates in the United States increasing completion rates within the
community college sector is essential.
The California Community College system is one of the largest systems in the nation and
located in a state that is also becoming increasingly diverse. Completion rates at community
colleges have been known to be historically low, in particularly for low-income and minority
students (Melguizo, 2007; Melguizo et al., 2008; Moore & Shullock, 2010). Given the national
context to increase institutional accountability and completion rates within community colleges
this study seeks to answer the following questions:
1. Is there a relationship between ethnicity and transfer rates to four-year institutions?
COMMUNITY COLLEGE ACCOUNTABILITY 68
2. Over the past two decades, what changes in transfer rates have occurred in California
Community Colleges for underrepresented minority students and non-underrepresented
minority students?
3. Are annual changes in transfer rates stable enough to be used for institutional
accountability purposes?
4. Are input-adjusted transfer scores stable enough to be used for institutional accountability
purposes?
Discussion
Filipino students experience the highest transfer rates (2.2% average) more than Asian
(2% average) and White (1.9% average) students. African American and Hispanic have
maintained lower transfer rates than their peers and this has remained relatively consistent over
the last twenty years. As observed in previous studies (Jenkins, 2007; Melguizo, 2007; Moore &
Shulock, 2010; Porchea, Robbins, & Phelps, 2010) there is a relationship between ethnicity and
transfer rates. African American students faced the lowest annual transfer rates to CSU's with
1.3% being their average rate of transfer over the last twenty years. The low rate of transfer
experienced by URM within this study is similar to previous studies (Moore & Shulcok, 2010;
Melguizo, 2008), indicating that race is still a factor in student reaching successful outcomes.
Moore and Shulock, (2010), indicate that African American students were the least likely
minority group to transfer to CSU's and over a six-year period 26% of African American
students transferred. They found that African American students had higher rates of transfer to
out of state public and private universities. It is possible that their choice to attend out of state or
private institutions could impact the transfer rates seen in this study. Moore and Shulock (2010),
also state 50% of Latino students transferred to a CSU and indicated they were most likely to
COMMUNITY COLLEGE ACCOUNTABILITY 69
transfer to a CSU. However, within this study the average annual transfer rates for Latino
students was 1.6%. It is evident that the way transfer is measured has a direct impact on how and
what outcomes are reported as it relates to transfer.
Over the past two decades, the transfer rates of URM students have consistently trailed
behind non-URM students. The highest annual transfer rate for URM students in the present
study was in 1997 (1.9%), and the highest rate of transfer for non-URM students was in 2013
(2.3%). In 2010, there was a significant decrease in the number of transfers experienced by all
students, due to the recession and budget cuts, that impacted both enrollment and course
offerings at all community colleges (Taylor, 2014). During that year non-URM students
transferred at a rate of 1.4% and URM students transferred at a rate of 1.1%. However, the
following year the transfer rate for non-URM students nearly doubled to 2.3% and the transfer
rate for URM students increased to 1.7%. With the exception of the 2009-2010 school year there
have not been any other significant changes in transfer rates for URM or non-URM students, but
it is evident that URM students continue to transfer at rates that are lower than their peers. Most
importantly, the overall transfer rates in 2013 are not much different than they were in 1993.
It is evident that raw transfer rates cannot be used for accountability purposes because
they are not correlated with factors beyond the institutions control. Thus most educators advocate
looking at annual change scores. Annual changes in transfer rates are considered “difference
scores” such scores are not reliable enough to be used for accountability purposes. The
variability in community college enrollment and student characteristics create sampling error,
thus further causing variability in year-to-year gain scores. Within the present study, when
disaggregated by year with the exception of one year (99-00), gain scores were not positively
COMMUNITY COLLEGE ACCOUNTABILITY 70
correlated with the previous year. Thus raw gain scores are not stable enough to be used as a
measure of accountability for community colleges.
Within higher education student completion is believed to be the most stable measure of
accountability that will allow for institutional comparison for accountability purposes. However,
given the variability in student populations served, using raw transfer rates has been established
as insufficient for institutional comparison (Bahr et al., 2005; Bailey & Xu, 2012; Pacheco,
2012). Student completion or in this case transfer is believed to be a reflection of both
institutional and student inputs. In this study, the inputs were income, educational attainment,
college size, non-traditional students, and percent of non-URM students. The percentage of non-
URM students, college size, and the number of non-traditional students were positively
correlated with increased transfer rates. Larger schools, schools with less non-traditional
students, and fewer URM students were more likely to have higher transfer rates. The impact of
student age on completion rates and the percentage of non-URM students correlates with
previous studies (Bahr, Hom, & Perry, 2005; Porchea, Robbins, & Phelps, 2011; Wasmer,
Moore, & Shulock, 2004). However, there are conflicting perspectives regarding the correlation
between college size and student completion, within this study it is positively correlated others
have found it to negatively impact student completion rates (Bailey et al. 2005). This further
solidifies the impact extraneous facts have on student outcomes. This is an important factor to
consider in the development of PIRS and desire to create peer group comparisons.
Because raw gain scores are not reliable enough to be used for accountability purposes
input-adjusted scores are a potential alternative. The year over year stability coefficients for
input- adjusted scores between 2009-2013 range from r= .590 to .748 indicating that input-
COMMUNITY COLLEGE ACCOUNTABILITY 71
adjusted scores are stable enough to be used for accountability purposes. This also correlates
with a previous study conducted by Pacheco, (2012).
Based on input-adjusted scores Table 11 depicts the top ten over performing community
colleges in regard to annual transfers to California State Universities for 2012 and 2013.
Table 11
Over-Performing CSU Transfer Colleges
College Name Residual 2012 Residual 2013
Hartnell 1.98 2.88
Orange Coast 1.80 2.51
Butte 0.92 2.27
Napa Valley 1.38 2.11
Cuesta 0.81 1.91
Porterville 0.25 1.44
Golden West 0.82 1.37
Redwoods 0.72 1.37
Fullerton 1.93 1.25
Evergreen Valley 1.47 1.12

Table 12 displays institutional and student characteristics for the over performing colleges during
the 2012-13 school year.
Table 12
Over Performing Colleges’ Characteristics

College Size
2013 Non-Traditional 2013
URM
2013
Hartnell 13,074 39% 65%
Orange Coast 26,629 29% 32%
Butte 16,924 38% 18%
Napa Valley 8,996 41% 38%
Cuesta 14,686 32% 29%
Porterville 4,987 34% 68%
Golden West 15,796 28% 29%
Redwoods 7,285 34% 16%
Fullerton 27,204 25% 50%
Evergreen Valley 14,920 47% 37%

COMMUNITY COLLEGE ACCOUNTABILITY 72
Table 13 depicts the community colleges that are under-performing in terms of annual
transfers to California Community Colleges for 2012 and 2013.
Table 13
Underperforming CSU Transfer Colleges
College Name Residual 2012 Residual 2013
Barstow -.849 -1.49
Copper Mountain -1.09 -1.53
LA Trade Tech -1.71 -1.53
San Diego Mesa -2.47 -1.53
Santa Barbara City -1.55 -1.62
Victor Valley -1.21 -1.66
Foothill -1.83 -1.79
Sacramento City -1.44 -1.82
Mt. San Antonio -1.80 -1.96
LA Mission -1.36 -2.02

Table 14 displays the institutional and student characteristics for the under-performing
colleges during the 2012-2013 school year. Over half of the under-performing colleges have
more the 40% non-traditional students and at least half of them have more than 50% URM
students. The institutional characteristics of underperforming colleges correlate with the findings
within this study on the impact of percent of non-URM students and number of non-traditional
on student outcomes.
Table 14
Underperforming Colleges’ Characteristics
College Size 2013 Non-Traditional 2013 URM 2013
Barstow 5,152 50% 51%
Copper Mountain 3,103 46% 31%
LA Trade Tech 23,129 51% 83%
San Diego Mesa 32,779 39% 36%
Santa Barbara City 26,257 31% 37%
Victor Valley 16,168 41% 55%

COMMUNITY COLLEGE ACCOUNTABILITY 73
Table 14, continued
Foothill 27,521 46% 24%
Sacramento City 34,389 40% 38%
Mt. San Antonio 53,829 34% 56%
LA Mission 13,559 38% 77%

Research was conducted to gain publicly available information to form possible
hypotheses of reasons colleges landed within the underperforming category (see Table 15).
Table 15
Additional Information on Underperforming Institutions
Barstow Predominately serve military families and 45% non-traditional
students. Completion rate for remedial math (31.9%) English (36%).
Majority HSI (Hispanic students-37.5%).
Copper Mountain Transfer /completion rates for unprepared college students 22.8%.
High number non-traditional students (31.2%) and students over 40
(14.7%.)
LA Trade Tech Predominately serves non-traditional students (34.5%). Completion
rates of remedial math (8%) and English (19.3%). Serve majority
URM students (Hispanic 58% and African American 28%)
San Diego Mesa Completion rates for remedial math 32.2%. Recently began offering
four-year degree health info management. 30% non-traditional
students
Santa Barbara City Significant number of students enter college prepared (78.2%) may
have higher transfer rates to local U.C. Santa Barbara.
Victor Valley Hispanic serving institution (46%). Transfer/completion rates Hispanic
students unprepared for college (29%). General completion rate for
English (42%) and Math (39.6%) for all students
Foothill Significant number of Non-URM students White (35.5%) Asian
(23.9%). Completion rate for math (43.2%) Could have higher transfer
rate to U.C.'s.
Sacramento City Transfer/completion rates for students unprepared for college 41%.
Completion rate for remedial math 21.2% and English 38.5%
Mt. San Antonio Completion rate of remedial math 31.3%. Completion/degree transfer
for unprepared college students 41.4%.
LA Mission HSI (76.2%). Completion rate for remedial math 29%.
Transfer/completion rate for unprepared college Hispanic students
(30.7%)
HSI=Hispanic serving institution
COMMUNITY COLLEGE ACCOUNTABILITY 74
Further exploration on few institutional websites led to additional information that is
beneficial to consider when looking at the rankings. For example, LA Trade Tech predominately
serves URM students, and developmental education appears to be a challenge for Hispanic and
African-American students. When tracked over a six-year period 11.3% Hispanic students and
3.7% of African-American students passed developmental math at LA Trade Tech. In regard to
developmental English tracked over the same period 24.4% of Hispanic students and 11.9% of
African-American student passed (CCCCO, 2015). This is a major concern in terms of student
transfer and completion rates. Likely in response to these outcomes in 2014 LA Trade Tech
implemented a student equity plan to improve outcomes for African American students, students
receiving support from disabled students program and services, and non-traditional students.
As an additional illustration, a similar situation could be seen LA Mission, which is
considered a Hispanic serving institution, 76.2% of the students self-identify as Hispanic.
Although LA Mission serves a significant number of Hispanic students, non-URM students
outperform Hispanic students as it relates to the transfer/completion rates. Students who enroll in
college having to take developmental math or English are labeled as "unprepared for college"
within this category Hispanic students have transfer/completion rate of 30.7% versus Asian
students, 50% and, White students 37.3%. In 2013, LA Mission received a grant to support
persistence and retention of Hispanic and other underrepresented students.
These are just a couple of examples of the circumstances surrounding colleges listed as
underperforming and are not meant to be definitive conclusions as there could be several
additional factors that impact student outcomes that are not mentioned here. For example,
schools could have greater rates of transfer to private institutions or to the University of
California system. They could also be a school that does not have transfer as a primary aspect of
COMMUNITY COLLEGE ACCOUNTABILITY 75
their institutional mission thus making it more likely that students are pursuing career or
technical education. Furthermore, the distance between a community college and the nearest
CSU can also impact transfer rates.
Implications and Conclusions
Annual transfer rates for all students were found to very low for all students and this has
been a consistent trend over the last twenty years. Previous studies have measured transfer rates
over a period of three to six-years; this provides the indication that transfer rates are considerably
higher. The gap in the number of annual transfers experienced by URM vs. non-URM students
has been consistent over the last twenty years. Furthermore, the percentage of African American
and Hispanic students transferring annually has consistently trailed behind their peers.
Over the last twenty years African American students consistently had the lowest transfer rates.
With the proposal of PIRS the low annual transfer rates experienced by all students to CSU’s is a
great concern as the CCC’s are supposed to provide access to a four-year degree for a majority of
students in California.
Implications
This study indicates that annual transfer rates for California community college students
to CSU’s are generally low and that the situation has not changed much in the last twenty years.
Previous studies have explained the need for transfer rates to be measured between three-six
years, in particularly as a means of effectively capturing the outcomes of students from
underrepresented backgrounds (Jenkins, 2005; Bahr et al., 2005). However, even when
measuring transfer rates annually ethnic disparities were still apparent for URM students. Input-
adjusted scores account for the unique populations and institutional factors that impact student
outcomes at community colleges. Input-adjusted scores have also proved to be a reliable form of
COMMUNITY COLLEGE ACCOUNTABILITY 76
measurement for the purposes of accountability (Bahr, Hom, & Perry, 2005; Pacheco, 2012;
Bailey & Xu, 2012). Based on the findings of this research and current literature the following
implications were identified:
Transfer. In considering the national effort to increase accountability for community
colleges the development of a universal measurement for transfer is essential. Currently, within
California transfer is defined by using a cohort model and students are tracked over three to six
years to determine transfer outcomes. There are also suggestions to extend the measure to ten
years in California (Harris, 2014). The way transfer is measured varies by state and if institutions
are not normed on the definition of transfer this could prove to make accountability difficult. A
move toward a national definition of transfer rates would also support the PIRS goal of having
cross state peer institutional comparisons.
Using inclusive measure of transfer (three to six years) could mask the underlying
disparities faced by students from underrepresented minority backgrounds. The annual rate of
transfers from CCC to CSU’s was extremely low for all colleges and there were evident
disparities between URM and Non-URM students. Through PIRS, the federal government would
like to acknowledge institutions that demonstrate change over time, but, within this study, there
have not been any significant improvements in annual transfer rates over a twenty-year period.
Using annual data also has been shown to be beneficial for benchmarking purposes to measure
student progression and close the achievement gap for URM students (Baldwin, Bensimon,
Dowd, Kleinman, 2011). Annual input-adjusted transfer data should be used as an accountability
measure for transfer rates.
Developmental coursework is one of the leading barriers to obtaining transfer to a four-
year institution. Consistent placement tests should be used for developmental English and math.
COMMUNITY COLLEGE ACCOUNTABILITY 77
Within California the need for such assessment is on the list of items to be addressed through the
Student Success Taskforce. When examining the math success rates over the last twenty years
Hispanic and African American students consistently perform behind White and Asian students.
While Hispanic students have shown minimal improvement in their success rates, African
American students have made almost no improvement in their success rates in 1993 they were at
43% and this remained consistent in 2013 (See Figure 6) from Scott, (2015).

Figure 6. Math Success Rate 1992-2013
There should also be consistent paths to complete developmental coursework to ensure
students have the opportunity to pursue transfer. It would be ideal if such paths were consistent
nationwide but even statewide consistency would be beneficial for community college students.
Accountability. California is making great strides to improve accountability within
community colleges. In February 2015 the Institutional Effectiveness Partnership Initiative
(IEPI) was established in an effort to improve accountability and student outcomes within
California community colleges. Through this initiative community colleges will be charged with
developing institutional goals as they relate to accreditation status, fiscal viability, student
performance and outcomes, and programmatic compliance with state and federal guidelines
0.00

0.20

0.40

0.60

0.80

1.00

1.20

African
Americans

American
Indians

Asians

Hispanic

White

COMMUNITY COLLEGE ACCOUNTABILITY 78
(State Goals For California’s Postsecondary Education System [66010.9 - 66010.95], 2013. Once
institutional goals are established, they will be used to develop system wide goals for community
colleges districts and the entire state. The legislature calls for these goals to be both "challenging
and quantifiable" and to address the needs of underrepresented students and align with
California's needs in regard to educational attainment and workforce needs.
Establishing realistic goals to improve outcomes for underrepresented minority groups
will be a challenge without looking at annual data and utilizing input adjusted scores. As it
relates to transfer, the inclusive measures of three to six-years mask the disparities and the lack
of improvement in annual transfer rates URM that was evident within this study. In addition,
given the variability in both institutional and student characteristics, utilizing input-adjusted
scores would prove to be a valuable tool to aid institutions in accounting for such factors and
setting realistic goals. Without utilizing annual data or adjusted for extraneous factors,
institutions will likely set unrealistic goals making it increasingly unlikely they will reach their
targeted outcomes. While the development of both institution and statewide goals aims to
improve outcomes and accountability within California Community colleges goals should be set
to account for both institutional and student factors.
Conclusion
There is no doubt about the value of community colleges and their role in serving as
gateway to for a four-year degree for a majority of the underrepresented and disadvantaged
students in America. It will be impossible for the United States to reach the goals set forth by
AGI without improving the outcomes of students in America’s community colleges. Kotamraju
and Blackman (2011) stated, "…If higher education is to meet the AGI goal a confluence of
variables both internal and external will determine how successful community colleges will be at
COMMUNITY COLLEGE ACCOUNTABILITY 79
improving graduation rates and completion numbers". Closing the achievement gap for the
underrepresented students within community colleges is the one way to meet that goal.
There are both internal and external variables that impact the measurement of student
success specifically as it relates to transfer (Bahr, Hom, & Perry, 2010; Bailey & Xu, 2012;
Jenkins, 2005; Kotamraju & Blackman, 2011; Porschea, Allen, Robbins, & Phelps, 2010). The
use of input-adjusted scores is a promising practice in accountability and will meet the unique
needs of community colleges by measuring transfer in a way that accounts for both internal and
external factors (Goldrick-Rab, 2010; Bahr et al., 2010; Bailey & Xu, 2012; NASFAA, 2014). If
community colleges are to reach the goal of increasing completion rates using input-adjusted
scores will serve as a valuable to tool to allow community colleges to use data to develop an
accurate picture of student completion given the relative student characteristics at their
institution. These data also will allow institutions to formulate an action plan to improve student
transfer rates through benchmarking and comparison with peer institutions.
California is one of the largest community college districts in the nation and also one of
two states in the nation where the majority of the population is no longer White (Chase et al.,
2012). The disparities in transfer outcomes for URM students will greatly impact CCC ability to
increase the number of students transferring from community colleges to four-year institutions.
The CCCCO has reservations regarding the suggested data to be used through PIRS but they
have supported the use of input-adjusted scores for the development of peer groups for the
purpose of accountability (Harris, 2014).
The fact that disparities between URM and non-URM students exist and have been a
long-standing challenge in higher education suggest that utilizing the results of this study as well
as previous studies urge the conversation to move forward beyond the existence of disparities,
COMMUNITY COLLEGE ACCOUNTABILITY 80
and on to utilizing data to explain why they exist, and develop strategies to close the
achievement gap for URM students.
Future Research
Further qualitative research could be done to gain a better understanding of the unique
challenges and institutional practices that impact transfer rates to CSU’s by each of the colleges
listed as over and underperforming. Additional research could also be conducted on the
implications of how transfer is measured and explore if institutions identified as over and
underperforming within this study have been identified in other studies that utilize alternative
measures to input-adjusted scores.
Research should be conducted in community colleges in other states on the use of input-
adjusted scores as an accountability measure as it relates to transfer rates. In addition input-
adjusted scores could be disaggregated by race to examine predicted and actual transfer rates by
ethnicity to gain further understanding of equity in student outcomes when other factors are
controlled for. Disaggregating input-adjusted scores by race would also provide insight to
institutions that are over and underperforming as it relates to outcomes of URM. Additional
research could also be done in utilizing input-adjusted scores to form peer groups among like
institutions for accountability purposes.
Specifically, in California more research could be done to examine the current number of
juniors in California State Universities that have transferred from a community college to gain an
understanding of the impact of the transfer admission guarantee (TAG) agreements between
CSU's and California Community Colleges. Research could also be done to explore the role of
impaction of CSU's. Currently, 60% of universities within the CSU system are labeled as
impacted, which means they currently have significantly more applications than they have spots
COMMUNITY COLLEGE ACCOUNTABILITY 81
available. Being labeled as an impacted CSU often changes the admission criteria thus likely
making it more difficult for a transfer student to gain admission to a CSU.
Research should also be conducted to examine student transfer rates to schools within the
University of California (U.C.) and private school systems. It would be beneficial to understand
student transfer rates to these institutions disaggregated by race to gain an understanding of
transfer patterns to more selective institutions.
COMMUNITY COLLEGE ACCOUNTABILITY 82
References
Accrediting Commission for Community and Junior Colleges (2014). Eligibility requirements
for accreditation. Accrediting Commission for Community and Junior Colleges Western
Association of Schools and Colleges. Retrieved on December 18
th
from:
http://www.accjc.org/wp-
content/uploads/2014/06/Eligibility_Requirements_Adopted_June_2014.pdf.
Argyris, C. & Schon, D.A. (1996). Organizational learning II: Theory, method, and practice.
Reading, Mass: Addison Wesley.
Bailey, T., Calcagno, J. C., Jenkins, D., Kienzl, G., & Leinbach, T. (2005). The effects of
institutional factors on the success of community college students. Community College
Research Center, Teachers College, Columbia University.
Bailey, T., Jeong, D.W., & Cho, S.W. (2010). Student progression through developmental
sequences in community colleges. Community College Research Center Brief, (45), p.1-
6.
Bailey, T. & Xu, D. (2012). Input-adjusted graduation rates and college accountability: What is
known form twenty years of research. Community College research Center, p.1-35.
Bahr, P.R., Hom, W., & Perry, P. (2005). College transfer performance: A methodology for
equitable measurement and comparison. Journal of Applied Research in the Community
College, 13(1), p.73-87.
Baldwin, C., Bensimon, E.M., Dowd, A. C., & Kleiman, L. (2011). Measuring student success.
New Directions for Community Colleges, 153, 75-88.
Beno, B. A. (2004). The role of student learning outcomes in accreditation quality review. New
directions for community colleges, 2004(126), 65-72.
COMMUNITY COLLEGE ACCOUNTABILITY 83
Bensimon, E. M. (2005). Closing the achievement gap in higher education: An organizational
learning perspective. New Directions for Higher Education, 2005(131), 99-111.
Burke, J.C. (2005). The many faces of accountability. Achieving accountability in higher
education: Balancing, public, academic, and market demands (pp.1-24). San Francisco,
CA: Jossey-Bass.
California Community Colleges Chancellors Office (2015). Student success scorecard. Retrieved
from: http://scorecard.cccco.edu/scorecardrates.aspx?CollegeID=851#home.
California Community Colleges, (2007). Focus on results: Accountability reporting for the
California Community Colleges ( A report to the legislature pursuant AB 1417).
Sacramento, Calif.: California Community Colleges System Office.
Community College Chancellor’s Office. (2014). Post secondary institution rating system (PIRS)
RFI. Community College Chancellor’s Office.
Cornbach, L.J. & Furby, L. (1970). How should we measure change or should we?
Psychological Bulletin, 74, (1), 68-70.
Chase, M.M., Dowd, A.C., Pazich, L.B., & Bensimon, E.M. (2014). Transfer Equity for
“minoritized” students: A critical policy analysis of seven states. Education Policy, 1-49.
Chase, M.M., Dowd, A.C., Pazich, L.B., & Bensimon, E.M. (2012). Transfer Equity for
“minoritized” students: A critical policy analysis of seven states. Education Policy, 1-49.
Cunha, J. M. & Miller, T. (2012). Measuring value-added in higher education: Possibilities and
limitations in the use of administrative data (Working Paper). Retrieved from: Calhoun
Institutional Archive of the Naval Post Graduate School
http://calhoun.nps.edu/handle/10945/37255.
COMMUNITY COLLEGE ACCOUNTABILITY 84
Denhart, C. (2013). How the 1.2 trillion dollar college debt crisis is crippling students, parents,
and the economy. Forbes. Retrieved February18 from:
http://www.forbes.com/sites/specialfeatures/2013/08/07/how-the-college-debt-is-
crippling-students-parents-and-the-economy/.
Dowd, A.C., Pak, J.H., & Bensimon, E.M. (2013). The role of Institutional agents in promoting
transfer success. Education Policy Analysis Archives, 21(15), 1-40.
Dowd, A.C. (2003). Community colleges as gateways and gate keepers: moving beyond the
access “saga” toward outcome equity. Harvard Education Review, 77(4), 407-419.
Dowd, A. C. (2003). From access to outcome equity: Revitalizing the democratic mission of the
community college. The Annals of the American Academy of Political and Social
Science, 586 (1), 92-119.
Dykas, M. J., & Cassidy, J. (2011). Attachment and the processing of social information across
the life span: theory and evidence. Psychological bulletin,137(1), 19.
Eagen, K.M., & Jaeger, A.J. (2009). Effects of exposure to part-time faculty on community
college transfer. Research in Higher Education, 50, 168-188.
Espinosa, L.L., Crandall, J.R., & Tukibayeva, M. (2014). Rankings, institutional behavior, and
college and university choice: Framing the dialogue on Obama’s rating plan (Issue brief
1-23). American Council on Education & Center for Policy Research and Strategy.
Boeke, M., & Ewell, P. T. (2007). Critical connections: Linking states' unit record systems to
track student progress.
Ewell, P. T. (2001). Accreditation and Student Learning Outcomes: A Proposed Point of
Departure. CHEA Occasional Paper.
COMMUNITY COLLEGE ACCOUNTABILITY 85
Ewell, P. T. (2011). Accountability and institutional effectiveness in the community college.
New Directions for Community Colleges, 153, 23-36.
Gall, M.D., Gall, J.P. & Borg, W.R. (2007). Educational research: An introduction. Allyn &
Bacon, Inc. Boston: MA.
Goldrick-Rab, S. (2010). Challenges and Opportunities for Improving Community College
Student Success. Review of Educational Research, 80(3), 437-469.
Goldrick-Rab, S., & Pfeffer, F.T. (2009). Beyond access: Explaining differences in college
transfer. Sociology of Education, 82, 101-125.
Harbour, C. P. (2003). An institutional accountability model for community colleges. Community
College Journal of Research and Practice, 27, (4), 299-316.
Harris, B.W. (2014). California community college chancellor’s office. PIRS Request for
information.
Harris, F., & Bensimon, E. M. (2007). The equity scorecard: A collaborative approach to assess
and respond to racial/ethnic disparities in student outcomes. New Directions for Student
Services, 2007(120), 77-84.
Hayward, C. (2011). The transfer velocity project: A comprehensive look at the transfer
function. Journal of Applied Research in the Community College, 18, (2), 1-12.
Hernandez, L., & Naccarato T. ( 2010). Scholarships and supports available to foster care
alumni: A case study of 12 programs across the U.S. Children and Youth Services
Review, 32, 758-766.
Hines, A.M., Merdinger, J., & Wyatt, P. (2005). Former foster youth attending college:
Resilience and the transition to young adulthood. American Journal of Orthopychiatry,
75(3), 381-384.
COMMUNITY COLLEGE ACCOUNTABILITY 86
Integrated Postsecondary Education Data System. (2014). About IPEDS. Retrieved from:
http://nces.ed.gov/ipeds/about/.
Jenkins, D. (2011). Redesigning community colleges for completion [electronic resource]:
Lessons from research on high-performance organizations. New York, N.Y: Community
College Research Center, Teachers College, Columbia University.
Jenkins, D. (2007). Institutional effectiveness and student success: A study of high and low
impact community colleges. Community College Journal of Research and Practice, 31,
945-962.
Kotamraju, P., & Blackman, O. (2011). Meeting the 2020 American graduation initiative (AGI)
goal of increasing postsecondary graduation rates and completions: A macro perspective
of community college student educational attainment. Community College Journal of
Research and Practice, 35(3), 202-219.
Long, B.T., & Kurlaender, M. (2009). Do community colleges provide a viable pathway to a
baccalaureate degree?. Education Evaluation and Policy Analysis, 31(1), 30-53.
Melguizo, T. (2007). Latino and African-American Students' transfer pathway to elite education
in California. Change: the Magazine of Higher Learning, 39(6), 52-55.
Melguizo, T., Hagedorn, L. S., & Cypers, S. (2008). Remedial/developmental education and the
cost of community college transfer: A Los Angeles County sample. Review of Higher
Education, 31(4), 401-431.
Miller, B. (2013). “Essay on how President Obama’s rating system should work: Real numbers.
Real goals.” Inside Higher Ed. Retrieved from:
https://www.insidehighered.com/views/2013/09/03/essay-how-president-obamas-rating-
system-should-work.
COMMUNITY COLLEGE ACCOUNTABILITY 87
Moore, C., & Shulock, N. (2010). Divided we fail: Improving completion and closing racial gaps
in California's community colleges. Sacramento, CA: California State University,
Sacramento, Institute for Higher Education Leadership & Policy.
Mullin, C.M. (2010). Rebalancing the mission: The community college completion challenge,
(Policy Brief 02PBL). Washington, DC : American Association of Community Colleges.
Myers, D. (2007). Immigrants and boomers forging a new social contract for the future of
America. New York: Russell sage Foundation.
National Association for Student Financial Aid Administrators (2014). Peers in PIRS:
Challenges and considerations for rating groups of postsecondary institutions.
Washington, D.C: National Association for Student Financial Aid Administrators.
Okypych, N. (2012). Policy framework supporting youth aging out of foster care though college:
Review and recommendations. Children and Youth Services Review, 34, 1390-1396.
Pacheco, R. J. (2012). Assessing and addressing random and systematic measurement error in
performance indicators of institutional effectiveness in the community college. Los
Angeles, California.
Porchea, S. F., Allen, J., Robbins, S., & Phelps, R. P. (2010). Predictors of long-term enrollment
and degree outcomes for community college students: Integrating academic,
psychosocial, socio-demographic, and situational factors. Journal of Higher Education,
81(6), 750-778.
Scott, M.B. (2015). Best Parctices of Developmental Math in California Community Colleges.
Los Angeles, CA.
State Goals for California’s Postsecondary Education System [66010.9 - 66010.95]. California
Education Code, (2013). Retrieved from:
COMMUNITY COLLEGE ACCOUNTABILITY 88
http://leginfo.legislature.ca.gov/faces/codes_displaySection.xhtml?lawCode=EDC§i
onNum=66010.91.
Shaffer, D. F. (2008). The states and their community colleges. Albany, NY: Nelson A.
Rockefeller Institute of Government.
Shin, J.C. (2010). Impacts of performance-based accountability on institutional performance in
the U.S. Higher Education, 60, 47-68.
Student Success Task Force (2012). Advancing student success in the California community
colleges: Recommendations of the California community colleges student success task
force. Retrieved from
http://californiacommunitycolleges.cccco.edu/Portals/0/Executive/StudentSuccessTaskFo
rce/Student_Success_Presentation_CCLC_2011-11-29.pdf
Taylor, M. (2014). California community colleges: A progress report on the student success act
of 2012. Legislative Analyst Office, 1-20.
Taylor, M. (2014). The 2014-15 budget: Analysis of the higher education budget. Legislative
Analyst Office, 1-43.
United States Census Bureau (2014). State and county quick facts. Retrieved from
http://quickfacts.census.gov/qfd/states/06000.html.
United States Department Education (2014). “College ratings and paying for performance.”
Retrieved from http://www.ed.gov/college-affordability/college-ratings-and-paying-
performance.
United States Department Education (2014). “A new system of college ratings-invitation to
comment.” Retrieved on February 18th from: http://www2.ed.gov/documents/college-
affordability/framework-invitation-comment.pdf.
COMMUNITY COLLEGE ACCOUNTABILITY 89
Wassmer, R., Moore, C., & Shulock, N. (2004). Effect of Racial/ethnic composition on transfer
rates in community colleges: Implications for policy and practice. Research and Higher
Education, 45(6), 651-672.
White House Office of the Press Secretary (2013). “Factsheet on the presidents plan to make
college more affordable: A better bargain for middle class.” Retrieved on August 2
nd

from: http://www.whitehouse.gov/the-press-office/2013/08/22/fact-sheet-president-s-
plan-make-college-more-affordable-better-bargain-
Whitehouse (2015). College affordability and transparency center: college scorecard. Retrieved
from: https://www.whitehouse.gov/issues/education/higher-education/college-score-card.

COMMUNITY COLLEGE ACCOUNTABILITY 90
Appendix
Data Collection Methodology
[1] Go to http://datamart.cccco.edu/Outcomes/Default.aspx

[2] Under "Transfer Volume" selected the blue highlighted field “CSU Analytic
Studies” This will open a new tab for California State Universities
(http://www.calstate.edu/as/ccct/index.shtml).

[3] The data sets will be listed according to year I opened each year individually
i.e. “College year 1985-1986”.

[4] Under the red heading “List of Tables” select “CCT1985-86 word document”.
This document will then be down loaded on your computer. Data was
downloaded between 1992-2013.

[5] Under the red heading “List of Tables” select “community college transfers
by ethnic group.

[6] At the top of the data set click download this table in Excel. The document
will then be downloaded to your computer and can be imported into SPSS.

Abstract (if available)

Abstract In recent years, due to issues of the increased cost of education, challenges with student retention, and significant increases in the number of students defaulting on federal student loans, accountability within higher education has gained national attention. Currently, the United States ranks 14th overall among developed nations in educational attainment, and the Obama Administration has charged the United States to work back to the top by increasing the number of college graduates by the year 2020. With the increased calls for accountability, the Obama Administration has proposed the post-secondary educational rating system (PIRS) to assess and rank institutional accountability as it relates to access, affordability, and student outcomes. These rankings will be tied to federal financial aid awards and thus impact funding received by both institutions and students. The goal of the rating system is to improve institutional performance and increase educational outcomes. Being the leading access point to higher education, community colleges play a significant role in increasing the number of college graduates via the transfer route. ❧ California has the largest community college system in the nation and is one the most diverse states. The purpose of this twenty-year study was to examine the relationship between ethnicity and transfer rates from California Community Colleges to California State Universities. Input-adjusted scores will be used as an accountability measure to adjust for external factors that have been known to impact student success. This study aimed to provide an accountability measure that allows for fair institutional comparison of transfer rates and provides increased awareness on equity in student outcomes as it relates to transfer in California Community Colleges. ❧ In the last twenty years, there were no significant changes in the annual transfer rates from California Community Colleges to California State Universities across ethnic lines. Both Hispanic and African American students consistently trailed behind their peers in annual transfers to California State Universities. Annual changes in transfer rates were not found to be a stable measure enough measure for accountability purposes. However, input-adjusted transfer rates were found to be stable from year to year and are recommended as a tool to accurately measure transfers rates and allow for fair institutional comparisons.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Accountability models in remedial community college mathematics education

PDF

Assessing and addressing random and systematic measurement error in performance indicators of institutional effectiveness in the community college

PDF

An exploratory, quantitative study of accreditation actions taken by the Western Association of Schools and Colleges' Accrediting Commission for Community and Junior Colleges Since 2002

PDF

A longitudinal study on the opportunities to learn science and success in science in the California community college system

PDF

The influence of executive leadership on community college completion rates

PDF

Relationships between a community college student’s sense of belonging and student services engagement with completion of transfer gateway courses and persistence

PDF

Use of accountability indicators to evaluate elementary school principal performance

PDF

Achieving equity in educational outcomes through organizational learning: enhancing the institutional effectiveness of community colleges

PDF

Developmental math in California community colleges and the delay to academic success

PDF

Community college transfer student involvement experiences at a selective, private four-year university

PDF

Understanding the barriers to college access for former foster youth at the Los Angeles Community College District

PDF

The 2003-2012 impact of Algebra When Ready on indicators of college readiness across California school districts

PDF

A formative evaluation of the student support services TRIO program for low income and first generation college bound students self-efficacy at Butte-Glenn Community College District

PDF

Community college transfer student involvement experiences at a selective, private four-year university

PDF

Concept mapping of the sources of perceived impact on community college students' identity development: a students' perspective

PDF

The effects of a math summer bridge program on college self-efficacy and other student success measures in community college students

PDF

Perspectives on accreditation and leadership: a case study of an urban city college in jeopardy of losing accreditation

PDF

State policy as an opportunity to address Latinx transfer inequity in community college

PDF

A gap analysis to find best practices in philanthropy to support California's community colleges and offer potential solutions to close performance gaps

PDF

Assessment, accountability & accreditation: a study of MOOC provider perceptions

Asset Metadata

Creator Jones, Josephine (author)

Core Title Input-adjusted transfer scores as an accountability model for California community colleges

School Rossier School of Education

Degree Doctor of Education

Degree Program Education (Leadership)

Publication Date 06/04/2015

Defense Date 04/27/2015

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

Tag accountability,community college,equity,input-adjusted,OAI-PMH Harvest,outcomes

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Hocevar, Dennis (committee chair), Keim, Robert G. (committee member), Pacheco, Robert J. (committee member)

Creator Email josephinejones.edd@gmail.com,josephlj@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-569117

Unique identifier UC11299026

Identifier etd-JonesJosep-3456.pdf (filename),usctheses-c3-569117 (legacy record id)

Legacy Identifier etd-JonesJosep-3456.pdf

Dmrecord 569117

Document Type Dissertation

Format application/pdf (imt)

Rights Jones, Josephine

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